your Commodore 128 or 64, If you wont to get the most out we hove goods news for you. The Pocket 128 and 64 Series of Software both offer you serious, professional quality software packages that ore easy to use and inexpensive.
Pocket 128 or 64 Software is so easy, you're ready to start using it as soon as it's loaded into memory. Even if you've never been in front of o computer before, you'll be up ond running in thirty minutes. In fact, you probably won't ever need the reference guide ... 'help' is available at the touch of o key. That's how easy.
Pocket Writer 128 or 64 Processing you see Is what you get
--
With Pocket Writer 128 or 64, there's no more guessing what text will look like when you print it. What you see is what you get ... on screen and in print. There ore no fancy codes to memorize, no broken words at the end of o line. Easy to learn and so p histicated. Pocket Writer 128 or 64 offers standard wor d processing features plus ... •
•
•
Pocket 128 or 64 packages hove all the power you're ever likely to need. They hove all of the features you'd expect in top-of-the-line software, and then some. The good news is that Pocket 128 or 64 Software Packages ore priced way down there ... where you con afford them. Fast, powerful, easy to learn and inexpensive. Soy, that is good news!
Pocket 128 or 64 Software Packages offer you something else you might not expect ... integration. You con combine the output of Pocket Writer, Pocket Filer and Pocket Planner into one piece of work. You con creole o finished document with graphs, then send individually addressed copies.
The word solutions is our middle nome and bottom line. When you purchase Pocket 128 or 64 software, you con count on it to solve your problems.
•
on-screen formatting end wordwrop
o n-screen boldface, underlines end italics
•
• •
no complicated formot commends to clutter text on-screen help ot oil levels
spelling-checker lets you odd words to your dictionary 40 or 80 columns on screen files compatible with PaperClip'" or other word processors
Pocket Planner 128 or 64 software lets you make fast work of all your bookkeeping chores. Cheque books, household accounts, business forecasting and bookkeeping ore just some of the jobs that Pocket Planner 128 or 64 packages make easier. You con even create four different kinds of graphs. Accurate, sophisticated and easy to use. Pocket Planner 128 or 64 offers standard spreadsheet features plus ... • accuracy up to 16 digits, about twice os many os most s p readsheets for the Commodore 128 or 64
• 80 column on-screen option for the Commodore 64 in addition to the standard 40 columns
•
•
•
sideways printing available on dot matrix printers, for oversized spreadsheets that won't fit on standard paper on-screen help ot all levels
•
ompa tible with VisiColc'" fi les
• c
graphics include bar, stacked bar, line end pie graphs that con also be used in word processing files smart evaluation of formulae for accurate complex matrices
With Pocket Filer 128 or 64, you con organize moiling lists addresses, inventories, telephone numbers, recipes and other information in on easily accessible form. Use it with Pocket Writer 128 or 64 (or other word processors) to construct individually customized form letters. Pocket Filer 128 or 64 p ackages ore fast, sophisticated and truly easy to use. In addition to standard database features they offer ... •
•
•
•
f.,.-PopttCf•p ' o rtg11ftred frode,.,.or� of 8of1tr••• 'tt(luded '
... v,,.colt 11 o �,uerHI "odel'ftOtft; ol Sohwort Am
use up to 255 fields per record (2,000 characters per record) sorts by �p to 9 crit eri a con seve 9 differen t sorts
•
,
•
print labels in multiple columns
•
flexible re port formatting including headers and footers
•
optional password protection including limited access viewing or updating on-screen help ot oil levels
print from ony record to ony record arithmetic end trigonometric functions in reports using up to 16 digit accuracy
SINGLE/DUAL NORMAL COPIER
GCR EDITOR
Yes disk fans, a full blown sector by sector or track by track GCR Editor. Includes TRUE Bit Density/Track Scan.
Copies a disk with no errors in 32.68 seconds. dual version has graphics & music.
SINGLE/DUAL NIBBLE COPIER
Nibble Copies a disk in 34.92 seconds. Dual version has graphics & music.
3 SUPER DOS FAST LOADERS
Over 15 times normal DOS speed. Super DOS Files are still Commodore DOS compatible. Imagine loading 150 blocks in 10 seconds.
SINGLE/DUAL FILE COPIER
7 times normal DOS speed. Includes multi-copy, multi-scratch, view/edit BAM, & NEW SUPER DOS MODE. In Super DOS Mode, It transfers 7·15 times normal speed, copies 150 blocks in 23 seconds.
SUPER NIBBLER/ SUPER DISK SURGEON
Quite frankly, these will provide you the user with the backup you needl Even copies itself.
TRACK & SECTOR EDITOR
Full editing of t&s in hex, dec, ascii, bin. Includes monitor/disassembler with printout commands.
$29.95
u.s.
PLUS $3.00 SHIPPING/HANDLING CHARGE - $5.00 C.O.D. CHARGE • -
PRISM
SOFTWARE SUPER KIT/1541 IS for archival use only! We do not condone nor encourage p1racy of any kind.
I
I
I
401 LAKE AIR DR., SUITE D • WACO, TEXAS 76710 ORDERS (817) 757-4031 • TECH (817) 751-0200 MASTERCARD & VISA ACCEPTED
See center page for mail order card.
Onlylhe NalnelsNew The p rofessional; full-featured software line from Digital Solutions is now called Pocket Software. Pocket Writer 128/64. Pocket Filer 128/64. Pocket Planner 128/64. The names are new, but this super software is still the same. From now on, wl}.n YOU it heor the
You wont inte grotea software- wore P-rocessing, aotobose on d.spreaasheet applications- at 0 sensiole price. But , you also want top-of-the-line features. Well, ouďż˝ Pocket 128/64 software goes one better.
CoVVlVVlodore RefereVlce Oiar�
JOIN TPUG The largest Commodore Users Group Benefit from: Access to library of public domain software for C-64, VIC 20 and PETICBM Magazine
Jim BV�tterfield
(10 per year) with advice from Jim Butterfield Brad Bjomdahl Liz Deal
• •
TPUG yearly memberships:
•
Regular memb er (a ttend s meetings) -$35.00 Cdn. Student member (full·tlme, attends meetings) -$25.00 Cdn. Associate (Canada) -$25.00 Cdn.
-$35.00 u.s.
Associate (Overseas- air mall)
• •
•
• •
• • •
•
• •
• • •
1986
-$30.00 Cdn.
Associate (Overseas- s ea mall)
•
•
•
-S25.00 u.s.
Associate (U.S.A.)
•
-$45.00 u.s.
COMPIIITER OIARY
FOR FURTHER INFORMATION:
Send $1.00 for an information catalogue (tell us which machine you use!} To:
The 1986 Commodore Reference Diary
TPUG INC. DEPT. A, 101 DUNCAN MILL RD., SUITE G7 DON MILLS, ONTARIO CANADA
M3B 1Z3
BOOKS: • • • • • • • • • • •
COMAL From A To Z, $6.95 COMAL workbook, S6.95 commodore 64 Graphics With COMAL, $14.95 COMAL HandbOOk, $18.95 Beginning COMAL, $22.95 Structured Pro<;�ramming With COMAL, $26.95 Foundations Wrth COMAL, $19.95 cartridge Graphics and Sound, S9.95 Captain COMAL Gets Organized, S19.95 Graphics Primer. $19.95 COMAL 2.0 Packages, $19.95 Library of Functions and Procedures. S19.95
OTHER: • • • • •
COMAL TODAY subscription, 6 issues. S14.95 COMAL 0.14, Cheatsheet Keyboard overlay, $3.95 COMAL Starter Kit 13 disks, 1 bOOk), S29.95 19 Different COMAL Disks only $94.05 Deluxe COMAL cartridge Package, $128.95 <includes 2 books, 2 disks. and cartridge>
ORDER NOW:
Call TOLL·FREE 1·800·356·5324 ext 1307 VISA or Mastercard ORDERS ONLY. Questions and lnformat•on must call our Info line: 608·222·4432. All o rde rs prepaid only-no C.O.D. Add 52 pe r book shiPP ing send a SASE for FREE Info Package or send check or money order In us Dollars to:
COMAL USERS GROUP, U.S.A., LIMITED S501 Groveland
Ter.,
Madison,
A 65 page reference section that includes: •
All hardware specifications including the C128 and PC10/20
COMAL INFO If you have COMAL We have INFORMATION. •
From The Guru Himself!
Wt
S 3 716
TRADE MARKS· c o mm ooore 6� of Commooore Electrontcs Lto: Captam CO MAL Of COMAL Users GrOUP. U S.A. ltel
•
Useful memory locations
•
Useful programs
•
SuperCharts
•
BASIC and machine language hints
•
Hexadecimal conversion
•
Sound, video
•
and more
The full calendar and date book includes: •
National holidays in ten countries
•
Personal notes
•
1987 forward planner
•
Name, address, telephone section
Just $5.95
(plus 50¢ postage and handling)
Order Your Copy Today! Canada The Transactor 500 Steeles Avenue Milton, Ontario L9T3P7
USA The Transactor 277 Linwood Avenue Buffalo, New York 14209
Dealer Orders:
Canada Norland Agencies
251 Nipissing Road Milton, Ontario L9T 4T7 (416) 876-4774
USA MicroPace Distributing 1510 North Neil Street Champaign, Illinois 61820 1 800 362-9653
Progranuning The Chips
Volume7 Issue 03 Circulation 64,000
Start Address
Editorial
Bits and Pieces
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
News BRK
6
C-128 Program Entry Tips Stop Amiga Disk-Death Easy PAL to CBM Source Conversion Micro STP Border Cross Disk Protection Some C-128 Tips Disabling The C-128 RUN Key C-64 Upper/Lowercase Timescroll Free C-128 CP/M Manual Non-Random Surprises Watchacallit? Quicker Assembly on the Amiga Changing Baud Rates For The C-64 with 1650/6240 Modems Borrowing Money?
Letters
•
. 3 77
Transactor Writer's Guide Finally Finished Subscription Timing Free Transactor Ts with Mag+ Disk Subscription Transactor Disk Price Increase Refund Policy Allow 2 to 6 Weeks Transactor Mai l Order News Transactor Disks, Back Issues, and Microfiche Sending Cheques For Tra nsactor Products The Transactor Communications Disk CompuServe and Quantum Link The Commodore Show, September 20 and 21, 1986 Twin Cities 128: The Commodore 128 Journal iNet2000 High-Powered FREE Terminal Program for the C64 Hardware Interconnection Products Aegis Ships CAD Program for the Amiga
10
TransBloopers
Column For Machine Language Sub-Routines? New Testament Available On 1541 Disk Format Flexible Vector Problem G-Unk Info The SID Chip: Does It Really Exist? Transactor Disk Directorytions Super Empty SuperPET Verifizer for the Cl28 Disk Un Assem bler Once Again Super Kit Users Notes Games From The Outside Out Another Vote Against The Atari Artic.le Rx For Grammar Spel Cheker
•
•
•
Amiga Startup-Sequence Eliminating SAVE@ Improving The SYS Command SYMASS 3.1 Fixes Cl28 Memory Maps
14
-
TransBASIC Installment #11 EXPO 86
•
•
•
•
•
•
a computerist's report from the Vancouver World's Fair
The AX2000
a 2 megabyte review
•
•
•
•
•
•
A Peek At Amiga File Structure Disk Copier Comparison Who Do You Trust? THE CLOCK
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
6 backup utilities pitted head to head
a hourly synopsis of a tun-filled evening
a dual alarm for the 64CIA TOD clocks
VIC-II Chip Interrupts Switcheroo Pivot
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Z80 and 8500 CPUs working together in the Cl28
Program Trace Monitor
a machine language profiler
•
•
•
•
•
•
•
•
•
•
•
•
•
The 8563 VDC Chip
allabouttheCI28VideoDisplayControllerchip
Keyboard Expander
the last keyboard utility you'll ever need!
Compu-toons
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
15 22 28 30 33 40 42 46 48 52 57 64 81
Note: Before entering programs, see
1
"Verifizer" on page 4
Nov. 1986: Yolume 7,1Aue 03
Program Listings In The Transactor
Editor In Cblef
All programs listed in The Transactor will appear as they would on your screen in Upper/Lower
Karl J. H. Hildon
case mode. To clarify two potential character mix-ups, zeroes will appear as '0' and the letter "o" will of course be in lower case. Secondly, the lower case L ('I') is a straight line as opposed to the number 1 which has an angled top.
Editor Richard Evers TechnlcaJ Editor
Many programs will contain reverse video characters that represent cursor movements, colours,
Chris Zamara
or function keys. These will also be shown exactly as they would appear on your screen, but
Art Director
they're listed here for reference. Also remember: CTRL-q within quotes is identical to a Cursor
Down, et al.
John Mostacci
Administration & Subscriptions
Occasionally programs will contain lines that show consecutive spaces. Often the number of
spaces you insert will not be critical to correct operation of the program. When it is, the required
Anne Richard
number of spaces will be shown. For example:
Kathryn Holloway Contributing Writers Jan Adam Daniel Bingamon Neil Boyle
Anthony Bryant Tim Buist
flush right
•
Jack Lothian
Stacy Mcinnis
Gary Cobb Jack Cole
Jim McLaughlin Steve Michel
Jeffery Coons
Chris Miller
Elizabeth Deal
Frank E. DiGioia
Yijun Ding
Paul T. Durrant Michael J. Erskine
Jack Farrah
R. James de Graff
-
Up
Right
-• - [Lit]
Left
RVS -1 RVSOtf- .
Michael Mossman
Gerald Neufeld Noel Nyman
Black
White
Dave Pollack Richard Perrit
Red
Cyan
Gary Royal
Bob Hayes
Fred Simon
-� -
-
-
Blue
-
P. A. Slaymaker
Aubrey Stanley
Robert Huehn
David Jankowski
Chris Johnson
Richard Stringer Nick Sullivan
Karel Vander Lugt
Mark Jordan
. Delete -I Clear Scm- I -
ID.ert
-�
Home
STOP
.
Orange
Brown
Lt. Red Grey I
-
-� -·
-
-
-
.
Grey2 U. GreenU. Blue Grey 3 - (Gr3)
Ft-11
F5-
F3-II F4- IJ
F7-
F2-
David Stidolph
Tom Hughes
•
Function Keys For VIC I 64
Darren J. Spruyt
Tomas Hrbek
.
Yellow- {Vel]
Edward Smeda
David Hook
[10 spaces]flush right
Colour Characters For VIC I 64
[Cyn] Purple [Pur] Gre-en -8
Terry Pridham Raymond Quirling John W. Ross Louis F. Sander
John HoiUum
-Ill •
Down
Terry Montgomery
Jim Grubbs Tom Hall Andy Hochheimer
•
Cursor Characten For PET I CBM I VIC I 64
Scott Maclean Don Maple
Betty Clay
Pierre Corriveau Robert V. Davis
James A. Lisowski
David Martin Steve McCrystal
Jim Butterfield
-would be shown as-
•
a
f6-
F8-
Audrys Vilkas Steven Walley
Clifton Karnes Jesse Knight
James E. LaPorte
Jack Weaver Evan Williams
William Levak
Chris Wong
Please Note: The Transactor's
ProductJon
phone number is: (416) 878-8438
Allie TypeseUi ng Ud.
Printing Printed in Canada by MacLean Hunter Printing
The
Tra n$6CtOr is publi$1>ed bi·monthly by Trans,acto r I'\Jbli$1>ing Inc., 500 Steeles Avenue, Millon. Ontario, L9T 3P7. Canadian Second Class mail regiS� ration number 3 42 USPS 725-050,Second Class postage paid at Buffalo. NY. for U.S. subscribers. U.S. PoSimasters: send address changes to The Transactor. 277 Lin w ood Ave n u e, Buffalo, NY, 14209 ISSN' 0827·2530.
6
Tr
Quantity Orden:
.
U.S.A. OistriiMor: Capita l Distributing Charlton Bu lding CT
r
The ansacto os in no way connected with Commodore Business Ma<hlnes lJd. or Commodore lncorporat<d Commodore and Commodore product names (PET. CBM, VlC. 84) are register<d trade marksolCommodore Inc. Subscnptoons· Canada $15Cdn U.S.A SISUS. Allother$2IUS Air Mail (Overseas o nly) $40 US. ($4.15 postaae/issue)
auboc:rlpllou l.9 T
Derby, 06418
(203) 73S 3381 (or your local wholesaler:)
Coqu
y
address magazine.
llltdr IN ""o: $4 . 50 each. Orde r all
Ed it ori� l conlr lbullons are al· ways welcome. W ti tt-r s are enco u ra ged 1 0 prepare ma terial acco r ding to themes as show n in Ed i to rial Sch<dule (see list near the end ol this Issue� R emunerllion is $40 per rint<d � · P r elerr<d media is 1541, 2031 . 4040 , 80SO. or 82SO disk et tes with Wo rdPro, WordCrah, over 20 lines should be provided on disk or tape. o r SEQ text fi les. Progum list in gs M anuocripts should be doub� � . w i th speciAl chal'lltl ers or f ormats clearly marked . lity . Authors submott ing diskettes Photos or ill ust rat ions be inc luded with anod es depending on qu a will the Transact the or Disk for issue cont atnJng their cont ribu ion t
rectvi e
O a kvi lle.
Norland Communications 25 I NlpissingRoad, Unit 3 Milton, Ontario L9T4ZS 4168764714
SOLO 0\IT: The Besl oiTheTr•n$6Cto r Volumes I & 2 & 3 ; Vol4 1sses u 04 ,OS. 06 , and Vo l5 bsuesOJ, 04 are available on mkrolic he onl SIJJJ Available: Vol. 4:01.02,03. Vol 5: 01,02.04 . 05. 06. Vol. 6:01.02,03, 04, OS. 06. Vol. 7: 01,02, 03
Seed all to: The Tran$6Ctor, Subscriptions Department, 500 Steeles Avenue, Milton, Ontario, Canada, 3P7, 416 878 8438. Note: SubscriptioN are handl<d at this address ONLY. Subscriptions sent to our Buffalo (above) will be forward<d to Milton HQ. For be$1 results. use postag epaid card at center ol
p Su per ocr\pt ,
i
Master Media CompuUt 261 Wyec r oft RQaBox 352 Ontarllx>rt iUam, BC VSC 4K 6 L6J 584 (416)842 1555 604 9 41 7911 (or your local wholesaler:)
b ac k Issues f rom Milton H Q.
A ll materia l acce pted beco mes the property of The Tran sact or. All material i s copyright by Tran sac to r Publk ations Inc. R ep r od uc t ion in any f orm w ithoot perm issio n is in viol ation of a pp licable laws. Please re<onfir m any permiss ions grant<d prior to this notice. Solidt<d material Is acc epted on an all rights basi s only. Write to the Milton add r ess lor a writers package. The opinions expr essed in con tribut<d 1 rtkles are not necessari those ol Th e Traruactor. Althoo gh ata�riC)' is• !llaJOf ob jective. The TrarlSI<lOr can not assu me liabi ity lor erron In 1rtid es or progra ms. to to Prog ra ms list<d in The Transaao are pu blk domai n; f r ee copy , not sell. r
typewntt en. will
ly
l
2
Nov.1Ut:�7,1nt•OI
I'm not saying you shouldn't copy the programs. In fact, if you would like to put them into your club libraries, please do. Most club software libraries are split into categories. All we ask is that you "de-unify" the disks into their respective categories in the true sense of a "library". Much like you can copy an article from any magazine at your local public library, ev�n at a nickel a page you wouldn't copy the entire mag - it would cost less to go buy one. Same for the disks. If the programs are split between two or three library entries, members can still obtain machine-readable versions of the programs, but the uni fied collection will cost less.If someone makes the odd dupe, from the
Copy Copy My camera-ready art for page three this issue is a mixture of brain overspills, mainly because I couldn't produce just one that would cover the entire page. With so many projects on full burn over the last two months (Writer's Guide, Bits Book, Q-Link/CIS, iNet, T-Shirts, Communications Disk, Micro Sleuth, disk ROMs, typesetter formats, plus others), I must admit I've been a little out of touch with the market outside my immediate neighborhood of magazine dealings. So much so that I even made some phone calls to help refresh my
pages or sectors, so be it. But in the context of my second paragraph, our product is software with inexpensive (usually required) documen tation. Alternately, we offer a magazine with an optional inexpensive service that can easily be justified in dollars per hour. Either way you
opinionatedness. And today, with all front burners on full for so long, my neurons aren't holding as strong a charge as usual. I suppose every writer has this experience at one time or another -you start to ramble just so you don't have to expand your type size and leading in order to fill a page and appear like a uninspired door-knob that can't.
look at it, wide spread duplication will only help to shorten our life as a supplier. Remember, we wouldn't be the first computer publication
. . er, see what I mean? Ok Karl, get on with it.
to become another page in Chapter I I , although we're probably the first to meet you half way.
First I'd like to retract a statement from page three last issue. I suggested there probably isn't one computer hobbyist out there without possession of at least one copy of a some piece of proprietary software that they haven't paid for. I suppose I exaggerated a little, but only to make a point: manufacturers need to offer more in package
Did I hear someone say, "you hypocrites - your SuperKit/1541 will make copies of protected packages, and you're selling 1541 upgrade kits that make only minor changes to the original."? No? Good. Because once you buy a piece of software, you also buy the rights to personal use as you see fit. Super Kit is for making personal backups handing them out is against the law, and that's not why we sell it. Secondly, we're not cloning disk drives- we're upgrading the operat ing system that 1541 owners have every right to use.Case in point: an upgrade ROM is available for the onboard computer of new Corvettes. It adds about 60 h.p. at around $10 per horse. Apparently the ROM
than just a program to accrue sales. By including a "batch of goodies" with the software, the customer who sincerely wants the item will be less willing to settle for the bare-bones, unprotected version. Besides, getting the occasional freebie isn't the worst crime in the world. But large-scale distribution of a pirates de-protecting efforts is, quite simply, detestable. That was my main point, which brings me to my next.
contains only slight mods to the original, but owning the car means you have the right to install those mods. It's just like adding headers they're the same at both ends, different in the middle. Nobody is breaking the law, but if a second manufacturer makes headers of identical design, room exists for a conflict.
We've been receiving a number of letters regarding local duplication of Transactor Disks. Here's a few clippings: "...I would like to sell Transactor Disks through our club ..." " ... we felt that since the programs are considered public domain that the disks could be included in the club library..." "... we we
My last spill is open for discussion, and responses are welcome. When the CBM 8096 was released, Visicorp decided they would not produce a version of Visicalc that would access the extra 64K. I don't know who, but someone else did. The 96K version powered up with an incredible 70K of workspace (IOK on the 8032 version) and it was even offered to Visicorp free of charge! They still didn't want it, but that did not make it ok to give out. Now that Visicorp has folded, what about Visicalc. Is it public domain? I'm sure the "YIN" fraction contains many arguments here. Some would say nobody will suffer..
would also like a subscription to The Transactor Disk, which
would make available to club members, with your permission."
"allow members who can show us a copy of the magazine to copy the . Sk..." dI " ... it seems to be a nuisance to mail off for them ..." " ... the prc i e doesn �seem to be a problem ..." " ...do not believe our members would be willing to purchase sub scriptions at $45 even though that is a reasonable charge for 6 disks that save typing the programs in. .. "
. except their existing competitors of course. What about other suppliers who may be gone, but whos software is not forgotten.And
Get the picture? Since introducing The Transactor Disk, we've re ceived the same request no less than 50 times. Our policy is, "free to copy, not to sell" (see bottom of opposite page). Although the disks are marked "(c) copyright. .. ",there is no copy protection and we couldn't even afford any attempts at thwarting copy activity. But consider this: we spend a lot of time on those disks -testing, editing, commenting, sometimes de-bugging -Chris, Richard, and myself spend between one and two days to complete each. Even with the price increase, they
what about the law? Does anyone know? Personally, I could go either way on this one. So readers, any ideas? There is nothing as constant as change, I remain
�---
still cost less than disks from many of the other magazines. And what little profit w e take (important revenue for us) doesn't even put a dint in our total expenses.
Karl J.H.Hildon, Editor In Chief and I thought I wouldn't need to scrunch this page this time. huhg
3
No¥. 1986: Volume 7, laue 03
Using "VERIFIZER" The Transactor's Foolproof Program Entry Method VERIFIZER should be run before typing in any long program from the pages of The Transactor. It will let you check your work line by line as you enter the program, and catch frustrating typing errors.· The VERIFIZER concept works by displaying a two-letter code for each program line which you can check against the corresponding code in the program listing. There are five versions of VERIFIZER here; one for PET/CBMs, VIC or C64, Plus 4, Cl28, and Bl28. Enter the applicable program and RUN it. If you get a data or checksum error, re-check the program and keep trying until all goes well. You should SAVE the program, since you'll want to use it every time you enter one of our programs. Once you've RUN the loader, remember to enter NEW to purge BASIC text space. Then turn VERI FIZER on with: SYS 634 to enable the PET/CBM ver sion (off: SYS 637) SYS 828 to enable th eC64NIC version (off: SYS 831) SYS 4096 to enable thePlus 4 version (off: SYS 4099) SYS 3072,1 to enable theC128 version (off: SYS 3072,0) BANK 15: SYS 1024 for B128 (off: BANK 15: SYS 1027) Once VERIFIZER is on, every time you press RETURN on a program line a two-letter report code will appear on the top felt of the screen in reverse field. Note that these letters are in uppercase and will appear as graphics characters unless you are in upper/lowercase mode (press shill/Commodore on C64/VIC). Note: If a report code is missing (or "--") it means we've edited that line at the last minute which changes the report code. However, this will only happen occasionally and usually only on REM statements.
PET/CBM VERI FIZER (BASIC 2.0 or 4.0) Cl CF Ll
HC DH GK OG JO AF IN ON IB
CK EB HE 01 JB PA HE EL LA Kl
EB OM
KE JF
With VERIFIZER on, just enter the program from the magazine normally, checking each report code after you press RETURN on a line. If the code doesn't match up with the letters printed in the box beside the listing, you can re-check and correct the line, then try again. If you wish. you can LIST a range of lines, then type RETURN over each in succession while checking the report codes as they appear. Once the program has been properly entered, be sure to turn
BE DH GK FH KP AF
VERIFIZER off with the SYS indicated above before you do anything else.
EC EP OC
VERI FIZER will catch transposition errors like POKE 52381 ,0 instead of POKE 53281 ,0. However, VERIFIZER uses a "weighted checksum technique" that can be fooled if you try hard enough: transposing two sets of 4 characters will produce the same report code but this should never happen short of deliberately (verifizer could have been de signed to be more complex, but the report codes would need to be
MN
Technical info: VIC/C64 VERIFIZER resides in the cassette buffer, so if you're using a datasette be aware that tape operations can be dangerous to its health. As far as compatibility with other utilities goes, VERIFIZER shouldn't cause any problems since it works through the BASIC warm-start link and jumps to the original destina tion of the link after it's finished. When disabled, it restores the link to its original contents.
1080 data165, 253, 41, 3, 133, 254. 32,236, 2 1090data198,254, 16,249,232, 152,208,229, 165 1100data251, 41, 15, 24,105,193,141, 0,128 1110data165,251, 74, 74, 74, 74, 24,105, 193 1120 data141, 1,128, 108, 163. 2,152, 24,101 1130data251,133,251, 96
VIC/C64 VERIFIZER
Ll
longer, and using it would be more trouble than checking code manually). VERIFIZER ignores spaces, so you may add or omit spaces from the listed program at will (providing you don't split up key words!). Standard keyword abbreviations (like nE instead of next) will not affect the VERI FIZER report code.
10 rem•dataloa der for' 'verifizer4.0' ' • 15 rem pet version 20 CS=O 30 fori=634 to754:rea d a:poke i,a 40 cs= cs+a: next i 50 : 60 ifcs<>15580 then print' '•••••data error •••••' ':en d 70 rem sys634 80 en d 100 : 1000data 76, 138, 2, 120, 173, 163, 2, 133, 144 101 odata 173, 164, 2,133,145, 88, 96,120. 165 1020data145, 201, 2,240, 16, 141,164, 2,165 1030data 144,141,163, 2,169,165,133,144,169 1040data 2, 133,145, 88, 96, 85,228, 165,217 1050data201, 13,208, 62,165, 167,208, 58, 173 1060data254, 1,133,251,162, 0,134,253,189 1070data 0, 2,168,201, 32, 240, 15,230, 253
IN
10 rem•dataloa der for ' 'verifizer' ' • 15 rem vic/64 version 20CS=0 30 fori=828 to958:rea d a: poke i,a 40cs=CS+a:nexti 50: 60 ifcs<> 14755 then print' '• ••••data error •••••' : end 70 rem sys828 80 end 100 : 1000data 76, 74, 3,165,251,141, 2. 3,165 1010data252, 141. 3, 3, 96,173, 3, 3,201 1020data 3.240. 17,133,252,173, 2, 3,133 1030 data 251,169, 99, 141, 2, 3, 169, 3,141 1040 data 3, 3, 96, 173, 254, 1,133, 89, 162 o. 2,240, 22. 201 1050 data 0, 160, 0,189, 1060data 32, 240, 15, 133, 91,200, 152, 41, 3 1070data133, 90, 32,183, 3,198, 90, 16,249 1080 data232, 208, 229, 56, 32,240,255.169, 19 '
MG OM CA NG OK AN 1090data 32,210,255,169, 18, GH 1100data 89, 41, 15, 24,105, JC 1110data165, 89, 74, 74, 74, EP 1120data 32,210,255,169,146, MH 1130data 32, 240,255,108,251, BH 1140data101, 89,133, 89, 96 VIC/64 Double Veriflzer
32,210,255,165 97, 32,210,255 74, 24, 105, 97 32,210,255, 24 0. 165, 91, 24
Steven Walley, Sunnymead, CA
When using 'VERIFIZER' with some TVs, the upper left corner of the screen is cut off, hiding the verifizer-displayed codes. DOUBLE VERIFIZER solves that problem by showing the two-letter verifizer code on both the first and second row of the TV screen. Just run the below program once the regular Verifizer is activated. Now.1M6: � 7,1ttue03
KM BC Dl GO IN EN KG KO FM LP
a , da:n et x ad 100 for ad= 679to 720:readda :poke d 110 sys 679: print: print 120 print·doubleverifizer ac tivate d · :new
01 1140data LK 1150data GJ 1160data ON 1170data GJ 1180data CB 1190data CB 1200data
130data 140data 150data 160data
120, 169, 180, 141, 20, 3 169, 2, 141, 21, 3, 88 96, 162, 0, 189, 0, 216 157, 40,216,232,224, 2 0 170data 208,245,162, 0, 189, 180data 4,157, 40, 4,232,224 190data 2,208,245, 76, 49,234
PE 1210data DO 1220data BA 1230data BG 1240data
20,133,208,162, 0,160, 0,189 0, 2. 201, 48,144, 7, 201, 58 176, 3,232,208,242,189, 0, 2 240, 22,201, 32, 240, 15,133,210 200, 152, 41, 3,133,209, 32,113 16,198, 209, 16,249, 232, 208, 229 165,208, 41, 15, 24,105,193,141 0, 12,165,208, 74, 74, 74, 74 24,105,193,141, 1, 12,108,211 0,165,210, 24,101,208,133,208 96
Tom Potts, Rowley, MA
VERIFIZER For Tape Users
C128 VERIFIZER The following modifications to the Verifizer loader will allow VIC and 64 owners with Datasettes to use the Verifizer directly (without the loader). After running the new loader, you'll have a special copy of the Verifizer program which can be loaded from tape without disrupting the program in memory. Make the following additions and changes to the VIC/64 VERIFIZER loader:
CF HA OH HL CB CP CB ME FG
fori= 850to 980: read a: pokei,a NB 30 ifcs<>14821then pr int .....data error**•** ·:end AL 60 IB 70 rem sys850 on, sys853 off -- 80 dele te li n e -- 100 dele te lin e 3, 165 OC 1000data 76, 96, 3,165,251,141, 2, MO 1030data 251, 169,121,141, 2, 3, 169, 3,141 EG 1070data 133, 90, 32,205, 3, 198, 90, 16,247 ·
FK
MD OJ MF OM El ON NH IJ ML DE ON
BD 2000 a $ = verifizer.sys850( s pace]" KH 2010 fori= 850to980 GL 2020 a$ = a$+chr $ (peek(] i : n et x DC 2030 open 1,1, 1,a$: clos e1 IP 2040 end •
Now RUN, pressing PLAY and RECORD when prompted to do so (use a rewound tape for easy future access). To use the special Verifizer that has just been created, first load the program you wish to verify or review into your computer from either tape or disk. Next insert the tape created above and be sure that it is rewound. Then enter in direct mode: OPEN 1 :CLOSE!. Press PLAY when prompted by the computer, and wait while the special Verifizer loads into the tape buffer. Once
LM
LE HC
1000 rem •data loader forverifizer128 1010 rem • commo dorec128- 40 a nd 80column mo de 1020cs=O x n et x 1030forj = 3072to3226: read x:pokej,x : cs =cs+ : 1040 ifcs<>19526then pri nt ·che cksum error!·: t s op 1050 print "sys 3072,1: remto en able· n sys 3072,0: remtodis able 1060 pri t 1070 rem 1080data 201, 0,208, 13,120, 165,253, 141 1090data 20, 3,165,254,141, 21, 3, 88 1100 data 96,120, 173, 21, 3, 201, 12,240 1110data 17,133,254,173, 20, 3,133,253 1120data 169, 44,141, 20, 3,169, 12, 141 1130data 21, 3, 88, 96,165,240,201, 13 1140data 208, 94, 165, 22,133,250,162, 0 1150data 160, 0,189, 0, 2. 201. 48,144 1160data 7. 201, 58,176, 3, 232,208,242 o. 2,240, 22. 201. 32,240 1170data 189, 1180data 15,133,252,200,152, 41, 3,133 1190data 251, 32,147, 12,198,251, 16,249 1200data 232, 208, 229, 56, 32, 240,255, 169 1210 data 19, 32,210,255, 169, 18, 32,210 1220data 255,165,250, 41, 15, 24,105, 193 1230data 32,210,255,165,250, 74, 74, 74 •
1240 data 74, 24, 105, 193, 32,210.255,169 OF 1250data 146, 32,210. 255, 24, 32,240,255 NC 1260data 108, 253, 0,165,252, 24,101,250 LF 1270data 133,250, 96 KE
loaded, the screen will show FOUND VERIFIZER.SYS850. To activate, enter SYS 850 (not the 828 as in the original program). To de-activate, use SYS 853.
Elizabeth Deal, Malvern, PA
8128 VERIFIZER If you are going to use tape to SAVE a program, you must de-activate (SYS 853) since VERIFIZER moves some ofthe internal pointers used during a SAVE operation. Attempting a SAVE without turning off VERIFIZER first will usually result in a crash. If you wish to use VERI FIZER again after using the tape, you'll have to reload it with the OPEN 1 :CLOSE 1 commands.
1 rem s ave· @O:verf i zer i b128 ,8 10 rem•data loaderfor verifizerb128" • 20cs=O 30b a n k15:fori =1024 to 1163:read a: pokei,a 40cs cs +a:n ext i 50 ifcs<>16828 the n print· • • data error • • : end 60 remb an k15: sys 1024 70 end Plus 4 VERIFIZER 1000data 76, 14, 4, 165,251,141,130, 2,165,252 1010data 141,131, 2. 96,173,130, 2.201. 39,240 Nl 1000 rem • data loaderfor verifizer+ 4 1020data 17,133,251,173,131, 2. 133,252,169, 39 PM 1010 rem • commo dore plus/4vers ion 1020 gra phc i 1: scnclr: gra phc i 0: rem ma ke roomforco de 1030data 141,130, 2,169, 4, 141,131, 2, 96,165 EE 1,202. 165, 27. 133 NH 1030 CS=O 1,134, 1040data 1. 72,162, 1040 forj =4096to 4216: read x : pokej,x :ch=ch+ x : n ext 1050data 233, 32, 118, Jl 4, 234, 177. 136. 240, 22. 201 AP 1 050 if ch<> 13146then print ·che 1060data 32,240, 15,133, 235,232,138, 41, 3,133 cksum error· : t s op NP 1 060 print sys 4096: remto en ab le· 1 070data 234, 32. 110. 4, 198,234, 16, 249,200. 208 1080data 230,165,233, 41, 15, 24,105,193,141, 0 JC 1070 print sys 4099: remtodis able" 1090data 208,165,233, 74, 74, 74, 74, 24, 105, 193 10 1080 end 1100data 141, 1,208, 24.104,133, 1,108,251, 0 PL 1090data 76, 14, 16,165,211,141, 2, 3 CA 1100data 165,212,141, 3, 3, 96,173, 3 1110data 165,235, 24, 101,233,133,233, 96,165,136 1120data 164, 137, 133, 133, 132, 134, 32, 38, 186, 24 OD 1110data 3,201, 16,240, 17,133,212,173 1130data 32. 78, 141, 165, 133, 56,229, 136, 168, 96 LP 1120data 2, 3, 133, 211, 169, 39, 141, 2 1140data 170,170, 170,170 EK 1130data 3,169, 16, 141, 3, 3, 96,165 •
•
=
•
•
•
•
•
5
Nov. 1986: Volume 7, Issue 03
Bits and Pieces
Got an interesting programming tip, short routine, or an unknown bit of Commodore trivia?Send it in- ifwe use it in the Bits & Pieces column, we'll credit you in the column and send you a free one year's subscription to The Transactor
C-128 Program Entry Tips
David J. Tajkowski
the light on and it looks like it won't be coming back up with a "guru meditation 11" or by re-booting on its own, you should eject
Baldwinsville, NY
the disk before you re-boot with Amiga/Amiga/CfRL. A forced re-boot seems to zap the disk on occasion.
Here are some tips for C-128 users who enter magazine program listings using line verification utilities such as Transactor's VERIFI ZER. These tips can help make program entry less painful and time-consuming.
If the above advice comes too late and you already have a bad disk in your collection, you may be able to recover it with "DiskEd", a
The first is almost too easy - if you are in eighty-column mode
direct-access disk editor that comes with the developer's package for the Amiga. DiskEd requires that you understand the format of
simply type the command FAST. This will reduce the noticeable delay from the time you hit enter until the cursor returns or the
Amiga's file structure (Explained in an article next issue), but an automatic disk-fixer called DiskDoctor is on the new version 1.2
next line number appears (if you are using automatic line number ing).
software support disk. Just say diskdoctor dfO: (or dfl :), and the program does its best to automatically fix up a bad disk the best it can. Having a disk restoring program on hand may save you a lot
The next tips concern the use of the function keys. The placement of the C-128's function keys just above the numeric keypad and
of re-typing time.
the programmability of the keys can can greatly ease the entry of programs. For example, when entering BASIC loaders which require many lines of DATA statements, re-define Fl as 'DATA' and F2 as','; then it will not be necessary for your hand to leave the area of the keypad at all.
Easy PAL to CBM Source Conversion A CBM assembler format source file is just a sequential text file with no line numbers, while the PAL assembler uses a format identical to a BASIC program. A surprisingly easy way to create a CBM-format file on disk from a PAL file in memory involves just one POKE to list the program without line numbers. On the C-64, the entire conversion process can be accomplished like this:
It is sometimes necessary to "go back" and re-check the verifica tion codes of parts or all of a program. Changing the definition of F7 to eliminate the CHR$(13) will make listing ranges of lines easier by eliminating the need to type in the word UST.
open 1,8,2, filename,s,w·: cmd 1: poke 22,35: list (then poke 22,25 to return to normal listing mode) •
Lastly, you can automate the entire process even more by creating a disk just for program entry. Set up the disk to auto-boot the
On P E T machines, the POKE is 19,32; and 19,22 to get back to normal.
verification program (VERIFIZER) and add the function key re definitions to the program itself. Then when then disk is booted you will be set for fast, efficient and accurate entry of programs.
MicroSTP
Frank DiGioia, Stone Mountain, Ga
Stop Amiga Disk-Death Chris Zamara published a nifty ML program in Transactor Vol. 5, One of the most common causes of Amiga disk failure is pressing the eject button while the drive light is on. This can mess up
Issue 6 called STP. Later, in volume 6, issue 4, Jack Weaver submitted a I 0 line BASIC program to Bits & Pieces which worked
important pointers on the disk, making it totally unreadable. Ejecting a disk at the wrong time is easy to do, especially since versions 1.0 And 1.1 of the operating system often put up disk
similarly to Chris's STP. Well,l couldn't resist adding my two cent's worth. Here is a one-line BASIC program that I wrote which has all the power of Chris's original program:
change requesters before the drive light has gone out - oblige the prompt speedily and your reward can be a zapped disk. (This has been corrected in Kickstart 1.2.)
open 2,8,2,·file·: poke 781,2: poke 812,73: sys 65478
The general rule, then, is never press the drive eject button when
When you use it to execute a sequential file of BASIC direct mode commands, it will execute all the commands in the file and then
the red light is on. Well, almost never. If the system crashes with
attempt to execute blank lines until you hit STOP/RESTORE.
6
New. 1916: Wume 7,111LJe 01
$0530 LOA $0101 $0533 BCC $0538 $0535 LOA $0102 $0538 BEQ $053F
When used to merge or tokenize a sequential program listing from disk, it will perform the desired function and then terminate at end of file with a syntax error (caused by the "READY." produced by US1). Not bad for a BASIC one-liner.
Load DOS type for drive # 0 Branch to $0538 Load DOS type for drive # 1 If DOS type is 00branch to write job
$053A CMP $FED5 If not compare to type 65 $0530 BNE $0572 If neither branch to error routine Continue write job $053F TXA
(Not bad indeed! Whats more, you can go one step further and execute any series of commands from a sequential file without having to RESTORE. Just end the command file with CLOSE2:PO KE812,47:SYS65484, and the sequence will terminate properly, closing the file and returning the system to normal! Other than the fact that it displays the lines being entered, the "nifty" STP program has nothing on this one! -CZ)
Some C-128 Tips
Andy Hochheimer, Wallaceburg, Ont.
I found that you can enter the I 28's Machine Language Monitor by holding down the RUN/STOP key at power-up. I've also found that some 80-column display programs don't take
Steve Lofton, Centralia, IL
advantage of the FAST command. They instruct you to insert the disk, then power up with the 40/80 key down. Instead try this:
Computing around one day, I was wishing, "It sure would be nice if I could afford a Transactor subscription. But they're BIG league, I said to myself." My hopes of winning a Bits & Pieces scholarship crashed. Then my Commodore 64 whispered fo me.
Depress the 40/80 display key, power up, type FAST, then insert the program disk and enter the command BOOT. The program will
Border Cross
boot just as if you had the disk in as they instruct. Some programs really benefit from this.
"Human unit, don't feel chr$(17) and chr$(31), send in that cute IRQ routine you were working on yesterday."
George Leotti
Disabling The C-128 RUN Key
Glenolden, PA
"Okay", I said, stroking her keys, "I'll give it a try, Connie Have you 128'ers ever pressed SHIFT-RUN/STOP by mistake, and
Commodore."
lost what was in memory? Here is my solution to this mis-key: So here it is. This program will scroll a multicolour bar in the border whenever a key is being held down. This adds new fun when you cursor to and fro.
poke 4104,0
disables RUN key
poke 4104,9
enables RUN key
This works by changing the number of characters in the RUN key's
10rem border cross- steve Iofton 20fori= 49152to 49173: reada: pokei, a : next 30data 169, 11,141, 143, 2,169,192,141 40data 144, o. 206, 32, 208 2, 96, 160, 50data 200, 208, 250, 76, 72, 235
definition to zero, effectively disabling the key. For more on defining the RUN and HELP keys, see last issues Bits & Pieces section. -CZ
C-64 Upper/Lowercase Timescroll
Jimmy Watters,
Ric O'Neill, Calgary, AB
Sussex, NB
Hiding on track 18, sector 00 of disks formatted on 1541 drives are many important byte values used by DOS. Manipulating these values can have a dramatic effect on how DOS interprets the data
Here is a neat "time scroll" for the C-64. Type it in and RUN it, making sure to enter the correct number of spaces. The more text on the screen, the better the result.
Disk Protection
read from the disk. 10print· ·;: [ 16spaces)print" ";: (16spaces)goto 1o 20rem each ·: " is followed by 16spaces
One particular byte of interest is the third byte on sector 00. This byte indicates to DOS what type of drive was used to format the disk. DOS uses this byte to determine if it can write to the disk. If it
Try changing the number of spaces to get different effects. Pressing a key changes things too. To really notice what is happening, add the following line:
finds an illegal value for the DOS type it will not allow the drive to write to the disk. This so-called "soft protect" has been used by some to prevent writing to a disk. This is done by changing the value of byte three to almost any value other than 65. It seems to be an accepted fact that the 1541 disk drive cannot write to a disk
5fori= 1 to 999: print chr$(169);: next
formatted with any DOS type other than 65. Free C-128 CP/M Manual
Paul Reeves,
This is not correct. Disassembly of the code that checks the DOS type reveals that a DOS type of 00 can be written to also. DOS
Hamilton, Ont.
evidently checks for a DOS type of 00 first and if found continues the write job. Here's the section of code:
I have stumbled across a very helpful piece of information which will help all of you who hate spending 20-30 dollars on computer
7
No¥.1986: � 7,...,. 03
manuals. If there was only a way to make one yourself. . . Now there is; to get a "FREE" 40 page CP/M manual for the C-128, put the CP/M System Disk in drive A and type the following:
certain exact rules. Hence if two series start with the same seed,
For a C-128 and a 1541:
Watchacallit?
(A>) pip 1st:
=
they will be identical through their entire length.
J.G. Krol, Anaheim, CA
The symbol • is widely used in computer programming, as to indicate multiplication. The name of • is not "star''. It is called
help.hlp
asterisk.
For a C-128 and a 1571:
The symbol I is also widely used, as to indicate division or optional selection. The name of I is not "slash". It has two different names depending on its usage. When I is used mathematically to indicate
(A>) pip e:= a: help.• (A>) help [extract] (A>) 1st:= help.dat
division, like 4/5 = 0.8, it is called solidus. When I is used to With the 1541, the manual will have a few pages of junk at first but after that all will be fine. When using the 1571 version, it will ask you to insert disk E into drive A; when it does, insert a blank disk formatted double-sided. You need to do this because all the information will not fit on one side (the Help.HLP file is 83K).
indicate optional selection, like his/her, left/right, or and/or, it is called virgule. A unit of time that is I /60 (that's a solidus) of a second is widely used within Commodore computers. It is called in Commodore documentation a "jiffy". That is an unnecessary neologism. An cient Babylonian astronomers divided an hour of time or a degree of angle into 60 equal parts, each called a minute (MIN-ute}, which name is cognate to minute (my-NOOT) meaning very small. For a smaller unit they divided a minute into 60 equal parts. That took a
Harold Anderson,
Non-Random Surprises
Oakvllle, Ont.
Type this into your C64 and run it:
second division by 60, so they called the unit a second. For a yet smaller unit, they divided a second into 60 equal parts. That took a third division by 60, so they called the unit a third. This "sexagesi
10 n rnd(0)•1024 20 poke 1024 + n 24 : poke 55296 + n,1 30 goto 10 =
mal" system is still widely used 5000 years later for measuring time and angle. It can be extended to "fourth", "fifth", and so on as needed. Thus a so-called "jiffy" is actually a third, except that the news hasn't reached Commodore's tech writers yet, not after 5000 years, which shows just how slow the mail is these days.
,
You would expect to have the whole screen filled in a random manner with white X's. Surprisingly, what you will find is that 3/4 of the locations are never filled and you end up with a pattern of vertical stripes on the screen. Clearly, the number returned by RND(O) is not very random. If you change RND(O) to RND(I) the problem disappears.
Mr. Krol is right - I looked it all up in a dictionary. But please, programmer's jargon is confusing enough. You want to start using Ancient Babylonian? -CZ
I investigated this at some length and found that the number returned by RND(O) is of the form:
Quic.k.er Assembly on the Amiga
Norman MacDonald Medicine Hat, Alberta
RND(O)= A X 1/256 + B X 1/(256 X 256 X 256) Where A is any integer 0 to 255 and B is any integer 0 to 63
While using the Amiga Assembler I was disappointed at the amount of time it took to assemble a short program when some of
This means that there are only 256 x 64 (16384) different values of RND(O) Also notice that the second term in the above formula is much smaller than the first. The result of this fact is that the possible values of RND(O) are grouped in tight bundles around the
the 'include' tiles were included in my source file. I have come up with a simple solution to the problem that cut the assembler's workspace and the assembly time in half. Here is what I did: I wished to use some of the Intuition structures for setting up a screen and window, so I made a source file as follows:
values indicated by the first term of the formula. This bunching explains why the screen is not filled evenly in the test program.
include 'exedtypes.i' include 'intuition/intuition. i'
The values of A and B in the formula are apparently derived from the time, since messing with the length of time between successive calls for RND(O) seems to affect the randomness even further.
You may wish to add other include tiles than the ones I used, but be sure to have 'exec/types.i' as the first include file. When my source file was set up, I assembled the source file above with the
The RND function on the C64 should be regarded as follows: A call for RND(O) gives one of 16384 possible starting "seed" values for
following lines from CU:
16384 possible random series. A call for RND(I) gives a value derived from the last value found by the RND function. While RND(I) returns values which are truly randomly spaced, each successive value is derived from the previous value according to
assem : i nf ile
-c w250000 -i :include -e : outfile.equ
This assembled an "EQU file" of all symbols that were present in the source file. This "EQU file" may then be used in your source
•
Nov. 1916: Volume 7, 111 ae 03
and will perform exactly as the include files would. You may even wish to examine the file and trim some more fat off it by deleting the definitions that you do not use.
If the programmer's reference guide is consulted, it is obvious how the memory locations are used. Locations $0295- $0296 are "RS232 non-standard BPS". Locations $0299 - S029A are "RS-232 Baud rate". With about 40-50 terminal programs, along with the
The advantages of doing this is a large decrease in the assembly time and less workspace is required to assemble a program. The
programmer's reference guide and The Transactor, we came up with this bit of painless BASIC code. I hope it helps ease the pain of
main disadvantages to doing this is that you do not have access to the macros defined in the 'include files'. Another disadvantage is
those who have had a chance to try and swallow the Programmer's Reference Guide's advice on Baud change.
that if you include two of these "EQU files" in an assembly and they contain duplicate symbols, you will end up with some errors during assembly. One solution to this problem is to change all the
Borrowing Money?
EQUs in the EQU file to SETs; this will allow the symbols to be defined more than once.
After those fun programs, here's something that's actually useful. Given the rate of interest of a loan and number of payments, it prints a table with monthly payments from $10 to $1000 per
This is the fastest method of decreasing assembly time on the Amiga that I have found. I hope it will be of use to anyone doing assembly work on the Amiga.
month in one column, and the amount of the loan in another. With this table at your disposal, you can find out how much you want to borrow based on what you are willing to pay each month. Alterna
A further increase in speed can be obtained by making an "in clude" directory in the RAM-disk ("makedir ram:include'J, copy ing the include (or EQU) files that you need to "ram: include': and
tively, you can look along the right-hand column until you find out how much you need to borrow, then look across to the left column to see what it'll cost you for each payment. So, if you're
using "ram:include" instead of ":include" as the "-i" assembly
buying a new car, you can scan from Honda up to Porsche until
option. If you have expansion RAM beyond the built-in 512K, you can put everything in the RAM disk (source, libraries, include files, object, etc.) and get super-fast assembles. The same applies for compiling. Even if you don 1 have extra RAM (a 2 Meg box is nice, but expensive), try squeezing as much as possible into RAM before
you
see what
you can afford.
The program formats the numbers so that the columns line up at the decimal point regardless of the values. As listed, it prints table to the screen - change the '3' in line 120 to '4' to get a print-out. The payments on the left side go from $10 to $1000 in steps of $10
assembling. You'd be surprised at how much faster things go when the system doesn t spend its time waiting for the read/write head to seek. -CZ
- you can change this with the for-next loop in line 200
Changing Baud Rates For The
James Shelley
PO AD OK
C-64 with 1650/6240 Modems
South Bend, ln.
CJ
100 remu loan principal - cz '86 . 110 s$="[13 spaces]':l=10 120 open 1 ,3: rem 3 =screen, 4=printer 130 input· annual interest (%)[4 spcs)12 [4 crsr lefts] ;ai 140 input· number of payments[S spcs)48 [4 crsr lefts]· ;n 150 i= ai/1200 :rem mo nthly interest 160 print#1 ,chr$(13): rem blank lines 170 print#1,·interest= ai·% , # payments= n 180 print#1 payment· amount of loan 190: 200 for pv = 1 0 to 1000 step 10 210 rem compute amount of loan 220 p = pv•(1-(1 + i)t(-n))/i 230 rem format amount field 240 p$= right$(s$ + str$(int(p• 100 + .5)),1) right$(p$,2) 250 print#1,pv,left$(p$,1en(p$)-2) 260 next pv 270 print#1: close1 .
•
After reading Tony Valeri's article in Volume 6 Issue 2 and the tip
DO
from Daniel Bingamon on modem speed-up to 450 baud, I thought a different approach might be of interest. Several of my
CO
friends and I have been working on a terminal program to use with our BBS. We found it desirable to change the baud rate in our
00
terminals and BBS while on-line. As the 1650/6240 modems can't communicate at 450 baud while set at 300, a way was needed to
OF
JF CD CM BG MN
change the baud rate and not re-run the terminal program or BBS or close and re-open the RS-232 file. We use the four lines of BASIC below. It will change the baud rate on the terminal and the BBS without any ill effects. This method has been tested to 1200 bps with good results. This does not change any parity, word
CM
length, etc. When the user logs off the BBS, a GOSUB in the log-off routine passes through these lines with the variable 'baud' set to
KA NK AF
300, and the BBS is back to 300, ready for the next call.
LL
•
, •
•
, •
•
• . •
10 open 5,2,3,chr$(6) + chr$(0) 20 : 30 rem - more pgm. lines 40 : 50 input· baud rate> ;baud 60 a= int(1 022730/baud/2-1 00): c= int(1 022730/baud) 70 d= int(a/256): e = a-d•256: poke 661 ,e: poke 662,d 80 d=int(c/256): e = c-d•256: poke 665,e: poke 666,d •
9
No¥.1916: � 7,1nue 03
Letters
should supply Randall with an equal number of diskettes (4for the New and approximately 10 for the Old) plus return postage. For
Column For Machine Language Sub-Routines?: I think that it is fair to say that The Transactor caters somewhat to the machine language programmer. How about a column devoted to machine language subroutines such as division, multiplication, base con versions, memory moves, etc. Published routines could be divided into two catagories: I) The shortest - for programmers trying to cram as much into a
people outside of the US, remember that only US postage is good in the USA. Ask your post office for help. They will gladly oblige. One additional thing to consider: Randall will have to pay for mailers to mail your diskettes. Any small amount that you could send towards the cost of the mailers would surely be appreciated. To contact Randall:
small place as possible.
Randall J. Bernard Box630
2) The fastest - measured in machine cycles, for the programmer who needs blinding speed.
MorenCI: Arizona, 8 5540 (602) 865-3550
Any routine could be updated if it's faster or shorter than its predecessor. The result of all of this would be a fantastic library of
One final thought. Many different religious backgrounds make up the world. As such, I feel that Randall's efforts should only be the beginning. It would be beneficial for so many people to also have English translations on disk for the Torah, Talmud, Koran and other religious scriptures. If anyone would like to begin efforts
be useful to any ML programmer. Who knows, these routines could even be compiled into a book at a
subroutines that would
future time! There are plenty of clever ML programmers out there sitting on hundreds of useful routines - why re-invent the wheel? Rick Nash, Millersburg, Ohio
similar to that of Randall, please drop us a letter. If you are fully willing to expend the effort, we will be more than willing to spread
Sounds interesting enough. At the moment we tend to include these in the Bits and Pieces section -perhaps a sub-section of Bits could be used to host your idea. Of course it all depends on what kind of
the word.
mail we receive, so, starting now, if we get enough fast and short MLs, we'll group 'em like you say. Same deal as Bits free subscriptions for those published. Thanks for the input!
Richard Evers
Flexible Vector Problem: Murphy's law has intervened in my
-
efforts to use your Flexible Vector program (Transactor Volume
6
Issue 02 page 64).
Disk Formal: Once in a
Actually this same law has a corollary originated by Theophilus
while, I receive a telephone call or letter that makes my day. Today
Cole during the last century and it has become quite well known in
was such a day. Randall J. Bernard in Arizona called to thank us for running his letter with regards to the help he required to key in the New Testament (see Volume 7. Issue OJ). He said that he now has the entire New Testament on four diskettes and has 60 people
some circles as Coles Law which, with the spaces removed repre
New Testament Available On 1541
sents a statement much more palatable than Murphy's law, even though it has the same meaning.
working on keying in the Old Testament. All that was still required was final editing of the New Testament and waiting for the keying in of the Old Testament to be completed by all the people who
Although
I have copied the code with due vigilance and some
attention to detail, I am unable to obtain the results expected. The code assembles well with PAL and responds to SYS by placing the
volunteered. He was overwhelmed by the response generated by his single letter. In truth, it also overwhelmed me.
40-digit message on the top line. The difficulty is with the non erasability which doesn't exist in my rendition of the program. Scrolling North erases lines, ultimately including the top-line
According to Randall, the final editing stage would be the most time consuming thing to come. One suggestion was to ask the help of a few Bible study groups to check for errors and omissions. If anything, this would ensure an accurate transcription of the Testa ment. Randall asked if I could mention that anyone who wanted the New Testament on disk could have it for a phone call. Now, I
message. Run-Stop/Restore has the same effect.
As an earnest novice at assembly language, I would greatly appreciate your advice in pointing to the error I must have made which prevents the non-erasure option from working.
don� know about you but I feel that Randall has already put enough into this venture. Between the time and effort that he has already expended, and the additional time and effort from his
Bob Tischer, Starkville, MS
Below is a corrected version of the 'F7exible Vector Management' program. The problems you encountered were not your fault. At the time, Chn·s forgot one important fact of life with the64; the older
volunteers. I feel tfiat it would be unjust to expect Randall to pay for all the expenses. Randall mentioned to me, though, that it would be morally wrong to charge anyone for the Bible. In this respect I
versions of the 64 had problems with colour and screen memory. Line #'s 410-420 will solve this problem. Colour and screen memory are both updatetFduring each interrupt to display the
completely agree. A nice way around this problem could be that anyone who wants the New Testament, and later the Old Testament, on diskette,
message in white at the top of the screen.
lO
100 rem save· O:irq message· ,8 110 rem •• chris zamara - volume 6 issue 02 page 64 120: 130 sys 700
Recently, I wrote an in-depth article about the C-Link IEEE-
140.opt oo 150.
the decision to cut my article really hurt.
160; 170 irqvec
488 interface to run in this issue. Unfortunately, it was decided that our reading audience contained too few people who would benefit from this article. I can honestly state that
$c000 To make amends to everyone who would have liked to team
=
a bit more about the C-Link, we are offering it to you for the
$0314
180; 190 .. ; initialization routine • sei 200 Ida #<newirq 210
price of a letter, phone calf or visit to our office (we'll also be including it with future C-Link orders). If you are at all
•
220 230 240 250 260
sta Ida sta eli
irqvec #>newirq irqvec+ 1
jsr jmp
interested in the inner workings of the C-Link, or just would
;irq vector low ;hi byte of destination address
a line. It would be nice to find that my efforts were not
370; 380 sfil 390 400 410 420 430 440 450
ldx = Ida sta Ida
#39
The SID Chip: Does It Really Exist?
I 0/4/85 - I sent a money order for $28.00 to C ommodore Business Machines, C-2659, I200 Wilson D r., West Chester,
;do disp routine
;normal irq dest 310 320; 330 .. ; irq vector-driven routine • • 340 ;displays message on top scrn line 350 displin • 360
wasted.
A long and frustrating story made as brief as possible:
••
displin $ea31
like to learn a bit more about the IEEE-488 bus, then drop us
;irq vector hi ;enable interrupts ;back to the calling prg or basic
rts
270; 280;•• new irq routine = 290 newirq • 300
;disable interrupts ;low byte of destination address
PA I 9380 to cover purchase and shipping of I 658 I SID chip, pn 906I I 2-01. Result: 6 weeks later, no chip. C. II I I 5/85- I send a letter inquiring about the status of the
;message= 40 char. (0-39)
•
above order. Result: nothing.
message,x ;next char from message $0400,x ;store in next scr mem byte
C. 12/15/85- Another letter amplifying first.
#1
;•• char colour white • •
sta dex
$d800,x
;.. store it in colour memory ;do all 40
bpi rts
sfil
Result: silence. ••
I /3 I/86- I send tracer to Postal Service on money order. Result: money order was cashed by Commodore Business Machines 11/8/85. Status of chip: nowhere to be found.
;byte #0 is the last one
460; 470 message .asc 0123456789012345678901234567890 123456789"
I am sending this letter to Commodore and also several
•
magazines that serve Commodore users because I am hop ing that if I can't get a response out of CBM, I can at least air my complaint in public.
G-Unk Info: I think I want a G-Link Interface, as described in the May, I986 Transactor, but need more information.
Calling (2 I5) 436-4200 is a waste of time. The phone is almost always busy and when it rings, no one answers.
My purpose is to connect a C-64 computer to an MSD SD-2 dual disk drive, using the parallel lEEE port to shorten disk access time. Questions are:
This particular part cannot be purchased from any third party that I know of. I can almost live without sound but some software uses the random number generator in voice
I. Does the G-Link connect to the C-64 expansion for parallel operation? 2. If so, what speed advantage can I expect?
3. Roy M. Randall
3. What additional cable must I supply? 4. Is necessary software included? (The article stated complete transpar ency.) 5. Has the G-Link been used with the MSD drive with success? Just what can I expect? Willard E. Wright, Bellevue, Washington
6032 Edsall Road "203 Alexandria, Virginia 22304 What can I say? Obviously, your order went missing and nobody has been designated to trace it back. I know for a fact that quite a few people at Commodore read The Trans actor. Perhaps somebody from CBM will contact Roy and
The C-Link IEEE-488 interface connects to the C-64's cartridge port to
help straighten this out. If all else fails, try dialing (215) 431-
provide true, trpnsparent IEEE-488 bus access. The only extra that you
9100. This number at 1200 Wilson Drive, West Chester does
will need to make use of our interface is a single Pet to IEEE cable. You
work and is usually not busy.
can expect a data transfer rate increase, in relation to a normal/54/, in the area of 50 to 650 percent, depending on the drive used. The C-Link has been used with the MSD drives with quite a degree of success.
n
Transactor Disk Dlrectorytions: I just received the disk for
the SuperPET, but whether they still stock programs may be another matter.
Volume 5, Issues 1-3. When I put the diskette in my disk drive, I write on my computer:
LOAD" •. ,8
Verlflzer for the
and
verifizer going. Is it me or you? I double double checked every thing twice but I still could not get it to work.I finally disassembled
LOAD. •. ,8,1
C128: I couldn't get the C128 version of
it and found that it starts with a branch. Is that right? Then when I write RUN the computer comes back to READY. When I LIST, I get...
Mike Iafrate, Parkersburg, WV Verifizer for the C/28 works as listed- to make sure, I keyed it in verbatim from the pages. Remember, you must be in 40 column mode. To activate, "SYS3072, I" will do the trick (actually, any thing from 1-255 will work as the parameter after the comma). To
10 REM--- VOLUME 5 ISSUE 1 -- READY. Please Help! What is the problem? My computer is a Commodore C-64. It is not the first time that I LOAD a diskette. Could it be that
de-activate, SYS3072,0 does it. I have noticed one problem, though. SYS3072,0 does not always do it on the first shot. Twice in a row always makes sure that it is turned off.
the program is not written on the diskette. I am waiting for your answer. Roger Laplante, Chateauguay, Quebec
The branch at the beginning of the code is co"ect. The C I 28 allows passing of the accumulator, x andy registers plus status register via the SYS call. The code begins with - BNE INSTALL - In simple terms, if the accumulator is greater than zero then activate the
There are 118 directory entries on that particular diskette. Of these entries, 115 are usable programs for some type of Commodore computer. Of these programs, many are Bits and Pieces items,
routine else de-activate. A little bit simpler than the 64.
others are simply to be Loaded and Run, then the balance are source code for the concepts covered in the related issue. You really need to read each issue before attempting to use the programs in
Disk Un-Assembler Once Agaln: Thanks to Transactor and J.
such an elementary manner. The program that you Loaded was the first program on disk. However, it is not really a program but
Lothian for the Disk Un-Assembler for the Commodore 64 in The Transactor (Volume 6 Issue 04). Thanks also to John R. Menke for
rather a marker that merely ads as a header to the files that come
his letter in The Transactor (Volume 6 Issue 06).
after it. If you were to look at the directory, you would have noticed that the filename listed is the Volume and Issue number covered on that portion of the disk. If you were to look through the directory even further, you would find that similar markers exist for the start
The program of Mr. Lothian and letter of Mr. Menke both contain small errors. They both recommend adding line 61:
of entries for Volume 5 Issue 02 and Issue 03.
61 MD(36)
There was a second reason for saving this marker as the first entry in the diredory: since it serves no other purpose it can be scratched
O:MD(44)
=
0
This cannot be done as the arrays are first dimensioned in line 120, and values are read into the array in line 130.
and replaced by some other more useful program that you would rather have as the first program in the directory. Using LOAD • •
=
I agree with Mr. Menke that BIT must be converted to BYTE with the Commodore assembler. He is incorrect, however, when he says:
•
would then give you access to your "fauourite" program without entering the entire filename.
"I note that lines 2020-2060 are never accessed in the un assembler, which is a shame because they would signal that a BIT operation was converted to a .BYTE. Also, lines 310-360 in the program are excess baggage, although it would be nicer if that weren't the case."
Super Empty SuperPET: We bought a SuperPET last fall, which
helps us in our office.We do not have any programs to go with it, so we have to create every program for use in the.office. Is it possible to get from you or some other people any programs such as: inventory, mailing list and programs for otrice use?
The simple answer to the BIT to .BYTE problem is DON'T ADD LINE 61. The best solution is to change line 1430 and line 2050.
Bruno Cote, 80 Leroux *104, Granby, Quebec, J2J-ISI
1430 if n>1 0 the on(n-1 0) gosub 1950, 1980, 2010, 2040 There are hundreds of programs available for the SuperPET The tn'ck is to find them. In my own personal collection I have many of the word processors, spread sheet programs, data base packages, accounting systems and such that have appeared in past. The
In line 2050, reduce the (8 spaces) to (I space) and as Mr. Menke mentioned in line 1480, be sure to include a space between .BYTE and $ in line 2050 also.
trouble is, passing them about gratis is not too kosher. My sugges tion is to locate a copy of the 'Commodore Software Encyclopedia' by Sam s Books. It lists and describes many of the packages written for all of the Commodore computers plus provides addresses for the suppliers. A few good company names to remember in your
With these modifications, you can un-assemble using EITHER BIT or .BYTE and if (BIT to .BYTE) is selected in lines 330-360, the program will signal that a BIT operation was converted to .BYTE. The MD(36)= 14 and MD(44)= 14 in line 360 accomplishes this in
quest is Batten'es Included, BMB Compuscience, BPI Micro Systems, Handic Software, ProLine Software, and Softwerx. These are but a few of the good software manufacturers and suppliers for
line 1430, which accesses line 2040 to signal the change. Any other .BYTE directives are listed normally. Patrick P. Malloy, Milwaukee, WI 12
Nov. 1916: '<lialume 7,1UTAe 03
Super Kit Users Notes: We have received quite a few letters and
Games From The Outside Out: I look forward every other
phone calls from people who have bought SuperKitlI 54I through
month to receiving my Transactor. For the most part, it is far above
the magazine. It seems a few people have had problems getting
average. Mostly, this letter is not as much of a "complaint" as it is a "reader's point of view".
SuperKit to duplicate itself as stated in the manual. Well, according to the people at Prism, SuperKit is quite capable of this task. The trouble may be that you have more than one Commodore 64, one
The cover for September said "The Tech/News Journal For Com modore Computers". It also said "Games from the Inside Out" showing pictures of demons, fighter planes, tanks, etc.
1541, and one monitor in your system, or that some hardware mod or plug in carrt idge is causing problems, or your disk drive is slightly out to lunch. As a final might-be, it could be that the disk you received is indeed bad. Many different factors could be causing
Mentally, I combined the slogan of Transactor, the image of the dark secrets of Zork exposed, or Enchanter, or if not that then a close look at multicolour, fast action of Zaxxon, or a clue as to the
you problems. SuperKit, with its vast speed and power seems to be very particular about system configuration. According to Prism, a re-write is under way that will straighten this problem out.
fabulous programming techniques of Beach Head, or Seven Cities of Gold. Oh, the thrill, the anticipation...
Now, to cover each problem sequentially. If you have a printer, extra disk drive, modem or virtually any extra peripheral device
Then, eagerly, I scan from cover to cover: One (I) Game! Animals? I had that 12 years ago (and a better version to boot) and IN BASIC!(?)
connected in any way to your computer or drive (even if it is not tumed on), this could cause problems. The Epson and Gemini printers are the biggest trouble makers for SuperKit. Disconnect everything extra completely from your system. Absolutely remove
Four valuable pages which could be spent telling me about Commodore were used for a useless ATARI review. Oh, the pain. If I want reviews, which I don't, I'll buy Compute!, which I don't.
the connectors and your problems might vanish. Please remember that extra peripherals also include Fast Load cartridges and such that are always taken for granted and forgotten. SuperKit is fast enough without any additional help.
Four more pages for Animals, and another wasted on silly car toons. In those NINE pages you could have told a lot about
Hardware mods could include a new set of ROMs for your 64 or
Commodore.
1541 that have made improvements on Commodore's design. There are lots of super ROMs out there that beef up the 110
I am especially interested in how things are done and why they are
handling plus provide all sorts of programming goodies, but
done in ML. A source of various ROM routines with comments could be a great teaching tool for a Transactor column, as could more information regarding the disk drive.
SuperKit hates them. To get all the neat goodies for you to play with, the ROM must contain substantially altered code from the original. Normal ROM entry points are no longer, some features have been completely removed to make way for the neat stuff and
Please don't misunderstand- Transactor is by far the best, but it is the best only when it is different! Please, no more reviews
pieces of code that are involved in critically timed routines are no longer so critical. Any number of mods to your ROMs could affect SuperKit, and for that matter any software package, with unpre
(especially NON-COMS) and no more cartoons to waste valuable space!
dictable results. Ifyou know that your system and SuperKit do not
Wayne Gurley, Wills Point, Texas
cohabitate well together, then return the package to
us
(with sales
receipt included). It makes little sense for you to retain a package that cannot live with your system.
Another Vote Against The Atarl Article: At the risk of being
labeled an Amiga jingoist, I am a little distressed as to the inclusion of the "Atari S T Notebook" in the latest issue. If the article was included because the Atari is a 68000 machine, then by that logic, I expect we will start seeing articles on the Macintosh. One reason why I subscribe to your magazine is because of its narrow focus on Commodore products.
The final problem that we know about could be that your drive is not as tight as it was when conceived. The head could slap around a bit too much, the belt could have a bit of slack and any number of other factors could turn your drive into something SuperKit has difficulties dealing with. When trying to duplicate SuperKit using the Super Nibbler on side two of the diskelle, set the maximum track number to 40. By telling SuperKit to bit copy as far out as
My chief complaint is that the pages used for the Atari article could have been used for an article on the Amiga operating system. The Amiga operating system is woefully lacking in documentation.
track 40, a slight bit of internal drive slack will not be a problem. If your drive problems are worse than being just a lillie bit out to lunch, then perhaps a service call is in order.
Despite this mild complaint, I think you are producing an excellent One other note: after using the Super Nibbler, it's best to send an
magazine. Keep up the good work.
Initialize command to your dn·ve. This brings the head back from the "far reaches·:
J. Richard McEachern, Saint Louis, Missouri
Let's hope that these few points make life with SuperKit easier. It is
Ouch. Being hit twice in a row about the Atari article really hurts!
truly a fine product, one that we gladly stand behind in every way.
We ran the article because it was exceptionally well wrillen,
It's just a shame that it cannot live in peace with more than its
without any biased Alan· hype. It did not contain any program
immediate family. Perhaps the much desired re-write taking place
listings, so it didn � require a machine in order to learn something
in Texas right now will cure it of its unsociable attitude. We're pretty
from it.
sure it will. You may also want to look at the two articles in this issue that address other aspects of SuperKit and other copy uitlities.
Although it went against our morals to cross over to the other side for that article, we felt it 13
was
worth it. Jack Cole is such a good No¥.19t6: � 7, ..... 03
author, and a master in R + D, that we had to run the article. At one
Improving The SYS Command: Volume 71ssue 01 page 64
point, we had been tampering with an idea for a 68000 magazine as
an insert within the Transador every issue. The extra magazine
porit on was abandoned, but Jack had already been working on the article to help us out. To finish the article Jack fought against a nasty flu bug to make deadline. Jack did pull through, and his Atari Notebook was the result. With such devotion, would it have been morally right for us to reject his effors t , especially after we had
The description of the routine at $ AD9E, described under the heading Routines used in passing parameters: said that the routine evaluates an expression and if the result is a string, puts it at $0100. In reality, the string doesn't go here at all; the routine returns a pointer to the string descriptor in locations $64 and $65 (these point to three bytes containing the string's address low, address
initially solicited his aid? We say NO!
high, and length}. Another point that should have been mentioned is that you should call $86A6 after $ AD9E to clean up the string
It might also be interesting to note that the ST is more like a
descriptor stack. After calling $B6A6, a pointer to the string itself will be in locations $22 and $23 (also in .X and .Y). and the
Commodore than anything Atari has built. Furthermore, the Amigo is more like an Atari than anything Commodore has produced. In fact, the Amigo project was originally commissioned
accumulator will hold the length of the string. If you call $B6A3 instead of $B6A6, it will first check that the expression was of a
by the pre-Tramiel Atari Corp. Jay Minor, who is mostly responsi
string type.
ble for those three workhorse VLSI chips in the Amigo, was also the chief designer of the "Antic" chip in the early Atari machines. And f d mostly by former Commodore employees the ST projed is stafe it's even possible that they stare t d the ST before becoming
SYMASS 3.1 Fixes
ex
sure.
The Symass 3.1 Assembler on Transactor Disk 1112 has a bug that makes it corrupt the symbol table when it encounters an expres
Referring back to Mr. Gurley's letter, it would take months to
sion defining a label as a function of another label (for example LENGTH ENDTAB-TABLE). To fix Symass 3.1 and turn it into
Commodore. A twist of events like this leaves one pondering to be
unravel the secrets of some games without source code from the author(s). Time like that we unfortunately just don� have. Quite
=
3.13, LOAD in SYMASS 3.1, make the follwing POKEs, and S AVE the updated version as SYMASS 3.13.
simply, the bulk of any issue is a collection of submissions, sprinkled with the odd item that we have time to write ourselves. And lastly, although we can� promise to exclude the cartoons, we
POKE 3304,56 POKE 2965,234
can promise that comparisons will take priority over reviews. The
POKE 5057,51
odd review may show up when it's important, but we will try to keep them to a minimum.
(The above changes turn SYMASS 3.1 into 3.12, as described in the article "Save SYMASS Symbols" in Volume 7,lssue 02)
Rx For Gram mar: There ain't no such word as "irregardless".
POKE 3312,98 POKE 3316,99
Dr. James N. Little, Walnut Ridge, Arkansas Too bad. .. It really was such a nice word to have around.
POKE 3318,98 POKE 3323,98 POKE 3328,99
Spel Cheker: Through the magazine in Vol 7
I, you misspelled
POKE 3332,99 POKE 3370,98 POKE 3389,98
"colonel" as "kernal". It's "churnal", isn't it? You even misspelled it on the kovar.
POKE 3393,98 POKE 5057,51
11
Jeff Jourard, Santa Monica, California
TransBloopers Bits & Pieces: Volume 71ssue 02 page 12
In the Amiga Startup-Sequence on page 12, the first greater-than (>) in the DATE function argument should be a less-than(<).
C128 Memory Maps: Volume 71ssue 01 page 29
Mike Mellinger from Connellsville, PA wrote in to draw our attention to a few more errors in these maps. Below is a list of the incorrect lines. with the corrections to the right. D800-D8E7 should be D800-DBE7 1178-1197 should be 1178-1179 1271-1274 4721-4274 should be 1271-1274 4721-4724
EUoinating SAVE@: Volume 7 Issue 02 page 33
Decimal addresses from 2595 through 2602 should be 2597 through 2604 (they're all off by 2).
The article did not mention that the complete ROM exists on the Transactor disk for that issue. The file is 16k long and contains a dump of the entire S AVE@-fix ROM for the 1 541. It's a "PRG" type file with a start address of $2000. Remember to strip the first two bytes if your EPROMer doesn't- otherwise the start address will be sent to the EPROM and throw everything off by two bytes!
Nov. 19t6: ¥olurne 7, .... 03
TransBASIC Install•nent #II
Nick Sullivan
Scarborough, Ont.
TransBASIC Notes TransBASIC has been a regular Transactor feature for almost two
the same system vector, however, a potential crash situation exists.
years. Those who have been following the series know all about it.
Part 7 deals with avoiding this problem.
Recently, however, we've received letters to the effect of "what is Part 8 - Describes the five modules for Part 8.
TransBASIC?". Quite simply, TransBASIC is a method of adding new commands to BASIC (see "Part 1:" below). The commands come in 'modules' which may contain one or more commands OR functions.
Part 9- Describes the six modules for Part 9, and makes first mention
After merging the modules of your choice, the entire lot is assembled
of The TransBASIC Disk.
and linked into BASIC. The new commands can then be used just like any of the other commands that are already in the BASIC ROM when
Part 10
the C64 is powered up.
minor bugs in the modules "MC GR APHics·: "MOVE & FILL': and
-
Descn'bes the six modules for Part 10, and details some
"PRG MNGMNT': The TransBASIC Disk The TransBASIC Disk contains all of the modules published so far and
TransBASIC Installment "11
it comes with its own assembler, SYMASS 31 . . Any combination of modules can be linked into BASIC with only a few simple steps. From
In this issue you'll find only one module, but it's a big one. The author
start to finish is usually no more than a couple of minutes...even less
is Paul Adams of Revelstoke, British Columbia, who wrote it to fill a
once you get the hang of it. It comes with a handy reference for just
personal need. Paul writes:
$9.95.See the order card at center page. "I was writing a program to log my daily running (by foot, not by
Note: A bug has been found in the SYMASS Assembler that shows up
computer) and needed some commands that would allow better
when you define a label as a function of another label. So far we
graphing capabilities than I could find. Of course, being an Engineer, I
haven't noticed it while using the TransBASIC Disk, only with other
like to graph everything..."
unrelated source code. Check the Transbloopers column in this issue for the fix.
Well, if you like to graph everything too, this is the module for you. Paul has come up with the interesting concept of combining two hi res screen areas into one double-height drawing area for his graphs.
TransBASIC Parts I to 8 Summary:
His plotting commands have the capability of addressing this area Part 1: The concept of TransBASIC- a custom command utility that
with y-coordinates from 0 to 39 9 (to make things Cartesian, he has
allows one to choose from a library only those commands that are
put the zero y-coordinate at the bottom rather than the top of the
necessary for a particular task.
screen).
Part 2: The structure of a TransBASJC module - each TransBASJC
On this big plane you get the capability of drawing lines and rectan
module follows a format designed to make them simple to create and
gles, either solid or in a dot pattern of your choosing. You can also
"mergeable" with other modules.
write text to the screen, including a variety of special characters that come in handy in graph work.
Part 3: ROM routines used by TransBASIC- many modules make use of ROM routines buried inside the Commodore 64. Part 3 explains
Because the hi-res screens take up 8000 bytes apiece, plus another
how to use these routines when creating new modules.
I 000 bytes of video matrix for each one, you sacrifice a bit of BASIC memory when you use this module. In fact, before you start using the hi-res, you should give the command:
Part 4: Using Numeric Expressions - details on how to make use of the evaluate expression ROM routine.
poke 56,140: poke 55,0: clr Part 5: Assembler Compatibility- TransBASIC modules are written in PAL Assembler format. Techniques for porting them to another assem
to protect the BANK 2 video matrix at $8COO and your BASIC strings
bler were discussed here.
from writing all over each other.
Part 6: The USE Command- The command 'ADD' merges TransBA
A couple more things: if you should commit a syntax or other error
SIC modules into. tex;,t space. However, as more modules are ADDed,
while you're in one of the hi-res screens, you will have to type the
merging gets slow. The USE command was written to speed things up.
DEFAULT command blind to get back. Alternatively you could in
USE also counts the number of statements and functions USEd and
clude William Turner's TRAP command from last issue in your
updates the totals (source line 95) automatically.
dialect, and make the DEFAULT command part of your TRAP routine.
Part 7
Usually TransBASJC modules don� need to worry about
Drawing large rectangles can take quite a while in hi-res, especially
interfering with one another. When two or more modules want to alter
when they're completely solid rather than patterned. I found myself
-
ltlellaneodor
15
Nov. 1986: Volume 7, ·- 03
New Commands
wanting a way to switch between the upper and lower sections of the drawing area so that I could monitor what was going on, so I added a few lines to the plotting routine that will cause the screens to be
PLOT (Type: Statement Cat
swapped when you press and release the CTRL key (the swap actually
Line Range: 12694-12972
happens on the release). This feature is only operational while a plot is
Module:
GRAPHCMDS
actually in progress. Sometimes I also wanted to see debugging
Example:
PLOT X,Y
information that I was printing to the text screen during a series of
This statement sets the specilied pixel in the hi-res drawing area. The
plots, so I also added a trap for the LOGO key. The text screen is made
pixel is cleared if erase mode has been selected. X and Y can range
visible whenever this key is held down during a plotting operation;
from 0 to 319 and 0 to 399 respectively.
the current hi-res screen is restored when the key is released.
•:
175 )
GCLR (Type: Statement Cat •: 176) Line Range: 12974-13016
If you look in the code for DEFAULT routine (in the neighbourhood of program line 13080), you'll see a patch I had to put in to cure a very odd hardware-related bug that occurred in the very standard routine
Module:
GRAPHCMDS
Example:
GCLR
The hi-res drawing area is erased.
that switches video banks from BANK 3 to BANK 0. The net effect of the bug was that the RTS at the end of that routine was frequently
GCOL (Type: Statement Cat •:177)
executed as a JSR (a $20 rather than $60). which in this case was a JSR
Line Range: 13018-13036
into the never-never. The patch, which consists of synchronizing the
Module:
GRAPHCMDS
bank-switch to the position of the raster beam, does lix the problem,
Example:
GCOL 16 •6 + 14
but I have no notion why and would love to hear from someone who
The hi-res foreground and background colours are set by this com
does.
mand. The foreground colour is set by the upper nybble of the parameter, and the background colour is set by the lower nybble.
The unpatched code works correctly on a number of other 64s that I
UPPER (Type: Statement Cat ,. : 178)
have tried. Mine is a standard-issue machine except that it has a
Line Range: 13038-13094
Cockroach TurboROM installed for fast disk operations. It seems unlikely that the Cockroach is the cause of the bug, which occurs whether or not the fast DOS routines are enabled, but perhaps it's not
Module:
GRAPHCMDS
Example:
UPPER
The upper portion of the hi-res drawing area is displayed (y
impossible. If anyone out there wants to test and report on the
coordinates 200 through 399).
following code, particularly if you also have a Cockroach ROM
LOWER (Type: Statement Cat 11: 179)
installed, I'd be interested to hear what comes of it:
Line Range: 13042-13094 Ida $dd02 ;set up CIA #2 DDR
Module:
GRAPHCMDS
ora #3
Example:
LOWER
sta $dd02
The upper portion of the hi-res drawing area is displayed (y coordinates 200 through 399).
loop Ida $dd00 ;switc h to bank 3 and #$fc
DEFAULT (Type: Statement Cat": 180)
sta $dd00 Ida $dd00 ;switch to bank 0 ora #3
Module:
GRAPHCMDS
Example:
DEFAULT
The screen is restored to the default text mode, with the ROM's
sta $dd00
uppercase/lowercase character set, and screen memory at $0400.
Ida $028d ;exit if SHIFT down beqloop
UNE(Type: Statement Cat
rts
Line Range: 13114-13548
•:
181)
Module:
GRAPHCMDS
If you don't have the bug, this routine will just make your screen go
Example:
LINE X1 Y1 X2,Y2
bananas till you press SHIFT. If you do have the bug, something else
A line is drawn between the specified coordinates using the pattern set
will happen.
in the last PATTERN command. The default pattern is $FF (a solid
,
,
line). Back to graphing: the print command GPRINT in the GRAPHCMDS
SUNE(Type: Statement Cat": 182)
module is set up, as I mentioned above, for the Epson MX-80, and
Line Range: 13096-13548
won't work on anything not compatible with that machine. If you want to drive another printer with this module, however, there's no reason why another version of GPRINT couldn't be dropped into the
Module:
GRAPHCMDS
Example:
SL I NE XL,YL,XH,YH
A line is drawn between the specilied coordinates using a single
module. Because another printer might need more code than Paul
dotted pattern (1 dot on. I dot off). This has the effect of setting the lirst
required for his MX-80 (the 6 by 7 print-head matrix of the 1525/
PATTERN parameter to $55.
MPS801, for example, is a bit tricky to write hi-res dumps for) I have left some unused program lines that you can spill over into if
DUNE(Type: Statement Cat
necessary (14614to 14998).
Line Range: 13100-13548
•:
183)
Module:
GRAPHCMDS
Just so the news, when it linally breaks, won't come as a life
Example:
DUNE X1,Y1 X2,Y2
threatening shock, I would like to officially start the rumour now that
A line is drawn between the specified coordinates using a double
next issue's TransBASIC column will be announced as the last of the
dotted pattern (2 dots on, 2 dots off). This has the effect of setting the
series. You don't have to believe it if you don't want to, but it just
lirst PATTERN parameter to $33.
,
might be true. See you then. 16
Nov. 19t6: Yolume 7, ..... 03
NUNE (fype: Statement Cat 11: 184)
SBAR (fype: Statement Cat 11: 189)
Line Range: 13104-13548
Line Range: 14096-14220
Module:
GRAPHCMDS
Module:
GRAPHCMDS
Example:
NUNE XL,YL,XR,YR
Example:
SBAR X1 ,Y1,X2,Y2
A solid line is drawn between the specified coordinates. This has the
A bar is drawn in the specified region using a single-dotted pattern
effect of setting the first PATTERN parameter to SFF.
(one dot one, one dot off). This command sets the pattern parameters to $55 and SAA.
QUNE (fype: Statement Cat 11: 185)
DBAR (Type: Statement Cat " : 190)
Line Range: 13108-13548 Module:
GRAPHCMOS
Line Range: 14104-14220
Example:
QUNE X1,Y1 ,X2,Y2
Module:
GRAPHCMOS
A line is drawn between the specified coordinates using a quadruple
Example:
DBAR X1,Y1,X2,Y2
dotted pattern (4 dots on, 4 dots off). This has the effect of setting the
A bar is drawn in the specified region using a double-dotted pattern
first PATTERN parameter to SOF.
(two dots one, two dots off). This command sets the pattern parame ters to $33 and S99.
CHAR (Type: Statement Cat " : 186) Line Range: 13550-13956 Module:
GRAPHCMDS
Example:
CHAR X,Y,CHR$(5) +
NBAR (fype: Statement Cat 11: 191) Line Range: 14112-14220 Module:
GRAPHCMDS
The string is printed at the specified location in the hi-res drawing
Example:
NBAR X1,Y1 ,X2,Y2
area. Most control characters are ignored. However, the cursor control
A solid bar is drawn in the specified region. This command sets the
characters have their normal effects, and the following have special
pattern parameters to SFF and SFF.
•
GUMBOOT"
meanings, as follows:
GSAVE (fype: Statement Cat 11: 192) CHR$(18)
(CTRL-9) Reverse on
Line Range: 14222-14384
CHR$(146) (CTRL-0) Reverse off
Module:
GRAPHCMDS
CHR$(31)
(CTAL-7) Upp ercase/lowercase
Example:
GSAVE
CHR$(30)
(CTRL-6) Uppercase/graphics
The drawing area (or the portion of it specified in the SIZE command)
•
BIG-GRAPHS.
CHR$(144) (CTRL-1) Erase mode on
is saved as a sequential file to drive 0, unit number 8. The characters
CHR$(5)
.g
(C TRL-2) Erase moe off
•
•
are appended to the first fourteen characters of the specified
filename. The two hi-res screens that make up the drawing area are Note that the last two of these are equivalent to setting the erase mode
stored contiguously in the file, which can hence be easily re
with the ERASE command in this module.
examined only with the GLOAD command. The GSAVE routine makes use of logical file ,.2 and secondary address ,.2 -- make sure these are available before attempting a save.
SCHAR (fype: Statement Cat 11: 187) Line Range: 13958-14094 Module:
GRAPHCMDS
GLOAD (fype: Statement Cat ..,: 193)
Example:
SCHAR 141 ,277 ,5
Line Range: 14230-14440
A special character (specified as the third parameter) is drawn at the
Module:
GRAPHCMDS
specified coordinates. The special characters are either 3 by 3 or 5 by 5
Example:
GLOAO
pixels in size. Those available are:
This restores a saved graph to the drawing area from drive 0, unit number
"0
:
•
HUGE-GRAPHS.
8. If only a portion of the screen was saved to this file, the
GCLR command should be given before reloading. As with GSAVE,
a 5 by 5 diagonal cross (x shape)
5 orthogonal cross ( + shape) by 5 hollow square by 5 hollow diamond by 3 diagonal cross by 3 orthogonal cross
"I : a 5 by
logical file ,.2 and secondary address ,.2 must not be currently in use
"'2 : a 5 "'3 : a 5 "'4 : a 3 "'5 : a 3 "'6 : a 3 by 3 hollow square "'7 : a 3 by 3 hollow diamond "8 : a 5 by 5 solid square "9 : a 5 by 5 solid diamond "I0: a 3 by 3 solid square "II: a 3 by 3 solid diamond
when this command is given.
SIZE (fype: Statement Cat 11: 194) Line Range: 15008-15068 Module:
GRAPHCMDS
Example:
SIZE 240
An upper boundary is set in the hi-res drawing area, for use by the GSAVE and GPRINT commands. The boundary is set as a y coordinate (up to 399, which is also the default). The specified number is then rounded up to the nearest 4-character (32 scan-line) boundary for actual use. The purpose of this command is to save disk space and printing time when part of the drawing-area is unused.
BAR (fype: Statement Cat": 188) Line Range: 14122-14220 Module:
GRAPHCMDS
LMAR (Type: Statement Cat ..,: 195)
Example:
BAR X1,Y1,X2,Y2
Line Range: 15000-15006
Plots (or clears, in erase mode) a bar in the area bounded horizontally
Module:
GRAPHCMDS
by XI and
Example:
LMAR N
X2, and vertically by Yl and Y2. The bar is drawn with the
The printer left margin is set to N columns. The default is II.
pattern last set by the PATTERN command. The default pattern is solid (both parameters SFF).
17
Nov. 1916: Yolume 7,1103 1u!
GPRINT (Ty pe: Statement Cat 11: 196) Line Range: 14442-14612 Module:
GRAPHCMDS
Example:
GPRINT
The current drawing area (to the limit set by the SIZE command) is printed to an Epson MX-80 print. Logical file '3 is used by this command, and must be available
••• Cl FH MP HH MN JH FF LH
B O
NH AF
when the command is issued.
KO
ERASE (Type: Statement Cat •: 197)
CL IH IB
Line Range: 15070-15082 Module:
GRAPHCMDS
Example:
ERASE 0
If this command is given with a non-zero parameter, erase mode is turned on. This results in all drawing commands clearing the affected pixels instead of setting them. Erase mode is turned off by using a zero parameter, as in the example.
PAtTERN (Type: Statement Cat 11: 198) Line Range: 15084-15154 Module:
GRAPHCMDS
Example:
PATTERN N1,N2
The plot patterns are set for the BAR and LINE commands. LINE uses only the first of the two patterns; BAR uses both. For BAR, the N 1 parameter is applied verti· cally, and the N2 parameter horizontally. The patterns effectively create a mask over the region being drawn to: bits that are set to 1 in the mask will be affected (set or cleared, according to the erase mode), while
0 bits are not affected. The default patterns are SFF and SFF, select
ing all bits.
Editor's Note
This incredibly huge TransBASIC module is listed here only for reference. Although the Verfizer codes are shown, I'll be shocked to find out someone actually entered it by hand. My apologies for the type size - its the only way we could justify including it.
IG
DB
MD GF GK OK AB KB DB
NN IC JM IJ OM GH Jl 00 KC GM BN AN FG LM JG FN 01 AJ MA BH ON HM NH AM FJ GO GO NA IE NG KE Nl KF CB OK EJ Ml GB MB PE ID
JK EP CP AH AJ LH
IF JG MA IP LK HG GM GA El NP NK
PK
HF NO PH NO
OL
NM KJ AG
IB
OJ
HA CK JM LM OM
0 rem graphcmds (paul adams) 1: 2 rem 24 s a t tements. 0 functions 3: 4 rem keyword characters: 113 5: 6 rem keywords #175 to #1 98 7. 8 rom 9 : $Hba 39 sedfs • 40selnam $ffbd 41open $ficO 42chkin $ff(:6 43dose $ffc3 44 drchn • $ffcc 45 getin $ffe4 46 chkoul $ffc9 $ffd2 4 7 chrout
• •• • •• • •••• •• • • •• • • • •• • • • •• • • •• • • •
=
=
=
=
=
48:
49tmp 501mp2 51 tmp3
52
=
-
•
tmp4
=
:more zp st()(age
$57
$59 $Sa
OG
$5c
KA BA
400.asc "ploTgciRgcoLuppeR" 401 .asc "loweRdefauiTiinE • 402 .asc stinEdlinEnlinE • 403 .asc • qlinEchaRschaR • 404 .asc • baRsbaRdbaRnbaR • 405 .asc gsavEgloaDsizE 406 .asc "lmaRgprinTerasE 407 .asc ·patterN , co-1,uppe-1 1400 .w ord plo-1,gcl-1 g 1401 .word lowe-l.deflt-l.plin-1 1402 . word sfin-1,dlin-1,nlin-1 1403 . word qlin-l,cha-l,scha-1 1404 .word b ba-1,sba-1,dba-1,nba-1 1405. word gsav-1.gloa-1.siz-1 1406 .word lma-1,gprin-1,eras-1 1408 . word patt-I 2620 usfp ldx 10 2622 SIX $Od sta$62 2624 2626 Sly $63 ldx f$90 2628
oc
EK HK AI OM NH AN AJ HA NM GL HE HN
•
•
•
•
2630
2632
FK BJ OA
sec
jnp $bc49
2634: 3694 powers.byte 1,2,4,8,16.32.64.128 3696 .
12694 plo 12696 plot 12698 12700 12702 12704 12706 12708 12710 12712 12714 plo2 12716 12718 12720 12722 12724 12726 12728 plo3 12730 12732 plo4 12734 12736 12738 12740 12742 12744 12746 12748 12750 plo5 12752 12754 12756 1275 8 12760 12762 12764 12766 12768 12770 12772 12774 12776 12778 12780 12782 12784 12786 12788
jsrgec2 sec :entry Ida l$8f sbc y
:read x.yparams :redef y, ty 399-y •
staty
Ida II
sbc y+1
sta ty+1
bcs plo2 rts ldx 1$c0 ldy l$c0 Ida y+ 1 bne plo3 Ida 1$c7 cmpy bcs plo4 ldx 10 ldy 11$a0 s t x tmp sty tmp+ 1 ldx x Ida x+l
beq plo5
:y>399 - too bog
;set base to upper
: or lower hi-res ;base SaOOO (upper) :base $cOcO (lower) ;lowery• 0-199 ;u pper y • 200-399 •
•
:lest x<320 :yes
cmpl1 bne plo2 cpx #$40 bcs plo2 ror
:no ;divide x by 8
r()(
;3 left with hi bit
txa 1st 1st
stacol
tax
Ida
:calc 7-(bit f) :conv to bot mask
17
powers,x ; in blip
sta bitp ldx ty Ida ty+1 ror txa
r()( lsr lsr sta
:no
:column m.m ber ;find bit , in col
txa and 17 eor
LJ NL BP AM PK AI KE GB FF JL NP KL FJ MH EO OH LO BP PK JL GK NL EN
row
;find row and line
:divide by8
: row number
11
LD KC HO PE MK NO PI HN EG EF KN JN JH CN EF AO GC NN ML
KP
KO FB
OA
BA PK EP GO EA ML KA GJ LJ
JO EA MH lA IF HC AM
GC
EP 01 BF OM AD ON 01 GJ cc EO
FF HJ
JA HO HF
FK
OL
;�;�
12794 12796 12798 12800 12802 12804 12806 12808 12810 12812 12814 12816 12818 12820 12822 12824 12826 12828 12830 12832 12834 12836 12838 12840 12842 12844 12846 12848 12850 12852 12854 12856 12858 plo6 12860 12862 12864 12866 12868 12870 12872 12874 12876 12878plo7 12880 12882 12884 plo8 12886 12888 12890 12892 12894 plo9 12896 12898ploiO 12900 12902 12904plo11 . 12906
txa and 17 sta rast Ida row ldx 116 jsr cmhad ldx 12
;add row·2t6 to
: base
;add rowo2t(6+2)
jsr mltadd Ida col
: f()( total rowo320 ;add Colo(2t3)
jsr cmhad ldy 10 sty tmp2
;.y1s ondex la t e r
i$radd sei Ida 1
;disable irq
ldx
Ida
13
rasl
and ISle
sta
1 Ida (lmp),y eor erflg ora bill> eor erflg sta (lmp),y Ida 1 ora 13 sta 1 eli ldy $ddOO Ida $028d tax
and 12 beq plo7 jsr deflt Ida S028d tax and 12 bne plo6 tya and 13 cmpl3 beq ploll eorll
bee plo9 txa and 14 beq ploll
Ida $028d
and bne tya and bne ;tlp
14
plo8 13 plo10
uppe
cmpl1 bne ploll JTIP lowe liS
•
12908 cmhad pha
12910 12912 12914 12916: 12918mltadd 12920 12922 12924 12926: 12928add 12930 12932 12934 12936 12938 12940 12942 12944 12946: 12948gec2 12950 12952 12 954 12956 12958 12960 12962 12964 12966 12968 12970 12972; 12974 gel 12976 12978 12980 12982 12984 12986 12988g<t: 1
:unaline huo
Ida 10 lmp 2
sta
;switch out roms ;gee byte to change ;onvertl erasing ;add point
:invertl erasing ;write to screen ;switch in roms
:enable irq ;get current screen :test logo key ;(save shill reg) : not pressed :text screen on ;wait for release
: not released ;get old screen :test text :yes
;switch screens
;branch always :test ctrl key
: not pressed ;wait f()( release
;sw�ch screens :test �ich scrn on :not lower ;turn on upper scrn :not upper ,turn on lower scrn
:entry to pre-de81
:tmp2
pia asl
rol
lmp2
dex bne mltadd
pha clc adc sta Ida adc
tmp Imp tmp2 tmp+ 1 sta tmp+ 1 pia
rts
jsr jsr jsr jsr Sly
sta
$ad8a
$aefd comx
$b7f7 X
X+1 IS' $ad8a IS' corny IS' $b717 sty y sta
y+l
:mun by2t(xreg) ;high byte tmp2 :low byte .a
;save low byte :add totmp
;restore low byte
;gee numeric arg ;check comma ;check range ;conv to integer ;save 1st word ;get2nd arg
;check range :conv to integer ;save 2nd
word
rts
Ida ISbl sta13 Ida ISff sta t5 ldx 1$20 ldy 1$40 Ida 10
pha
;cle81 32 pages : of memory at
; $a000 and $eOOO : (upper and lower : hi-res scr eens) ;i ncompl. top page from gcol
CG OB EC ME KD GF
OF
HM LO BP Jl PM CM OE
BC
FO PF
BB HD NH FO KM OL
CG JF LM CF PM NL MA BN IN CB ON EN
GO GB
EO
DK IJ
GJ
AO FD
00 EC GK GA
CJ
LA AA
JO FJ
MJ MN FA KE JA OK
NA KD KL
OK
GB NM GG EN GN PG
CN lA MG GO PN MH LK NK GN EO LB PH FE BP Kl FP HP FJ LN
oc CJ
FK DA CA OK
MA FO FP HP
OA NG
I
12990 12992 12994 12996 12998 gd2 13000 13002 13004 13006 13008 13010 13012 13014 13016; 13018gco 13020 13022 13024 13026 13028 13030 13032 13034 13036' 13038 uppe 13040 130421owe 13044 1 3046 defll 13048 13050 13052 13054 13056 13058 13060 13062 13064 13066 13068 def1 13070 13072 13074 13076 13078 13 080 13082 13084; 13086 scrdal 13088 13090 13092 13094 ; 13096 slin 13098 13100 dlin 13102 13104 nln 13106 13108 qhn 13110 13112; 13114 plin 13116 13118 13120 13122 13124 13126 13128 13130 13132 13134 13136 13138 13140 13142 13144 13146 13148 13150 13152 13154 13156 13158 13160 13162 13164 13166 13168 13170 13172 13174 13176 13178 13180 13182 13184 13186
lhe 1'1G.ac�
Ida NO
sla Sla
12
: rouline (below)
14
pia
dey
(12 ).y (14).y bne gcl2 dec 13 dec 15 dex bne gd 2 ns S18 Sla
JSr $b79e Ida 1$81 SIB
13
Ida ISci Sl8 15
txa
ldx 14
,gel arg pages ol
.r• 4
. memory at S8c:OO and Scc:OO , (upper and lower VIdeo matriCeS) •
•
ldy 1$e8 bne gd1
.bfanc/1 always
ldl< 10
,diSplay upper scr
ldx
,diSplay lower scr
.byle$2c
13 .byle $2c ldx 16 Ida scrdat.x Sla $d011 lOX Ida scrdal,x Sla $d018 Ida $dd02 ora N3 sta $dd02 inx
,diSplay deflt scr
and
ISle
sta
$dd00
DC
KO EF GF OK
BL PA JE EA EP CF HH OF FF MG BM DE
GO
sei
Ida $d012 bne dell Ida $dd00
AL HB JB KL IB BF FL HM FC FC FJ DP MA LB NB ED FP BH FM
,compensate lor hardware bug r ?) •
ora scrdat.x
ell
ns
PB PJ DP BF FB Cl El KN GB EH KJ
-·
.byle $3b,$38,$01 ,upper byle $3b,$38,$00 .lower .by!e $1 b,$17 ,$03 ,default
Ida 1$55 .byleS2c Ida 1$33 .byleS2c Ida
ISH
.byle$2c
Ida 1$01
sta
jsr jst
jsr
ldx ldy
jsr jsr sty
s1a
jsr jsr fSr
dol
$ad8a
Saeld comx H<yo l>yo $bbd4 $b717 x1 x1 +1
$ad8a
Saefd corny ldx l<tmpf ldy l>lmpl jsr $bbd4 jsr $b717 sty y1 Sla y1+I
JSI
jst JSr
$ad8a Saefd
comx
l<xo ldy l>xo JSr $bbd4 JSr $b7f7 Sly x2 Sl8 x2+1 Ida l<xo ldy l>xo ldx
$bba2
JSr Ida l<yo ldy JSI ldx ldy
jst JSr
l>yo $b850 l<dx l>dx 4 $ bbd $ad8a
.siOQie dolled ;double dolled .solid
.quad doned
.gel x1 ,arg
;check comma :check range .x1 1 0 yo lemp :conv to 1n1eger
:get y1 arg
:cneck comma ;check range :y1 1 0 tmpflemp ;oonv to 1nteger
,gel x2 arg .cneok convna .clleck range .x21o xo temp
MC
DO BE GK AI
Cl JJ MO NG AB AG DF
FF
BJ BC PH cc
HM FA DG
MF JE NK HP NO DC
HB pp
CA MM KD IL cc
KA NA LG IL GN PF
OP OM
,recall x2 :mem-lac, x1-x2
:s1ore dx .get y2 arg
HL HI El
co CL FH DN FK PL EN
13188 13190 13192 13194 13196 13198 13200 13202 13204 13206 13208 13210 13212 13214 13216 13218 13220 13222 13224 13226 13228 13230 13232 13234 13236 13238 13240 13242 13244 13246 13248 13250 13252 19254 132561in1 13258 13260 13262 13264 13266 13268 13270 13272 13274 13276 13278 13280 13282 13284 13286 13288 13290 13292 13294 13296 132981in2 13300 13302 13304 13306 13308 13310 13312 13314 13316 13318 13320 13322 13324 13326 13328 13330 13332 133341in3 13336 13338 13340 13342 13344 13346 13348 13350 13352 13354 133561in4 13358 13360 13362 13364 13366 13368 13370 13372 13374 13376 13378 13380 13382 13384
jsr
ldx ld y
jsr jsr
sty sta
Ida ldy
jst Ida ldy jst ldx ldy jst
Ida
beq Ida
ldy jst
corny
l<xo l>xo $bbd4 $b7f7 y2 y2+1 l<xo l>xo $bba2 l<tmpl l>tmpl $b850 l<dy l>dy $bbd4 $61 n h 1 f<dx l>dx $bb01
ldx l<sx
ldy l>sx
jst
$bbd4
Ida l<lmpf ldy
f>tmpf jsr $ba2 8 jsr $blb4 Ida l<yo ldy l>yo jsr $b867 ldx l<xo ldy l>xo
jsr $bbd4 Ida l<dx ldy l>dx jsr $bba2 Ida
$61
beq lin2 Ida ldy
l<dy l>dy jsr $bb01 ldx l<sy ldy l>sy jst $bbd4 Ida l<yo ldy l>yo
jst
jst
Ida
ldy
jst
ldx
ld y
jsr
Ida
ldy
jst lsr
$ba28
$blb4 l<tmpl l>tmpl $b867 l<yo l>yo $bbd4 f<dx l>dx $bba2 $66
ldx l<tmpf
ldy js r
Ida
ldy
jsr lsr
Ida ldy jsr bmi Ida bmi jsr
Ida sta
Ida
sta
Ida
l>tmpl $bbd4 l<dy l>dy $bba2 $66 l<lmpl l>tmpf $bc5b lin7 dy+1 lin3 swap y1 y y1+1 y+1 dy
bne lin4
Ida x1 Sl8 X Ida x1 + 1 sta x+1 jmp lin6 ldy y Ida y+1
:cneck range :y2 to xo temp ;conv 10 i nleger
sec
jst
;Siore dy
HH MG EF HF
;memlfac. dx/dy
CA
NB KF IM
;slore slope, sx ;lac= lacotmpf- sxoy1 ;lac•-SX•Y1 ;lac • x1-SX•y1
;lac exponen1 :lac=O
OG
;memllac, dy/dx
;store slope, sy ;lac=rac.yo-syox1 ;fac -sy.x1 •
;lac•y1-sy.x1
;store yo
:abs(lac)
•
abs(fac)
;lesl dy<dx :yes ;m1 of dy :y1-y2<0
GF BA GN GE MM HG AN Gl MB LE AP EL CP HG IP LG
co
;dx lo fac
;lac
BP IL GB EO EB JO CH ID GN NL
;dx 10 lac
:dy 10 lac
GC
pp
;slore xo
•
EK AD FJ CK GD JE BE GB OD IF LE EE
01
;clleck dy•O
HM AO EO HG AC CF NP MC NK El
;set X•X1
HK
:y=y1
;plot point, exil ;conv y 10 lloal
Ida l<sx
19
DE
. lac=O
;store abs(dx)
IC
MH EE AO
OF
;lac exponen1
$bc49
l>sx $ba28 jst Ida l<xo ldy l>xo jsr $b867 jsr $b849 jsr $b717
GD HB CM JJ EB IH
:mem-fac, y1-y2
1$90
ldy
OH
;recall y2
Sly $63 sta $62 ldx
IK BD IP LJ
co
00 CG JB Bl PM BO 01 GN Bl HM IB
OG
;y.sx ;xo+Y•Sx •y ;addO.S :conv 10 inleger
GL LA FE AB AH HM AB
13386 13388 13390 13392 13394 13396 13398 13400 13402 134041in5 13406 13408 13410 134121in6 134141on7 13416 13418 13420 IonS 13422 13424 13426 13428hn9 13430 13432 13434 13436 13438 13440 13442 13444 13446 13448 13450 13452 13454 13456 13458 13460 13462 13464 13466 1346811n10 13470 134 72 13474 1347611011 13478; 13480swap 13482 swa1 13484 13486 13488 13490 13492 13494 13496 13498; 135001320 135021400 13 504 ' 13506comx 13508 13510 13512; 13514 corny 13516 13518; 13520 chek 13522 13524 13526 13528 che1 13530 cl1e2 13532; 13534 dplot 13536 13538 13540 13542 13544 13546 13548. 13550cna 13552 13554 13556 13558 13560cl1a1 13562 13564 13566 13568 13570 13572 13574 13576cl1a2 13578 13580 13582
X x+1 jsr dplot Ida y+1 cmpy2+1 bne lin5 Ida y cmpy2 beq hn6 lllC y bne lin4 InC y+1 bne hn4 ,np plo1 Ida dx+1 sly Sla
bmi hn8
JSr
Ida Sl8
Ida
Sl8 ldy Ida JSr
Ida
ldy
fSr
Ida ldy
fSr
jsr jsr sty
swap x1 X x1+1 x+1 x x+1 usfp l<sy N>sy $ba28 l<yo l>yo $b867 $b849
$b7f7
y SIB y+1 jsr dplol Ida x+1 cmp x2+ 1 bne lin10
Ida x
cmpx2 beq lin11 inc x
bne lin9 inc x+1 bne lin9 jmp plo1 ldx 13 ldy
x1.x
Sla
x1,x
;po l t dotted point ;lest Y•Y2
;no ; yes :ncremenl y
;plollasl poinl
;m1 o f d x ;x1-x2<0 : x= x1
;conv x 10 float
;x•sy ;yo+X•SY=X :add0.5 ;conv1 0 inleger
;plot doned point ;tesl x. x2 ; no :yes :increment x
;plollasl poonl ;swap x1-x2, y1-y2
Ida x2,x lya
Sl8 x2.x dex bpi swa1 rts
.byle $90,$20,$00.$00,$00 .byle $90,$48,$00,$00,$00 Ida 1<1320 ldy 1>1320 bne chek
:test x range
Ida H<l400 ldy H>l400
;tesl y range
ldx $66 bmi chel jsr $bc5b bpi che2 rts
jmp $b248 ror
dot
rol
dot
php
; 'illegal qly'
;plol dotted line ; (rolale pattern ; register. don'l
: plol ij low
pip ror dot
bee cl1e1
: bi1 clear.)
,np plo1
JSr
JSr JSr
JSr
ge12
Saeld Sad9e
Sb6a3
Sl8 len ldy 10 Ida ($22).y tax and 1$71
;get X.Y location :cneck comma
;eval expressiOn
:make descriptor
;store lenglh
;get characler
cmpl$20
lxa
bcs cha6
ldyl9
cmp ctrls,y beq cl1a3
;I or ctrl cnars ;ctrl char ident
dey
bpi cha2
Ncw.1916: � 7, lllue03 •
I
18 CB PL BO HG IH FC
OH PF BA At MJ BH JD CK IH HI IK LC BB KH JO KE
EL PH NB JK KD AO GO CM BK LC LE MC CE OM CE HD AL LH IN GO BE BG El
co GG GJ KC IC HP BL FE FF FO
FP
BF KP OH JA 00
ML
JF
13584 13586 cha3 13588 13590 13592 13594 13596 13598 13600 13602ch84 13604 chaS 13606 ; t3608 ctrls 136t0 t36t2: 13614 ccrtns t36t6 t36t8: t3620 ccr t36 22 t3624 eel 13626 t3628 t3630. 13632 ccrx 13634 t3636 ccu 13638 t3640 ccut t3642 13644: t3646 cce 13648 t3650ccex t36 52 t3654 13656. t3658ccdn 13660 t3662ccup t3664 13666. t3668ccrt 13670 t3672cch 13674 13676: t3678 cha6 13680 t3682 13684 t3686 t3688 t3690 t3692 t3694 cha7 t3696 13698 t3700 13702cha8 t3704 13706 t3708 t37 1 0 t37t2cha9 137t4 t37t6
GB MB EO KB 137t 8 GO t3720 chatO OE t3722 OF 13724 KK t3726 EP t3728 OB t3730 K B t3732 NJ 137 34 HK t3736 JB t3738 co 13740 E l t3742 FB t3744 cha11 PC t3746 FN 13748 Nl 13750 Fl t3752 OL t3754 JC t3756 BK t37S8 HJ t3760 EK t3762 00 t3764 CE 13766 FL t3768 GE t3770 PO t3772 M K 13774 chal2 OJ 13776 HI t3778 CL t3780
I
Tile TIGiloXIoor
bmi cha4 tya asl
;unknown ctrl char ;get routine addr
tax Ida ccrtns. x
Sla t2
Ida ccrtns+l . x
sta 13 JSl
cha5
,np cha20 ,np (t2) . byteS t 2.S92.$1f,S 9e, S 90 .byte sos.stt.S t d,$9t,$ 9d . word ccr.ccrx, ccl,ccu.cc e .word ccex,ccd n ,c crtcc . up,cclt ;reverse on
Ida 14 . byte S2c Ida 18 ora chaff bne ccut
;uppernowercase
Ida ISdb
.reverse off
byte$2c Ida l$d7
,uppercase/graphoc:s
and cholf Sla cholf rts Ida ISIf byte $2c Ida NO SIB erflg rts
Ida N$18 byteS2c Ida 18 ,np yoff 18 byteS2c
Ida
;erase on :erase off
;cursor down :curSO< up ;add offset ;cursor right
Ida f$18 ,np xolf
;curSO< lelt :add offset
and #SeO cmp#$60 bne cha7
. retain bits 5-7 :test $60-$71 ;no
txa sec
sbc #$20 ta x bne cha9 cmpl$80 and 1$40 bee cha8 adc 1$31 Sla
;branch always ;subtr $40 Ia< ; either b6 or b7 ; 581
t2
sec txa
sbct2 tax Ida
10
sta
Imp
txa ldx 13 asl rol tmp dex bne chatO sta tmp4 Ida cholf
etc adc tmp Sla tmp4+t Ida 1$19 yoff JSl 17 ldy seo
Ida I and ISib SIB
pha txa
ldx
:save carry
:scr code offset :add high offset . h o byte oHset . set offset ; y = y-7
;char byte counter ;block rnterrupls ;o/o ram out
t
Ida (tmp4).y tax Ida t ora 14 sta t· eli tya
pha
:clear f o r rotate ;get scr code ;times8
; s ave char byte :Vo ram in
;enable interrupts :save byte counter :savech81 byte
17
�
;char bot counter :rSII char byte
asl
;bot to carry
pha
:savech81 byte
bee c ha t ?
; skip d bot = 0
Ida X+l ON 13782 bmi cha14 N N 13784 OP 13786 cha13 Ida ISle sta tmp NC 13788 Ida #$cO KD 13790 OE 13792 jsr xoff2 13 794 bpi cha13 JC HC 13796cha14 Ida 11 Sla Imp DE 13798 HB 13800 Ida #$40 GF 13802 jsr xoff2 ED 13804 bmi cha14 Ida y+ 1 IP t3806 HP t3808 bmi cha16 EC t38t0chatS Ida ISle sta Imp HE t3812 Ida N$70 oc 138t4 jsr y off2 PH 138t6 IF t38t8 bpi chatS HB t3820chatS Ida , NF t3822 Sla Imp 00 t3824 Ida 1$90 HI t3826 jsr y off2 bmi chat6 88 OE 132 FB 13830 IX8 JN 13832 pha AP t3834 jsr plot PO 13836 � JN 13838 tax JE t3840chat7 dex bmi chat8 N G t3842 Ida 11 L P t3844 jsr xoff 10 t3846 jmp cha12 AJ t3848 MA t3850 c ha18 pta pta H N t3852 tay NO t3854 dey LN t3856 bmi chat9 EE 13858 Ida 1$19 OJ t38 60 FF t3862 jsr xoff Ida II PA t3864 OF t3866 jsr yoff HN t3868 ;np cha11 F H 13870 ch at9 Ida II CF 13872 jsr xoff IG t3874 cha20 inc S22 bne cha21 FE 13876 inc $23 PN t3878 KA t3880cha2t dec len beq cha22 BG t3882 jmp chat AC t3884 P M 13886cha22 rts GL t3888: t3890 xoff pha II cmpl$80 OP t3892 bcs xofl D H t3894 Ida #0 LE 13896 FC t3898 . byte S2c PE t3900 xoft Ida ISff sta tmp CN t3902 IN t3904 pta IM 13906; FM 13908 xoff2 etc l A t39t0 adc x GF 139t2 sta X AK t39t4 Ida tmp FP t39t6 adc x+ t sta X+t DE t39t8 ME t3920 rts IN t3922: MK t3924 yell ph a cmp#$80 AC t3926 bcs yoft GJ t3928 Ida 10 N G t3930 HE 13 93 2 .byte S2c DH t3934 y o f t Ida ISH E P t3936 Sla tmp KP 13938 � KO t3940. KO t3942 yoff2 de adc y LC t3944 sta y 46 J H t39 Ida tmp CM t3948 adc y+t IB t3950 sta y+t t3952 GG OG t39S4 rts KP t3956 ; AC t3958 scha jsr get2 jsr Saefd JB t3960 JC t3962 Ida ISle jsr xoff BL t3964 Ida ISle N C t 3966 M L t3968 jsr yoff IP t3970 jsr Sb79e IE t3972 txa MP t3974 Sla tmp3 EM t3976 asl asl GM t3978
20
;test x < o :yes :subtract$0140 : unti x is neg
CM KH GK AP
Ml :x•x-140 :repeat while pas ;add $140 ; untiX IS pas ,X•X+140 :repeat whole neg :test y<O ; yes :subtract SOt 90 : until y is neg :y-y-t90 ;repeat while neg :add$0t90 : until y is pas
NJ
CG NP LD FJ IH
01 AN GH MC MJ
MF ON Bl FE
GK OJ
:y•y+ t90 :repeat while neg ;restore bot dr :and save ;plot single poont :restore bot dr
:fonished byte :set offset ;x•X+1 ;get next bit :waste char (done) ;byte counter
. f onoshed char ;set offset ;x•x-7 : set offset ;y•y+ t ;get next byte :set offset :x•x+1 ;inc strong ptr
:decrement length :get nextchar
:push olfset ;test negative ; yes
;hogh byteO
EE Bl IE DC JK BK MF BK Jl MP
PL AC IF PM NA OJ DB IN KC KN GG
AI NM
AG BB GP FB AB CL EP JN LM EH EA
JO El BG
:hogh byte Sff
; add o ffset to x
.NL LP Ml DB NH DA EJ IF NO EF
NN OJ
:push oflset ;test negative
; yes
;hogh byte 0 ;hogh byte SH
: a d d offsetto y
PC
MH JD MK EB MH JF IP KJ DO AG
BN HC El OD
LK
BD :get x. y :checkcomma ;set offset :x • x-2 :581 offset :y•y-2 ;char I to . x ;save- mull x 5 ,X 2 ;X 2
LK ED HO GN DH JD Kl BA AF
co LA
13980 adc tmp3 ;+1 13982 sta tmp3 :save address 13984 ;byte counter ldy 15 ;restore address Ida tmp3 13986sch1 :addrto .x 13988 tax 13990 inc tmp3 ;IOCf addr 13992 SCX , X Ida ;get data byte 13994 ldx 15 :bit counter 13996sch2 asl ;btl to carry 13998 pha :save byte 14000 :no plot bee schS Ida X+ I ;test x in bounds 14002 t4004 beq sch3 ; y es t4006 Ida X t4008 cmp#$40 t4010 :no bcs sch5 t40t2 sch3 Ida y+l :test y in bounds t40t4 beq sch4 ; ye s 14016 Ida y 140t8 cmp#S90 bcs t4020 schS :no 14022 sch4 IX8 :save counters t4024 pha t4026 tya 14028 pha 14030 ;plot x,y JSl plot t4032 ;restore counters � 14034 tay 14036 � t4038 tax 14040 schS Ida , jsr xolf ;x•x+1 14042 pia :resta<e byte 14044 14046 dex ;deer bit counter bne sch2 t4048 ;next bit t4050 Ida , t4052 1sr yoff ;y y+ I t4054 Ida #Sib :x• x-5 t4056 1sr xolf t4058 de y :byte counter t4060 bne scht :next byte t4062 rts t4064; t4066 ;chars - 51elt bots ol 5 bytes t4068scx 14070 .byteS88.S50.S20.SSO.S88 ;5x5 x t4072 .byteS20.S20,Sfc. $20,$ 20 :SxS + 14074 .byteSic, see.see.see.stc ;SxSsqr t4076 .byte$20,$50, $88,$50,$20 :5x5dmd 14078 .byteSOO.SSO.S2o.sso.soo ;3x3 x t4080 .byte$00,$20,$70,S20.SOO ;3x3 + 14082 . byte SOO.S70.SSO,S70.SOO ;3x3sqr t4084 .byte$00.$20.$S0,$20.SOO ;3x3dmd 14086 .byte Sic, Sic, Sic, Sic. Sic ;SxSsqr t4088 .byteS20.S70.Sic. S70,S20 ;5x5dmd t4090 .byte$00,$70.S70,$70,SOO ;3x3sqr t4092 .byteSOO.S20.S70.S20.SOO ;3x3dmd •
•
t4094;
t4096sba t4098 t4t00 t4t02 ; 14t04 dba t4106 t 4 t 08 14110; t4112nba t4114 14116nbt t4118 14t20; t4t22 bba t4t24 t4t26 14128 t4t30 141 32 t4t34 t4t36 t4t38 t4t40 t4t42 t4t44 t4t46 t4t48 t4150 t4t52 bat t4t5 4 14156 t4158 t4t60 t4t62 t4t64 ba2 t4t66 t4168 t4t70 t4172 t4t74 t 4176
ldx #$55
Ida I S a a
:single dol ven ;songle line hO<iz
bne nbt ldx
1$33 Ida IS 99 bne nbl
;double dol vert ;double tone horiz
Ida ISII tax SI.X do t sta ldot
:no p attern
Ida dot sta tmp3 Ida ldot pha
:save y p attern :save x pattern
jsr get2 sty yt
:get xt,yt
ISr S a eld JSr get2 sty y2 sta y2+ 1 ldy y2 Ida y2+t sty y sta y+t Ida tmp3
, c heck comma :get x2,y2 :store y parameter ;x2 saved in 'x' ;init y countdown
:store parameters
Sla yt +t Ida X Sla XI Ida X+t Sla XI+t
sta
dol
sec
Ida
X
sbc xt Ida X+t sbc Xt+t
bee baS ror
:restore y pattern
ldot
;test outer (x) . loop complete
. yes ;rota te x pattern
Nov. 19M: Yl*.me 7, 11111e 03
I
OK 14178 GO
ON JO
OG MG CF O K
KB
JG
HP Hl FD
CD JP
DG HD FP JM PO
GH CA
AB 08
JD KA HD PJ AB Dl GE
FG PI HE MD
KN
FH Gl JP
AM IG EJ NG CK EE
AI
co AK
OJ
JC Dl
GB AE FO
PF BG OJ
AD
OD OF EK
AF
FK
01
BM BF KM 10
LG
10
FN
AN IE LM EF PM
IG KG LF DC
JN
EK Cl A F
MO
DO AB LJ
PI EC JA IP AB LF MC PO HC KB KF
I
: save low bit
php
14180 141 8 2 14184 14186 t4188ba3 t4 t 90 14192 14t94 14196 14198 1 4 200 14202 14204 14206 1420Bba4 14210 t42t2 t4214 baS 14216 14218 14220; 14222 gsav 14224 t4226 14228: 14230gloa 14232 14234: 14236gd1 14238 14240 1 4242 14244 14246 14248 14250gd2 14252 14254 t4256gd3 14258 t4260 t4262 14264 t4266 1426Bgd4 14270 t4 272 t4274 t4276 14278 t4280 14282 t4284 14286 t4288 1 4290 14292 1 4294 14296 14298 14300 14302 14304 14306 14308gd5 t43t0 t 43t2 14314 14316 14318 14320 14322 14324 14326 t4 328 14330gd6 t4332gd7 14334 t4336 t4338 14340 t4342 t4344 14346 14348 14350 14352 14354 14356 14358 t4360 1 4 362 14364 14366 14368 1 4370 14372 14374 •
n.ll'al-lot
rei
Idol
pip
ror ldot bee ba4 sec
Ida y
: recover low bit : complete rota11on :no draw if bit•O ;test inner (y) : loop complete
sbc yt
Ida y+1
sbc y1 + 1
bee ba4 dplot JSl Ida IS If
JSl
yolf
bcs ba3
Ida l$ff JSl xofl bcs bat
pia
sta liS
:yes
;draw rt pan bot•1 ,countdown y
.oc means y•-1 .countdown x .ccmeansx•-1 ;restore x pattern
Idol
LB 14376 OE 14378 gd8 LB 14380 Jl
GK AD BF GM 01 EO
00 NG
MN
OL
KH
AA
CP MP
CP CP KJ 81
ldx
bne gd1
:save to diSk flag ;'wr1te' :blanch always
l dx , Ida I. r.
;load trm disk flag :r' ead'
110
Ida l'w'
sta gnsuf+S stx tmp2 :eva I filename jsr Sad9e :make descriptor jsr $b6a3 :test len > 14 cmpiSOf :no bee gd2 Ida IISOe : ma xlen pha :save len tax :c oun te r ldy 10 Ida ($22),y ;get char sta glnam+ 2.�opy to buffer on y dex bne gd3 :add g.s.rtw· ldx 10 gnsuf.x Ida sta gfnam+2.y ny nx cpx 16 bne gd4 •
;get length
pia
clc adc 18
ldx l<gtnam ldy l>gtnam JSl setnam Ida 12
tay l dx
jsr
18 settts
.adJUSt
pia
JSl chrou1 Ida $ 90 bne gd8 ny bne gd7 one tmp+ 1 beq gd8 Ida tmp+t cmpfScO bna gd7 Ida #$eO
PN GN IJ ED
10
ON Nl NC EG
NL
DO
00 KO
AF
lA FM FG
EF
LA
Fl
NO BN BF AS MO NO LM
HC
DO K II
.
addr :device II
:sec
JSI open bee gd5
cmp 114 bne gd8 ldy 110 sty tm p ldx 12 Ida tmp2 bne gd9 jsr chkout Ida hite tay rsr chrout Ida hote + t JSr chrout sta tmp+ 1 seo Ida t and ISle sta 1 Ida (tmp).y pha Ida t ora 13 sta 1 di
DH LF MP PO Jl
:testlolenotfound : no - real error ;ctr save startaddr ;tole I :save/load llag :open output chan ;graph height ;lb addr counter
AI
BF NO GM AP
OH NO BH JJ Bl EE
GA AJ
NO DP
KG BB HI
OK FJ ;hb addr counter .dosable irq ,swotch out roms
.get byte10 save ;SWitch n roms
;-blewq
:output byte
;check status : ex� cond�oon
OE Hl
AE AN NB NB KD Fl FD
DA PI
HD
0 1 JK
NF FN
OG BN .top ol upper block
DA CP EK
ED
: set lower block
HB
14382 t 4384 : 14386 gd9 t 4388 14390 14 392 14394 14396 14398 14400 144 0 2 14404 gd10 14406 14408 14410 t44 t 2 t4414 14416 14418 t4420 14422 14424 14426 t4428 14430 t4432 ; 14434 gfnam 14436 14438 gnsuf 14440; t4442 gprin 14444 t4446 14448 14450 14452 14454 t4456 14458 14460 gpr1 t4462 t4464 t4466gpr2 14468 14470 14472 14474 14476 14478 t4480 14482 14484 gpr3 14486 gp<4 14488 t4490 1 4492 144 94 14496 14498gpr5 14500 14502 14504 14506 14508 gpr6 t45t0 14512 t45t 4 14516 145tB gpr7 20 145 t 4522 t 4 524 1 4526 14528 14530 14532 14534 14536gp<8 t4538gp<9 14540 14542 t4544 14546 14548 14550 t4552 14554 t4556 14558 14560 14562 14564 14566 t4568 t4570 14 572
b n e gd6 clrchn j sr Ida 11 2 jmp close jsr chkin get n i jsr sta Me lay Ida $90 bne gd8 jsr getin sta hite+ 1 sta tmp+ t
Ida jsr Ida ldx ldy
jsr jsr
GE
$90
:no
good
BH
good
;bump addr counter
•
MK Kl MJ Kl
OL
;bump addr hogh ;done ;top of upper block ;setbonom lower
;branch always
IM MD EB HA OM AD EK EA
NA
•
.
BN
DL EH BH IP NH MH DB JK
:hb addr counter ;byte from diSk ;store ;test status
bne gd10 inc tmp+1 b e q gd8 Ida tmp+ t cmpiScO bne gd10 Ida l$e0 sta tm p+1 bne gdtO
.asc
;open input channel :get lb height
;get ub height
iny
.asc
NE NO BB JD
:no
bne gd8
=
;close channel
;lb addr counter ;test status
jsr getin sta (lmp),y Ida
MN
O:Ot23456789abcd.g,s, w .g,s,w
•
•
NO satn am
#3 14 1$11 setffs
:no name :tole #
;p<inter :no secondary addr
open
bee gp<t jmp g pr t 2 ldx 1 3 jsr chkout ldx 16 Ida plrs.x jsr chrout
dex bpi gpr2 Ida w sta tmp
:no error
bplgp<9 jsr chrout
dey bpi gp<8 pia lay
dec tmp3
bne gpr5 cpy 1$40 bne gpr3 Ida tm p+ 1
cmpfSff
b e q gprtO cmpf$bt bne gpr3
21
JD CM AB NK EN
;set tmp to start
AM
;lb addr counter .set p<t t o met row ;send et. ht. and ; 320 do!s
;column counter
cc AB GC OJ
CN
FO LM Jl
CF
KA BA
JO NP PM EN
;get 8 t ones data
A G
;switch out roms
CK OM 81
IN
g ; et byte :to b u ffer
HA
CD OG
EP
dex
rei dex
DL
OM MB
sta
bpi gpr6 Ida 113 ora 1 sta 1 eli tya pha ldy#7 ldx 117 rcl buff.x
NO EN BK EO NK
;f i l e 11 ;open output ch an ;count 7 chars :get char :to printer
Ida hile+ t tmp+1 ldy hite ldx 15 Ida p!. rr x jsr chrout dex bpi gpr4 Ida f$28 sta tmp3 ldx 17 sei Ida fSfc and 1 sta 1 Ida (tmp)y . sta buff,x n i y bne gpr7 in c tmp+1
EK MM
;switch in roms
IP HA LA
08 DC AE
;save addr lo ;byte counter ;bit counter :bit to carry ;bit to .a
;to p<nter ;restore add r lo :test row dcne
:no :test screen done :no
;low scr bottom :re s t ore
prt ;upper scr bottom
FE
ED 10
FF
JD NO HE PO
CP LE
MK HG CE OP 18 GF BJ KO FE AD JD IK
Ida l$e0 ;set to low scrn 14574 14576 sta tmp+ 1 ldy 10 t4578 14580 ;branch always beq gp<3 :restore p<inte r 14582gp<tO ldx #3 ;cr,line space 14584 gp<tt Ida ptrl ,x r chrout 14586 rs t4588 dex t4590 bpi gp<11 ;close channel t4592gp<12 jsr clrchn Ida 13 ;f � l 14594 14596 ;closet� jmp close t 4598 ' t4600 ,p<tconditoonng data strings t4602 p!rS 14604 .byte $00,$0d,$44 ,$1 b. $08.$41,$1 b 14606 plrr 14608 , 0d .byte SOt,$40,$ 4b,$1 b.$09$ t4610 plrf .byte $32,$ 1b,SOd 146t2. ;set left margin 150001 ma jsr $b7 9e 15002 SIX p!rS+1 ; for p<inting t50041mat rts t5006, ; get graph height 15008 SIZ jsr SadBa ;check range jsr comy 150 t0 jsr $b717 t5012 ;conv to i nteg er t 50 t 4 ;save height sty tmp3 t 50 t 6 sta tmp3+ 1 ldx nao ;max height t5018 t5020 SIX hite+1 15022 SZ1 sec ;test height>367 Ida 1$61 t5024 t5026 sbc tmp3 Ida Nt 15028 sbc tmp3+1 15030 ;finished bmi lma1 15032 ;tmp3 tmp3+ 32 et c 15034 15036 : (4 char heights) Ida 11$20 t5038 adc tmp3 15040 sta tmp3 t5042 Ida 10 15044 adc tmp3+t t5046 sta tmp3+ 1 etc ;hite=hie t + 5·256 15048 15050 Ida 15 ; (4 char heights) 15052 adc h�e+ 1 15054 cmpiiScO :test off SaOOO scr t5056 :no bee sz2 cmpfSeO 15058 :test on SeOOO sa 15060 bcs sz2 ; yes 15062 ora 120 ;advance pointer 15064 sz2 sta Me+1 ;branch always 15066 bne szt 15068. ;condrtion t5070 eras JSl $b79e txa t5072 :erase flag 15074 b e q era1 t 5076 ldx 1$11 15078 era 1 SIX erflg 15080 rts 15082; g ; et line pattern t5084 patt rsr Sb 79e SIX dot 15086 15088 JSr S aeld .check comma ;gel bar pattern t5090 jsr Sb 79e 15092 SIX Idol 15094 rts 15096: 15098 X1 .word $0040 .word $0040 t5100y1 15102 x2 .word $0080 t5t04y2 .word $0080 ·=· +5 15106dx 15108 dy • +5 ·-·+5 15t10 sx · · +5 15t12 sy .word$0080 15tt4x 15116y word$0080 ·• ·+5 15t18 tmpf ·-·+5 15t20 XO 15122 yo ·•·+5 ·-·+2 t5t24 ty +1 15t26 bttp •• •+ 1 t5128row .=.+1 st t5130ra -. + 1 15132ccl .byte Sff 15134 dOl 15t36ldol byteSH = +2 15t38 addr t5140erllg .byte $00 •=•+2 15142tx 15t44ty2 ·•·+2 15146hite .word $aOOO +8 15148 buff 15150choff byte $dO 151521en .b yte $10 15154; •
•
=
•
•
·
. =.
•
•
•
• • •
Nov. 1916: Volume 7, ..... 03
I
EXP086 Vancouver, British Coltimbia Commodore and IBM head-to-head.
•
•
by lan Adam, P.Eng., Vancouver, BC. Abstract
Any World's Fair promises a unique blend of fun and education, and Expo 86 being held in Vancouver until Octo ber of 1986 is no exception. The World's Fair theme of Transportation and Com munication offers a number of items of interest to computerists. Computers are used for everything from running the transportation system, to coordinating displays and finding lost children. Where there are computers there is the Transactor, and your reporter selected two items of particular interest. The World's Fair visitor information system is a useful package of applied IBM tech nology, each terminal combining a per sonal
computer
with
touch-screen
technology and an interactive videodisc to create a friendly and useful informa
Expo Centre, one of the theme pavilions at Expo 86. Its displays include the Futures
tion kiosk. Not to be outdone, Commo
Theatre, in which the audience can vote on issues affecting the future using illuminated
dore is well represented with the Amiga
buttons in the arms of the seats. Background: the Ontario pavilion includes a miniature
Studio Theatre, a sophisticated package
replica of Niagara Falls.
of arts and computers. A Trip To EXPO
Now maybe I'm giving away a lot about my age, but I still harbour many fond memories of a trip to Montreal in 1967 to see Man and his World, the famed Expo '67. Simply the mention of a World's Fair still evokes such a wide range of images and expectations for me. On the one hand, like any fair, it suggests an
occasion for people to
-
gather for a good titne: the proverbial
••
fun and games, but in an unusual, ex otic context. At the same time, how
\1······· ····�·····
ever, a World's Fair promises so much
.... .
more. By virtue of its international na ture, it represents an opportunity for each country's best ideas and products to be brought to one place, and put on display for all the world to see. Montreal's fair lived up to all of those expectations, and did so admirably. It combined fun and world-scale informa
The Canada Pavilion, with its five-sailed fabric roof, located on a pier jutting out into
tion
Vancouver Harbour. The rounded portion to the right contains the Amiga Studio Theatre.
into
an
overall
package
that
'clicked'. So it was with some mixed
Foreground: the Seabus, a harbour ferry. 22
Nov. 1986: 'IIIDiume 7, ..... 03
feelings that I first heard of the plans for a World's Fair in Vancouver.
a multi-media presentation. These votes are then tallied by computer,
Yes, the mention of a Fair brought back all those great memories, and
and reported back to the audience.
what could be more convenient than right here in Vancouver? At the same time, however, Montreal had set such a high standard away
Contributions to a fair can take many forms, ranging from formal
back in
1967, that I feared Expo 86 might somehow tarnish the
national pavilions to the 'official supplier' of everything from soup to
concept by not quite living up to that image. After all, there have been
nuts and bolts (the latter having been developed ad infinitum at the
some recent disappointments in the Exposition arena.
Los Angeles Olympics). At Expo
86, two computer companies have
made substantial contributions, IBM Canada and Commodore. While However, these fears turned out to be groundless. Expo
86 has been
both demonstrate new installations of their equipment, the contrast
able to stand on its own two feet, and is extending the Canadian
between the two applications is fascinating.
reputation established in Montreal for putting on a good show. Since this will probably be the last fair of the century in North America, I
Information System:
thought a bit about the aspects that are contributing to its success. IBM's contribution takes the form of applied big-blue technology. First, physically, it has a beautiful waterfront site in downtown
Although I didn't go to the fair looking for IBM material to document, I
Vancouver, very much reminiscent of Montreal's island fair. With
felt the system was so fascinating that you might like a peek at it. The
ocean and harbour alongside, set against a backdrop of city and
company has developed and is supplying a computerized visitor
mountains, the site alone is enough to draw the attention of one's
information system that will assist fair-goers in finding their way
psyche away from the humdrum of day-to-day life, and open up a
about the site. True to the tradition of World's Fairs, this system offers
broader vision of the world. And that, remember, is one of the primary
a demonstration of new technology, while at the same time it is a
roles of a World's Fair.
source of help and entertainment lor the user.
Second, the fair has attracted considerable international attention. Some
52 countries are represented at Expo 86, along with a dozen provinces and states, better than 40 corporate participants, and a
number of special theme pavilions. This alone is sufficient to ensure a wide variety of exhibits and a
good level of interaction between
nations. Third, the World's Fair theme of Transportation and Communications is particularly topical, and offers ample latitude for exploration of high technology. Vancouver is a city that was founded on transportation, being the western terminus of the Canadian railways, and the second busiest port in North America (eclipsed only by New York).
1986 is the I OOth
anniversary of the City, as well as of the transcontinental railway.
1986 is also a time when communication is rapidly taking over many of the functions of transportation, promising to provide an alternative to many types of personal travel. Computers will be a big part of the communications revolution, just as they have played a major role in transportation. As a result, the fair's theme offers a great deal of material of interest to all computerists. A World's Fair should naturally demonstrate the leading edge of technology, and every memorable fair has had its own particular specialization, an area of technology where notable advances were
A typical IBM Information System kiosk containing
brought forward and identified. At Montreal, much of the exploratory
7 touch
screen terminals all coordinated by a master PC/AT.
work was devoted to film, and the results of that effort are still with us today: giant screens, split-screen images, and all-enveloping film
At first glance, the system could pass lor a video game, with the
experiences have become familiar elements of our society, all traced
familiar colour monitor mounted in a cabinet. Faced with a series of
back to the Montreal fair. The
1985 exhibition in Tsukuba, Japan, was
menus, the user simply touches the screen in order to select the
devoted specifically to the theme of technology and its impacts.
information desired. After a short pause, videodisc footage of the
Robots that drew portraits, spun tops, and played sheet music were
selected item is shown on the screen. This combination - a sort of
the most memorable exhibits.
'Dirk the Daring goes to the lair' - is particularly well-suited to the application, and offers trouble-free access to the computer, even for
In Vancouver, a major legacy will be in computers, and Expo
86 may
the casual, non-computer-oriented user. Behind this seemingly sim
well be known as the lair where computers came into their own, both
ple facade, however, lurks an impressive collection of the state of the
running the fair and rela�ing to the visitors. In the theme area of
art in interactive computing technology.
transportation, the rapid transit line to carry you to the site is fully computerized, requiring no driver. Even the City's traffic signals are
The touch screen is an essential element of the overall concept,
coordinated by a new computer system. In the field of communica
providing trouble-free input from the user. A standard keyboard was
tions, there is the Futures Theatre, in which illuminated buttons built
considered for input, but raises the obvious concerns of intimidating
into the arms of the chairs allow the audience to vote on issues during
users not familiar with computers, as well as the potential for damage 23
No¥.1916: \lob'ne7, .... Ol
from spills, chewing gum, and so on. A gummed-up keyboard would
screens. Each cluster, or kiosk, is linked internally by PCNetwork to
mean a non-functional information system.
its master PC/AT, as well as to the 4381 mainframe on a batch basis.
Similarly, the laser videodisc is critical to the overall plan, providing
Program Mate.rial
television-quality moving images, and contributing to the overall system image. The videodisc is a standard consumer item, requiring
Aside from the technology, putting together the program discs has its
only that the data be addressable by track. There is no program code
own challenges. One problem is just assembling the program mate
on the disc.
rial, when it has to come from 50 different countries and many other
sources. Many countries are unable or unwilling to release informa tion about their pavilions, or simply will not conform to the required format and schedule. Another challenge is language. Bilingualism is a fact of life in Canada, with both English and French entrenched as official national lan guages. The stereo capability of the videodisc matches this perfectly, with one channel used for each official language. However, the system has not taken up the challenge of the 20 or more languages that could be appreciated by World's Fair visitors! The Potential The World's Fair information system is strictly a prototype at this point, assembled specifically for this purpose. However, one immedi ately conjures visions of its potential. With little modification other than a new videodisc, each console could be adapted to such tasks as education, catalog shopping, or night information at an airport. The handicapped, if they have the mobility to reach the screen, could also benefit. As evidenced by the usual lineups, the IBM information system is a definite hit at Expo 86. This suggests strongly that it has good potential for public acceptance in these other applications. Commodore's Amiga Theatre In contrast to IBM's straight display of high-tech, Commodore has chosen an artistic forum to highlight the capabilities of the Amiga computer. Commodore is an official corporate supporter at the Can ada Pavilion, and has assumed sponsorship of the Amiga Studio Theatre, a unique facility that complements live stage performances with computerized special effects. This sponsorship consists of two elements: as well as the usual financial support, Commodore has supplied a number of Amiga computers for artistic use and theatre operations. The Canada Pavilion is a short train ride away from the rest of the site, on a pier jutting out into Burrard Inlet, part of the Port of Vancouver. Its five-sailed fabric roof is one of the trademarks of Expo '86, also covering a cruise ship terminal. (Just so you hosers don't get confused Intermingled with the video footage are screens of computer graphics,
about where you are, the portal to the host pavilion features the
used primarily for menus for the touch screen and for some elemen
world's largest hockey stick and puck!)
tary animation (for example, to show amusement rides). Even a 20Meg hard disk has its limitations of speed and capacity (at up to 112K
The Amiga Studio Theatre is a 380-seat facility located near the
for a digitized photo), so this animation will never be mistaken for
harbour end of the pavilion: in the accompanying photo, it is located
videodisc imagery. Nonetheless, it is an additional feature that intro
in the round structure near the right side of the pier. Doug Welch, the
duces some variety. The system also has the capability to overlay
proud Theatre Manager, filled me in on some of the details. Billed as
computer graphics onto the output of the videodisc, for example to
Canada's most technologically sophisticated theatre, this studio has
superimpose current messages about pavilion conditions or special
an intimate atmosphere with a distinct techno-mystical edge. While
events.
the live performances themselves are clearly artistic in nature. the stage is augmented by 6 projection screens, each 7112 by I 0 feet in size
Quite naturally, IBM has used the PC/AT as the heart of each display.
(approx 2.2 by 3 metres). The three screens above the stage are fixed
IBM's representatives indicated that the system could as easily run on
in place, and illuminated by ceiling-mounted video projectors; the
an XT or PC, but the ATs improyed speed and other capabilities
remaining three are at stage level, movable, and use a self-contained
provide for a superior user interface. (One can't help but wonder what
rear-projection system. The six projectors, in turn, are backed up by a
sort of system could have been created if the heart were Amiga gold
technical extravaganza of Amigas, video cameras and recorders,
instead of blue!)
lasers, and other special effects.
The consoles are arranged in clusters of seven 2A
No¥.1916: � 7, ISiue03
During the course of the fair, the theatre has a wide variety of scheduled performances, includ ing dance, small orchestras, and stage plays, up to nine events a day. The majority of these use Amiga computers to generate performance mate rial to complement the live performances. For •
.
.
.
.. .. ...
,.
-
anyone seriously interested in this machine, or
.. ::,..
even just in the potential of computers in the arts,
'
the theatre is worth a trip to the fair all on its own.
,..
•
,
,..
- ,.
:-
,. .,
:
The number of Amigas and the jobs they look after in the theatre are truly impressive. First of all, Commodore is supplying computers to the performing artists for advance preparation of performance material. At any one time, 18 to 20 machines are out on loan in this manner to groups who are commissioned to produce artistic works. Later, when these groups arrive at the theatre, there are two more Amigas in service in the control room. Graphic material which the performing groups have prepared previously and stored on tape or disc can then be displayed; in
The lobby of the Theatre features a colourful neon and sculptured mural display.
addition, the Amigas are applied to routine tasks such as projecting title credits onto the screens. A third Amiga in the control room is reserved for interaction with the audience in a current show, which I'll discuss further in a moment. Finally, yet another Amiga is used to entertain visitors waiting to enter the theatre. This computer pre pares graphics displays (some pre--packaged, others developing as they run), and shows them on some 25 monitors arranged in the anteroom.
-
During a short visit to the control area, I tallied six videodisc players, six tape players. chroma key and Fairlight equipment. some 27 video monitors. and two windows (real ones) offering the best stage view in the house. Horrors - the four Amigas even had to share shelf space with three Apple If's! Sample Applications
Your intrepid reporter selected a couple of per Interior of the Amiga Studio Theatre - seating area. The theatre is in the process of
formances to view the results - all in the line of
being set up for another act.
duty, of course. The technical capabilities of this setup are difficult to describe in words,
so
per
haps it's better simply to explain how they are applied in these two performances. The first is a production by The Great Canadian Theatre Company of Ottawa, entitled 'Alii Get is Static'. At its core. this is a fairly traditional stage play concerning the lot of Reginald Fessenden, a contemporary of Marconi who was the first to transmit voice by radio, but was forgotten by history. The performance begins with minute graphic-and-sound
a
five
introduction:
the
graphics were prepared on the Amiga and other studio equipment, and include a mixture of digi tized images, pure graphics, and video wallpa per. This is followed by the main performance. a live half-hour play by three actors. This would Some of the 25 monitors used to entertain waiting patrons in the lobby. 25
be a worthwhile performance in its own right, Nov. 1986: Yolume 7, ....... 03
but is made far more entertaining through the layering of additional
control-room Amiga contains a list of scenes stored on videodisc, and
characters, as well as background graphics, from the video system.
an appropriate background is selected to represent the desired locale.
With this performance, all graphics were prepared in advance, and
Cast members then improvise a scene in an off-stage studio, which is
displayed from videotape.
superimposed on the video footage and projected for the audience. Later in the show, another volunteer is taken from the audience and included in the composite scene. The location, props, and plot direction can all be varied in accord with audience input. While these concepts are not entirely new, they are freshly presented in this production, and made far more realistic as a result of the many videodisc backgrounds available through the Amiga. Even more innovative use of the Amiga is made when a number of audience members, up to 52 in total, are given touch-sensitive input devices to influence the progress of the game. Each of these devices consists of two electrodes embedded in small Velcro cuffs that are attached to two fingers of the subject's hand. According to the story line, these 'fingerbinders' are used to measure and collect energy to power the theatre: in fact, they are measuring the galvanic resistance of the subjects' skin, an indicator of their mental state. This is where the extra computers in the control room come into the picture - an Apple 11 collects this information and passes it to the Amiga. The Amiga then constructs a graph showing the 'energy level' using different coloured bars for the 52 subjects; this graph is projected in real time for the audience, and fed back into the story to influence the outcome. A little corny, perhaps, but very effective. Other productions I wasn't able to see include dancers on-stage accompanied by video images of themselves, Amiga-synthesized voices welcoming visitors to the studio, and many more fabulous shows promised for later in the summer (after press deadline, unfortu nately).
Behind the Scenes Putting on innovative shows of this type involves the participation of many people besides the performers. As well as the many assistants, control room technicians, and so on, innovative work requires some one to make it happen. I already mentioned Doug Welch, the Theatre The control room of the Amiga Studio Theatre: the foreground
Manager. In charge of the several theatres and other performances is
controls are for switching and special effects, while the Amiga
John Cripton, Director of Cultural Programming for the Canada
in the background is reserved for the TheatreSports event.
Pavilion. It was his original vision at the inception of the pavilion that provided the opportunity for a theatre combining artistic perform
It would have been all too easy to detract from a worthwhile play by
ances with high-tech effects. While John is responsible for overall
superimposing unnecessary techno-pop graphics, merely to demon
coordination and lining up performers, it is Diana Bockus whose
strate the computers' capabilities. However, that is certainly not the
experience and creative forces have brought the Amiga Studio The
case with this performance. The graphics are very carefully inte
atre to life. Over the past eighteen months, she designed the studio
grated, woven into the production and throughout the plot, so that
facilities, oversaw their installation, and worked with each of the
they seem to be a natural part of the play. Oh yes, there's one more
performers in adapting their material. Any expert in this business
role for an Amiga - being used as a stage prop.
must be self-taught, since authoritative books and courses
are
only
now becoming available. Originally from Toronto, Diana's back The second performance makes much more dynamic use of the
ground includes five years in development of interactive video; in a
Amiga. This production, by an improvisational group called Vancou
young art-form such as this, five years looks a lot like forever!
ver TheatreSports League, actually uses members of the audience to construct a sort of life-sized video game; through twists of plot too
Despite the imposing title of Video Development Manager, Diana
contorted to detail here, the audience eventually has to rescue the
enthusiastically gave me the inside look at the technology, starting
theatre from an alien threat. In the course of achieving this, good use
with her own office. While a PC/XT sat idle to one side, she proceeded
is made of a variety of special effects, mostly computerized.
to demonstrate on a fully-equipped Amiga system, with Tecmar memory expansion and 20 Meg hard disk, an extra 3112" drive,
While the basic plot of this story is established in advance, it unfolds
printer and mouse, all hooked up to a Sony LOP IOOOA videodisc
differently with each performance. At various points, members of the
player and extra monitor. A venue like a World's Fair is a mixed
audience suggest bits of information - names, places. and concepts -
blessing - while it provides the opportunity for new techniques to
that are then worked into the developing plot. For example, someone
blossom, the tight five-month schedule leaves little room for experi
in the audience is asked to name a famous time or place in history. A
mentation. Nonetheless, Diana indicates that there will continue to be 26
No¥.1916: 'Mime 7, ...... 03
progress throughout the fair. She also sees ample area for continuing growth of the medium, with the logical next steps being into artificial intelligence, and into total automation of the theatre, with manual override as necessary. As for software, much of the work is done by custom programming, either totally written for the task at hand, or else modified from standard material. Software to retrieve videodisc imagery, for exam ple, is actually a two-part job; one is the supervisory task of control ling the machinery itself, while the other is the interface with the user, who must be able to select from among hundreds of images quickly and efficiently. The software for this purpose was written by Bill Henning, a programmer at nearby Simon Fraser University, and handles the different tasks well. One example of software requiring modification is the use of chromakey to superimpose live actors on the background scenes. The frame-grabber and genlock could be applied to this task, but the present software configuration simply does not respond to the requirements of live performance, primarily
Your intrepid reporter takes a self-portrait from
in terms of the time needed to access these functions.
the frame-grabber display's monitor Well, I'm sorry to talk at such length, but the IBM information system and the Amiga Studio Theatre seemed to offer such optimistic views of the fu ture of computers, that I just had to elaborate.However, they only scratch the surface of the computer activities at Expo '86. Computers are used rou tinely for just about everything, from running the traffic lights and train sys tem, to finding your lost children after the computerized fireworks. Just one last example - not far from the Amiga Studio Theatre, there are two computerized robot arms that me thodically manipulate a piece of pa per. This activity quickly draws a curious crowd, anxious to learn what the robots are up to... until they fin ish folding their paper airplane, and launch it into the crowd! That amusing demonstration perhaps
The office of Diana Bockus, Video Development Manager of the Amiga Studio Theatre.
epitomizes the World's Fair - while
This Amiga is fully equipped with 2 megabytes. a 20-meg hard disk, extra 3112" disk, and
you're having fun, you hardly notice
videodisk player and monitor. A prairie scene from the videodisk player is displayed on
that you're being educated as well.
the second monitor.
Judging by the enthusiastic response of the crowds, the many computerized
Other software currently being used includes DPaint, DPSiide, and
displays at Expo 86 have quickly earned public acceptance in their
DPVideo for titling and other graphics effects. These also had to have
applications, thus contributing actively to broader role of a World's
some modifications to improve access time.
Fair and, in the process. to a better understanding of our world.
Other Applications
For further information, contact the author at:
Beyond the Studio Theatre, Amigas are also in use in the interactive
4425 West 12th Ave., Vancouver BC V6R 2R3
displays in the main pavilion area.In a viewing area, a visitor stands
or:
in front of a video camera, while his or her face is continuously
IBM Canada Ltd
displayed by the frame grabber. At the touch of a button, the image is
PO Box 10132, Vancouver BC V7Y IGI
frozen: the Amiga then produces that image at several different
or:
resolution levels, and displays them on a series of video monitors.
Expo 86 Canada Pavilion, Vancouver BC V6C 3CI
604-224-5879
604-664-6600 604-666-2000
This seemingly simple process is sufficient to draw a crowd and keep them entertained.
Nov. 19t6: � 7, .... 03
The Comspec AX2000: 2 Megabyte RAM expansion for the Amiga Chris Zamara, Technical Editor
Price: $1276 ($899 US) (plus appropriate taxes) Warranty: I year parts and labour Manufacturer: Comspec Communications Inc. 153 Bridgeland Avenue, Unit 5 Toronto, ON. M6A 2Y6 (416)787-0617
Think about two Megabytes for a moment. That's 2,048 Kilobytes. Putting that much memory on a micro sounds akin to strapping a Jet engine to a Chevette, but believe it or not, it fits the Amiga more like a high-boost turbocharger. You see, the Amiga, by its very nature, loves using lots of memory. Its internai512K can quickly fill with screens and windows opened by programs in the system. The multi-tasking nature of Amiga makes it only too convenient to bring in another program on a whim and have it sitting around in a window for whenever you need it. 50K here and lOOK there can quickly add up. (If anyone is trying to run an Amiga with the minimum 256K, you know what I mean.) Another RAM-eater is the built-in RAM disk, which can be used to speed up program development, or cut down disk swaps with a single-drive system. All of these factors contribute to
Since non-graphic information such as programs are best stored in
making Amiga a machine which greatly benefits from added
fast RAM, the operating system tries to use fast RAM whenever a
memory. That's a nice way to call it memory hungry.
programs asks for some memory. If a program just asks for any old memory and doesn't specify what kind, the system will allocate
Before getting into the review about the Comspec board specifi
fast RAM . That creates problems for some programs, because they
cally, I'd like to say a few things about memory on the Amiga in
were written on non-expanded Amigas and their authors didn't
general - one 2 Meg expansion will function much like another.
think of allocating chip RAM for graphics information (for exam ple, a gadget or sprite may be coded as part of the program itself,
Amiga holds two flavours of memory: "chip" ram, and "fast" ram.
which will reside in fast RAM). To the user, that means that some
Chip ram is accessible by the video chips, sound chip and DMA
programs that worked fine on the standard Amiga won't look right
channels, and cannot be expanded beyond the internal 512K. This
when you run them on the expanded machine. That may lead you
512K comprises the built-in 256K plus the optional 256K ex
to believe that the fault is with the added RAM, but the fault is
pander which plugs in behind the front panel. All graphics infor
actually with the program itself. Even version 1.0 of Amiga's
mation, such as screens, gadgets, sprites, windows, etc. must be
operating system had a little trouble with RAM expansion. So
stored in this chip ram, so the number of graphic-intensive
beware - the software you're using now may not run properly with
programs you can run at once is limited no matter how much expansion RAM you add. The video and sound chips access chip
extra RAM on the system, but the problems will probably be fixed on program updates and all future programs. The changes aren't
RAM directly, and have priority over the CPU. If you are in hi-res
that hard to make.
mode (640 X 200 or 640 X 400)and using 16-colours, for example, the CPU will only be able to access chip RAM during the screen's
So, if added RAM doesn't necessarily let you run more programs,
vertical and horizontal retrace - about 24% of the time. This will
and some of your existing software won't work with it, is it really
obviously slow things down if you're trying to execute a program
worth the significant amount of cash that you'll have to put out for
residing in chip RAM while displaying a hi-res screen.
it? That all depends on what you're using your Amiga for. If you are developing programs, in C, assembler, or another language, the
Fast RAM, on the other hand, does not conflict with DMA access from the chips, so a program executing from fast RAM will run at
use of a large RAM disk can speed up compiles greatly. If you use
full speed regardless of any graphics or sound activity. A stock
all commands to RAM and have them execute instantly. If you
Amiga has NO fast RAM, and any RAM expansion beyond the
only have a single drive, bolting on a super-fast 2 Meg RAM drive
512K chip RAM is "fast". Thus, Comspec's box reviewed below
can definitely make life easier for you. And even for just running
gives you 2 Meg of FAST RAM.
application programs, you can have more data in memory at once
AmigaDOS commands from CLI frequently, you may want to copy
Nov. 19t6: � 7, lllue03
for your project, cutting down disk access and allowing longer
system" line of products. The Transactor has been convinced
documents, bigger spreadsheets, faster database access, etc.
enough about Comspec's reputability to offer their RAM board on our mail-order card. The AX2000 goes for $1276 (Canadian, plus
The greatest demand for expansion RAM, though, will probably
federal and provincial taxes), and the 1-Meg AX I 000 is $1035. (US
come from software developers, who will want to keep all of their
prices are $899 and $729 respectively.) If you're a little wary of a
CU commands, software tools, editor, compiler, libraries, source
review about a product sold by the magazine reviewing it, keep in
files, and include files in RAM. You just can't do that with 512K. In
mind that we reached our conclusions about this RAM board
fact, for this kind of programming, even running just an extra I
before deciding to take it on.
Meg would be tight. The added 2 Meg, along with the internal 512K is just about right for a good "turbocharge", and isn't as much
The AX2000 is a small steel box which plugs into the expansion
of an overkill as it sounds.
slot on the right side of the Amiga. It has an identical slot on its side for further expansion, and you can snap in the plastic slot cover
To get an idea of what the added RAM could do for compile times, I
from the Amiga if you don't need to plug anything else in. The box
tried compiling Commodore-Amiga's program "dotty.c" using the
is Amiga-White, and measures 9 1/4" by 4 1/4" by I 3/4" (not
Lattice C Compiler under three different conditions. The Compile
including the protruding edge-card connector). The extra I 3/4"
and link times for the three tests are given in minutes and seconds.
that it adds to the Amiga's width isn't too demanding of desk space, and the box is the same height as the Amiga, creating a perfect spot
In the first test, the source code, compiler, linker, libraries and
to the right of the monitor to place a second drive. Quite a compact
include files are all on a single disk, and the .q, .o and executable
package for all that RAM, really.
files are being written to the same disk. The second test involved putting just the source and object files in RAM and using every
Besides the unit's small size, there are other factors which give the
thing else from a single disk. In the final time test, everything -
impression of a well designed piece of hardware. The two
Compiler, linker, include files, libraries, and source - was stored in
Megabyte board draws only 500 milliamps of current, the same as
RAM. Using all of the includes from the developer's disk, plus a few
competing one-Megabyte units. Because of the low power draw
favorite DOS commands costs just under a Meg of RAM.
and the design of the board, you can plug in two RAM boards, one into the other, and add a whopping four Megabytes. There are no
Test I: Everything on a single disk:
ventilation slots anywhere on the box, since the little heat that is
Compile: 02:50
internally generated is dissipated through the steel case. And a big
02:02
plus circuit-wise is that the board causes no wait states, never
Link:
making the CPU wait for the RAM to catch up to it. Test 2: Source and object in RAM: Compile: 01:27
The RAM will "auto-config" itself, providing that you're using at
01:44
least release 1.2 of Kickstart and Workbench, which have auto
Link:
config implemented. This means that once you plug the box into Test 3: Everything in RAM:
the Amiga, the RAM will automatically be added to the system
Compile: 00:44
when the Amiga is powered up and fed its usual two-disk diet. The
01:12
auta-<:onfig process lets you plug in any configuration of RAM,
Link:
devices or whatever without worrying about setting DIP switches As you can see, in this example compile time was over 3.8 times
to avoid memory conflicts. If you are using older operating system
faster when everything was in RAM. For source consisting of many
versions, you can add the RAM to the system using a program
files, the RAM advantage would probably be even greater. To even
included on a disk which comes with the RAM box. This program
further reduce your waiting times, there's still room left in RAM for
will do a RAM test, then add it to the system. You can include a call
your favorite editor and any programming utilities you like to have
to this program in your Startup-Sequence to automate the proce
kicking around. No wonder software developers like extra RAM so
dure, so running the RAM with older system versions isn't really
much.
any more hassle.
There are disadvantages to using RAM as a disk like this, though.
The disk that comes with the AX2000 is actually a Workbench 1.1
First of all, setting up for your programming session can take quite a while, as copying a disk full of files into RAM is no instant
disk with a special Startup-Sequence, two extra commands in the
operation. And the worst part is that every time you have to re
cleverly designed to first test the RAM and print installation
boot (can you say, "Guru Meditation Time"?) you will lose every
instructions if the test fails. This way, you can boot the disk when
thing in the RAM drive, meaning re-<:opying again. So you'll still
you first get the RAM and learn how to install it, then use the same
have to save any work you're doing on disk once in a while. Of
disk once it's properly plugged in.
C directory, and two "Read_Me" files. The Startup-Sequence is
course, no amount of RAM expansion is a substitute for a hard drive, but it would be so nice if there was some way for the RAM's
I have been using the AX2000 for several months now and have
contents to survive a re-boot.
had no problems with it. The difference it makes to programming productivity and the amount of "work" it seems to save is incredi
Comspec's 2-Meg RAM Box
ble. Now that I've been spoiled, life with the Amiga just wouldn't be the same with a measly 512K. Then again, once an 8-Meg
Comspec Communications is a Toronto-based company known
board comes out, I'll probably be saying the same thing about only
among many Commodore users for their "Microshare multi-user
2.5 Meg. Now, about installing that Jet engine. . . 29
No¥.1916: � 7, .... 03
Betty Clay Arlington, Texas
A Peek At Amiga Disk Structure
In order to understand the Amiga file structure, it is necessary
Hash Tables
to learn some terms that have not been familiar to Commodore users. We are accustomed to a directory track; the Amiga has a
Offsets six through seventy-seven of the directory blocks
hash-chain and a root block. We speak of the BAM; the Amiga
contain the 'hash-table'. 'Hashing' is a mathematical process
has bit-map pages. We are accustomed to a single directory on
through which the name of a file is changed to a numerical
the center track, and the Amiga can have many directories in
code, using a formula that will ensure that every possible name
various locations, and even of different types. We are accus
will be encoded to a number within the range allowed for a
tomed to program, sequential, and relative files, with most
particular use, in this case a number between six and seventy
sectors looking very much alike. The Amiga has root directory
seven, inclusive. The formula is built into the DOS and is
blocks, user-directory blocks, header blocks {also called keys),
automatically calculated for us, so we do not need to know
list blocks, extension blocks, and data blocks, not to mention
exactly what the formula is. To understand the hash table, we
corrupt blocks and deleted corrupt blocks! Once the new words
need only to understand that every possible file name will hash
are learned, however, there are some similarities.
to a value of six through 77, the same numbers as the slots in the hash table.
The Amiga directory begins with the root block, which stores the location of other directories on the disk, and of files not
If your file name has a hash value of twelve (and if it is the first
assigned to one of the other directories. Just as we are accus
file in the directory that has that hash number), then the
tomed to the directory track being at the center of the disk, so is
number of the block in which your file begins will be stored in
the root block on the center track of the Amiga disks.
slot twelve of the directory block. Each slot contains the block number of a file whose name hashes to the same value as the
On the current Amiga drives, a disk is formatted with eighty
slot number. Were this the end of the matter, that would mean
tracks, each having twenty-two sectors, and each sector hav
that a directory would be limited to seventy-two files and that
ing 512 bytes. Each sector is divided into 128 slots containing
no two could have the same hash-value. This would be an
32 bits each, and each slot may hold one 32-bit value, two 16-
intolerable limitation, so we must consider the hash-chain
bit values, or four 8-bit values. The Amiga calls the tracks
also.
'cylinders' and the sectors are called 'blocks', which is already a familiar term. There is also a 'surface' number, which is always given as zero by the DiskEd disk editor, but as either zero or
The Hash-Chain
one by DiskDoctor!. Suppose that you have saved a file called "filename" in the root The root directory on each Amiga diskette is at block 880 (track
directory. "Filename" has a hash-value of 59. Suppose this file
40, sector 0) on the current drives. The block contains 128 slots
begins with its header block in block 884. Slot 59 of the root
for information, numbered 0 through 127. The zero slot holds
directory will now contain the number 884. Suppose that later
the file type identifier (a '2' for 'short file'). Slots one and two in
you wish to save a file in the root directory and call it "File4",
the root block always contain zeros; slot three currently holds
which also has a hash value of 59. This file will begin on, say,
the value 72, but on other drives might hold other values.
block 985. When the root directory is examined prior to storing
Seventy-two is the value obtained when you subtract the
"File4", it will be discovered that slot 59 already contains the
number of slots that hold identifying information (the first six
number 884. Block 884 will then be read, and offset 124 (the
and the last fifty offsets) from the 128 that are available. If a
hash-chain slot) of that block will be examined to see whether
future DOS is set for blocks longer than 512 bytes, then there
it contains a block number or a zero. Should it contain a block
will be more than 72 slots left for the hash table, and slot three
number, that block would be examined to see if it contains a
will hold a different value. Offset four is unused on the root
zero in slot 124, and so on until a block in the chain is found
block, and number five holds the checksum on all types of
with zero in this offset. When the zero is found in the hash
sectors.
chain slot, the zero will be replaced with 985, placing "File4" No¥. 1M6: Yoeume 7, lnue 03
on the proper hash-chain. Then "File4" will be written to the
value of 32, and had stored "Filename" in the user-directory.
disk, beginning with its header-block in block 985 . Slot 4 of
Then the root block will store the block number assigned to the
block 985 will hold the number of the next block used for
user-directory in offset 32, and offset 59 of the user-directory
"File4",and slot 124 will contain a zero, showing that this is the
block will contain block 884. This is why the full directory path
last file on the chain at this time. It is this hash-chain that
must be spelled out when you call for a file. The root block
permits the Amiga to have an apparently limitless number of
must be read to find the user-directory, the user-directory
directory entries, yet the drive has a relatively small number of
block read to locate the key (block number) of the file, and then
items to read in order to locate a specific file.
the header-block of the file read to find the blocks in which the data is stored. If you want a printout of the keys to your files,
When you give a command for "File4" to be read back, a
you can get it by typing:
similar process is followed. That is, the name will be converted back to its hash-value, 59. Then offset 59 of the root directory
LIST TO PRT: directoryname KEYS.
will be examined, and the 884 stored there will cause block 884 to be read. The two filenames will be compared and, since they do not match, slot 124 will be read to find the number of the
File Header Blocks
header block of the next file with a hash-value of 59. In our case, offset 124 will contain the number 985. Block 985 will
A file-header block contains 512 bytes, 128 slots, but they are
then be read, and the file name there compared with "File4".
used differently here. Offsets zero and one contain the usual
These will match, so the file will be loaded.
information (file type and number of this block), but on these blocks offset two contains the number of blocks used for the file, three tells how many of the slots (6- 77) are used to store numbers of data blocks, and slot four holds the block number
The Other Fifty Slots
of the first data block. Slot five holds the checksum, as usual. At the end of the root block are fifty more offsets which contain
The next seventy-two offsets hold the numbers of the data
information about the disk. Slot 78 holds a flag to indicate
blocks used for the file. The block numbers begin at the end of
whether the bitmap is valid, and is followed by twenty-five
the table, slot 77, and move upward to number six. On most
slots reserved for the bitmap pages. The following three slots,
cases, the files are stored in eleven consecutive sectors, skip
numbers I 05-108 on current disks, contain the date and time
twenty-one sectors, and then use eleven consecutive sectors
the disk was last altered. The date is held as a single number,
again. The Amiga reads an entire track at one time, making this
and is the number of days between Jan. 1, 1978 and the date of
an efficient method of storage.)
alteration. (Don't forget to include leap years.)
The time is
protection bits, and number 81 contains the size of the file in
stored as the number of minutes since midnight and as the
bytes. There are twenty-two slots for your file comment, and
number of seconds since the hour. The next twelve slots are
the rest of the information is the same as in the user-directory
reserved to hold the name of the disk, which is limited to thirty
except for number 126. It is used only if the data-block table is
characters. Slots 121 - 123 hold the date and time of the disk's
too large to fit into the seventy-two slots in this block. When a
creation, the next three are not used on a root block, and the
file contains more than seventy-two blocks, another block is
last one holds a number that indicates the secondary file type (1
designated to hold the data about the overflow. It is called an
for a root block).
extension block, and its number is stored in slot 126.
User Directory Blocks
File List Blocks
A user-directory block is quite similar to the root block. The
The extension (file list) block begins with the now-familiar six
first six slots are the same except that slot one holds the block's
items of information: the file type, the block's own number, the
own number. The hash-table is just like that in the root block.
number of offsets in this sector used for the block list, the
Instead of the bitmap, however, there are protection bits and
number of the first data block, and the checksum. The data
twenty-five slots in which your file comments are stored. Slots
block numbers again begin in slot 77 and are stored upward
105-107 contain the time and date the disk was formatted, held
until they reach slot 6. If necessary, yet another extension block
in the same format as those in the root block, numbers 108-
is assigned. The area reserved for comments and the date are
123 are reserved for the name of the directory, and slot 124 is
unused on these blocks, and there is no hash-chain offset, but
for the hash-chain. Slot 125 of the user-directory holds the
the last three slots are used to indicate the number of the parent
header-block number of its parent directory, number 126 is
directory, the number of the next extension block (if any), and
not used, and 127 contains a '2' to indicate the secondary type
the secondary file type.
Offset number 80 holds
'user directory'. If you indicate a file path when you store your files, the header
Data Blocks
block number is stored in the proper offset in the user directory rather than in the root block. Suppose that you had
On these blocks, only the first six items of file data are used.
made a user-directory called "Datafiles", which has a hash
They hold the type of block, the block's own number, the
the bal11odor
31
No¥.1986: YoUne '· ..... 03
sequence number in the file for this block (is it number 5, etc.),
You will see a notice that DiskEd has been activated, and each
the number of bytes of data in this block, the number of the
line after that will begin with a prompt,
next data block in this file, and the checksum. The remaining
you will need to use are:
"#".
Some commands
bytes are used for the data. Normally, all data blocks will be completely filled except the last, so the number of bytes will be
488.
g
The last block will contain fewer bytes, and slot 4 will
contain zero for a forward pointer.
- to read the block you want to change or read (example: g
880) i
- to display information about the block just read
b
- to get the hash-value (and thus the slot number) of the filename. Type h and the file name.
Editing Disks
t
-to display the slot number or range of numbers you want to examine. Put a space between the numbers in the range- not a hyphen, as in 't 6
Of what use is this information? Well, it answers some of the questions that are asked about disk organization, of course. But
I
20'.
- to change a number in the slot. Type the slot number, a
it is essential that the hash-chain and the hash-table be
slash, and the number you want to place in that slot.
understood if you are to edit your disks. File recovery is an
Example:36/5 will change the value in slot 36 to 5.
important part of disk maintenance, and files cannot be recov
k
- to correct the checksum after you have made a change
ered if you don't understand the structure.
x
- to set o r unset the write-protect mode
p
- to write the corrected block to the disk
q
- to get out of the disked program
A disk came to me recently that was almost full, but the root block contained only one key number, and that one was incorrect --a completely corrupted disk. Every time we tried to
If you follow this sequence of commands, you should be able to
read a directory, or even insert the disk in the normal way, we
edit your disk. It is not a complete explanation of DiskEd. There
had to reset the computer. Yet we recovered that disk!
are fourteen other commands available, but the nine given here are sufficient. They will permit you to change any byte of
The disk editor we used was DiskEd, from the WACK disk
information on the disk. I have found DiskEd to be reliable and
included in the developers' package. This disk is now available
easy to use.
for anyone to purchase. In order to recover the disk, we first had to locate the header blocks of all the files we wanted to
Other disk editors? I just received a pre-release version of
resurrect. Most user directories and header blocks are stored in
Kickstart and Workbench v 1.2, and there is a DiskDoctor in the
the blocks between
The blocks with smaller or
new "c" directory. For my first effort with it, I duplicated a disk
larger numbers are usually data blocks. By using DiskEd's 'g'
known to contain an error. DiskDoctor read the disk looking
command to read the blocks in this range, we noted the header
for all the world like a formatting operation, except it said
blocks and hash-value of each of the files we needed. With our
"Reading cylinder
information complete, block
880 was read, and the number of
showing the files that were being replaced on the disk. When
each header block was restored in the offset that matched the
ever a serious error was encountered, it was reported on the
hash value of its file name. AmigaDOS is pretty smart. It
screen as in "Hard error on cylinder 67." Multiple efforts were
refused to accept the wrong block number in a slot! When all of
made to correct each error. Finally, I was asked whether
the headers had been restored to their proper positions in the
corrupt files in a named directory should be deleted. The files
root block, the checksum was corrected and the block written
were not named - just the directory. I answered 'yes', and
back to the disk. The directory was called, and the entire disk
found that the programs were not only deleted, the blocks in
was reclaimed. Only the root block had been damaged, but
which they had resided were completely reformatted as though
without it the remainder was useless.
they had never been there. On another disk, I answered 'no',
880 and 1100.
1",
etc. Then came a series of reports
and the files were not deleted. I tried it on a disk that had Had we wanted to eliminate a file that was making it impossible
managed to get the directory into an endless loop, and Dis
to validate a disk, we could have put a zero into the slot that
kDoctor corrected the error with no effort at all on my behalf.
referenced the bad file. This would make the remainder of the
Remember that I was using an alpha version of an early pre
disk usable, but that file would be permanently lost once
release version. Perhaps when this version is complete, there
something was written to the disk. On the other hand, it could
will be no need for knowledge on our part to repair disks. but
also have been restored by placing the header-block number
for now, it is better to learn the file structure and do it ourselves.
back in the root block. Perhaps a reading of the file would make
With DiskEd, we have complete control over every item writ
it possible to correct its error and save the file.
ten out to the disk. With DiskDoctor, we just sit and watch it do its thing. It is easy as formatting a disk or making a backup.
The documentation for DiskEd is sketchy, and it has not
Each of these tools will fill a need for most of us. DiskDoctor is
reached the popularity level it deserves. To use DISKED, place
fast and easy, but the control offered by DiskEd will permit us to
the WACK disk in the drive and, at a CLI prompt, type:
change any byte on the disk. We'll want to do it both ways.
cd df1 :wackstuff Disked df1: (or dfO:). Nov. w.6: Volume 7,1aue 03
Disk Copier Comparison
David Ma•·tin
Seabrook, Texas
Diskmaker V3.3 DiskMaker was introduced several years ago and has since seen very little improvements in its arsenal. However, earlier this year, Basix released V3.3, only a small improvement over earlier versions, that incorpo rates modular design and speed increases. The modular design comes in the form of updates to the disk nibbler called "Masterkey" modules. These up dates are available individually or through subscription at a very high price. Recent updates have also included speed increases during disk copying. I advise Commodore users to avoid this program until it sees vast improvements. The program should see the following improvements to justify it's high cost of
$49.95, which even then will cost too much. It needs the addition of programs that will allow a fast copy on two drives as well as a fast file copier. Also, the addition of disk utilities would be wonderful. Basix has adver Copy protection is becoming more and more complex month
tised a disk utility package called "Toolkit" for over a year that
after month, often making it extremely difficult for a person to
retails at $39.95 and offers the features that DiskMaker sorely
protect his newest software investment by keeping a duplicate
lacks. However, the reader can see in the comparison chart that
copy on hand.
expenses run high with products from this company, when other vendors offer more features for under $30.00. Again,
In fact, some of the newest techniques push the 1541 to its
until Basix prices change or features are added, this is one of
limits or beyond. They will often go where no 1541 has gone
the titles to avoid at all costs. I rank this title a one on the 1-10
before, recording data well out past track 35, even out as far as
scale due to it's high cost, no added features and updates that
track 40. These schemes are apparently very sensitive to drive
are not vastly improved since version 1.1.
alignment problems, which means that if the alignment is off, the program may not Load and Run.
Keymaster
Today users face a better opportunity in creating backups of
Keymaster is MegaSoft's newest disk copier, which retails at
their valuable software through the use of state-of-the-art disk
$29.95 and does incorporate some of today's newest disk copy
copy programs which can considerably speed up copy times
ideologies (parameters, fast DOS, etc.).
and provide many other features to enhance disk drive usage. The program supports various disk drive configurations (single The scope of this review is to help the reader decide which
or dual) using a 1541 or 1571. Also, the program is fast and will
copier is best in terms of features and cost while leaving the
copy disks on a single drive in under three minutes and under
final decision up to him. Therefore, I am basing my recommen
one minute on dual drives. Additionally, it supports a fast
dations on cost and available features on a comparison basis,
format that formats disks in about I 0 seconds.
since not every program offers the same features at the same prices. So, make yourself comfortable as we begin to explore
Keymaster also offers a method for fast file copying using a
the wide world of disk copiers.
rather unique method of transferring files from disk to disk. 33
No¥.1916: � 7,....,. 03
This is done by fast loading a file into memory and then fast
in the near future from a company that supports better products
saving it back to disk. A drawback to this method is that, since
for other machines. At this time I do not recommend the
the program does not use a buffer, which could hold more than
purchase of this program because it is not cost effective and
one program, the user is forced to swap disks many times (at
offers little added features, unlike other programs. This pro
least 2 or more for each file he wishes to copy). Keep in mind
gram is awarded a rank of five on the 1-10 scale.
that this method will support one file at a time no matter how large or small the file is.
Disector V3.0
A further enhancement to Keymaster is the ability to copy disks
Disector is the program from Starpoint Software that started it
using parameters for backing up some disk titles easier than
all a few years back.The program has grown quite a bit from
before. The program c�mes packaged with 50 parameters, and
version 1.0 that we all started with at One point or another.
more are available on disk at $10.00 per disk from MegaSoft.
Recent improvements to the program in version 3.0 have been a welcome sight to the copier world with improvements that
Here again, 1 advise readers to avoid this program since it is not
have followed what I hope to see as a continuing trend towards
cost effective due to the lack of features and lack of support
programs that carry more features at a very good price.
from MegaSoft (I'm a registered owner who was not advised of recent updates from them). This program does perform as
Disector is a program that has improved by supporting fast
advertised, under strict hardware configuration conditions, as
DOS routines for both single and dual drive owners. These
mentioned in its two page manual. My dual drive fast file copy
programs are rather fast and do function on both 1541's and
test failed because of Keymaster's lack of 1571 support, a
157l's as well. I did, however, experience some trouble with
problem that the other programs did not face. Lack of support
the dual disk copier when it got wind of certain protection
and features as well as high cost ranks this program a low three
schemes (the single disk copier didn't experience any prob
on a scale of 1-10.
lems). I would expect that an update will clear these problems up.
Copy 1164/128 One disappointing feature of Disector is its lack of a fast file Copy II is Central Point Software's entry as a disk copier,
copier contrary to the packages advertisement of an "Ultra fast
produced by a company that is well known for producing disk
file copier....". I found this to be very disappointing after
utility software for other name brand computers.This program
examining the program's other features.
does offer high speed copying and a few utilities in its arsenal. It even offers a different method of presenting parameters via an
The program also provides a parameter copier which only
enclosed leaflet that tells how to set the programs parameters to
supports about 40 parameters {this is also disappointing - see
copy certain disks easier and accurately.
the comparison chart).This utility is basically like the others, which require you to do a nibble copy then perform some type
Currently, Copy II does support the C64 and Cl28 as well as the
of special function using the program.
1541 and 1571 disk drives. It does provide a working copier on these drives in single drive or dual drive set-up. The copier is
For the most part, I do think that Disector is going in the right
able to examine the hardware present to determine the single
direction and found that its disk drive monitor utility, that is yet
or dual mode of operation, and uses the information to load the
to find a rival, well worth the purchase price of $39.95. The
appropriate program.
monitor itself will disassemble unimplemented opcode, and I find it to be a very invaluable tool in my programming and
The utilities that Copy II currently supports are a delete file
tinkering with the C64 and 1541 DOS. It does not currently
program as well as a fast load for BASIC. It would be nice to see
support printed output, but I have been told that a new version
more utilities for this program that could help a user in
of the program now supports printer output from the monitor.
determining how to set the programs parameters. This cur
A note here: A majority of the programs in Disector do not
rently is impossible, and requires that the user refer to their
support printer output where it is sorely needed. For example,
enclosed parameter sheet Thus, Copy II can very quickly
in the Format Editor you cannot print the results of a disk scan.
become outdated.
Currently, users are required to hand-write information about the I to 40 tracks on the diskette - this can lead to writer's
Again, as a copy program, Copy II does not offer anything that
cramp, which is quite painful, believe me! Hopefully, future
makes it worth a cost of $39.95 (and half price for updates).
versions will clear up this very important problem.
Here, the program is similar to Diskmaker since it offers very little for such a high cost •
Here again, I can see a program that comes from a reputable company and would suggest its purchase for the drive monitor
Although compounded by the lack of support outside the copy
alone. Hopefully, some of its flaws will be repaired and some
process, the program does work well, but would work better
additions made to its features in a future update. Although this
with some form of disk scanner with which users could check
program is awarded a seven on the 1-10 scale, it is still not the
disks not listed on the parameter sheet for protection to set the
most cost effective. However, read on and find the cure.
program up accordingly. Hopefully, we will see these changes Nov. 1916: � 7,1111» 03
Fasthack'em V3.0
Basement Boys produces such a great MSD utility product. Also, the auto copy feature that does not require the computer
Fasthack'em, one of the most popular disk copiers, by Mike J.
is nice for us lazy users, or people who wish to make multiple
Henry of Basement Boys Software, is the program that started
backups (sigh, if only we could get it to put the disks in!).
current trends in disk copiers. In a manner of speaking, this is the one the others are trying to beat, and currently, no one has
Although I greatly enjoyed the usage of Fasthack'em, I must say
come close (but read on).
that its lack of utilities is giving some of its competition an edge, and hopefully we will see it expand and be more valuable in
The program is quite good and does come with some rather
the near future (I am told that Version 4.0 will support some
interesting programs supporting Commodore drives as well as
utilities).
the MSO dual drive. At a retail cost of $29.95, I found it very good value for the following three reasons.
For MSD dual drive owners, I do recommend the purchase of this product since it currently is the best and only supporter of
First, the program does support a fast load program that is the
the MSD. However, for 1541 users, it does not really support
only one currently available, to my knowledge, that will work
them to the full extent that it should, utility wise (which would
on both the 1541/71 as well as the MSD. The program is
extend its ability to function). I am ranking this program a nine
intelligent and, when activated, selects the type of drive in use
on the 1-10 scale due to its lack of available features.
and works accordingly. This is nice because I can now load programs faster on my MSD, which also allows more storage
Super Kit/ 1541
then the 1541. Super Kit/1541, from Prism Software, is an innovative new Second, the program supports a method in which the user may
product that supports a majority of features for 1541I1571
disconnect the computer from the drives (MSD or 2 1541/71's)
users for a low cost of $29.95.
and still copy disks. Fasthack'em will then take over the drives and allow copying without the computer. Now, you can fool
Having recently received this new product, I find it to have the
your friends with C64 multi-tasking as you use a word proces足
flavour and flare that I have been looking for for sometime. At
sor at the same time, or something like that (I usually use my
last, someone has released a full featured disk copier and disk
modem while all this is going on). However, remember not to
explorer for the Commodore 64/128. It uses most of the
let them see the disconnected cables from the drives.
current methods of disk copying as well as enhancing some of the old stand bys.
Third, the program not only supports the use of I 54 1171's in C64 mode, but will support them in C128 mode as well. I hope
SK is currently the fastest around, with Fasthack'em following a
to see the C128 support expand more and more in the near
close second. I found that version 1.1 was easy to work with
future. Currently, the program will copy a double sided 1571
although a few problems do arise that become annoying after a
disk in a little over a minute. A little note here about speed; the
while. For instance, a user is unable to return to the main menu
program will copy most disks in under two minutes, and
from any of the other program modules. In later versions, this
sometimes under three, depending on the amount of data and/
feature will be added, if I know my users out there (and I do
or degree of protection. However, Fasthack'em does have a few
because- wink! -I am one). For the most part, this is the major
drawbacks from an otherwise well thought out program.
complaint with SK.
First, the program currently does not support any form of disk
Another minor problem is the inability to view a directory from
utility such as disk editors or scanners. These would be very
the disk copiers (entire disk copiers, NOT the file copier). This is
helpful in exploring the disks that Fasthack'em will not copy.
also something that is now being incorporated into a future
At times, when copying disks, a parameter change within the
version, and has been promised to appear in version 1.2 or
nibbler might help a user create the backup, hence the need for
above this summer. For the time being, keep those disks
disk scanners and editors. Currently, users are unable to do so
labeled!
and must experiment with the various copy programs on the disk. Secondly, the program parameter copier, of which a
Super Kit (SK) is truly the most cost effective copier to date. It
majority of programs have duplicated, will often display in足
supports both single and dual drives (but not the MSD). The
structions for copying disks and then exit to BASIC. It would be
program comes on a double sided disk featuring a total of ten
nice if the parameter copier would execute as a copier without
programs. Seven of these appear on side one and are composed
having to load the various other programs in the Fasthack'em
of dual and single nibblers as well as normal disk copiers. The
arsenal.
rest of the programs deal with file copying and disk editing. Side two, composed of the last three files, includes another
The program Fasthack'em comes from a very reliable author
nibbler (single drive only) as well as a scanner and disk
who plans on expanding Fasthack'em in the near future. I
parameter copier.
highly recommend this program for MSD dual drive owners, as it supports that drive quite well. This alone makes it well worth
The copiers included on the disk on side one are the fastest to
the purchase price for MSO owners, as no one other than
date, and involve the usage of recent advances in programming Mov. 1916: YoUne 7, ..... 03
methods. The programs will operate at the maximum speed
The Sector Editor is basically a standard sector editor with a
possible without major data loss or corruption on Commodore
built in machine language monitor, which does support the
drives. The single drive copiers prove to be very fast and
printer for printed outputs of displayed data (the editor itself
reliable, giving the user a choice of whether or not to verify the
also supports the printer). The monitor is a standard ML
copies being made (which can speed up the copy process if
monitor that supports assembling/disassembling as well as
turned off). The dual versions offer the same type of options as
number conversions. However, it does not support a method of
well as something interesting to do while you wait, which, by
editing memory in the drive or unimplemented opcodes,
the way, doesn't last very long - the dual versions are so fast
things that I hope future versions will support. In this module,
that it makes it worth going out and buying a second drive,
the BAM editor is also supported (see above).
especially if you do a lot of disk copying (i.e. users groups). The thing that I miss here i.s the automatic option of Fasthack'em that one can become dependent on. However, the program
The best disk utility published to date, that others have pre
does provide some entertainment to ease the pain of losing the
or hardware, has finally surfaced in SK with the debut of the
automatic option. During the dual copy process, the program
Group Code Recording Editor included in SK, for the basic cost
plays music and displays graphics. Although it ain't Mr. Jack
of $29.95 (with all the above and more!). Basix has advertised
son, it does suffice, and seeing those 1541's on the display
such an addition to its DiskMaker product for over a year now
working as hard as the ones on my desk was truly cute, to say
at a price of $39.95 (along with some other utilities), however,
the least.
such a product has yet to be delivered (l'm still waiting ... will
sented in hardware or have attempted to do in either software
someone tell Basix to cancel my order?). The SK program is also the first program to provide a dual file copier as well as one for single drives. The copier program
The GCR Editor is a wonderfully innovative addition to any
operates in two different modes of operation; normal mode and
one's software library, be they novice or hacker.This will really
Super DOS mode. The program will copy or scratch files from
allow you to access the DOS in a way you could not do before. It
the disk in the same manner as other programs of this type. It
helps users to actually study and devise means or methods of
also supports a fast format that makes it easier to prepare disks
recording data on the 1541 disk drive.As a tool for exploring
to receive files. The program will copy SEQ, USR, PRG, and
different disk formats, it has proven to be an invaluable tool
even DEL files from disk to disk. Operation is the fastest to date
and has helped me backup disks that I was not able to do
in either single or dual drive modes.Please Note: According to
before. Plus, it has brought out the adventurous hacker out in
an inside contact at Prism Software, they are currently revising
me and taught me a few things I didn't know before about
the file copier to include features like fast verify and fast
standard DOS formats or the unnatural ones.
scratch, among other things.Watch for these in future updates. Well, enough raving. On to the details. The program provides a The secret of the Fast File copier's speed is in the two modes of
means of scanning headers as well as viewing and editing
operation that it supports: normal and Super DOS. The normal
tracks and data blocks. A user can then access a buffer area and
mode uses standard 1541 DOS, except it does speed it up a
edit these sections, viewing and changing the information in
whole lot. It only uses the standard DOS method of storing
the buffers. Later, he may re-save the new information onto
information on disk where the files are linked normally. In
disk.
Super DOS mode, the sector blocks are linked every other fifth block instead of every other tenth. For example, links usually
Please Note: I must honestly not recommend this program for
follow like this: 17,0 to 17,10 .... With Super DOS, they link
the novice user. However, you're only new once, so tinker
like this: 17,0 to 17,5 etc .. This can speed up load times
away at it and experiment on non-important disks. The man
considerably as well as speed up copy times. This is because a
ual offers some help for novice users by offering a list of books
normal 1541 has a very slow rate of data transfer. When data is
to read (see the end of this article) that will help tutor you in
retrieved from disk, it normally takes about ten sectors of
really getting to know the DOS.They are highly recommended.
movement of the diskette in relation to the head at 350 rpm to
Also, the manual offers a small "book" in the back by Rob
pass the data out of the bus. With Super DOS, data transfer is so
Vaughn that is called "The History of Commodore 64 Program
quick that by changing the spacing to every five sectors, a
Protection", a very nice change of pace, giving the reader a few
noticeable improvement is provided over its already fast load
hints and tips.
and copy times. The only problems that I encountered with the GCR editor was The File Copier also provides a BAM editor which will allow a
the lack of printer support for dumping the data buffers contain
user to view and edit the Block Availability Map on the disk. A
ing the GCR information. One reason for this is quite possible:
user enters a plus or minus to allocate or de-allocate disk
when dealing with the DOS, a programmer often cannot clearly
blocks, then can re-write the edited version back to disk.
access the printer due to the havoc caused by some of the DOS
•
routines. So, for the time being, this is something that I would The remainder of side one provides the user with several
like to see added to future versions as well. But, to date this has
utilities, a sector editor and a GCR editor. These programs are
only slowed my poking around a little bit and I am still enjoying
invaluable tools for those who like to tinker with new disk
myself.
protection schemes or develop their own ideas for such things. 36
Side two of the SK disk provides even more innovative fea
software using the dual or single drive nibblers. In other words,
tures. A Super Nibbler, a Super Disk Surgeon and the Super
you can set some program's parameters according to the
Scan.
information you receive from a disk scan.
The Super Nibbler is a very fast nibbler copier for single drives
This is a utility that is most important in exploring disks, and
only, that does not support any parameter changes except track
has yet to be bested by another vendor (although Disector does
ranges for copies. It is fully automatic and will detect most copy
come close). The format of this programs pictorial graph is a
protection schemes. This program is used to create backups of
form of BAM which shows all the disk tracks and how they
SK that will only copy to one generation. In other words, the
look. It helped me find out the exact tracks I needed to copy for
copies will not copy. The manual mentions that backing up
duplicating Diskmaker V3.3. The "Wall" track (Track
your original SK disk is recommended not only to protect the
peared as a pi sign that was not documented in the manual.
original but also to "customize" the program to best work on
Perhaps it is something that might possibly change in its
your particular drive. This program was tested on what I
appearance. My advice to users is that if you see anything other
consider to be the hardest to backup program that I own:
than the standard I, 2, 3, 4, then it's most likely a "Wall" track,
Diskmaker V3.3. It would copy the program only if I skipped
so skip it when copying or the drive will "hang."
6} ap
track 6 which I call a "Wall" track. A "Wall" track is one that has never been formatted, thus no sync. A drive attempting to read
This utility was truly marvelous except for its lack of printed
such a track will "hang". "Hanging" means that the drive will
output (Note: Disector would also not print out its reports for
spin continuously and freeze up on that track.
errors and density), and the inability to set a range of tracks to test. My advice is this: When scanning a disk, do a density scan
However, the program created a backup of Diskmaker in under
first because if you find a "Wall" track on the disk, doing an
a minute that worked just like the original. Except for the
error scan may cause the drive to "hang" in the error checking
parameter change above, I did no other edi. ting. The program
mode of operation (density scan is not affected and I have no
automatically detected the disk protection and acted accord
idea why}. The information provided is not too lengthy in
ingly, producing a working copy in no time at all. I would try
detail, but printed outputs for reference would be a nice
this one first on all your programs before trying the others. It
addition. However, currently the lack of printer output of
works rather well.
density and errors is only annoying; future updates will fix this problem. Note: the display that SK uses during a scan
is
easier
The Sector Surgeon is the doctor to see if your disk backups still
to record than Disector's and is also shorter. Also, the addition
will not work properly. Although it isn't the cure for everything,
of selectable scanning ranges has been promised in a future
this little program contains 270 parameters for copying a good
upgrade. I was told that version 1.2 would also correct many of
amount of todays most popular programs. Perhaps I should
the scanner's problems.
correct myself and say "large" program. But no matter what the size, the program does do a wonderful job.
A final part of SK's arsenal is the Autobooter and Super DOS disk fast load modules, which are the fastest around. These
The Surgeon even approaches disk parameter copying in a new
utilities are placed on the SK disk to provide users a means of
and innovative method that, by far, surpasses all other pro
getting their own programs to fast load in various methods.
grams around. This is accomplished after the user selects the parameter he wishes to use. After selection, the program turns
The Autobooter allows a user to create the appropriate fast
itself into a fast disk copier (single drive only), therefore
loader for his program with a selection of which Super DOS to
removing the need to load another program to copy the disk.
use. The Super DOS programs offer much support of various
During the copy process, the parameter copier will use the
modes of operation having to do with selecting whether or not
parameter provided to make a working backup of the program.
to have the screen on or off as well as working around the computer's interrupts. The Super DOS modules, at their best,
The copies made with Sector Surgeon may often run better
are able to load ISO blocks of data in about I 0 seconds. One
then the originals, and will often run on drives they did not run
note here is that Super DOS formatted files are still standard
on before. This program is designed to remove those nasty
DOS format compatible; they are made differently so that Super
head rattling protection schemes that so many of us have
DOS can load them faster (see above). The manual advises file
learned to hate. Also, once again, the copies are "customized"
copying the original files in Super DOS mode to get the best
to run better on your particular drive.
speed out of the fast boot.
I readily approve of the Sector Surgeon doing all my delicate
Super Kit/1541 is the newest kid on the block of disk copiers,
disk surgery and hope to see a dual drive version soon. This
and is to become the standard that the others will seek to
program also comes with the largest amount of parameters to
follow. On a cost/feature basis, I highly recommend this
date, and is yet to be beaten in ease of use.
program because of its very visible and innovative techniques used. At the cost or less of other copiers, you can get a whole lot
The last program on side two, Super Scan, provides users a
more in Super Kit/1541. This program is awarded a rank of a
method of scanning disks for DOS errors and standard or non
perfect ten, because as the advertisement truthfully states
standard densities. This can aid in making backups of most
Super Kit/1541, "Has it all!" No¥.1986: � 7, ..... 03
Benchmark Table
Conclusions The programs reviewed here follow two basic characteristics that make them important; cost and the number of features. Out of the group of six, the programs can be divided into two
Normal Copy
- Standard DOS Disk 0 Blocks Free.
Protected Copy- Attempted Backup of DiskMaker V 3.3. - 4 Standard Files each ISO Blocks long.
File Copy
groups; the ones to buy and the ones not to buy. The first three programs reviewed, Diskmaker, Keymaster and
Equipment Used
Copy II fall into that last category. These programs do not provide enough t o sustain their high costs. An examination of
Single Drive- Commodore 64 and 1541
the comparison chart will show you what is meant by this.
Dual Drive - Commodore 1281 and two 1571 's
I
MSD Drive - Commodore 64 and two MSD SD-2's2
truly feel that these products should cost less than they do until more additions are made to their arsenals.
Footnotes: I - In Commodore 64 Mode 2- FastHack'em Only
The following programs, Disector, Fasthack'em and Super Kit/ 1541 should be purchased for a variety of reasons. Disector's machine language monitor is very well put together
FastHack'em MSD SD-2 Benchmarks
and can be an invaluable aid to the programmer. A person in need of such a utility should run out and get Disector if your are desperate for such a monitor. However, pretty much of the rest of Disector proves to be a disappointment (Disector V2.0 was a
Program Name
Normal
Protected
Standard Nibbler
01/04
01100
Deep Nibbler
01/09
01/06
whole step forward - V3.0 was a half step), especially the file copier which is said to be "Ultra Fast" on the program's packaging. Unfortunately, this is an outright mistake on their
Note: See Benchmark Table for Disks used for testing. Time is Minutes/Seconds
part. The program Fasthack'em is a wonderful piece of software from an author who could , and should have offered more with his
Disk Copier Publisher Information
package . The program is invaluable to us MSD fans and C128 owners (as it supports the C128 in C128 mode). Fasthack'em is
DiskMaker V3.3
Disector V3.0
MSD or C128 owner, but as a C64 and 1541 owner, it does not
Basix
Starpoint Software
yet support you as fully as it should.
3463 State Street Suite 1541 L
122 S. Broadway
Santa Barbara, CA 93105
Yreka, CA 96097
The program Super Kit/1541 is truly a light in the dark, and is
Orders :(805)687-1541 ext. 34
(916)842-6183
currently unrivaled in the cost/feature area and is the one that
Tech
a reliable product that I would advise purchasing if
I were an
I
: (805)682-4000 ext. 33
recommend investing in as a user's first disk copy utility. Being the first product from a new company, Super Kit/ 1541 is truly
Keymaster
Fasthack'em
amazing, and I'm sure we will see more from Prism in the
MegaSoft, LTD.
Basement Boys Software
months and years to come as they provide more and more
PO. Box 1 080
PO. Box 30901
C64/128 support. Watch the software waves as this is the one
Battleground, WA 98604
Portland, OR 97230-0901
that the others are going to start trying to beat! Currently, at
(800) 541-1541
(503)7 61-1 1 1 4
$29.95, no one surpasses it.
(206)687-5205 Super Kit/1541 Copy IJ 64/128
Prism Software
Central Point Software
401 Lake Air Drive Suite D
I would suggest that readers interested in learning more about
9 7 0 0 SW Capital Hwy 11 I 00
Waco, TX 7 67 1 0
the DOS and disk protection methods pick up the following
Portland, OR 9721 9
Orders: (817) 7 5 7-4031
books:
(503)244-57 82
Tech
Suggested Readings
:(81 7 )751-0200
I) Inside Commodore DOS by Richard lmmers and Gerald Neufeld 2)1541 User's Guide by Gerald Neufeld 3)Program Protection Manual
I by CSM Software 4) Program Protection Manual II by CSM Software
Nov.19t6:ďż˝7,111â&#x20AC;˘03
Disk Copier Comparison Chart DiskMaker
Keymaster
Copy II 64/128
Disector
Fasthack'em
Super Kit/1541
3.3
1.0
2.6
3.0
3.0A
1.1
Publisher
Basix
MegaSoft
Central Point
Starpoint
Basement Boys
Prism Software
Level (I)
N
N
N
N and E
N
N and E
Docs (2)
I
3
6
8
7
10
Yes
Yes
Yes
Yes
Yes
Yes
3to 5
3to 5
3to 4
3to 5
3to 5
3to 4
Backs up self?
No
No
Yes
Yes
Yes
Yes(C)
2 drive option(s)?
No
Yes
Yes
Yes
Yes
Yes
Works with MSD?
No
No
No
Yes (6)
Yes
No
Price
$49.95
$29.95
$39.95
$39.95
$29.95
$29.95
Backup Cost
$15.00
No
backupable
backupable
backupable
Backupable
Upgrade Cost
$24.95
$10.00
$20.00
$5.00
$12.00
$10.00
Name Version
Easy to Use? 11
of Disk Swaps(3)
Speed
(I drive- 1541X4) V =Verify VF =Verify Off See Benchmark Table for Disks/Files Used for Testing
Normal Copy
01/45
01/24
02113
01/31
V02/31 VFO II46
V02/13 VFOI/37
Protected Copy
03/40
01/22
01/59
01/41
V02/09 VFO I/39
V00/59 VFOO/45
No
02/30
No
16/00
02/33
Nrm:02/17 SD:01/52
File Copy
(2 drive- 2 1571 'sX4) See FASTHACK'EM MSD Benchmark Table for MSD Information
Speed Normal Copy
No
02/03
02/43
00159
V00/57 VFOO/ 45
VOI/14 VF00/46
Protected Copy
No
01/57
03/40
Failed
02118
00126
File Copy
No
Failed
No
16/00
No
Nrm:02/14 SD:OI/47
Sector Editor?
No
No
No
Yes
No
Yes
GCR Editor?
No
No
No
No
No
Yes
Fast Format?
No
10 Sec
IS Sec
8Sec
10 Sec
10 Sec
Fast File Copier?
No
Yes(8)
No
No
Yes (9)
Yes (A)
Parameters?
No
Yes 50
Yes 200 (7)
Yes 40
Yes 100
Yes 270
Fast Loader? (5)
No
No
Yes
No
Yes 1541/MSD
Yes (B)
Utilities
KEY: I - Novice/Expert
5 - Fast Boot for BASIC
9 - I00 Blocks/20 Seconds
2 - Scale I to I 0
6 - Not All Features
A- ISO Blocks/23Seconds
3 - Single Drive Only
7 - Not on Disk
B - 150 Blocks/I 0 Seconds
4 - Minutes/Seconds
8- One File at a Time
C- Copy Protected
NOTE: Programs that failed did not work to specs or would not duplicate the Benchmark Disk. Super Kit/1541 NRM = Normal Fast DOS SO = Super Dos Mode for File Copiers. Copyright I986 By David W. Martin
39
Nov. 1916: VOlume 7, 1111â&#x20AC;˘ 03
WHO DO YOU TRUST? by The Disk Orderly
This is mainly a tale of anguish and woe -- describing twelve
hope that maybe the copies made there will run at my place.
hours in my life that I'd rather not relive. From the frustrations
My second hope is that those same copies will allow me to
experienced a few weeks ago, I ought to be sent to the funny
make further copies on my own equipment. My third hope is
farm. The pseudonym is to protect the guilty -- I'd rather not
that my dinner will still be edible (some say I haven't missed a
be known as the perpetrator/victim of these misadventures.
meal in my life).Well, it was one...two.. .three strikes I'm out. As hour four comes to a close, I'm no further ahead and losing
Oh, about t.he name...
whatever patience I had at the outset.
If Jim Butterfield is the guru, and Dick lmmers/Gerry Neufeld
Hour Five
can lay claim to the (deserved) title of being the Disk Doctor(s), then I suppose my qualifications rank me just above the
On doing some deep thinking, I recalled running a speed test
hospital volunteer "candy-stripers". If I could ever reach the
from an American copier program. It reported some months
exalted heights of "nurse-hood", I would rejoice and probably
ago that I was running 309-310 r.p.m. If alignment wasn't the
buy a real computer. Suffice it to say that the term "veteran"
bugbear, then maybe speed was killing me. After all, I rea
could apply to my experience. I've done lots of programming
soned, the SUPERKIT must be using its own carefully-timed
(including machine language), taught BASIC and written a few
routines to re-create all those fancy sectors from protected
articles in the past eight years. None of this experience quali
disks.
fied me for what that fateful day brought! To set the scene:
So I located a drive speed test program from RUN magazine It was a dark and stormy night. ..Oops, apologies to Snoopy for
(July 1985), typed it in and found my drive to be precisely on
that lapse.
speed. The same issue told me how to disassemble my drive for alignment/speed adjustments, but I couldn't picture myself
One day I purchased an Accolade game, since the SUPERKIT/
performing such a drastic action. Yet.
1541 package would enable me to obtain a backup. I once proclaimed that I'd never patronize software publishers who
Who do I believe now? The article did hint that if the drive was
abuse legitimate owners as well as thieves, but that noble
way off speed, then the program might erroneously report all
principle has been abandoned, albeit reluctantly.
was well. Do I hang my hat on that?
I raced home with the SUPERKIT like a kid with a new toy. I
Hours Six and Seven
should have known what was in store for me when I was unable to make a backup copy of the second side of the disk.
These were totally wasted, as I pretended that the elapsed time
Reset, reload, recopy. Futility, through three full cycles. Now
may have healed all the wounds. I repeatedly fired up the
try the SUPERKIT original to make a game backup. No luck
computer, tried to copy both disks again, loaded the copied
here either, as the three hour mark came and went.
programs and watched the computer crash. Time after time. With the same agonizing results.
Drag out the 1541 alignment utility from Transactor Volume6, lssue2; the light doesn't blink at all while confirming my
Hours Eight and Nine
alignment is dead centre.On to Plan B... It was time to call for help. The chap who had shown me the Hour Four
other drive speed program lives just across town, and a nine p.m. phone call would not threaten our friendship. He and his
So I call up my friendly dealer and ask him if he has a well
64 welcomed me. Now I had three sets of programs (both
aligned drivein the store.A 1571 running in 1541 modewilldo
SUPERKIT and game): originals, copies made on my drive of
the trick. I hustle down there with a few blanks and make
each (none worked), and copies made on the dealer's equip
copies of both the SUPERKIT and the Accolade game. The
ment. One by one, minute by (passing) minute, they were fed to
copies run perfectly on his equipment. Now, I dash home and
my friend's equipment. All failed. Even a full attempt to copy
No¥.!916: � 7,111TJe Ol
everything on his system was a dismal failure. Then I loaded
But what will happen when I flip the drive over, and seal the
the original game disk, and it didn't even run properly! So
case?
much for the scientific method--if I'd tried that one first it would have saved two hours.
Hour Twelve
When we located his drive speed program, lo and behold, my
Gulping even more, I dust off the RUN instructions on how to
friend's drive clocked in at 311 r.p.m. According to that copier
drill a hole in the 1541's case. This makes the drive speed
program, both our drives were out-of-whack. (His alignment
adjustment screw accessible (once you have re-installed about
was also fine). Now I was firmly convinced that speed was the
a thousand tiny screws). A 5/8-inch diameter hole in the
problem. After all, I knew of no serious complaints with the
bottom of the case. Now my friends KNOW that I've totally
SUPERKIT package. Maybe the RUN program WAS wrong in its
flipped.
OK status report. Now if anything goes awry, I can crank the speed back to what it was. I'm in the home stretch, as I put the chassis back into the
Hour Ten
case, turn it over and try everything over again: speed (both So I go home, with disks spilling from every pocket. Sure
programs), clone the game and the SUPERKIT again, and run a
enough the borrowed speed test program gave me the 310
few other disk programs. The RUN program now tells me that
r.p.m. verdict, as it had reported a long time ago. With that RUN
my speed is off by 3 ms, whatever that means. At least I have a
article in hand, I swallowed hard (three times) and started to
calibration mark for when this all stops working for me!
disassemble my drive. Me, the mechanical klutz, actually taking my drive apart! Friends say, "You did what?", when I
You should be able to determine by now that it is past two a.m.,
relate this part of the saga.
with Tuesday morning's alarm looming. After this marathon, I'm too wired to sleep, so I play a few rounds with my game
With the drive upside down, and the chassis resting in its cover,
clone.
I notice that there's strobe markings for 50 Hz and 60 Hz. When But what a price!
the drive is set spinning, the strobe markings are completely stationary. (I woke up my wife after midnight to help me find an old fluorescent desk lamp, so that there'd be no mistake). The markings didn't budge, not a millimeter per minute. Nothing.
Conclusion
OK, so who do I beiieve now? The RUN speed test and the
Sometimes persistence pays off. My scientific training helped
strobe marks confirm 300 r.p.m. speed. I'm sure Ontario Hydro
me to accomplish my goal. How many of you would have
knows enough to deliver 60Hz electricity! The other drive
trashed both disks earlier in the adventure? Was it worth the
speed program says 310 r.p.m. And the SUPERKIT won't make
effort? Probably not.
copies. What's going on here? I blame Accolade for one. Their "warranty" is a postcard, which offers a $3 rebate on another Accolade purchase if you answer
Hour Eleven
their market research questions. By now we've all seen the Taking a screwdriver in hand, I nudge the adjustment screw a
"non-warranties" on software that don't even promise that the
little, until the copier drive speed program reports it's now 300
disk will load. Accolade sinks to new depths--they don't even
r.p.m. Meanwhile the strobe marks start spinning around like
mention that there is anything in the package that you just
the label on a record. With the drive upside down, I proceeded
bought! Maybe that postcard is what I paid $40 for?
to make my backup of the SUPERKIT and then the game. These questions disturb me: Crossing all my fingers and toes, I reload the clones and try them out. Miracle of miracles, everything works! But what
How is it that two separate speed test programs, both reputable,
about all the stuff recorded at the "old" speed? Having a
give wildly different readings? Which do I believe?
hundred disks that won't work now, just to have a game and a Why did I have to disregard the accurate strobe markings, and
disk utility, is not a fair trade-off.
the RUN speed test results? In adjusting my drive speed to the I decide that an intermediate speed, of say 305 r.p.m., accord足
copy program's version of 300 r.p.m., then the SUPERKIT
ing to the copier speed test, might let me work with both old
began to work. Why?
and new disks. After setting this speed, I again make successful clones of everything. A few other commercial disks seem to
If anyone can help to explain what has happened here, I'd love
load OK, though the testing is superficial.
to hear about it. Right now, I find it hard to know who to trust.
41
Nov. 1986: Volume 7, laue 03
1HECLOCK
Donald P.
Maple
Calgary' Alberta
for Commodore 64
It is very easy to become oblivious to all worldly things while deeply involved in some profound programming task...
This handy little program will make sure that you will never
described below. They will start up only after the time has been
again miss anything because you got carried away while
set. CLOCK 1 is controlled by holding the CTRL key in addition
working on your computer. It will enable you to display two
to the function keys and CLOCK 2 is controlled by holding the
time zones and set audio and visual alarms for both!
Commodore key in addition to the function keys. Or, if you wish, the F-Keys tell the computer what to do, and the CTRL or
How many times have you missed your favourite TV show,
Commodore keys tell it who to do it to. This way the normal
been late for a date ...etc, all because of your computer? It is
F1-F8 are undisturbed.
very easy to become oblivious to all worldly things while deeply involved in some profound programming task on your
Hold CTRL for Clock 1, C = for Clock 2.
64, such as solving the ultimate question of life, the universe and everything. (fhe answer to which, of course, is 42!) Well,
Fl/2- will toggle the display on and off. Sometimes the actual
despair no more because the solution to all that is nigh! The
display of one or both clocks is not desirable. It may get
following program will offer not only one, but two clocks,
in the way of whatever is listed on the screen. Pressing
totally independent and both with alarm features. These can be
the key will disable the display, then pressing the key
used to keep track of two time zones or one for the time and the
again will bring it back. The clock itself is unaffected by
other as a stopwatch or anything else you can think of...
this and continues ticking whether the display is on or
Further more, they will not interfere with anything you happen
off.
to be running at the time! All this interest free, and no money down! If you can't wait any longer and/or are not really
F3/4- will toggle the alarm on and off.The visual feedback for
interested in the technical nitty-gritty of the program itself read
the alarm is the asterisk following the AM/PM flag. If the
on and the next section will reveal all the "HOW TOs"
asterisk is present the alarm is on, if it's absent the alarm is off. Again if the display of the clock is off the alarm is
Shake Before Use
unaffected and will wake you up as expected.
First and foremost you need peace of mind! Having achieved it,
When the alarm is triggered the display color of the clock
type in the BASIC loader using the Verifizer. Having completed
rapidly changes and there is also an audible sound. The sound
the typing save the program first, and then run it.The screen
will be heard even if the computer happens to be playing a
shows the range of memory locations into which the machine
song! When the display is off, the alarm sound will still be
language program is being loaded. The counter in reverse
audible but there will be no rapidly changing colors. The
video will constantly change to reflect the memory locations
reason for not awakening the display at this time is to prevent
stuffed as well as to provide a visual due that the program is
the destruction of whatever is in the appropriate corner on the
running. The BASIC loader contains the checksum to assure
screen. After all if the display was turned off it must have been
that the data loaded is correct. If everything has gone well you
for a reason!
can issue the SYS 12â&#x20AC;˘4096 in line 60, otherwise the innocuous "CHECKSUM ERROR" will be displayed to indicate it is time to
Once activated the alarm will go on for a couple of forevers,
get agitated all over again, with optional 4-letter words...
unless turned off by pressing the same F3/4. At this time the display (if visible) returns to the default color, the buzzing
Assuming that everything has gone well, or you have regained â&#x20AC;˘
sound stops and the asterisk disappears.
your composure, you will now see two clocks in reverse video on the topmost line of the screen. The left one in white,
FS/6- sets the time
affectionately known as CLOCK I , and the right one in black, nicknamed surprisingly enough, CLOCK 2! Note that the
F7/8 - sets the alarm
clocks are not running.To get them going use the function keys
Nov. 1916: Wume 7,111:Je 03
When these keys are pressed a snapshot of the current time or
applied to minutes, turning 10 into I 1. This brief time, how足
alarm selling is displayed, using the default color, in the
ever, is enough to trigger the alarm. The same logic applies to
following format:
hours as well.
TIME
:
021055P
If you are unhappy with the audio alarm, the pitch can be
ALARM : 030000P
changed by poking assorted values in the following locations:
The cursor is on the first digit and the time/alarm can be set by
CLOCK I - 49785 ($C279) freq lo
overtyping. To make the program as short as possible there is
49787 ($C27B) freq hi
no error checking and anything typed in a format other than
CLOCK 2 - 49786 ($C27 A) freq lo
the default snapshot will set the timers to unusual settings.
49788 ($C27C) freq hi
However, this is easily remedied by reselling. Having set the time/alarm, type "A" for AM or "P" for PM.
Having completed the customizing (sound, colours) the whole
Finally before pressing RETURN the color of display can be
program can be saved using a monitor:
changed in the same fashion as in BASIC, that is by pressing
S. CUSTOM CLOCK. 08 COOO C26E
CTRL or Commodore key and numbers I -8. The color can
,
,
,
actually be changed any time before RETURN is pressed. Next time load your own clock using a non-relocating load: When setting the alarm, the alarm flag (asterisk) is turned on
LOAD "CUSTOM CLOCK",8, 1
automatically. Note also that when the 64 is initially turned on the two clocks are not running. They are started up by setting the time.
and activate it with:
THE CLOCK will not only work in direct mode (such as BASIC editor or SUPERMON) but from within most programs as well. The only programs that will disable it are those that change the
This command is actually a toggle and it alternates THE
hardware interrupt vector. Also, while the editor is in the quote
CLOCK on and off. When turned off the actual timers keep on
mode the setting of both time and alarm, are temporarily
ticking preserving the correct time. However, having disabled
disabled.
THE CLOCK program, the alarms will not be checked. Note that before saving your own version, THE CLOCK must be
Unlike the BASIC time functions obtained from the "jiffy足
turned off! This is so that when enabled it will set the interrupt
clock" THE CLOCK is unperturbed by 1/0 activity on the serial
correctly.
bus. The display may freeze temporarily but the clock con足 tinues ticking. This is obvious when the display resumes.
The Life And Times Of 6526 CIA
During the testing, a couple of idiosyncrasies of the 6526 chip,
CIA, as we all know contrary to unsubstantiated rumors, stands
that contain the clocks, have been observed.
indeed for COMPLEX INTERFACE ADAPTER. Your Commo足 dore 64 has two of these. But what is this adapter and what
When setting the time/alarm any time beginning with I 2 in the
does is do? For one thing, it lives to communicate and it loves to
hours will cause the chip to flip the AM/PM flag. For example
talk to peripherals in particular. As such it serves as a middle
to set 5 minutes past noon (if" I 20500P" is used), the chip will
person between the processor and the outside world. Outside
turn in into I 2:05:00 AM! Apparently the clock sets the flag first
world, being a fairly complex machine, could not be handled
and then when it sees " I 2" it advances the flag causing this
with only one interface so that's why your 64 has two. One
problem. To circumvent this, in the example above use
controls the RS-232 (modem for example), certain memory
"000500P" for correct setting. Strangely enough the clock will
management aspects and the serial bus (to which we connect
handle the" I 2" properly if it arrived at it by itself. For example
peripherals such as disk drive, printer, ploller, etc). The other
I I :59:59 AM will turn into I 2:00:00 PM and eventually into
samples the keyboard and everything that hangs off control
0 I :00:00 PM.
ports I and 2, i.e. joysticks, paddles, light pen, trackballs, Koalapad etc.
The second "bug" has to do with the alarm. For example if the alarm is set for 03: I 0:00 PM it will go off as expected when this
In order to chat with all of those, timing is essential to any
time is reached. We would now turn the alarm off. However, if
computer. That's why each CIA has three clocks on board. Two
the alarm is turned back on it will be triggered again when the
of them are I 6 bit timers and the third is a Time Of Day (call me
time reaches 03: I I :00 PM??? The apparent reason for this is in
'TOO") clock. While the operating system uses the first two
the way the internal clock handles the carry. When the clock
timers extensively it has totally forgotten TOO. And that's
rolls 03: I 0:59 it turns briefly into 03: I 0:00 before the carry is
where we come in.
Nov. 1916: vaun. 7, ...... 03
Each 6S26 CIA has 16 registers used for various purposes. CIA
denotes AM, and if it's one then is PM. To be fair, though, the
I lives at SDCOO (S6320) and CIA 2 at SDDOO (S6S76). For this
high nybble of this register is not used for anything else.
program we are only interested in 6 of those registers. Finally there is the Interrupt control register at SOD (13). It is used to enable and sense the occurence of alarm that was set as
$08 ( 8) - TOO I Oths of seconds
9) - TOO seconds
previously described. This register is also connected to two
SOA (10)- TOO minutes
internal registers, a read only interrupt data register and a write
SOB (11)- TOO hours
only interrupt mask register. When the alarm occurs is will
$00 (13)- Interrupt control register
create an interrupt. Since there can be several sources of this
$OF (IS)- Control register B
interrupt the program monitors bit 2 of this register. If this bit is
$09 (
on then the time in TOO matches the time in the ALARM! And To obtain the real addresses of the registers simply add the
that's all there is to it.
offset shown above to the CIA start address. For example to get the hours register for CIA
I, add SOB (I I ) to SDCOO (S6320).
To do all this THE CLOCK modifies the hardware interrupt vector. This interrupt occurs 60 times per second and at that
An interesting note here. The designers of the 64 have decided
time the processor goes to a predetermined location and
not to decode all of the address lines pointing to these registers.
executes the "interrupt routine". For 6S02/6SI 0 type proces
This leads to the occurence of so called "ghosting". What that
sors the address of where to go is at $FFFE-$FFFF. On
means in practical terms is that the 16 registers, in this case
Commodore 64 it ends up getting the next address at locations
actually occupy 2S6 locations. In other words they are repeated
$0314-$031S. Without getting into it in more detail THE
16 times. So, for example, $DCOO will also appear at $DC I 0 and
CLOCK changes this vector to point to THE CLOCK itself. The
$DC20 and SDC30 . .. etc!
program, however, politely makes note of where these two locations used to point to and, when done, jumps there to
The four TOO registers actually access eight internal registers.
resume regular processing. This means that if some other
In other words the same four addresses point to both time and
program has already modified the vector, this utility will main
alarm functions. Which one of those functions is accessed
tain it. Normally, however, $0314-$031S will point to $EA31.
depends on bit seven of Control Register B. If this bit contains a zero, writing to TOO registers will set the time, but if this bit contains a one, writing to TOO registers will set the alarm. bit 7 of SOF (IS) bit 7 of $OF (IS)
Conclusion
=
0 - writing to $08-SOB (8-11) sets time
This article
and
the
accompanying program have
=
I - writing to $08-SOB (8-11) sets alarm
scratched the surface of this, indeed, complex interface chip. If
only
you have any questions or suggestions you can contact me Note that the state of bit 7 of Control Register B is of importance
either through this magazine or directly at the address below.
only when writing to TOO. When reading from TOO we will always get the time.In other words TOO contains a write only
Donald P. Maple
ALARM and read/write TIME registers. Because of this THE
P.O.Box 23, Station M
CLOCK program saves the alarm information independently.
Calgary T2P 2G9 CANADA
The order in which the TOO registers are read from or written to also plays an important role. Upon reading hours register at SOB
(II), TOO clock stops. It will resume only after reading
Editor's Note:
register $08 (8) - I Oths of seconds. This does not apply to minutes and seconds. They can be read without sloping the
We asked Don for a copy of the source code for this program
clock. Actually the clock continues tick-locking internally but
but apparently he used (believe it or not) the simple assembler
the output registers are latched until the read on I Oths of
in Supermon to wn·te it, and unfortunately, time did not permit
seconds occurs. The same sequence is necessary when writing
us to disassemble and comment it for you.
to either time or alarm registers. The clock will commence only after a write to I Oths of seconds occurs. To make the life of programmers easy the designers of this chip have decided to store all of the TOO data in BCD (binary coded decimal) and that is how the data is read or written. On the other hand, to make the life of programmers difficult the same designers have decided to use the hour register at SOB
(II) to
store the AM/PM flag! If bit seven of this register is zero it
Nov.1916: � 7, ..... 03
THE CLOCK: BASIC Loader EG
10 f
LN
20 print·
49152:1
49837
=
ng from·f ·to·t
LJ
550data174,115,194, 93,109,194,157,109
Ml
560data194, 96, 160, 147, 44, 160, 152,165
01
570data212,208,246,169,193, 72,152, 72
01
580data
8, 72, 72, 72,165,206,174,135 2,160,
KP
30 for j
OK
40 pokej,x : ch
MA
d 50 if ch<>84866 thenprint checksum error· : en
LE
590data
BM
60 rem
OM
600data 76, 129,234, 162,
EG
70 rem
KA
610data162, 15, 56,173,134,
JG
100data120,174, 20,
3,173
GL
620data172, 115,194,185,113,194,141,134
GO
110data135, 192, 141, 20,
3, 173, 136, 192
KK
630data
OM
120data141, 21,
3, 142,135, 192, 140, 136
PK
640data220, 24, 109,115,194,133, 252, 133
0,181,251, 72
JG
650data254,160,
4,208,248,164,197,204
EN
660data
=
•
f to t : rea d x : print· =
ch+ x : next
•
J
•
sys12•4096 3,172, 21,
IM
130data192, 88, 96,162,
PJ
140data232,224,
IN
150data116,194,240, 38, 140, 116, 194, 192
7,177,251, 40,144,
NB
700data194, 133, 254,
32
IM
710data194, 41,
2, 240,246,160,
MC
190data194, 185, 122, 194, 141, 80, 192,
3
670data154, 194, 32,210,255,202.136, 16
1, 141,115
180data 17,169,
8
9, 128, 44, 41,127,145,251,189
690data129, 176,
GP
3
8,133,251,133,253,169
FN
170data141,
0, 44,169,
2, 72,
6, 74,144
PO
7, 176, 27, 173
2, 74, 74,176,
7, 24, 144,
680data246, 232, 240,
160data
3,144, 31, 192,
IP
2,169,
0,132,207, 32, 19,234
HB
GN
25,165, 254, 74, 169
2, 169,136, 133, 253,169 32, 69, 193,189,109 3,177
LJ
200data 69,193, 169,
8, 133, 251, 169, 220
KC
720data253, 16,
5,162, 80, 41,127, 44
JN
210data133,252. 169,
0, 133, 253, 169, 212
AA
730data162, 65,
44, 177, 253, 72, 74, 74
KP
220 data 133,254, 162, 12,172, 111,194, 173
GB
740data 74, 74,
9, 48, 32, 210,255,104
MC
230data 109, 194.
32, 137, 192, 230,252, 169
OA
750data 41, 15,
9, 48, 32, 210, 255,136
BE
240data
7, 133, 253, 162, 39,172,112, 194
BJ
760data208,233,138, 32,210,255,160,
LM
250data173, 110, 194,
3
PC
770data 169, 157, 32,210,255, 136,208,250
GJ
260data 104,149,251,202, 16,250, 76, 27
LP
780data160,
El
270 data192, 74, 144, 82, 72, 74, 169, 160
EK
790data 10,153,170,194, 32,207,255, 41
KA
280data 144,
FC
800data 15, 25,170,194,153,170,194,136
AE
290data 72,138, 72,160,
Nl
300data
JF
310data194,162,
HL
320data 74, 74, 74,
GJ
330data177,251, 41, 15,
32,137, 192, 162,
2, 169,170,141,153,194,152
7
3, 32,207,255, 10, 10, 10
3,177,251, 16
IF
81 0data208,232, 32, 207, 255, 74, 160,
3, 169, 144, 44,169, 129,141, 151
IN
820data185, 170, 194,144,
4, 41, 127,176
LF
830data
3,185,170,194
FH
840data145,251, 145,253, 136, 16,246,174
PG
850data 115, 194, 173,134,
0, 177,251, 41,127, 74 9, 176,157,142,194 9,176,232,157
7,
9,128,144,
3
2, 157, 111,194
Jl
340data 142,194, 232, 232,136,208,228,177
81
860data 157,113,194,104,141,134,
2,169
HI
350data251, 160,
FC
870data 13, 32, 210, 255, 96,
1,
1,
1
KP
360data157,
0,216
FN
880data
0,
1,
0,
0,
0,
0
HJ
370data202,136, 16,241,104,104, 74,144
OP
890data
0,
0,
0, 10,
5, 117, 43, 69
BM
380data 73,165,252, 74,162,
1
NF
900data 114,
0,
o,
8,
EM
390data232,189, 109,194, 10,176, 22, 160
GN
910data
0,
0, 145,
OL
400data
4,240, 51,189
PB
920data186, 176, 183, 186, 178, 183, 160, 129
HM
410data 109,194,
9, 128, 157,109, 194, 160
NC
930data141, 170, 58, 69, 77, 73,
KN
420data
0, 145, 253, 160,
0, 189
JA
940 data 32, 13, 58, 77, 82, 65, 76, 65
OP
430data121, 194, 145, 253, 200,189, 123, 194
MO
950 data 32, 13,
AE
440 data145, 253, 160,
co
450 data200, 169, 15,145, 253, 200, 169, 240
GP
460 data145, 253, 169, 15,141, 24, 212, 254
HE
470data111,194, 96,169,
Cl
I
=
0,
12,104,170,185,141,194 4,104, 72,157,
5,177,251, 41, 4, 169,
0,144,
0,
0,
0, 64,
o,
0,
0,
o
83, 160, 176, 176
0,
0,
84, 32
8
4, 169, 33, 145,253
1, 32,104,193
480data 74, 176, 17, 165, 209, 72,165,210
AB
490data 72,162,
FO
500data210, 104, 133,209, 96, 169,
ME
510data104, 193, 74, 74, 176, 25, 138, 74
0, 32, 255, 233, 104,133 2, 32
Ll
520data 160,
CF
530data2 · 12, 189,109,194, 41,127,157,109
KG
540data194,189,113,194,157,111,194, 96
1M Tlaneaclot
4,144,
2,160, 11,153,
0
Nov. 1916: Yolume '· ...,. 03
I
VIC-II Chip Interrupts
TomasHrbek Rocky Point, New York
This month's issue is about programming the Commodore chips. The most known and programmed chip by the average user is the SID chip, controlling sound and a few other things such as game paddles. Then there are the two CIA chips used for the housekeep ing functions of the C-64. They are the timers for data transfers
As mentioned in the previous paragraph, there are about 250
between the computer and different devices. They read joysticks, determine which key on the keyboard was hit and also generate
screen lines, just like in high resolution mode), and the border ends at about 280, although the electron beam continues on a little
the IRQ and NMI interrupts. There is also the VIC-II chip, which manages the Commodore 64 architecture. It switches between the
past that. This is why sprites can disappear behind the screen edges.
four 16k banks, displays different kinds of modes, as for example the text, hi-res and multicolor mode, controls sprites and other features. In this article I would like to discuss the lesser known features of the VIC-II chip.
The two registers that contain the value of the raster line are located at $0012 and $0011. Bit 7 of $0011 contains the overflow
raster lines displayed on the screen. However, the topmost visible line is raster line number 30, which is in the border area of the screen. The topmost line of the text area is raster line 40. The same is true for the bottom. The text area ends at raster line 240 (200
from $0012. That is because a memory location can only hold a value from 0 to 255. If a value is written to these registers, we instruct the VIC-II chip to generate an interrupt at that location. Before the interrupt can occur, however, we have to do some
As the title of this article suggests, the VIC-II chip is capable of generating its own interrupts. These interrupts are of four different natures. They can be generated by a lightpen, sprite to sprite or sprite to background collisions, or by a specific raster line being reached. The last interrupt mentioned is the least understood by the average programmer, and is the one that I will discuss further.
preparatory steps. To start, we have to tell the VIC-II that we are enabling one or more of its interrupts. This is done by setting some bits in the Interrupt Mask Register (IMR), location $00 I A. Bit one tells the VIC-II that we want to enable the raster line interrupt; bit two,
You might be wondering what this raster line interrupt might be and if you have ever seen it. You probably have, even though you might not know about it. If you own the popular word processor SpeedScript, the raster line interrupt generates the top message line. Other programs that also have this feature are PaperClip , KMMM Pascal, Cad-3D and others. Here, the raster line capabili ties are used to highlight a single line, but you will see that raster line interrupts are much more powerful than that.
sprite to sprite collision interrupt; bit three, sprite to background collision interrupt; and bit four, lightpen interrupt. The next three bits are meaningless. Bit seven is used to indicate that any one or more of the previous bits has been set. Now that we know how to enable the raster line interrupt, we have to adjust the IRQ routine so that we will be able to take advantage of the interrupt generated by the VIC-II. Whenever any chip on the C-64 generates an interrupt, it jumps through a vector at location
To gain an understanding of this interrupt, we have to begin by examining how the computer displays information on the screen. Attached to the back of your monitor's picture tube is an electron gun that is used to generate an intense and accurate flow of electrons towards the phosphor coated surface of the inside of the
$314-$315, which points to the address SEA31. However, this vector can be easily changed to point to our own custom routine. In our IRQ routine, we have to check if this interrupt has been generated by the raster line interrupt. This is done by reading the
picture tube. Monochrome monitors use one type of phosphor that allows graduations of intensity between white and black (on and
Interrupt Raster Register (IRR) located at $0019. This register has the same configuration as the IMR register. It, however, indicates by setting the appropriate bits which device generated the inter
ofQ, to provide an illusion of color and shading. Colour monitors use three types of phosphors throughout to produce three colours; Red, Green and Blue (RGB). These phosphors are grouped accu rately and very closely together to produce all possible colour combinations depending on the combination of RGB chosen for
rupt. Since w e only enabled the raster line interrupt, it is sufficient to check bit seven, which indicates that the VIC-II generated this interrupt. Thus, if this bit is not set, we simply proceed with the
display. Phosphor, for those who are not sure just yet, is a substance that will produce light when struck by an electron
usual interrupt routine. However, before we do that we have to clear the CIA! IRR by reading register SDCOD.
particle. Now that we know that the routine was generated by the VIC-II, we can proceed with our IRQ routine. Before doing that, we have to clear the VIC-II IRR register, or it will generate another interrupt
In order for the picture on the screen to be of any quality, the electron gun controlled by the computer has to generate over 250 visible lines and refresh them 20 times per second. Each line is also divided into about 180 visible points. Because of the speed with which each point has to be refreshed, the two VIC-II registers
as soon as w e exit the IRQ handling routine. This is done by writing the value we read from the VIC-II IRR back to the IRR register. Having done all of this, we can finally write our program.
that tell us the position of the electron gun, contain only the number of the raster line, not each individual point.
46
No¥.1916: � 7, llliJeC»
PAL Source Listing
The Program My example is a very simple one. It highlights a line on top of the text screen just like in SpeedScript. As I mentioned earlier, the raster line is much more powerful. We can display more than eight sprites on the screen at the same time, different kinds of modes, more character sets, etc. We can even divide the screen into more than one part (split screen). Of course, we can divide it only along the horizontal lines. After doing all of the preliminary steps, we decided that this particular interrupt was generated by the VIC-II chip. We compare the value of LINE2 to that of the RASLIN register, location $0012. If the value of LINE2 is greater than that of the RASLIN register, we store the value of UNE2 in the RASUN register, change the screen color and exit. However, if the UNE2's value is equal or smaller, we load the accumulator with the lower value of the highlighted line (LINE I), store it in the RASLIN register, change the screen color and exit.
Here are some hints on how to utilize the raster line interrupt in other ways. If you do not want text to scroll into the highlighted line on top of the test screen, clear bit three of $001.1. This will tell the computer to display 24 instead of 25 lines per screen. Then if you want to display something in the highlighted line, just poke the message there. Poke the first 40 bytes of screen memory usually starting at 1024 or $400. If you want to display more than eight sprites each time you switch between the two parts of the screen, just redirect the sprite pointers to the appropriate part of the screen. The same thing applies to the hi-res or multicolor screen. Just repeatedly switch in and out of these modes. To wrap this session up, I must say that raster interrupts are very· powerful feature of the Commodore 64 that is unfortunately almost unknown. It is certainly due to the fact that this kind of feature can be only explored and utilized if one is a machine language programmer. I hope that this article made it clearer how to use the VIC-II chip in other ways than the usual ones. I also hope that it will help you in further exploring the C-64 and in writing better programs.
Basic Loader
PE PO AO PA CA GL DH NH BP EJ OK PF DH AF IC AO KH 01 BL LM JO LN JA HC
pp NG
10 PL LB FA 08 Gl KB El DA BL OF ME DH AM PI EH CH PA AE LE
OE KF MM
100 rem save· O:raster irq.ldr ,8 110 rem • • by tomas hrbek rocky point, new york 120 rem • • raster demo for the c64
00 NA KH
130 for j = 828 to 903: read x: poke j,x: ch = ch + x: next 140 if ch<>8431 then print· checksum error!·: stop 150 sys828
GK KH KM
CN HL
160data 120,169, 91,141, 20, 3,169, 3 170data 141, 21, 3, 169, 50,141, 18,208
GO
LE El
180 data 173, 17,208, 41,127, 141, 17,208 190 data 169, 129, 141, 26,208, 88, 96,173 200 data 25, 208, 141, 25, 208, 48, 7, 173 210 data i3,220, 88, 76, 49,234,173, 18 220data 208,201, 58,176, 10,169, 1,141
IN OA IL GA 01 KA AB AD CL CD CK
CP EP EB FD GM IH
•
230 data 33,208, 169, 58, 76,130, 3, 169 240data 0, 141, 33,208,169, 50,141, 18 250 data 208, 76,188, 254
DO
100 rem save· O:raster irq.pal" ,8 110 sys 700 120.optoo ;cassette buffer 828 130. 140; 150 key $ea31 ;old irq handling routine 160 cinv $0314 ;irq vector 170 raslin $d012 ;raster line $d019 ;flag for video interrupt 180 irr 190 xmr $d01a ;video controller interrupt flag 200 colreg $d021 ;background color register 210 icr $dc0d ;flag for timer interrupt . 50 220 line1 ;start of highlighted line 230 line2 ;its end 58 ;raster line color 240 color1 1 ;background color 250 color2 0 260. 270 setup • 280 set 290 Ida #<irqnew 300 sta ClnV Ida #>irqnew 310 320 sta cinv+ 1 330 Ida #line1 340 sta raslin ;for interrupt Ida raslin-1 350 360 and #$7f ;clear hi bit sta raslin-1 370 380 Ida #%10000001;permi t irq by raster sta tmr 390 400 eli 410 rts 420. 430 irqnew = • 440 Ida 1rr sta 1rr ;clear interrupt flag 450 460 bmi vicrastr ;interrupt by raster line '
'
470; 480 490 500 510. 520 vicrastr 530 540 550
ICr
Ida eli Jmp
key
=
•
;clear interrupt flag ;allow raster interrupt
'
rastn ;read raster line Ida cmp #line2 bcs greater ;greater than or equal second value
560; Ida 570 580 sta Ida 590 600 Jmp 610 . 620 greater = Ida 630
#color1 colreg #line2 ;next interrupt at 2nd line exit
•
640 650 660. 670 exit 680 690 700; 710.end
sta Ida
•
#color2 colreg #line1 ;next interrupt
•
•
sta Jmp
raslin $febc
;to wrap things up and exit
Nov. 1916: Volume 7,111U. 03
Switcheroo Pivot
Chris Miller Kitchener, Ontario
I
-
An 8500 & Z80 Working Together In The C128
The C-128 is a two processor system. Inside are an 8500 and a
8-Blt Internal Registers
280. The 8500 is really just a 2mz. 6510. The 280 is one of the most advanced 8 bit processers around. The 8500 is a memory
A
A'
accumulator
based microprocessor. Even just to subtract the contents of one
B
B'
general purpose
register from another would require some free RAM. Indirect
C'
addressing modes and a sparsity of internal registers make the
c D
8500 very reliant on zero page pointers. The 280 is a register
E
E'
based microprocessor. It has two sets of general purpose
H
H'
registers. Each of these sets contains an accumulator, a status
L
L'
register and six, 8 bit, general purpose registers. The second set
F
F'
D'
flag (status)
can be used for the interrupt flip-flop (IFF) or by the exchange (EXX) command to remember and restore register contents.
Data registers can also be paired for 16 bit addressing and
16-Bit Register Pairs
arithmetic. In addition to these there are four other 16 bit registers: the PC (program counter), the SP (stack pointer) and
BC
B =hi byte C =low byte
the (IX) and (IY) (index) registers.
DE
D =hi byte E =low byte
HL
H =hi byte L=low byte
Nov. 19t6: Volume 7, ..... 03
True 16-Bit Registers
The
6. The 8-Bit Arithmetic And Logical Group
IX
index
These allow for manipulation of one byte values in pretty much
IY
index
the same way 6510 programmers are used to. Addition and
SP
stack pointer
subtraction are possible with or without carry.
PC
program counter
280 has several times as many commands as the 8500;
7. The 16-Bit Arithmetic And Logical Group
some therefore require more than one byte of opcode. These commands can be functionally divided into 13 groups, each of
Same as above but with two byte values being manipulated.
which is more extensive than its
The logical AND, OR and XOR are not found in this group.
8500 counterpart where
counterparts even exist. 8. The CPU Control Group I. The Eight Bit Load Group Processor and interrupt modes and status flags are handled. The
280 assembler load instruction, LD, might more aptly be
named MOVE. There is no store instruction. Every LD will be followed by two operands delimited by commas. The first
9. The Rotate And Shift Group
operand represents the destination and the second the source, so that the instruction LD($COOO),A means store the contents of
Many different types of shifts accessing both one and two byte
A at $COOO whereas LDA,($COOO) would mean load A from
values via a variety of addressing modes are available.
$COOO. In 280 mnemonics, parenthesis define a memory loca tion; otherwise an immediate value is assumed. 10. The Bit Set Reset And Test Group 2. The Sixteen Bit Load Group
These commands provide for complete bit addressing. Each takes two parameters. The first will specify which bit (0-7) is to
This includes all the commands which move two byte values
be set, reset, or tested; the second will designate the register or
either between registers or between registers and addresses.
memory location to be manipulated. For example SET3,(1X + 0)
Included here are the PUSH and POP instructions which is
would set bit 3 in the address pointed to by the IX register; ie
handy since addresses are what stacks are mainly for.
OR it with the number 8.
3. The Exchange Group
11. The Jump Group
Register contents can be swapped with the secondary set or
Conditional and unconditional jumps (direct) and branches
within the primary set. There's nothing like this on the
(relative) are supported. Anyone who has ever had to fake a
8500
although we often wish there was.
conditional jump in 6510 via BNE• + S:JMPFAR or an uncondi tional branch via SEC:BCS NEAR will appreciate the versatility of this
280 group.
4. The Block Transfer Group Set a few register pairs and use one of these to move or fill
12. The Call And Return Group
memory a byte at a time or in a 280 controlled loop. The short 280 routine which we will later call from Basic to copy its ROM into
8500 visible RAM uses an LDIR loop.
Subroutines may also be called and returned from condition ally or unconditionally.
5. The Block Search Group
13. Input Output Group
As above, the
280 can automatically control looping by count
These are specialized load and store instructions. In the C-128,
ing down the value contained in the BC pair and incrementing
when accessing 110 memory (DOOO-DFFF), IN and OUT com
the address pointed to by DE. Ranges of memory are compared
mands should be used instead of LD.
with the A register until a match is found or the BC pair decrements to zero.
Nov. '1916: � 7, lnue 03
LO A,$81
Programming The Z80 In 128 Mode
LO ($0505),A The 280 brings a convenience and conciseness to ML program ming that is sure to please and impress 6510 assembly lan
This is inferior. There is a bleed through condition in the Z80
guage programmers. I hope the above has whetted your
mode using this type of store. A $B I will also be written to
appetite for doing a little exploring. It will inspire you to know
underlying RAM. (fhis is where my 280 cross-assembler sits,
that this microprocessor can be used in conjunction with (not at
making this feature especially bothersome.)
the same time as) the 8500 in the C-128, even from Basic; switching between them is not much more difficult than
Here is the proper way:
switching between memory banks once you know how. LO 8C,$0505 LD A,$81
Working Together
OUT (C),A In order to figure out how the 280 and 8500 worked together, it was necessary to disassemble some ZBO ROM; in order to
Not only does bleed through not occur using OUT storage but
access this ROM it was necessary to enter, and get back alive, from 280 mode--sort of a catch 22 situation. One hundred
110 memory between $0000 and $DFFF can be written to. In our Basic coding sample the background ($D021) and border
cups of coffee and about a thousand crashes eventually pre
($0020) are poked via a 280 OUT. This would not work using
vailed. Disassembling the boot sector on the CP/M disk got the
LD($D020),0 which would put the value in RAMO.
ball rolling. Bit 0 at $D505 (54533) controls the microprocessor mode. If it is turned on then the 8500 becomes active; if it is off then the 280
Ordinarily you would have to bear in mind that the 280 might not necessarily take off at $FFEE the next time you activated it. It, like the 8500, wakes up where it went to sleep. The best procedure for switching back and forth is to try to always put
takes over.
the microprocessors to sleep in the same spots. These switches You can't just poke it off (believe me). A little housekeeping is
could be followed with jump commands. Before invoking them
first in order:
you could set the jump address for the other microprocessor to anywhere you like. 280 ROM puts a RET ($C9) command after
Disable 8500 interrupts via SEI because you are going to switch
the 8500 switch allowing the 280 to CALL the 8500 from anywhere and return when the 8500 switches back. Switch
to a memory configuration in which Kernal ROM is not visible.
eroo puts an RTS ($60) after the 280 switch so that the 8500 can Store a $3E (62) at $FFOO (the configuration register). This
JSR the 280. This also makes the Switcheroo routine com
leaves 110 RAM intact but switches everything else to RAM 0.
pletely relocatable.
You're still not quite ready. The 280 PC register holds $FFED
Now it just so happens that there are two routines high in RAM 0 through which the two microprocessors can invoke each
after 128 initialization. There is a NOP ($00) there. The first actual 280 command goes at $FFEE. If you look through the
other. The 8500 invokes the 280 at $FFDO. When the 280 returns control, the 8500 picks up at $FFDB. Leave the NOP
monitor you will see a $CF there. This is an RST8 opcode byte which will cause the 280 to jump (ReSTart) to its own ROM
($EA). You can take over at $FFDC (65500).
routine at 0008. You do not want this. After moving some 8500 code into place at $3000, the 280 would return control to the
The 280 invokes the 8500 at $FFEI. When the 8500 returns
8500. The 8500 wakes up exactly where it left off after you
control, the 280 picks up again at $FFEE--and so on and so on.
switched to the 280. If you followed this switch with a NOP (lets not wake it up too fast) and then a JMP $3000 (like the operating system does) you would go into the I 28's boot CP/M
The Switcheroo Pivot
routine. This is pretty useless from a programming standpoint, so don't bother. Instead, put your own 280 code at $FFEE.
Switcheroo handles the 280 stack, the user call, and controls the "sleepy time" program counters for the two microproces
Before you do any 280 subroutine calls, you should set its stack
sors while making use of the RAM routines at $FFEI and
pointer register (SP) to point to some area that will not interfere
$FFDO. The Switcheroo pivot thus allows you to easily execute
with your code or Basic. Switcheroo points it to $BFFF before
hybrid programs and, as our example shows, even call the 280
calling a user routine. The return address goes into memory at $BFED in this case.
from Basic.
The last thing the Z80 will have to do is to turn the 8500 back
The Switcheroo code sits at 3000, high in the I 28's tape buffer. The address of the Z80 code to be executed should be in the
on. There are two ways to do this:
8500's X (=low byte) and A (=high byte) registers. These can
50
No¥.1916: � 7, ...... 03
be passed directly from ML or even Basic via the I 28 s new
Switcheroo Pivot Source Listing
'
improved SYS command, which is exactly what our little Basic example does. The program pokes some Z80 code in at $6000,
OM LN
calls Switcheroo to execute it and then continues in Basic. The
100 rem save·O:switcheroo.pal" ,8 110 open 8,8,1, O:switcheroo •
280 code copies its ROM into RAM at $8000. Notice how easy it is to code this move (4 instructions, II byt es). The 280 then
JE
120 sys700
FK
130.opt o8
pokes the screen colours just to show off.
GO
140.
MA
150:
HN FA
160 z80pc
The Switcheroo pivot isn't long at all, and should pave the way for some serious exploring of the 280 language and environ ment in the 128 by 8500 buffs.
AL Bl EH
Switcheroo Pivot Basic Loader
•
-
3000
1 70 invokez80 180 c1 28pc 190 invoke128 200 z80stk
$ffee
;z80 wakes up here
$ffd0
;by 8500
$ffdc $ffe1
;wakes up here ;by z80
$cfff
:safe place for stack
IE
210;
AL
220 :••• 'switcheroo' for hybrid processing •••
KM
100 rem save·O:switcheroo.bas ,8
GJ
230 :••• by chris miller, may 20, 1986
01
110 rem
240;
CC FD
120 rem •• code into memory at $6000 and then
GG NO
250
sty $ff01
AB 8P
260
ldy #$31
NF
130 rem •• execute it via the switching 140 rem •• routine at 3000.
270
sty z80pc
KA
150 :
FF
280
ldy #<z80stk
OK
160 rem •• now put z80 code into memory
MO
290
sty z80pc+ 1
8K EA
170 for x
FG DA
300
ldy #>z80stk
310
sty z80pc+2
JH
190 print ·back alive and well!!!·: end 200:
CG
210 rem •• the z80 code to copy 4k rom to $8000 in ram 0 ••
EC NC
MD 88
•
**
=
this program will poke some z80
24576 to 24598: read b: poke x,b: next
180 sys 3000,96,0: rem call z80 at $6000
HI
220 data
33,
LN
230 data
17,
FD DL
240 data 1, 0, 250 data 237, 176
JP
260 rem •• z80 code to poke background and
GL
0 :rem ld hi,O 0, 128 :rem ld de,$8000
;set ram 0 config ;ld sp,($bfff)
320 ' 330
ldy #$cd
340
sty z80pc+3
350
stx
°
;call Gsr) opcode
z80pc+4
0,
(sys parameters)
16 :rem ld bc,$1000 :rem ldir
sta z80pc+5
360
ON 10 NH
380
Ida #$c3
NN
390
sta z80pc+6
KL
Ida #<invoke128
G8
400 410
KM
420
NC
430 440
Ida #>invoke128 sta z80pc+8
370
'
°
border colour •• LG
270 data 62,
GA
280 data
HM EM LM FO NC
0
:rem ld a,O
32, 208 :rem ld bc,$d020 :rem out (c),a 290 data 237, 121 :rem inc a 300 data 60 1,
310 data 12 320 data 237, 121 330 data 201
;subroutine address
:rem InC c
sta z80pc+7
:rem out (c),a
oc AO
:rem ret
DG
460
Ida #$60 sta c128pc
8M
470
JSr
GF
480
KO
490
Ida #0
MA Ml
500
sta $ff00
510
eli
EP
520
rts
51
;jp Gmp) opcode ;invoke c128 when can't continue
° '
450
:rts opcode
invokez80
° '
:back to basic
Nov. 1916: \llaUne 7, ....... 03
Richard Stringer
Progra1n Trace Monitor
Dallas, Texas
- A profiler for Commodore 64 machine language programs
This program will trace the execution of any program and show how often each section of the program is executed. It is primarily designed to be used with machine language programs, but can be used for programs in any language. The output, in the form of a chart, can be used either by another program or the programmer himself to provide information on the execution speed, memory use, location of called subroutines, and general efficiency of the code. The output of the program is in the form of a chart, which is sent to the printer due to the large amount of data generated. No printer codes are used, so any printer should work. Along the left side of the chart is the high byte of addresses in the traced area, and along the top are the low address bytes, in increments of 16. At the intersection of a high/low address is found a hexadecimal value which indicates how many times code has been executed within that 16-byte area during the trace period. With this layout, you can quickly look up areas of interest in your program to see if a section The actual value of the program counter is not stored. The data
is being executed, and if so, how frequently.
buffer is arranged as a two-dimensional array and the data is used This program works in an atypical manner for a trace program on
as an index into this array. On program initialization, this array is
65xx series computers. Since a great deal of programs use or alter
set to zero. The correct array element is simply incremented each
the system IRQ interrupt, 1 found that I could not insert the trace
time an address is read from the stack. The counts are stored as 16-
vector into this interrupt and still get acceptable performance.
bit values, allowing counts up to $FFFF, after which an overflow
Another more serious difficulty was the frequency of the IRQ
counter is incremented. Also, a frequency counter is maintained to
interrupt. On the Commodore 64, the IRQ frequency is set at 60
show how many samples were taken. This is needed to interpret
Hertz (every 1/60 seconds). This is far too slow to trace an
the individual counts. The frequency counter is stored as 24 bits,
assembly language program. Also, many programs disable the
and the trace shuts off automatically when this counter overflows.
system interrupt with a SEI instruction, which would shut off the trace. To avoid these problems, the program uses the NMI (Non足
The Trace cannot be active during Input/Output operations due to
Maskable Interrupt) as its interrupt source.
timing conflicts. For this reason I have implemented two jump vectors to stop and start the trace without resetting any program
The NMis can occur at a user-selected rate, allowing you to
variables or the accumulated counts. A call to KILL will stop the
determine how many times per second the program makes a
trace and turn NMI interrupts off. RESET will resume execution
count. You can select frequencies from about 9800 per second
and preserve any data in the buffer. A call to CHART at any time
down to less than 50. The frequency is varied by placing a value in
will print the chart showing the current counts. Output can be
location 828 decimal. (A value of zero will default to one.) This
stopped by pressing the SHIFT and COMMODORE keys together.
value is used as the upper 8 bits of a timer which generates an NMI
It is important to call KILL before calling CHART to reset the
whenever it counts down to zero. The lower 8 bits is supplied by
interrupt vectors.
the variable CCYCLE in the source listing, which is normally set to
Using The Program
I 00. The 16-bit timer value thus formed indicates the number of cycles between interrupts - on the C64, a cycle is I /977778
Assuming the program is assembled at $8000, the entry points
seconds.
look like this: The trace is performed by recording the data from the program
trace
counter of the interrupted program. This is left on the stack as part
kill
of the interrupt sequence, and can be accessed by using the stack
chart
pointer as an index into the stack and using indexed addressing.
reset 52
=
=
=
=
32768 trace+ 3 trace+ 6 trace + 9 Nov. 1986: VOlume 7, ...... 03
To start the program once it's in memory, do a SYS TRACE. This will zero
IC
1000data
the
IF
1010 data 76, 39,129, 76, 29.128, 32, 38
all counts in the buffer and begin the trace. When you wish to
see
chart, do a SYS KILL: SYS CHART. KILL and RESET are used as explained
'
above.
The start and end addresses for which the trace takes place are defined in the source code by the variables CBEGIN and CFINISH. As listed, these are set to SAO and SFF respectively, to trace all code from $AOOO to $FFFF: the ROMs and the often-used $COOO-CFFF block of RAM. Change these and re-assemble the source to trace other areas. When assembling,
o. 128. 76, 12,128, 76, 97. 128
AM 1020data 130. 169, NA
0. 141, 122. 131, 141, 123
1030data 131,141,124,131,141,125,131,173
IF
1040data 24,
3. 141,114,131,169,123,141
HA
1050data 24,
3,173, 25,
LF
1060data169, 128,141, 25.
IK
1070data208,
LE
1080data
HD 1090data
5,169,
6,221,169, 0,141,
3,141,115,131 3, 173, 60,
1,141, 60, 0,141,
3
3,141
7,221,169
5,221,169,100,141,
4
KH
1100data221,120, 169, 65,141, 15,221.169
The mathematical expression used in the subroutine ZBUFF may cause
18
1110data
KP
1120data221, 88, 96,169,127,141, 13,221
some assemblers trouble - modify it or compute the values by hand if
EE
1130data169,
necessary for your assembler.
LO
1140data173,114,131,141, 24,
note that this source was assembled using the Commodore assembler.
1,141, 14,221,169,130,141, 13 0,141, 15,221,141, 14,221 3,173,115
00 1150data131,141, 25,
3, 96, 72,138, 72
All output is in the form of hexadecimal numbers. Leading zeros are
FK
1160 data152, 72, 169,
2, 44, 13, 221, 240
suppressed. Zero values are indicated by blank spaces on the chart.
FN
1170 data
JC
1180data127,141, 13,221, 76, 76,254,169
Hexadecimal representation was chosen for the counts to fit more information on an eighty column printout.
3, 76, 149, 128, 32, 97. 128, 169
FM 1190data127,141, 13,221,169,
0,141, 14
AO
1200data221,141, 15,221,186,189,
BN
1210data141,121,131,189,
FH
1220 data240,
OL
1230data176, 10, 56,233, 160,141,119,131
NN
1240 data 88, 32,200,128, 32, 51,128, 76
JP
1250data 129,234, 32, 29,130,169,
FA
1260data120,131,173,121,131, 41,240, 74
taken most often is checked for first. There are a great many such small
FO
1270data 74, 74,141,121,131,173,119,131
AN
1280data162,
changes that can result in a remarkable increase in execution speed.
LC
1290data249,101,251,133,251,165,252,109
Interpretation of the chart can be done in several different ways. If the objective of the trace is to improve program performance, the amount of time spent in each section of code can be analyzed to determine what portion of the code is using the most processor time; the most-used routines can then be optimized. If certain branches are never being taken, they can be removed. Branches can be arranged so that the one
2, 144,
6,
5.
1
1,201,160
16,201,255,240,
2
0,141
4, 10, 46,120, 131,202, 16
O C 1300data120,131,133,252,172,121.131,177
If your program is not working at all, a trace chart can show what sections
PI
of ocde are being executed. This is usually all you need to know in order
IH EA
to get it running. Once the problem has been located, the fix is usually quite simple. If the program is going into an endless loop, changing the trace program to perform a JSR CHART before it ends due to frequency counter overflow will let you see where the loop is taking place. If you are jumping to a location out of the program area for some reason, you may be able to
see
where the jump is going. This is sometimes enough
information to determine how the jump took place.
LG JP
1310data251, 24,105, 1320data251, 105, 1330data
1,145,251,200.177
0, 145,251,144, 15,169
0,109,125,131,141,125,131,169
1340data255,145,251,136,145,251,238,122 1350data131, 208, 13, 238, 123, 131, 208,
HG 1360data238,124,131,208,
8
3, 32, 97,128
BM 1370data 96,169,160,141,117,131, 32,250 KD
1380data 129, 162,
4, 32,201,255, 32, 71
BH
1390data 130, 32, 29, 130, 169, 13, 32, 210
OJ
1400data255,173,117,131,240, 77,201,255
GL
1410data240,
how the interpreter works. It will show the location of ROM routines. It
LJ
1420data 58, 32,210,255,169, 32, 32,210
BN
1430data255,160,255,200,192, 32, 240, 32
will also show which routines are using the most time and resources. IRQ
FL
1440data173,141,
routines, both user and KERNEL, are are also traced. All of this informa
MP
1450data 13, 32,210,255,144, 37,177,251
tion can be useful to a programmer.
LK
1460data141,116,131,208. 45,200,177,251
KO
1470data208, 43, 32, 57,130, 76, 89,129
FA
1480data238,117,131, 24,165, 251,105, 32
DN
1490data 133,251.165, 252. 105,
Running a trace on a BASIC program will show you a great deal about
If anyone would like to correspond with me regarding this program or any topic related to assembly language programming, feel free to contact me at this address: 7805 Villa Cliff *I l l Dallas, Texas75228
der for trace 100 rem •data loa ·
·
d a:cs= cs + a:next OP 110 'fori=1 to 883:rea
GK
120 if cs<>95794 then print·d ! ata error! · :stop
JD
130 remdata ok, write to file
lA
140 open 1,8,1. O:trace.com
OK
150 restore
FO
170 close 1 : end
•
53
96
4, 32,195,255, 32,204
1530data172,116,131. 32,182,129,169, 32
AC
1540data 32,210,255,172,113,131,208,163 3, 32.197,129,173, 61
3, 32, 220, 129, 76, 212, 129, 32
CG 1570data220. 129,201, 48, 208,
8,192, 48
JB
1580data208,
3, 32,210
AN
1590data255,152, 32,210,255, 96,170, 41
PG
1600data240, 74, 74, 74, 74, 32,240, 129
Nl
1610data 72,138, 41, 15, 32,240,129,168
NH
GO 160 for i=1 to883 : rea d a: print/11,chr $(a);: next
0, 133, 252
JG
GD 1620data104,
•
7,169
1520data255, 96,200.177,251,140,113,131
HO 1560data
•
3,208,
EB
DC 1550data140, 61.
Program Trace Monitor: Generates PRG object file on disk
2,201,
1500data 76, 58, 129, 32, 71,130, 32,
MO 1510data130,169,
Richard Stringer
PE
FE
2,176, 71, 32,197,129,169
1630data
7,160, 32,208,
96,201, 10, 144,
3. 24, 105
7,105, 48, 96,169,255,141, 16
DK
1640data130, 173, 24, 208, 41,
JF
1650data169,
2, 240,
7,141, 16,130,169,
5
4,162
Nov. 1986: Volume'· ...,. 03
I
10
1660 data
LE BK
1670data
32, 189,255, 32,192,255, 96, 169 1680data 126, 133,251, 169, 131, 133,252, 96
CF
1690data
KK
1700data 145, 251, 200,208, 251, 230, 252, 202
HM JL AA
BN
32, 29, 130,162, 12, 160,
96,132, 53,133, 54,160,
1760 data 246, 96,160,234,169,130, 32, 79
HA
1770 data 130,172, 60,
01
1780data 129, 160, 21,169,131, 32, 79,130
KB
1790data 160,214,169, 130,
HO
1800 data 100,169,
0,
BH
1820data 131, 32,
79, 130,173,124,131, 32
MC
1830data 220,129, 32.212,129.173,123,131 1840 data 32. 220, 129, 32.212,129,173,122
79, 130, 160,
1850 data 131, 32, 220, 129, 32,212,129,160
HH
21,169,131, 32, 79, 130, 160, 92
AM Kl
cc NO JO
1860data
1930data
32,
lA
1940data
FO LD MJ
PM JB BC PC NO
1950data 1960data 1970data
32, 91,
2030data
48, 32, 48,
LF
2050data
48, 32, 32,
MM
2060data 2070data 2080data
32,
32.
57, 48,
86,
69.
82. 70, 76,
MC
2090 data
32.
32,
32.
cc
32,
BL
2100 data
91'
32,
0
cd ce. cf: dO: . dl: d2: d3: d4: d5: d6. d7 . d8: d9: da. db; de . dd de . . df ao. . e 1 : Ia e2: a3: e4: e5: 01 e6: e7:Ob e8: e9.6e ea: a2 eb .
32, 32
. .
110 load·trace.com· ,8,1
NK
130use of the trace program from basic
DO EG
170 trace= 32768: kill =trace+ 3: chart=kill+ 3
FM OJ
GF IJ
180 poke 828,10: 190 sys trace:
KA
260 sys chart
KC
270return
1he 'INaiiGCD
.
.
. .
.
. .
.
.
.
.
.
.
.
.
.
.
. .
. . . .
.
.
.
.
.
.
. .
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
01 Oa 05 15 05 Oa 52
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Ob 02 . . 04 : 01 31 28 . . 3639 Ob Oc 01 41 . 05 . .
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.. . ..
. .
.
.
.
.
.
.
.
..
. .
.
.
.
.
.
.
.. .. . .
.
. .
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
. .
. .
.
. .
. .
.
.
.
..
.
.
.
.
.
.
.
..
.
. .
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
. .
.
.
. .
.
..
.
. .
.
.
. .
. .
.
.
.
.
.
. .
. .
.
.
.
.
. .
..
.
. . .
.
.
.
.
.
.
.
.
.
.
.
. 02 . . 3b 30 . . 50 42 39 a3 :06 . . .
.
. .
.
. .
.
.
.
.
.
.
. .
.
.
.
.
.
.
. . .
.
.
.. .
. .
.. . . . .
.
.
. .
.
.
.
.
.
.
. .
.
.
.
.
.
. .
.
. .
.
.
.
.
.
.
. .
. . . . . .
. .
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
..
..
.
.
.
.
..
. .
. .
. . . . . .
.
.
. . . . . . . Ob . 0 6 03 02 . .
•
•
.
.
.
. .
05 17 oa 07 . 25 . . . .
..
. .
.
.
.
. . . . . . .
. . . . . . .
.
. . .
. .
.
. .
.
.
.
02 le 14 02
.
07
. .
37
.
. .
. . . . . . . . ; 01 67 .
. .
.
.
.
.
.
. 01 03 0 2 oa . . . . : 01 . . . :01 04 . : : 01 16 17 . : 01 02 05 07 01 Of oa Ia 02 OS :01 . . . :03 01 . .. . . . . . .. .. . . . . . . . . . . . . .
.
.
.
.
.
.
. . . . . . . . . . . . . . . . . . . . . . .. .
.. . .
.
.
.
.
. . . . . . . . . .
.
.
.
. .
.
.
.
. .
.
. .
.
.. . . . . . .
•
.
.
.
.
.
. .
. . .
.
.
.
. . . .
.
.
. .
. . .
. .
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
:Oc . . . . . . 1 2 09 . 02 . 02 02
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. ..
.
10 . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
ed.
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
. 01
.
.
.
.
.
.
. . . .
.
.
.
. oa . .
.
. .
.
. .
. .
..
. .
. .
. . .
. . . . . . .
.
.
.
.
.
.
.
.
.
. . .
.
.
.
.
41 . . . . 49 06 : 05 ld . 36 136d 89 98 03 06 03 .. . . . . . . . . . . . . .. . . .. . 28 14 . . . . . . . . . 01 . . . .
.
.
.
.
: . 02
.
.
.
. .
.
. .
•
.
.
.
. .
.
.
. .
.
. .
ef . . . . 10 . . . . . II . . . . . . 12 . . . . . . . 13: . . . . . . . 14 . . . IS : . . . . . 16 . . . . . 07 Of 06 Oc OS 05 02 17 . .. . . . . . . . . 18 ; . . . . . . . . . . 19 . . . . . fa : . . . . . . . . .. fb : . . . fc : . . . . . . . . . . fd : . . . . . . . fe : . . . . . ff . . . 10 08 Of 08 70 90 00 10 20 30 40 50 60 80 ao bO dldO eO 10 . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
220 print peek (53281): print d ,$ f
250 sys kill
.
.
.
210 poke 53281,peek(53281)and 15
PB
.
.
.
.
.
.
..
.
.
.
.
.
200 fort= 1to 110: d =sin(t): f$=str$(int(d ) + 1)
240next: end
.
.
.
.
.
.
rem start trace
CK
.
.
.
.
.
rem set frequency
230 ift= 100then gosub 250
.
.
. . . . : 01 . . .
.
.
.
rem protect program from basic
NM
.
. .
.
.
150an assembly language program
co
ec.
..
. .
. . ..
ee:
140the same technique is used totrace 160 poke 56,128:
LH
I
100 if peek(32768) + peek(32769)= 88 thengoto 160 120this program will demonstrate the
HE
.
.
. 02 lb 09 Ob . :03 02 04 . .03 oa 01 02 04 02 : 01 03 :01 Od 07 04 03 19 2d .. . . 01 04 03 01 01 Od 16 lb 28 02 13 : Oa 08 12 Oa Ob 2e 14 03 Oc 0 5 . : 01 : 02 :06 06 :08 : 02 06 20 Oc 03 0 1 06 : 01 Ob 08 : 02 02 . . . . . . . . . : 07 01 . . : 06 le 09 Oc 04 04 . . . . . . . . . : 01 0605 Od 03 07 . . . 01 . :06 06 2 1 5e 5b 34 48 12 20 55 30 2d 05 05 03 71 61 celeO . 16 18 3b 187 le9 2ce 38 32 21 Oc09 03 38 52 23 lc 03 19 10 10 Ob oa 2f 3a 2 4 3b 03 . Ia Oc Od 06 02 . :08 : 1 2 4a 3c . : 01 63 Ill ce 33 IC 02 07 04 01 2b 27 02 . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . ; . . . . . . . . . . . . . .. .
.
.
0, 13, 79 79, 87, 83
NE
.
.
. .
.
Program Trace Monitor: Demo Program (note line 110)
ON
. . . .
.
.
NN
.
.
•
.
32, 32, 65
32, 32. 32,
.
.
.
.
66, 48, 32, 32, 67 48, 32, 32, 68. 48, 32. 32. 69
48, 32, 32, 70, 48,
.
.
.
32, 55, 48, 32, 32, 56
2040data
OF
32, 91'
.
.
32
48, 32, 32, 53, 48, 32, 32, 54
32,
.
.
76
2020 data
2000 data
32,
.
.
.
32
2010data
1990data
32,
.
.
32
0, 32 93, 0, 32, 32, 32, 32, 32, 48 48, 32, 32, 49, 48, 32, 32, 50 48, 32, 32, 51, 48, 32, 32, 52
1980data
.
.
o. 13, 13, 13, 67
69, 83, 32, 32. 32, 32, 32. 40, 66, 41, 32, o. 13, 83, 65, 77, 80, 91' 69, 32, 67, 79, 85, 78, 84.
.
.
.
40, 65, 41
89, 67, 76,
ME JG
.
1900data 210, 255, 96, 32, 32, 32, 58, 0 1910data 13, 67, 76, 79, 67, 75, 32, 67
HG KL
. .
1890data 169, 131, 32, 79,130, 169, 13, 32
89, 67, 76, 69, 32,
.
.
.
32, 220, 129, 32,212, 129,160, 21
1920data
.
.
.
1870data 169, 131, 32, 79, 130, 173, 125, 131 1880 data
.
.
0, 169
GE
..
.
32, 182,129,160, 21
32.
..
.
.
32, 79, 130,160
1810 data 169,131,
.
.
.
0, 32, 182
co
.
. .
.
6, 32,210,255,200,208 3,169,
.
.
0,177
CM
53,240,
.
. . .
.
1720data 169, 130, 32, 79,130,172,113,131 1730data 96,160, 24,169,131, 32, 79,130 1740data
.
. . . . . . .
.
.
.
.
. .
1710data 208, 246, 96,140,113,131,160,209
1750 data
LF
80: . . al : . . a2: . .. . a3: . a4. II . . as . . . e6: . . a7. :oa e8. 0 7 a9: 02 03 OS aa: : 01 ab: :09 18 ac: .. ad: . : ae: 07 01 09 af . 13 07 07 b0. 03 03 01 b l: 24 23 b2: b3: b4: .. . . b5. . 01 . b6. . . b7 : b8. Of 06 02 b9: 72 7e 167 . Od be: bb:03 : 10 be: 67 239 IS bd: IS 03 be:03 14 Od bl:04 01 dl: . .. cl : . c2: .. c3: . C4: . . c5. . c6 .. c7 : . c8 . c9 . .. . ca . cb .
0, 152
KO
KO
Program Trace Monitor: Sample Ou�ut oo 10 20 30 4 0 50 60 10 eo 90 ao bO dl ao . . . .
0
4, 160,255, 32, 186,255,169,
.
.
.
.
.
.
.
. .
.
.
.
(b) loa 1 cycle$ clock cycle (a) l 64 I sample count l 0038cb I overflows
M
l oo I
Nov. '1916: ¥alllme 7,111LJe 03
I
Procrua Trace Mollltor: PAL Souru Code EG OM KA
MH
BG AJ A H
KM KJ
OK GM FN GN KA
IM
LG GP
lA LG HG
GH IM
AH CD NG IG El KL GE
BK EC
MA
OJ IK lA
KN KB
GE
GE BC
BH
FJ
BK
OF OP
MA
GC LH
ON PE
CK JP
AB HA
NC AC
HF JE
JH
cc
OE AM JF IH KF
1000 open 1,8,1, O:trace.com 1010 sys700 1020.opto1 1030; 1040;program nam&-tracelnmi 1050· 1 060;purpose -to trace and rflCOfd 1070 ;the data from the <pc> 111 order 1080 ;to analyze the perlormance ol 1090;a program 1100; 1110 :lasl update -12126185 1120 . 1130 ;authof- r.w.stringer 1140: 7805 via cliff 1111 1150; dalas texas 75228 (214)-327-3039 1160: 1170; 1180 ;special note-Vo canno1 operate 1190 ;with nmi interrupts lunctl()(ling 1200 ;!he unset routine (kil ;np 1210 ;vector ) should be called befOI'e 1220;any inpuVoutput 1230; -buffer si ze in the 1240 ;present configuration is lim�ed 1250;to 3 kbytes.this is room lor a 1260 ;trace of 24kbytes of address 1270;space. to trace more space the 1280;program will need to move the 1290;buffer or remove basic from !he 1300;current address space.no basic 1310;routines are used 1320: 1330.' 1340; system equates 1350 zpg $1b 1360temp 53 1370 skey 653 shfVctrVcmdre flag 1380cbegin ;trace start address SaO 1390dinsh $ff trace end address 1400 :above set range for uace 1410ccycle 100 .cycle count tiner a 1420 mli 828 :tiner b count :nm1 vectOf 1430vect04' 792 %01111111 ;111terrupt mask 1440 wqoll 1450 bmask %00010 000 :liner b mask 1460talo Sdd04 :tiner a latch low . 1470tahl hi $dd05 1480 tactrl ;trner a control reg $ddOe 1490tblo $dd06 ;trner b latch low . 1500 tbhi hi $cld07 1510tbctrl ;tmer b control reg $ddOf ;interrupt control reg 1520 ire $cldOd 1530 ; kernel equates 1540 setHs $ffba 1550setnam $ffbd 1560open $ffc0 1570 chrout $ffd2 •
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
'
•
•
•
•
1580 close 1590 ctrchn 1600chkout
•
•
$ffc3 $Hcc
$ffc9
1610; 1620 ;the difference between cbegin 1630 ;and cfinsh can be no larger than 1640 ;$5f.the buffer at !he present OF 1650 :location is only 3k IOftQ OJ 1660 :in Ol'der to trace the complete GO 1670 ;64k address space you will need MA 1680:to allocate an 8k buffer. the L H 1690 ;chart wil be 2561ines long the GF 1700 ;buffer space fOI'mula IS IP 1710 :buffer (#kbytes to trace)/8 oc 1720; 10 1730. $8000 ,program org111 CE 1740: OE 1750: JllfTlPvectOI's MJ 1760 trace :start ;np Setirq KA 1770kl ,np unset .Slop NM 1780char1 ,np pntcht ,printout CL 1790reset ;restart ,np settwo OH 1800: KK 1810 ;set up vectOfS zero variables MD 1820;set up and start tiners BP 1830 ;liner b counts underflows GP 1840 ;from timer a and counts down LB 1850;the number of underflows in •
•
I
lheTta.:w::U
MN
GB
El IN JP BG AI ME KF
FF JO oc LP HP KK IB PB
GA
BG
AH
JP IP LM FB
OK
HN AM
PO IP pp LA
LP
LB NA BD
OP KD CK
GO
oc ML pp
OF GO IF BP BJ
JJ
CG PF 10
KF
1860 ;milli.on timer b underflow a 1870;nmi is generat ed.vector points 1880;to tstprg 1890; 1900 setirq zbuff jsr 1910 Ida 10 freq 1920 sta sta freq + 1 1930 1940 S1a freq+2 1950 S1a vflag 1960 settwo Ida vector 1970 S1a oldvec 1980 Ida l<tstprg 1990 S1a vector 2000 Ida vector + 1 2010 S1a oldvec+ 1 Ida l>tslprg 2020 2030 S1a vector+ 1 2040settmr Ida milli bne sttinb 2050 2060 Ida II 2070 S1a mili 2080sttimb S1a tblo 2090 Ida 10 2100 S1a tbhi 2110 Ida N>ccycle 2120 Sta tahi 2130 Ida #<ccycle sta talo 2140 2150 sei 2160 Ida N%01000001 2170 S1a tbctrl Ida N%0001 2180 00 2190 sta tactrl 0 0 Ida N%100001 2200 2210 S1a ire 2220 eli 2230 rts 2240; 2250 :reset vectOfS to normal and 2260;turn interrupts off.needed fOf 2270 :Vo operations 2280· 2290unset Ida lirqolf 2300 S1a ire 2310 Ida 10 2320 S1a tbctrl 2330 S1a tactrl Ida oldvec 2340 2350 S1a vectOf 2360 Ida oldvec+ t S1a vector+I 2370 •
10 2380 rts MM 2390: FA 2400 ;stop interrupts.pull address OE 2410 ;of next program step from stack ON 2420 ;check �in range.subtract offset G l 2430 ;from hibyte and store.store El 2440 ;lobyte and jsr to array routine KG 2450 ;on return turn interrupts on KJ 2460 ;reset timers and exit. MB 2470; E H 2480 tstprg pha OG 2490 txa pha AE 2500 Fl 2510 tya pha EF 2520 FM 2530 Ida Nbmask BF 2540 bit ire HJ 2550 beq ouljmp LO 2560 jmp swdl 25 70 KP out,inp jsr unset IB 2580 Ida #irqoff K H 2590 S1a ire F H 2600 ; np $fe4c Ida lirqotf LE 2610 swdl IJ 2620 S1a ire B D 2630 Ida 10 S1a tacUI PM 2640 LN 2650 S1a tbctrl
FG CE
2660
2670 A B 2680 HF 2690 P H 2700 JJ 2710 00 2720 IP 2730 tryhi
!SX
Ida S1a Ida cmp beq
$0105,x loby1e $0106.x #cbeg1n tryhi bee bugout cmp #c1insh
55
OA ME
AG
AJ EG MF
KO ON HG El
GK JM NP
DH
2740 beq issame 2750 bcs bugout 2760 issame sec 2770 sbc #cbegin . sta hiby1e 2780 2790; 2800 ;we need to allow syslem irq 2810;requests here because we are 2820 ;stealing so much cpu time 2830; 2840 ell 2850 ;index array jsr mai'lpg 2860 bugout jsr settmr 2870 jnp $ea81
GL 2880; GO 2890 :calculate array index and OF 2900 ;increment the proper array var GM 2910 :array IS- array\0. . .xx.O. ..15) AN 2920;ol double integer [0. . .65535) HG 2930 :array index is calculated as NH 2940: (hiby1e-lcbegin)•32 + LJ 2950; (lobyte/16)•2 + co 2960: buffer start address GG 2970 :overflow is checked for and GG 2980 :sample count incremented.sample KG 2990;count is maintained as a 24 bit NB 3000;integer (2t24)-1.trace is NA 3010 ;terminated on sample overflow NH 3020 ;a flag is set on var overflow KF 3030;var's are kept in lolhigh format GF 3040; JE 3050mainpg jsr setbul PN 3060 Ida #0 IE 3070 sta hiby1e+ 1 Ida lobyte CG 3080 PA 3090 and 1$10 a GO 3100 lsr AE 3110 a lsr KE 3120 a lsr CN 3130 S1a lobyte Ida hibyte 01 3140 NJ 3150 ldx 14 FJ 3160miA32 asl a HM 3170 rol hibyte+ 1 dex oc 3180 DP 3190 bpi mul32 K adc 3200 J zpg pp 3210 S1a zpg E K 3220 Ida Zpg+ 1 M J 3230 adc hibyte+ I GP 3240 S1a Zpg+ 1 M G 3250 ldy lobyte Ida (zpg),y IJ 3260 MD 3270 etc PK 3280 adc #1 OF 3290 ;indexed sta (zpg),y CJ 3300 iny :element KM 3310 Ida (zpg),y FN 3320 ade NO MB 3330 sta (zpg),y NL 3340 bee tesll BA 3350 Ida NO Ll 3360 ade vtlag sta vflag BO 3370 BP 3380 Ida N$ff 3390 IF sta (zpg),y DA 3400 dey MG 3410 sta (zpg),y inc freq 01 3420 test1 BE 3430 bne test2 EB 3440 inc freq + 1 FF 3450 bne test2 M C 3460 II1C freq+2 JG 3470 bne test2 PL 3480 /Sf unset GL 3490test2 rts cc 3500, 35 I 0 ;pr111t results in chart fOI'm 81 El 3520 ;With PfOper fOI'matlllQ.Ieading EM 3530 .zeros are suppressed on two 80 3540 :by1e hexadecinal numbers EF 3550; OF 3560, JB 3570 pntcht Ida lcbeg1n OL 3580 sta fre2 p2scr GO 3590 jsr P F 3600 ldx #4 MN 3610 jsr chkout
Nov. 1916: VDIIIme 7, .. .-.03
I
AP 3620 FM 3630 CF
jSf
3640 pnch1
JB 3650 A N 3660 FC IH
3670
3680 BA 3690 FE 3700 N C 3710chks
Nl 3720 J G 3730 DC 3740 N H 3750 CH 3760 FH 3770 pnloop NL 3780 MG 3790 M F 3800 I B 3810 DE 3820 MH 3830 HN 3840 LJ 3850 OM 3860pnch AN 3870 GB 3880 CB 3890 IB 3900 IC 3910
MM AB CA EO PK
LP BA GK NH IP FH
FA HK FM BF
NO MN PB AO
JO
MO
LM DL NA Jl
HM KM PG ON KF OE FD IH HD
KB HL oc GM
IP KD JD
OH LN CJ MO cc
DP CB FC HD
OF IG CH MH NL
CP
ED NJ
I
3920 3930
3940 pnch2 3950
3960
jst
Ida
jst
Ida
phead setbuf 113 chrout tre2
beq pfini
cmp lcfiOSh beq chks bcs
!Sf Ida jsr Ida
jsr
pfini pnthex ... � chrout 1 32 chrout #255
ldy n i y cpy 132 beq pnch2 Ida skey cmp 13
bne pnch
Ida !Sf
113 chrout
bee pfllll Ida sta
(zpg).y
lent
bne bumpy
ony Ida bne jSf jmp inc clc Ida
3980 3990 4000
11w Trai-tor
ho2
jSf lay pia rts
cmp 110
bee nixadd de
adc 17 4580nixadd adc •·o·
01 KN 4590 rts OG 4600; LK 4610 ;open p rn i ter as device #4 Cl
JJ JF BN OM EA BE Ll
HM FL PO
EE PG
PE MA 01
4620; Ida #255 4630 p2scr sta p index+1 4640 Ida 53272 4650 and W2 4660 4670 beq ucase Ida 17 4680 sta pindex+1 4690 4700 ucase Ida 14 ldx 14 4710 4720 pindex ldy 1255 4730 setlfs jsr Ida 10 4740 4750 setnam JSf 4760 open jSf rts 4770 4780; 4790;reset buffer to start 4800; Ida f<trbuf 4810 setbuf zpg sta 4820 Ida l>trbuf 4830 z.pg+ 1 Sla 4840 rts 4850
cc CK GO AN JE NM
zpg
GD ON CH 4860; JE 4870;zerobuffer area Gl 4880; setbut JL 4890zbuff jsr ldx l(cfinsh + t-cbegin)>3 HJ 4900 JH 4910 ldy 110 PO 4920 tya MP 4930zloop1 sta (zpg),y M C 4940 iny FK 4950 bne zloop1 AJ 4960 inc zpg+ 1 dex BC 4970 OM 4980 bne zloop1 KG 4990 rts
4010 Sla zpg+1 4020 ,-np pnch1 JSr 4030 pfono phead 4040 JSr pntsta Ida #4 4050 JSr 4060 close jsr 4070 clrchn 4080 rts 4090bumpy iny Ida (zpg),y 4100 4110 pntn sty ysave ldy lent 4120 4130 h ex16 !Sf 4140 Ida 132 4150 chrout !Sf 4160 ldy ysave 4170 bne pnloop 4180; 4190;output hexnumber <two bytes> 4200; 42t0hext6 sty 829 4220 JSr pnthex 4230 Ida 829 4240 jsr hex out 4250 Jmp nozera 4260. 4270 ;output hex number <one byte> 4280· hexoul 4290 pnth e x !Sf 4300 cmp 1'0' 4310 bne nozera c p y ro· 4320 4330 bne nozery 4340 ldy 1132 4350 bne nozery chrout 4360 nozera jsr 4370 nozery tya 4380 chrout js• 4390 rts 4400 hexout tax 4410 and nto 4420 a ISf a 4430 ISl 4440 ISf a 4450 lsr a 4460 jSr ho2 4470 pha 4480 txa and nat 4490 •
4500 NG 4510 A D 4520 OJ 4530 L O 4540 ho2 HN 4550 GE 4560 FM 4570
(zpg ),y pntit pnt3 sp pnloop fre2
adc 132 Sla zpg Ida zpg+1 adc 10
3970
FO
OP 5000;
GG 5010;pr i n t 3 spaces and a colon CB 5020; GP 5030 p nl3sp Sly ysave AN 5040 ldy l<space Ida t>space GH 5050 jsr pstrng PK 5060 HB 5070 ldy ysave EM 5080 rts IF 5090·• JJ 5100 ;prin t top of chart MG 5110; FD 5120 phead ldy N<messg AM 5130 Ida N>messg pp 5140 p strng jSf KA 5150 rts OJ 5160; GB 5170 ;print string a > y <address CL 5180; BE 5190pstrng Sly temp sta temp+1 HA 5200 FK 5210 ldy 10 PM 5220 pstrg1 Ida (temp),y beq p strg2 KO 5230 chrout PE 5240 jsr iny CG 5250 bne pstrg1 BP 5260 HD 5270 p strg2 rts GB 5280; HB 5290 ;print sample slats KC 5300; IH 5310 pntsta ldy l<messg2 Ida l>messg2 GE 5320 NL 5330 jsr pstrng •
AO 5340 BN 5350 JJ 5360
10
5370
•
ldy
milli
jsr ldy
hext6
Ida
16
5380 JP 5390 O P 5400 EK 5410 HB 5420 0 0 5430 JN 5440 OP 5450 CE 5460 10 5470 OF 5480 M F 5490 CA 5500 BH 5510 IB 5520 Cl 5530 LF 5540 CD 5550 AK 5560 JH 5570 01 5580 OL 5590
Ida jsr
01
HJ IN
OH JO OP EK HA PJ
IB BP CD
IN
DE 10
DE
CG GP
LM KA OK
MJ IC
00 PE
EM AF FJ CM
MO
IH
BJ
OJ
EB AK KM co OL OK NG
Ida
t>messg5 pstrng f<messg3 l>messg3
jsr
pstrng
Ida
I>rope hex16 l<messg5 l>messg5 pstrng f<messg4
ldy
ldy I<OOfde jsr
ldy Ida
jsr ldy Ida jsr Ida jsr JSr Ida
jSf
jsr Ida jsr
5600
jsr
f>messg4
pstrng freq+2 hexout nozera f r e q+ 1 hex out
nozera freq
hexou t
nozera l<messg5 l>messg5 pstrng l<messg7 f>messg7 pstrng vtlag hexout nozera f<messg5 t>messg5 pstrng 113 chrout
5610 ldy Ida 5620 5630 jsr ldy 5640 Ida 5650 jsr 5660 Ida 5670 5680 jsr jsr 5690 5700 ldy Ida 5710 jsr 5720 Ida 5730 5740 jsr 5750 rts 5760; 5770 ;pro gram dala and message tables 5780; 5790space .asc .• 5600 .byteO 5810; 5820 messg3 . byte 13 5830 asc ·clock cycle (a) [" 5840.byte 0 5850; 5860 messg2 .byte13.13,13 5870 .asc ·cycles (bl 1· 5880 .byte0 5890; 5900 messg4 .byte13 591o .asc • sample count I . 5920.byte0 .
•
.
5930;
5940 messg5 .asc • I· 5950.byte0 5960; • 5970 messg 5980.asc • 00 10 20' EK 5990 .asc 30 40 50 60 DO 6000 asc • 70 80 90 ao • KB 6010.asc • bO oO dO eO to' IH 6020 .b y t e0 MB 6030 messg7 .byt e13 GH 6040.asc overflows I . GJ 6050.byte0 -
•
•
.
•
LK 6060; AB 6070ysave PM 6080oldvec LK 60901cnt OK 6100 fre2 FP 6110 hibyte BD JA
61201obyte 6130 freq LJ 6140 vtlag OF 6150 trbuf MO 6160.end
dala
.byteO .byte0,0 .byte0 .byte0,0 .byteo.o .byte0 .byte0,0,0 .byte o •
•
10
l<messg5
No¥. 1916: 'llbM'ne 7,1nn• 01
I
The8563 Video Display Controller
David Stidolph Madison, Wisconsin
The 80 column screen of the Cl28 is driven by a chip called the 8563 Video Display Controller, or VDC for short. The VDC is extremely flexible, and almost every aspect of the display (characters per line, lines per sc.reen, height and width of characters, etc.) can be changed to suit your preferences. Before describing how t o use these features I will explain how to access the VDC.
Accessing The VDC The VDC contains 37 registers (numbered 0 t o 36) which control how it displays information contained in its own bank of 16K RAM (this memory is not accessible directly by the CPU- only the VDC can get at it). Access to these registers is restricted to two memory locations - Sd600 and Sd60 I. The register number you wish t o read or write is placed in Sd600 and Sd60 I becomes "connected" with the register pointed to by Sd600. This complicated addressing scheme is made a little more difficult by the possibility that the VDC is not ready to connect the register you want immediately following a write t o Sd600. This is handled by making Sd600 dual purpose. Writing to Sd600 specifies the register number. Reading Sd600 returns status information. Below is a description of the bits in Sd600:
Location $d600 (VDC access registers) d7
VDC Registers Reg
Normal
0
%01111110
Horizontal Total
I
%01010000
Horizontal Displayed
2
%01100111
Horizontal Sync Position
3
%0100xxxx
Vertical Sync Width
%xxxxl001
Horizontal Sync Width
4
%00100110
Vertical Total
5
%xxxx0000
Vertical Total Adjust
6
%01010000
Vertical Displayed
7
%00100000
Vertical Sync Position
8
%xxxxxx0x
Video Mode
%xxxxxxx0
Interlace-Sync Mode
9
%xxx00111
Total Rasters/Character
10
%x01xxxxx
Cursor Mode
%xxx00000
Cursor Start: Raster Line
II
%xxx00111
Cursor End: Raster Line
12
%00000000
Display Start (HI byte)
13
%00000000
Display Start (LO byte)
14
%00000000
Cursor Address (HI)
Description
d5
d4
d3
d2
dl
dO
15
%00000000
Cursor Address (LO)
r5
r4
r3
r2
rl
rO
16
%00000000
Light Pen Vertical
verO
veri
ver2
17
%00000000
Light Pen Horizontal
18
%00000000
Address Pointer (HI)
When you read Sd600, the highest bit is set if the information at Sd60 I is valid (for a
19
%00000000
Address Pointer (LO)
read or write). Bit 6 is cleared when the light pen signal has been sensed. Pit 5 is set
20
%00000100
Attribute Start (HI)
during vertical retrace. The lowest 3 bits return the version number of the VDC. An
21
%00000000
Attribute Start (LO)
easy test to determine a C 128, in 64 mode, from a normal C64 is to PEEK this location.
22
%0111xxxx
Character Total Width
%xxxx1000
Character Displayed Width
23
%xxxx0111
Char Displayed, Ver:ical
24
%0xxxxxxx
Block Copy Mode
will be shown in 8502 assembly code. Furthermore, the 80 column chip is just as
%x0xxxxxx
Reverse Entire Screen
accessible from 64 mode, so all examples will work unchanged in either mode.
%xxl xxxxx
Character Blink Rate
%xxx00000
Vertical Smovth Scroll
%0xxxxxxx
Bit-Map Graphics Mode
register 'X' should have the VDC register number, and the accumulator is used to pass
%xl xxxxxx
Enable Attributes
the information. Register 'Y' is not affected.
%xx0xxxxx
Semi-Graphic Mode
%xxx0xxxx
Double Pixel Width
%xxxx0000
Horizontal Smooth Scroll
%llll xxxx
Foreground Color (RGBI)
%xxxx0000
Background Color (RGBI)
d6
Write Read
status
lp
vblank
A C64 will ALWAYS return a zero. Programmi11g the VDC seens impossible from BASIC (it is too slow), so all examples
The following are two routines for reading and writing the VDC registers. When called,
25
READ
STX SD600
;WRITE REG" TO VDC
WAITOI
BIT $0600
;WAIT FOR BIT 7 OF
BPL WAITOI
; THE VDC TO BE SET
LOA $0601
;LOAD REGISTER DATA
27
%00000000
Address Increment Per Row
RTS
; AND RETURN
28
%001xxxxx
Character Set Start
%xxx0xxxx
8563 DRAM Type
26
WRITE
STX $0600
;WRITE REG111 TO VDC
29
%xxx00111
Underline Raster Line
WAIT02
BIT $0600
;WAIT FOR BIT 7 OF
30
%00000000
Word Count
BPL WAIT02
; THE VDC TO BE SET
31
%00000000
Data Byte
STA SD601
;PUT I N REGISTER
32
%000000
Block Copy Source (HI)
RTS
;AND RETURN
33
%000000
Block Copy Source (LO)
34
%01111101
Display Enable Begin
The following is a list of the VDC registers and their functions. Many of the registers
35
%01100110
Display Enable End
provide multiple functions, so each function is given its own line. The default (upon
36
%xxxx0101
DRAM Refresh Rate
power up) is given for each setting. An 'x' means that the bit position it is in is not used for that function. For example, register 3 has two functions. One uses the upper four
Before we look at individual registers and functions,
bits to control the vertical sync width, and the other function uses the lower four bits to
we must look at how the VDC uses its 16K memory
control the horizontal sync width.
bank, and how we can access it. 57
Nov. 19t6: '\lloUne 7, Ill ue 03
VDCMemory
address pointers are incremented. This means that you should use one less than the number of characters you want to fill. The
The VDC has two 16K by 4-bit dynamic rams attached to it. These
following example clears the entire VDC memory.
rams can only be addressed by the VDC, so any access to them MUST be through it. You first place the address you want to read or
CLEAR
LOX '18
;SET ADDRESS
LOA '00
; POINTER TO START
or write the data byte register ('31 ). For example, to load VDC
JSR WRITE
; OF MEMORY
memory location $1Ode with the value $20, the following code
INX
could be used:
JSR WRITE
write into the address pointers (registers 18 and
19). Then you read
LOX '24 SET20
LOX '18 LOA 1$10
;CLEAR COPY BIT
JSR READ ;HI BYTE
AND ...$7F
JSR WRITE
JSR WRITE
INX
;NOW POINT TO
LOA 'SOC
;LO BYTE
19
JSR WRITE
LOX '31
;LOAD DATA BYTE
JSR WRITE
; WITH 0 (CLEAR)
LOX *30
; AND CLEAR REST ; OF FIRST PAGE
LOX '31
;DATA REGISTER
LOA 1255
LOA ...$20
;BYTE TO WRITE
JSR WRITE
JSR WRITE
LOY '47
;47 OTHER 256 BYTE
RTS
LOA *00
; PAGES TO CLEAR
JSR WRITE
;CLEAR 256 BYTES
Due to an undocumented "bug" in the VDC, you should always
DEY
;COUNTER ZERO?
load the address pointer high byte first. If you load the high byte
BNE CLRPAG
;NO, CLEAR ANOTHER
last, you may get the wrong address.
RTS
;ROUTINE DONE!
CLRPAG
One nice feature of the address pointers is that after each byte is
Another feature of the VDC is the ability to COPY one area of VDC
read or written through register 31, the address pointer is incre足
memory to another area (up to 256 bytes at a time). This comes in
mented to point to the next memory position. The following code
handy for scrolling the screen.
could be used to print my name to the VDC memory (we will use the same location as the starting point):
To copy memory the BLOCK COPY bit (register 24, bit 7) must be set. You first set the address pointers (18 and
PRTNAM
LOX ... 18 LOA '$10
address and the block copy source registers (32 and 33) to the ;HI BYTE
source of the memory to copy from. Then you write the number of
JSR WRITE
bytes to copy in the word count register (30). As with fill, a zero
INX
;NOW POINT TO
LOA ...$DC
;LO BYTE
a byte in the data register (31 ), so you do not need to write one less than the number you want to word count register (30). The copy
LOX '31
;DATA REGISTER
process goes from low to hi, so be careful not to overlap your
LOY '00
;BEGINNING OF NAME
copying.
LOA NAME,Y
;GET CHARACTER
BEQ DONE
; END CHARACTER?
JSR WRITE
;PRINT CHARACTER
INY
;POINT TO NEXT
19
JSR WRITE
PRTCHR
19) to the destination
value indicates 256 bytes are to be copied, however you do not put
Setting Up The Display Most video chips are set to show a fixed number of characters per
JMP PRTCHAR
line and lines per screen. The VOC, however, is more flexible than
DONE
RTS
most display controllers. The number of screen lines is controlled
NAME
.BYT 'DAVID STIDOLPH',O
by the vertical displayed register (6). As you increase or decrease the number of lines, you will note that the top line stays fixed at the
Of course the above example ignores the fact that the normal
same position. As you increment the vertical sync position register
definition of characters does not follow their ASCII equivalents
(7), the screen will "move up", and as you decrease it the screen
(unless you to set them up that way). The code also does not update
will "move down". A good formula for setting the register is to add,
the attribute memory.
or subtract, one number for every other line added, or removed, from the normal 25. A more mathamatical way to represent this
FUI and Copy Mode
would be: R7
=
32 + (R6-25)/2
If you want to clear the display by filling it with spaces (or any other data), there is a special register to write the same byte to successive
The VDC allows you to design your display to be a "window" that
VDC memory locations. This register (30) should be written AF足 TER you have defined the starting address in the address pointer
moves over a larger screen. Figure I shows the relationship of the display to the theoretical screen. You could, for instance, have an
registers and written the data to the data byte register. You must
80 column display with 25 lines that moves about a 200 column
remember, however, that as soon as you write a byte to the data
screen with 40 lines. The program examples here will assume only
byte register, the memory location is set to that value and the
the standard 80 columns. Nov. 1916: Valume 7, ..... 03
�----- E
Vertical Smooth Scrolling
A
The type of scrolling shown above works, but is rather jerky. The
�------�
VDC, however, has the ability to smoothly scroll, pixel by pixel. Since the display can be flexible in the number of lines that can be displayed, we don't have to reduce the display in order to smooth scroll. In order to smoothly scroll the display "down" a line, just
c
DISPLAY
follow these steps: I. Put the text that will be scrolled onto the screen when it moves
down a line into VDC memory. If the text is already there, skip 1----- C
this step.
-------+-1
2. Increment the vertical smooth scroll register (24, bits 4-0) until it is equal to the number of raster lines per character.
B·�------�
3. Add the number of characters per line (register 22) to the display start registers (12 and 13) and set the vertical smooth scroll register back to zero. Here is an an example routine to scroll the screen down one line
AT TRIBUTES
smoothly. It is assumed that the text on the new line is already there. SMOOTH LOX '24 JSR
READ
;GET VERT SMOOTH ;SCROLL REGISTER
AND'%11100000 ;AND SET TONO SMTH I
JSR WRITE
;SCROLL
JSR
;READ VALUE
READ
CLC
A
=
Start of Display (R 12, 13)
ADC 101 JSR WR ITE
B
=
Start of Attributes (R20, 13)
C
=
Horizontal Displayed (R 1)
D
=
E
=
SMTH2
LOY 8100
Vertical Displayed (R6)
DEY
Total Number of Columns
BNE SMTH2
;AND ADD ONE ;DELAY VALUE
AND 8%00011111 ;CHECK SCROLL AGAINST
R27 (Address Increment per Row) should be set to
CMP8 7
the Total Number of Columns (E) mmus
;BOTTOM RASTER AND
BNE SMTHI
;CONTINUE UNTIL DONE
the Number of Displayed Columns.
JSR
READ
AND,.. Il l 000000 JSR
WRITE
You can "scroll" your entire display up or down by changing the
JSR
MOVEUP
display star! registers (12 and 13). Add 80 to them and the display
RTS
;GET REGISTER AGAIN ;AND CLEAR SCROLL ;NOW MOVE UP ONE LINE
would start one line down, so the screen would appear to "scroll" down. The following code shows how to move your display one "line" down (this presumes that the new line bellow the display
Horizontal Smooth Scrolling
already has data you want shown .) In a similar manner, the screen can be scrolled smoothly left and MOVEUP
LOX
11
!3
JSR
READ
;GET LO-BYTE
right. This is a little more complicated because you have to set up
;OF DISPLAY
the VDC to believe that there are more columns per line than it will
CLC
display. This is done through the address increment per row
ADC ,.80
;ADD 80 AND
register ( 27 .) The value of this register is added to the end of each
PHP
;SAVE CARRY
line to get to the beginning of the next line. For example, if you
JSR WRITE
;STORE NEW VALUE
wanted a "virtual" display line of I 00 characters, with an actual
DEX
;NOW THE HIGH BYTE
display line of 80, you would have to put the value 20 into register
•
JSR
READ
27. Simple movement to the left or right one column is then a
PLP
;RECALL CARRY
matter of incrementing or decrementing the display start registers
ADC ... 00
;ADD IN CARRY
(12and l3).
JSR WRITE
;STORE NEW IIIGH BYTE
RTS
;AND RETURN
59
Nov. 1986: Volume 7, ..... 03
Smooth scrolling the screen horizontally is similar in concept to vertical smooth scrolling, except that Commodore has come out with different versions of the VDC which require different initiali zation, and use, of the smooth scroll register. When you read Sd600 for the status bit, you also get other information which includes the version number of the VDC. If the lower three bits of the VDC are all zero then the smooth scroll off setting is zero. If any of the three bits are on, then the smooth scroll register should be set to seven to be off. The difference between these chips is that the first type is designed to scroll the text to the right, and the second is designed for scrolling to the left. Either way it is a shame Commo dore had to cause these incompatibility problems.
Intensity Off Red
Green Blue
Color
0
0
0
Black
0
0
I
Blue
0
1
0
Green
0
I
I
Cyan
I
0
0
Red
1
0
I
Purple
I
I
0
Brown
I
I
I
Light Grey
Intensity On Attributes
Like the VIC-II chip, two bytes of memory are used for each character on the display. One byte to determine what character from the character set is displayed, and the other for the color it will be shown in. Unlike the VIC-II chip, every bit in ·the attribute byte is needed because each character has extra attributes that can be attached to it. In simple terms, the lower four bits determine the color, and the upper four bits determine its attributes. Attribute memory is set up just like the display memory, and has a one-for
Color
Red
Green
Blue
0
0
0
Dark Grey
0
0
I
Light Blue
0
I
0
Light Green
0
I
I
Light Cyan
I
0
0
Light Red
I
0
I
Light Purple
I
I
0
Yellow
I
I
I
White
one correspondence with it. One thing to keep in mind, is that the normal Cl28 default settings
Attribute bytes are defined in the following table:
for the display has the attribute memory start right after the display memory. If you increase the display size, or scroll down by Bit 7
Alternate character set
6
Show character in reverse mode
5
Underline the character
4
Blink the character
3
Red
2
Green
I
Blue
0
changing the display pointers, you must move the attribute pointer
Attribute
to make room. I suggest you place attributes at $1000 to leave plenty of room (4096 bytes for display and attributes each). If you need more memory than that, you can turn off the attributes (the VDC will no longer look for them) by clearing the attribute enable control bit (register 25, bit 6). This would give you 8K for display area, but limit you to monocolor and using the first character set (as well as no underlining, blinking or reverse field).
Intensity Character Definitions
Bit 7 determines which character set is used for displaying the is used. This allows each character to select from either set,
The VDC requires 16 bytes of definition for each character (verses 8 for the VlC-ll) so that it can show as many, or as few, raster lines
allowing 512 possible characters.
per character that you want. By default (on Cl28 power up), the
character. If the bit is set the second (or upper) character definition
VDC is set to display characters 8 pixels wide, and 8 pixels hight in. Bit 6 is used to display the character in reverse mode. If a pixel in
This can be changed to make characters 5 pixels wide and 10 lines
the character definition is on, then it will be displayed of. If it is
tall. The only limits are that characters cannot be taller than 32
defined off, then it will be displayed on.
lines (the VDC will switch to 32 byte character definitions if you define your characters to more than 16 lines) or wider than 8
Bit 5 allows you to underline any character (or combination) on the
pixels. You can also define a normal "space" between lines or
screen. The raster line used for underlining is set by register 29. It
characters by use of "interspace". The following figure shows the
could be set to strike through (overline) by changing its value.
effect of interspace on display characters.
Bit 4 will make the charact�r blink. The rate character blinks is set by register 24, bit number 5. If this control bit is cleared, then any character displayed with the blink mode set will be blink fast. The lower nibble (bits 0-3) selects the color. The following tables shows the result of "mixing" the colors:
60
Nov. 1916: Wume 7,111LJe 03
insert mode in a word processor. The cursor color is set by the color of the character it is over. Interlace and Video Mode
CHARACTER DEFINITION (16 bytes)
l.
The VDC provides three methods of displaying the video informa tion to the RGB monitor. The normal method (by default) is called non-interlaced. This means that the screen is sent 60 times a second and provides a normal display. Another method (used by normal TV's) is to send the even numbered raster scans first, followed by the odd numbered raster, which effectively doubles the resolution of the picture, but cuts the screen "speed" in half. By turning the interlace mode on, the VDC displays the same number of raster lines and duplicates them for the even and odd screens, but sends the second screen one half scan line down, so the lines are better "filled in."
F H
I- J ---1
The last display mode is called the interlaced sync and video mode. This causes the VDC to display characters in up to double the resolution, if your monitor can handle it. Try turning on this mode and setting the number of raster lines per character (register 9) to a value greater than 8.
CHARACTER DISPLAYED G
Graphics The VDC offers a graphics mode that uses the VDC memory as a bit map. Instead of taking the display area and using it as pointers to the character set, the display area is used to display straight binary data. It takes as many bytes to display a raster line as you have columns in the display (each byte equals 8 pixels). In normal display mode (25 lines of 80 columns) this would result in a 640 by 200 display screen. This screen requires 16,000 bytes, so no memory is left for the attributes. In fact, several graphics programs have already been written that allow you to draw on this graphics screen in mono-color mode.
K
F G
=
Char acter Total Horizontal (R22 bits 7-4)
=
Character Total Vertical (R9 bits 4-0)
H
=
Character Displayed Horizontal (R22 bits 3-0)
I
=
Character Displayed Vertical (R23 bits 4-0)
J
=
K
=
Horizontal Interspace
V er tical I nterspace
The bit map mode, however, follows the same rules as the text display, so you can shrink the display size (freeing memory) and make use of attributes or smooth scrolling. In order to shrink the display, lower the value of the vertical displayed register (6), or the horizontal displayed register (1). By reducing the number of columns to 72, you could have a 576 by 200 graphic screen, and still have enough room for the attributes. The attributes still work only on a standard character size (8 by 8), so you would only have the equivilant foreground/background selection of the hi-res VIC-II chip, but with 80% more display area!
The default settings of the VDC set no interspace between charac ters. By designing higher density characters, interspace might help make the display more readable. Interspace could also be used to provide a blank area under the character for the underline. •
-The Cursor The VDC provides a hardware cursor. This means that you provide the cursor position (by memory address), turn the cursor on, and if it is within the display area, the cursor will be displayed. The cursor has 4 modes of display:
For those of you wanting to write printer "dump" programs of the bit-map screen. remember that the address pointers (registers 18 and 19) are always incremented after a read of the data register (31) so if you want straight raster data, you only have to specify the start of the bitmap which is stored in display start pointers (registers 12 and 13).
Register 10- Cursor Mode Value 0 I 2
3
,
Meaning Solid Cursor (no blinking) No cursor Blinking Cursor (fast) Blinking Cursor (normal)
There are other features of the VDC, but more time will be needed to uncover the proper settings of the other registers to use them properly. For now, alii can suggest is to experiment with different registers. The following program was written for the COMAL 2.0 cartridge that allows you to interactively change the different register settings and see the results. For those of you who get The Transactor Disks, there is a package that adds commands to COMAL 2.0 for output on the 80 column screen.
The cursor can also be set to any height from a block to an underline. You set the starting raster line in register I 0 (bits 4-0) and the ending raster line, plus one, in register I I (bits 4-0). This could be used to show different types of typing modes, like an 61
No¥. 1916: Volume 7, ..... 03
0900 FORi:•OT01 DO
001011delete 'vdc'edotor' 0020 II save • vdc'editor • 0030 II by Dallld Stidolph
0910 0920 0930
004011
0940
0050 USE system 0060 textcok)(s(6.6.1}
0950 0960
0070 DIM value(O 4 7} 0080 DIM name$(47} OF 20. reg(47} 0090 DIM Slat1(47}, 1111(47}
0970
0980
0100 IF NOT superchop THEN 0110 IF NOT c128'tn'mem01y THEN
0990
1000 1010
0120 link'c128 0130 ENDIF 0140ENDIF
1020 1030
0150USEc128
1040
0160 popover 0170PAGE 0180 PAINT 'lnotoalozong VDC display...' 0190init80 0200 read'regiSters 0210 read'valoes 0220 display'screen 0230 PAGE 0240 PRINT ·Sample text has been written to the 80 • 0250 PAINT · colum n screen. ThiS ptogram requires· 0260 PAINT 'that you have two mon�ors 01 can swap· 0270 PAINT ·between the 40 and 80 column displays. 0280 PAINT 0290 helpscreen 0300 II 0310 PROC helpscreen 0320 PAINT ·These keys allow you to edit the VDC • 0330 PAINT Move pointer up one register· 0340 PRINT • CRSR-up 0350 PAINT • CRSA-down Move pointer down one reg.· Move pointer to lett column' 0360 PAINT CRSR-tett 0370 PAINT 'CRSR-roght Move pointer to right column' Make POPOVER saection' 0380 PRINT 'STOP 0390 PAINT + Add one to register value· 0400 PAINT Subtract one from register· 0410 PAINT H LISI ol commands' 0420 PAINT ' I INPUT numbell 01 VDC register' 0430 PAINT • S Show all VDC registers' 0440 PAINT ' W Wrote PROCedure to disk that ' will initialize the VDC to' 0450 PAINT • 0460 PAINT the current settings· 0470 PAINT • Restart VDC initialization· Quit program· 0480 PAINT • 0 0490 PAINT 0500 INPUT ·Press RETURN to continue: ·: dummy$ 051 0 ENDPAOC helpscreen 0520 dctne:•FALSE
1050 1060
•
•
•
•
'
•
0530 pos
•1
0540 PAGE
0550 diSplay'regiSiers 0560AEPEAT
0570 change'regosters 0580 UNTIL done 0590quil 0600 // 0610 PROC quit 0620 show'regosters 0630 textcolors(- t -1, 13} 0640 PAINT ·please send these settings to:· 0650 PAINT • C OMAL Users Group, USA Limited. 0660 PAINT ' 6041 Monona Drove' 0670 PAIN T . Madison. Wl 53716'. .
0680
textcok)(s(
t
•
-1.1)
END ''145' '145. ' 0700 ENDPROC quot
0690
0710// 0720 PROC diSplay'screen
-
0890 II
.
•
FORg:•OT01 DO FORb: •0 TO 1DO colot80(0 1 1,1) II k cyan ,
,
ptint80(row,i•30+10.state$(r)) print80(row.i•30 + 16.state$(g)) print80(row,i•30 +23,state$(b))
col0180(r,g,b.� print80(row.i•30+ 29." color·) row:+1 ENDFOR b ENDFOR g ENDfOA r ENDFOR i
co10180(1,1,0,1) II yellow attributes(1 0,0,1) 1120 ptin180(17.27."This teXl is blnking.) 1130 attributes(l.l.O.O) 1140 ptint80(19.27." This teXlos reversed·) 1150 anributes(1.0, 1.0) 1t 60 ptint80(21,26. ·ThisteXl is underlined·) 1170 attributes(1.1,0,1) 1180 print80(23,26, • Rever sed. Blinking text·) 1190 1200 1210 1220 1230
,
attributes(t,O,t,t) print80(25,25. Blinking, Underlined text·) attributes(! ,0,0.0) display80(1,25) ENDPROC display 'screen •
1240 II 1250 FUNC Slate${!)
1260 IF I THEN 1270 RETURN • ON • 1280 ELSE 1290 RETURN 'OFF. t300 ENDIF 1310 ENDFUNC stateS 132011 t330 PROC read'registers CLOSED 1340 IMPORT nameSQ,regO 1350 IMPORT SlartQ,IinO 1360 cou nt:•0 1370 WHILE NOT EOD D O 1380 count: + 1 1390 READ nameS(count) t400 READ reg(count} 1410 READS1at1(count) illcount) ( t420 READ f 1430 ENDWHILE 1440 II 1450 DATA 'H01izontal Tolal ' ,0,0 7 ', 1,0.7 t460 DATA 'H01iz Displayed ',2,0.7 14 70 DATA 'H01iz Sync Pos 1480 DATA 'Vert Sync Width • ,3,0,3 1490 DATA' H01izSyncWidth '.3.4.7 1500 DATA ·Vertical Total • ,4,0.7 1510 DATA 'Vert Fine Adjust • ,5,0,4 1520 DATA 'Vert Displayed ',6.0,7 1530 DATA 'Vert Sync Pos '.7.0,7 1540 DATA • Interlace Mode ,8,0,0 • Video Mod e 1550 DATA ' ,8,1,1 1560 DATA • Lines Per Char ',9,0,4 1570 DATA 'Cursor Begin '. 10,0.4 1580 DATA 'Cursor End ',11,0,4 ',t 0,5,6 1590 DA TA • C ursor Mode 1600 DATA • Text Start H I ',12,0. 7 t610 DATA 'Text Start LO , 13,0.7 1620 DATA • Cursor Addr H I • ,t4,0, 7 1630 DATA • Curs01 Addr LO t5.0. 7 1640 DATA • Memory Point HI • ,t8,0,7 , 19,0.7 1650 DATA • Mem01y Point LO 1660 DATA 'Attribute HI ',20.0. 7 1670 DATA. Attribute L O '.21,0,7 • ,22.4.7 1680 DATA 'Char Total Width 1690 DATA 'Char Disp Widt h 22 ,0,3 1700 DATA • Char Disp Vert ,23,0.4 '.24,7,7 1710 DATA 'Copy(1)/Fiii(O) 1720 DATA ·Reverse Screen .24,6,6 1730 DATA • Char Blink Rate ',24,5,5 24,0,4 1740 DATA 'V-SmoothScrol 1750 DATA • H-Smoo!h Scrol • ,25,0.3 1760 DATA 'GraphicsiTeXl • ,25,7,7 1770 DAT A • Mono/ ATR - Colot ,25,6,6 1780 D A T A Sem igraphic Mode • ,25,5,5 .
•
•
0730 background80(0.0.0.0) II back= black 0740 colot80(1,1.1,1)11char·white 0750 display80( I .42) 0760 page80 0770 FOR x •1 TO 80 D O 0780 IF x<10 THEN (x)) 0790 print80(1,x." Line 1 0800 ELSE 0810 ptont80(1.x.CHA${(x DIV 10)+48)) 0820 ENDIF 0830 pttnt80(2.x.CHA${(x MOO 10)+48)) 0840 ENOFORX 0850 FORx ·3 T040DO 0860 ptintBO(x.1 Ltne • + SffiS(x)} 0870 ptont80(x,71." Roght sde·) 0880 ENOFORx •
1070 1080 1090 1100 1110
color80(0.1,0,1)111t.green t • +state${�) print80(4.i•30 + 1o. • Colot Table - Intensiy prWlt80(S.i•30+ 10,. = • • ••••••••••• ) ptint80(6.i ·30 + 10." Red Green Blue COLOR. ) row:= 7 FORr.•O T01DO 1790 DATA • DBL-Pixef Width
•
•.
•
•.
•
•
•.
•
•
62
1800 1810 1820
• ,25,4,4 ,26 ,4,7
DATA • F01eground CoiOI DATA. Background Colot 26.0,3 • ,27, 0, 7 DATA • Offset Per Row •
•.
1830 DATA ·Character Base 1840 DATA ·Underline Raster 1850 DATA • Word Count 1860 DATA Data Register 1870 DATA Copy Block HI 1880 DATA Copy Block LO 1890 DATA 'Disp Enable ON 1900 DATA • Disp Enable OFF 1910 DATA • DRAM Refresh 1920 ENDPROC read'registers 1930 II 1940 PROC dispfay'registers
2 8, 5, 7 ,29,0 ,4
•.
•
30.0,7 ,31.0.7
•.
•
•
•
• ,32,0.7
•
•.
33.0.7
•
•
,30,0,7 ,35.0.7
• .
36.0,3
1950 PAGE 1960 FOR x: • 1 TO 47DO 1970 row:=x; col:= 1 1980 IF x>24 THEN row:-24: col.•21 1990 IF x •pos THEN 2000 PRINT AT row,col: • • 18 :na meS(x): • 146• • . 2010 ELSE PAINT AT row.col: + nameS(x): 2020 • •
•
•
•
2030 ENDIF 2040 ENDFOA x 2050 teXlcolors(-1 .-1 .10) 2060 PAINT AT 24.22: ' + ADD - SU B T RACT'. 2070 textcol01s(-1.-1.1) 2080 PAINT AT 25,1: ' ' 18' ' + nameS(pos) + ': ';value(pos): 2090 PRINT • '.TAB(31)." MAX:' : 2100 PRINT USING. 111'146' ': max(pos), 2110 ENDPROC dispfay'registers 2120 /1 2130 PROC change'registers 2140 REPEAT 2150 new' pos: • pos 2160 valid: TRUE 2170 CASE KEYS OF 2180 WHEN' q'."0' 2190 quit 2200 WHEN'' 2210 init80 2220 read'values 2230 display'screen 2240 WHEN • h • . H' PAGE 2250 •
"
2260
2270 2280 2290
2300 2310 2320
2330 2340
2350 2360 2370
2380 2390
2400 2410 2420 2430 2440
2450 2460 2470 2480 2490
2500 2510 2520 2530
2540 2550 2560 2570
2580
helpscreen dispfay'registers WHEN 'w' ."W' wrile'registers display'registers WHEN 'i'. 'I' REPEAT INPUT AT 25.24,3: : dal'num$. TRAP ok: • TRUE new'num: •VAL(dal'num$) HANDLER ok: • FALSE ENDffiAP UNTIL ok AND new'num>-1 AND new'num< • max(pos) ••
value(pos):- new'num: make(pos) WHEN's'."S' show'registers PRINT INPUT Press RETURNto continue: • : dummy$ display'registers WHEN • • 17' • IIcu rsor down IF pos<47 THEN new'pos:+ 1 WHEN • • 145 • • II cursor up IF pos>1 THEN new'pos:- 1 WHEN · '29' //cursor right IF pos<24 THEN new'pos: + 24 WHEN • '157' • II cursor lett IF pos>24 THEN new'pos:-24 WHEN • + • IF value(pos)<max(pos) THEN value(pos): + 1; make(pos) WHEN '-' IF value(pos)>O THEN value(pos):-1; mai<E(pos} '
•
2590 WHENCHAS(19)11home 2600
2610
new'pos; • 1 OTHERWISE Nov. 1916: � 7,, .. ,. �
I
valid: •FALSE 2630 ENDCASE 2640 IF new'pos<>pos THEN CURSOR (pos MOO 25)+(pos OIV 25),(pos OIV 25)•20+ 1 2650 PRINT ' ' +nameS(pos); 2660 2670 pos;•new'pos CURSOR (pos MOO 25)+(pos OIV 25),(pos OIV 25)•20+1 2680 2690 PRINT ' '18' '+nameS(pos);' '146''. 2700 valid·•FALSE PRINT AT 25,1. "18' ' +nameS(pos)+• . ':value(pos), 2710 PRINT ' '.TAB(31),'MAX:' ; 2720 PRINT USING 'Ill. 146' ': maJ<(pos), 2730 2740 ENOIF 2750 UNTIL v alid 2760 PAINT AT 25,1: •• 18 • •+ name$(pos)+ •:•;value(pos); 2770 PAINT ' ',TAB(31),' MA X :.; 2780 PAINT USING 'Ill' 146'': max(pos), 2790 ENOPAOC change'regislers 2800 II 2810 FUNC max(r) 2820 RETURN 21((1ln(r)-star1(r))+ 1) -1 2830 ENOFUNC max 2840// 2850 PROC read'values 2860 FOR X. ·1 TO 47 00 2870 byte. •read80(reg(x)). mask: 0 2880 FOR y• starl(x) TO ln(x) 00 2890 mask:+ 21y 2900 ENOFOR y 2910 byte: ·byte BITANO mask 2920 value(x): •byte/(2tstart(x)) 2930 ENOFOA x 2940 ENOPAOC read'vatues 295011 2960 PAOC rnake(r) 2970 byte • read80(reg(r)) 2980 mask.•O 2990 FOR y: •starl(r) TOiln(r) DO mask: +21y 3000 mask: •2S5BITXOR mask 3010 byte: •byte BITANO mask 3020 byte: • byteBITOR (vaJue(r)•(21start(r))) 3030 seiBO(reg(r),byte) 3040 ENOPROC make 305011 3060 PROC show' registers 3070 PAGE 3080 PAINT • • 18·Current VOC valu es I CTAL-P lor Hardcopy· 146•·, 3090 FOR x: • 0 TO 36 00 3100 CURSOR (x MOO 19) +2,(x OIV 19)•20+ 1 3110 PAINT USING 'Register II: Ill': x,read80(x) 3120 ENOFOR x 3130 PRINT 3140 ENOPROC show'regisl6fs 3150// 3160 PROC wrrt e'registers 3170 PAGE 3180 PRINT ·Do you WISh to write a PROCedure to· ' 3190 PRINT d!SIItha1 would set up the VOC regiS!IlfS , 3200 PRINT ·to produce the present display?· 3210 PRINT 3220 REPEAT 3230 INPUT AT 0,0, 1: '(Yes/No): y. 157••: answer$, 3240 UNTIL answer$ IN yYnN' 3250 PAINT 3260 PAINT 3270 IF answerS IN 'yY' THEN 3280 TRAP 3290 PRINT 'writong the procedure: SETVOC 10' PRINT 'the lie· •• proc.set'vdc••• 3300 3310 PRINT 3320 OPEN FILE 5,. proc set'vdc. ,WRITE PRINT FILE 5 .• 0100 II' 3330 3340 PRINTFILES: '0110procset'vdc' PRINT FILE 5: ' 0120 use c128. 3350 FOR v: • 0 TO 36 00 3360 3370 PRINT FILE 5: ' 0'+STA$(v•1 0 + 130); PRINT FILE 5: set80(' ,v,' , , 3380 3390 PRINT FILE 5: read80(v ),')' 3400 ENOFOA v 3410 PAINTFILE 5: 'OSOOeodprocset'vdc' CLOSE 3420 3430 HANDLER 3440 CLOSE 3450 PAINT '01511 error:' ,ERRTEXT$ 3460 ENOT R A P 3470 PRINT 3480 PRINT ' Press RETURN to continue:. ; 3490 INPUT AT 0,0,0' ' ' : 80SW6f$ 3500 ENOIF 2620
•
•
•
•
'
4400 3510 t:NOPAU<.,;wnte regosters .•. q ,·u 4410 3520 II quit 4420 WHEN 'h','H' 3530 FUNC superchip CLOSED 4430 3540 USE system PAGE 3550 setpage($84): chip: •TRUE 4440 helpscreen 4450 3560 RESTORE chipdata WHEN 's' ,'S' 3570 FOR x: a$8014 TO $8018 00 show'regiS!IlfS 4460 4470 3580 READ num PRINT 3590 IF PEEK(x)<>num THEN chop. FALSE 4480 INPUT AT 0,0,0. 'Press RETURN to 3600 ENOFOR X conttnlle' dummy$, 3610 RETURN chip 4490 WHEN 'w','W' write'reg iSiers 3620 II 4500 3630 chipdata: 4510 WHEN · '13'• llcarriagereturn RETURN 4520 3640 DATA 4,67,49,50,56 4530 OTHERWISE 3650 ENOFUNC superchip 3660 II 4540 done' popping: FALSE 3670 PAOC link'c128 4550 ENDCASE 3680 PAGE 4560 UNTIL done' popping 4570 ENDPAOC popup 3690 DIM a$ OF 100 . 4580 II 3700 a$: 'LINK ''pkg .c128 ''13'' 3710 aS:+ 'RUN'13'' 4590 II 4600 FUNC row 3720 FOR xl: 1 TO LEN(a$) 00 4610 current'row:+ 1 3730 POKE 49151 + xi,ORO(a$(xl)) 4620 RETURN currenl'row 3740 ENOFOR xl 4630 ENOFUNC row 3750 POKE $c866, 0 3760 POKE $c867, 192 4640 II 3770 POKE $c865,LEN(a$) 4650 PROC clear'keys 3780 STOP 4660 WHILE KEYS>•• DO NULL 3790 ENOPAOC link'c128 4670 dummyesc: •ESC llcJear stop key 360011 4680 ENDPAOC clear'keys 4690 II 3610 FUNC c126'in'memory CLOSED 4700 PROC setup CLOSED 3620 USE system 3830 setpage($00); is'c128: •TRUE 4 71o II setup by jesse knight 3840 RESTORE C128data 4720 TRAPESC3650 F O R x:=S800e T0$8012 00 4730 FOR xl: •0 TO 12 00 3660 READ num 4740 READ byte# 3670 IF PEEK(x)<>num THEN os'c128: =FALSE 4750 POKE Sc86a+ x1.byte# 3680 ENDFOR x 4760 ENDFOR x1 3690 RETURN is'c126 4770 POKE $c7e2,S6a 3900 II 4780 POKE Sc7e3,Sc6 3910 c128data: 4790 POKE $4d,PEEK($4d) BITOR $20 .3920 DATA 4,67,49,50,56 4800 DATA SaS.$4d,$29,$08,$10,$06,$a9 3930 ENOFUNC c126'in'mernor y 4810 DATA $04,$05,$4d,$85,$4d,$60 394011 4820 ENOPAOC setup 3950 PROC popover CLOSED 4830 II 3960 llcopyrigh1 1986 ten lindsay 4840 PAOC set'colors CLOSED 3970 II original by len lindsay 4850 USE system 3980 INTERRUPT 4860 USE graphics 4870 PAGE 3990 IMPORT quit,show'registers,read80 4000 IMPORT write'registers,helpscreen 4860 LOOP PRINT AT 3,1 · • 18 ·set colors now' 4890 4010 USE graphics 4020 TRAP ESC 4900 PRINT AT 6,1 'press '18' 11 '146' border color • 4030 setup 4040 USE system PRINT AT 8,1 'press '18' 13 '146' 4910 background color· 4050 DIM start'screenS OF 1505 4920 PRINT AT 1 0,1: 'press '18' IS '146' 4060 getscreen(start'screen$) 4070 clear'keys text color· 4930 PRINT AT 13,1: 'press '16' f7 '146' 4080 popup 4090 setscreen(start'screen$) or '18' q'146' quit colors' 4100 ctear'keys II optional hne 4940 CASE KEYS OF 4110 INTERRUPT popover WHEN '' 133' ' 4950 4120 II 4960 textcolors((inq(1) + 1) MOD 16, -1, -1) WHEN • '134'. 4130 PAOC popup 4970 4140 coi:=RND(3,15) 4960 textoolors(-1 ,(inq(2) +1) MOD 16,-1) WHEN '' 135'' 4990 4150 current'row: •AND(2,9) 4160 PRINT AT row,col: ' 18' ------------------------ ' 4170 PRINT AT row,col: •'18': ' VOC Editor Menu 4180 PRINT AT row,col: •'18' ----------------------- ' 18': C : SetColors 4190 PRINT AT row,col: .' .' 4200 PRINT AT row,col: '' 18': - Restore VOC and • 4210 PRINT AT row,col: ..18 ·: : eo column screen 4220 PRINT AT row, col. ' ' 18': H . Help Screen ' 4230 PRINT AT row,col: '18': S : Show VOC values .' 4240 PRINT AT row,col.•• 18': W : Write ME AGEable • 4250 PRINT AT row,col: ' '18': : PROCedure to 4260 PRINT AT row,col: 16': : initialize VOC ' 4270 PRINT AT row,col: ''18': 0 : Ou� Program 4280 PRINT AT row, col: ''18. ------------------------ ' 4290 PRINT AT row,col: ''16.: RETURN to continue • 4300 PRINT AT row,col: ' 16' 4310 REPEAT 5000 textcolors(-1, -1,(inq(3) + 1) MOO 16) 4320 done'popping:•TAUE 5010 WHEN' '136'','q','Q' 4330 CASE KEYS OF 5020 EXIT OTHERWISE 5030 WHEN c' .' C ' 4340 4350 sel'colors 5040 ENDCASE 5050 ENDLOOP 4360 start'screen$(1): ·CHRS(nq(1)) 5060 ENOPROC set'colors start'screen$(2): •CHRS(nq(2)) 4370 5070 ENDPAOC popover start'screen$(3)· • CHR$(inq(3)) 4360 done ' poppin g : •TRUE 4390 •
•
a
•
=
•
' •
· '
•
"
•
•
· '
------------------------'
Keyb�ard
Aubrey Stanley Mississauga, Ontario
der A Total Keyboard Utility For The Commodore 64
In our everyday life we use words to convey our thoughts. A
The program comes in the form of a loader which is run just like a
thought, perfect and spontaneous, must nevertheless be articu
BASIC program. It will relocate itself up against the top of available
lated by a far from adequate speech process. Unless we use
BASIC memory and then run transparently in the background while
telepathy! In a manner of speaking, the keyboard replaces our
you go about your normal work. At this stage all the enhanced editing
vocal chords when we are dealing with the computer. We type in
commands are available, mainly on the function keys.
character after character to express a BASIC statement which.had formed instantly in our mind to begin with.
If you wish to add more features, these are simply defined in a Profiling program as array assignments. Run this BASIC program and
We use the Screen Editor in Direct Mode to create program source
your keyboard will take on the profile you've given it. You also have
in memory. And by executing BASIC statements directly, we can
the option to generate another version of Keyboard Expander with the
control our development environment. Keywords are entered
new profile. This version can then be installed in future, instead of the
over and over, and certain commands are used frequently. A Key
original.
To-String capability would be of extreme value. We could pre define keywords, op-codes, pokes, commands, even complete
Repositioned Keys
BASIC programs; for instant recall at the touch of a key. And with an interactive feature which paused while we typed in variable
This feature allows you to re-define as many keys as you wish to. The
data, this function could be even more useful.
normal, shift, commodore, and control modes are treated quite independently, so you can move individual functions of keys around.
The Screen Editor would benefit from a few enhancements to take
Repositioned keys are always active, even in Run mode, allowing you
the toil out of editing programs. We could have an alternative
to use them in your programs.
Insert/Delete mode where we worked from the current cursor position instead opening up space in the text or moving the cursor
Strings
past the final character to be deleted. A Tab key to move automati cally back and forth through BASIC and DATA statements, and a
This feature allows you to assign an unlimited sequence of keystrokes
Clear from the current cursor position, are also worthwhile candi
to any key function. When you press the key, the string will instantly
dates.
be printed to the screen and actioned, as if you had typed it in
yourself. A special code can be encoded in the string to pause output If we were given the choice, most of us would re-arrange at least
while you type in some variable data such as a filename. Then press
some of the keys. Those shifted cursor functions might get relo
and release IRON/STOP] to continue. Pressing [RETURN] instead, will
cated to other, unshifted positions. And DVORAK fans would likely
terminate string output, actioning whatever is displayed on the line.
re-arrange their entire keyboard.
Or you can press
]SHIFT] ]RETORNI
to abort the process. The en
hanced editing features are temporarily disabled while entering varia We only have to browse through the back issues for a wealth of
ble data.
machine language routines designed to improve our development environment. Many of us could put our talents to good use in this
The string feature is temporarily disabled in Run mode
area. Trouble is that these individually good programs refuse to
interfere with your programs, and also when the editor is in Quote
co-exist under the same roof. They vie for system vectors, chase
mode so that graphics can be printed normally.
so
as not to
zero-page, run in the same RAM, etc. With a little adaptation, and a traffic cop to direct their execution, any one of them could be
Routines
mobilized for instant action. This feature allows you to tie a ML (machine language) interrupt Welcome Keyboard Expander. Perhaps not a total utility, but it
routine to a key function. The interrupt routine will run in the
could well become a cornerstone in your development environ
keyboard (timer) interrupt. It may perform a specific task of its own or
ment.
be used to initiate a background routine or a parallel program.
The Program
A Background Routine, when initiated, will run in every subsequent keyboard interrupt until it disables itself. It can also be started from
Keyboard Expander can make each key work for you in normal, shift,
within a running ML program in the BASIC environment.
commodore and control mode. Reposition any key, recall a sequence of keystrokes, or run a routine of your choice. You may create your
A Parallel Program normally multitasks with BASIC by running after
own configurations or use it in its pristine form for the many Screen
alternate interrupts. But it may elect to run exclusively for any period
Editor enhancements that it has to offer.
of time. It does not run interrupt code and therefore can do anything a
lhe Tlaneaclot
64
Nov.1916: � 7, .... 03
normal program can do. Like a background routine, it can also be
Tennlnate Program:
started from a running ML program.
ICTRLI C I =I ISHIFT!
KE can be entirely eliminated by pressing these three keys together.
You can use the routine feature to execute your own commands. The
The keyboard will revert to its usual state. The memory used b y the
command line is converted to ASCII and copied to your program
program will be released to BASIC memory space only if no other
space whenever [CT'RLI [RETURN! is pressed.
programs have been installed after KE.Otherwise the top of memory
Routines tied to keys will not be activated in Quote or Run mode. However, as mentioned, you may activate background or parallel
pointer is left untouched. Clear Quote Mode:
programs in Run mode.
ICTRLIIROMEI
This command does exactly what it says. It becomes especially useful
in KE, because when you're in quote mode all keys print to the screen,
even keys that initiate the built-in commands or those you may have
Compatlblllty Issues
assigned to your own routines. This command also clears any inserts
Keyboard Expander will co-exist with other, (properly installed},
left over after you have used UNSTI to open up space in the text.
programs. The word, Install, in the context of this article, refers to
programs that allocate room for themselves at the top of available
Set Auto Insert/Delete Mode:
Both POWER and PAL (Proline S/W Ltd} run without trouble when
In this mode theiiNSTI key is no longer active. Instead everything past
BASIC memory by modifying the top of memory pointer downward. installed with Keyboard Expander, and probably SYMASS will too.
C [ =I [DELl
the cursor on the BASIC line, (logical line of 80 characters}, moves
right as you type, and the new character is placed at the cursor
Keyboard Expander will not co-exist with programs that modify the
position. When the BASIC line is full, characters will start to get
keyboard and IRQ vectors (locations 655 and 788}. You may well find
overwritten.
allows you to incorporate new features quite easily.Except for these
The cursor and home keys behave as they normally do, so you can
other keyboard utilities practically redundant as Keyboard Expander vectors, Keyboard Expander leaves the BASIC environment intact.
move around the text at will.
to preserve this environment when creating your own configurations.
If repeat mode is on, then all keys will be repeated if held down.
And it does not use any zero page or free RAM.As far as possible, try
Strings assigned to keys, however, will not be repeated.
The memory it takes away from BASIC is IK of tables and approxi
The [DELl key behaves differently in this mode.Everything past the
mately I.SK of code.
cursor, up to the end of the BASIC line, is moved left one position and
overwrites the character at the cursor.In effect the key acts as a black hole through which you may drop unwanted characters.It will repeat
GETI1NG STARTED First type in and save the generator program,
to create
•
KE
•
•
if held down.
KE.GEN ·.Then run it
on disk. Now load and run KE just like a normal
Clear Auto Insert/Delete Mode:
BASIC program.
Returns the editor to the normal insert/delete mode of operation.
You'll see the start address and a copyright message.The start address
is the one you want to SYS to in order to restart after a RunStop/
Set Quote Mode:
Restore.
then don't have to enter the quote character.
IS HliT liC-I, (gRLI and [RUN/STOP! keys. Built-in commands have been mainly assigned to the function keys.
This only affects the next character typed. It is like being in quote
is
up}.This command will toggle the mode on and off. Repeat.s Mode:
One Character Quote:
ICTRLIIC=I
In this mode each key will click as you type (if your volume
mode for one character only.
turned
Forward Tab: This command is useful for tabbing through a BASIC or DATA line. Try
ICTRLIISHIF"fl
it to see how it works.
In this mode all keys will repeat. Again, the command toggles.
Back Tab: Disable/Enable Program:
IC=IIHOMEI
This could be useful when you are editing text between quotes as you
Four keys on the keyboard are intrinsic to the program.These are the
Click Mode:
ICTRLIIDELI
ICTRLIIRON/STOPI
ISRit"riiFII
This command works similarly to forward tab except that the move
KE can be temporarily disabled with this command. The keyboard
ment is reversed.
will then behave as it normally does until you press these keys again.
Use this command when you need to print an original key, one that
Tab To End:
string or program assignments you may have made are automatically
The cursor moves forward to the position past the last character on the
you've otherwise reassigned. However, if you are in quote mode, any disabled, so you don't need to disable the program in this case.
ltleTianeocb
BASIC line. 65
Nov. 1986: Vdume 7,1aue 03
ICTRL][£I)
Tab To Start:
Configuration 2
The cursor moves to the beginning of the BASIC line.
KE is again totally relocatable, but you are adding a profile to it, with your ML routines (if any) residing in free RAM. After installing KE, run
Clear To End Of Une:
ISHIFI'j(nJ
your profile program. Do not select the option to generate an absolute version. This configuration is useful for testing out a new configura
All characters on the BASIC line from the cursor onward are c.leared.
tion.
Clear To End Of Screen:
Configuration 3
All characters from the cursor to the end of the screen are cleared.
This is essentially the same as 2, except you want to install your ML routines. As the Profile program defines the absolute start address of
ISHit"l'j !Lll
Screen Colour:
each routine, the only way this is possible is for you to have assembled your routines into an absolute memory block at the real end of BASIC
Each time you enter this command the background colour of the
memory, which is normally $9FFF. Install your program (or programs)
screen will change to the next higher colour, i.e., black to white, etc.
before installing any other programs.
By using this command in conjunction with the following two for border and text, you should be able to set up an ideal colour
Configuration 4
combination. This is the absolute version of 2. KE must be the first program installed so that it sits at the real top of BASIC memory. Run your
Border Colour:
profile program and this time choose to generate the absolute version to disk. In future you can install your own version, instead of KE. Be
Increments the border colour.
sure it is always the first program installed. Of course you still have to
ICTRLIIF71
Text Colour:
load any ML programs into free RAM.
Increments the text colour. The entire text displayed on the screen
Configuration 5
changes instantly. This is the absolute version of 3, and one that will find the most use. It
ICTRLI!fiD
Ust Freezer:
is identical to 4 if there are no ML programs. ML programs must first be installed, then KE, thus concatenating the programs at the real end
Use this command to freeze the listing of a BASIC program as it flies
of BASIC memory. Now run the profile program and save the result to
past on the screen. After typing this command, if you hold the ICTRL]
disk.
key down, the listing will continue to scroll at a slow rate. Releasing
the key freezes the listing again. The IRON/STOP] key also behaves
similarly in this respect, except that the scroll will be faster.
KEYPOWER FROM KEYBOARD EXPANDER
It is really only useful for freezing a List, but will also freeze everything
The configuration process will now be explained in terms of Keypo
if you use it otherwise.
wer, a program we will generate along the lines of configuration 5.
Get out of a freeze by pressing the IC
I key on its own.
Type in
•
KE.PF ·, the Profile Program, and save it under this name.
This program is used to create profile programs by adding definitions Phosphor Saver:
ICTRL]In)
in the line range, I 00 to 799. It should never be modified. With KE.PF still in memory, type in
So named because it blanks out the screen completely, as if it were
•
KPOWER.DEFS' and save it under the
name, 'KPOWER.PF'.
powered off. It also does a complete freeze of anything that is running. Use the
IC-I key
We are going to include two useful ML routines. One is a BASIC Line
to get out of this one too.
Ruler for highlighting program lines and the other is a List Scrolling routine that I have adapted from an article in the TRANSACTOR, Vol 5, Issue 6, by Darren Spruyt. Type in and save the generator program,
CONFIGURATIONS
•
SCROLL.GEN·, then run it to generate
•
SCROLL.OBJ
•
on disk.
To unlock the door to Keyboard Expander and add your own key board profile, you need to run a Profile program. But first a word about
Type in and save "INSTALL· which is a general purpose program to
the possible memory configurations. There are several, depending on
install ML programs in memory.
whether you are defining ML routines. whether these will reside in free RAM or are to be installed, and whether you are going to generate
To start with, no program should be installed, not even KE. If KE is the
an absolute version to disk.
only program installed at this stage, you can use the "Terminate"
'
command, described before, to eliminate it. Otherwise power up or otherwise reset the computer. Load and run INSTALL. Enter the
Conflgu.ratlon 1
filename, SCROLL.OBJ, when asked, and press return. Now load and The original program, KE, you have already seen. It is totally relocat
run KE. Finally load and run KPOWER.DEFS. It will take about 10
able and generally useful to have around without the overhead of
seconds, then ask if you wish to create an absolute version. Follow the
strings and routines.
steps to do this, saving the file under the name,
•
KPOWER
• .
You can
use Keypower now, a s it is in memory. 66
New. 'IM6: � 7,111. 03
You must use INSTALL to install Keypower
Just loading it as an object program wont do the job. You can make INSTALL specific by replacing line 22 with:- F$ KPOWER. Also remember that Keypower must always be the first program installed so that it loads at the top of real BASIC memory.
Initialize Drive:
in future.
Initializes the disk drive.
=
Validate Disk:
In Keypower, all the KE commands are still there, but now you have some additional ones.
Rename File:
List Scroll Down:
[CTRLI�
This key, when held down, will scroll a BASIC listing on the screen (assuming the BASIC program is in memory). The listing will scroll downward and display BASIC lines in order of decreasing line num bers.If no BASIC lines are displayed on the screen, the listing will start with the highest numbered BASIC line.
Prints and executes a one line (line 8000) BASIC program to read the disk error channel. The line is automatically deleted afterwards.
[CTRLI�
WARNING: If a List is currently in progress, press lRUN/STOPI to
Prints and executes a BASIC program (lines 8000-8008) which lists the disk directory directly to the screen. Press any key to pause, or Q to quit. The program lines are not deleted, so RUN 8000 will execute the program again. To delete the program lines press ICTRLI [Z]. If these lines clash with your BASIC program in memory, you can select the "TEST" BASIC Partition (described below).
stop the List before you use this function. This also applies for scrolling up.
List Scroll Up: Like list scroll down, only the scroll is upward and lines are displayed in order of increasing line numbers.
This program was copied from an example in the POWER Manual.
ICTRLI[±]
Test Partition:
The Scroll routines have been implemented as a Parallel program.
This command creates a temporary BASIC partition immediately following any BASIC program currently in memory.You may then use the Test partition for whatever purpose.To release the TEST partition and return to your original BASIC program environment, press [CI'RLI B.
Load BASIC Program:
BASIC Line Ruler: This command will put a border (much like a transparent ruler) around the BASIC line at the current cursor row. It is useful for scrutinizing the BASIC line underneath it. The ruler remains on the screen and will move in conjunction with the cursor.
[CTRLI[!J
This function can be used simultaneously with all the other functions described so far.
This function recails an interactive command string. You enter a variable filename to a (LOAD· ,8) command. As for all interactive strings, press and release [RUN/STOP[ to continue after entering your data from the keyboard. Pressing lSHIFTllRETURNI aborts the whole procedure. If you complete the command yourself (e.g., you may want a ML load instead- ... ,8, I), then the [RETURNI key on its own will both action what is printed on the line as well as terminate the remainder of the string output. .
.
.
.
.
ICTRLI[N]
This interactive command lets you enter the new filename and then the old filename. Press and release IRON/STOPI after each filename is entered.
The Up and Left Arrow keys perform the Cursor Up and Cursor Left functions respectively. However, the shifted cursor keys retain their normal usage.
List Directory:
[CTRLIIYJ
Validates the disk.
Repositioned Keys
Read Error Channel:
ICTRLI[I]
•
The routine has been implemented as a Background routine. While running, it uses the Tape Buffer (location 828) to generate the sprite data.
•
Run BASIC Program:
This command terminates the Line Ruler function.
ICTRLI[RJ
BASIC Keywords
Like the previous command, but also runs the program it loads.
Save BASIC Program:
Press the IC I key in conjunction with the alphabet keys to print the following keywords:
[CTRLI�
Interactive command to save a BASIC program.
Load From Directory:
ICTRLI[fil
Kill Line Ruler:
ASC(
CHR$( DATA
FOR
GOTO INPUT LIST
MID$( NEXT OPEN PRINT RIGHT�TR$(
[CTRLI!!!]
Profile Update
This function loads the file displayed in a normal directory listing when the cursor is positioned at the beginning of the displayed line. It is a good example of how a tedious manual process can be replaced by a string.
With an absolute version like Keypower in memory, and no other program installed, it is possible to run a profile program as many time as you wish in order to add new definitions.You can even create disk 67
No¥.1916: Volume 7,111ue 03
versions, but if strings have been replaced by new ones, there still will
ZS$(CD,D) =ZS$(CD,D)+LN$ + CHR$(Y) ZS$(ND,Y)= L$ + LINKED STRING.
remain the memory overhead for the old ones. Then it's best, at some
•
later stage, to create a fresh version.
LN$, specifies a normal (unshifted) factor, and CHR$() has its usual sense, as in BASIC. On the second line, L$ is Note the linking value,
used
BUILDI.NG YOUR OWN PROFILE
I 00 to 799 and don't
define any variables. Whatever you need to use has been declared in KE.PF. Refer to KPOWER.DEFS for examples.
Each key has its own variable to define it as shown in Table
the first character in the linked string, instead of S$ which is
used to define a new string. This is always so for a linked key.
As stated before, you must add your profile to KE.PF to generate a Profile program. Use line numbers in the range
as
Any number of keys may be linked in this way. A key you link to does
not loose its original function, but you must not use it in another
string or routine definition. For this reason, it is advisable to use the
normal typing keys and save the shifted functions for strings or
I.
Qualifying variables for the shift factors are shown in Table 3. Which group to use will depend on the sense in which a key is defined, the
routines.
When defining an interactive string, you need to insert the variable,
source key in an array or the target key for reposition.
1$, at the positions you wish to pause at. For example,
Each key function has an entry in a two dimension array. The first
ZS$(SD,K)= S$ + "INSERT DISKETIE. + 1$+ N$ ZS$(SD,K)=ZS$(SO,K) + LOAD. +0$ + SYMASS. + 0$+ •. a· +A$+ "RUN" +A$
dimension specifies the shift factor and the second the key. Each
•
feature- reposition, string, routine- has its own array. There are also two variables for defining the start address of a "jump vector table"
•
and a "command line" for ML programs. See Table 2.
Routines
When defining strings you will need to use the string variables shown
These get defined in array ZR(). For example, ZR(TD,G)=49195, will
in Table 4.
run the interrupt routine located at 49195 when you pressiCTRLI[Q). If you are using background or parallel programs, then the variable, ZV,
must be assigned the address of a 36 byte area for the "jump table". If
Repositioned Keys These get defined in array
you are implementing your own commands, the variable, ZC, must
ZP(). For example, ZP(SD,2)=PO, moves
be assigned the address of a 40 byte area to hold the command line.
the QUOTE function to the unshifted POUND key. The QUOTE
function can still be used on its original key as long as you don't make
it the target of some other function. ZP(ND,UA)=LA+SF gives you
MACHINE LANGUAGE INTERFACE
the example below, the PLUS and MINUS keys have been completely
With machine language you can realize the full potential of Keyboard
the normal
UP ARROW display on the shifted LEFT ARROW key. In
interchanged.
Expander. The simplest
type of ML program is the lnte.rrupt Rou·
tlne which you assign to a key. It will be entered as a subroutine
ZP(ND,PL)=MI:ZP(SD,PL)=M l+ SF:ZP(CD,PL)= Ml + CF:ZP(TD,PL) MI+TF ZP(ND,MI) = PL:ZP(SD,MI) = PL+SF:ZP(CD,MI) Pl + CF:ZP(TD,MI) =PL +TO
within the Keyboard (timer) interrupt every time the key is pressed.
The key matrix code (0 to 63), in case you need it, is passed in the .Y
=
register and the shift state in the .X register. These values are also
=
stored by the Kernel in locations 203 and 653.
For Background and Parallel programs, you will need to interface into
Strings
Keyboard Expander through a "jump vector table". This table takes 36
These get defined in the array lS$U . There are some elementary
bytes and contains 12 jump instructions. Only the first six are of use to
Always start off a new string definition with S$. For example,
The jump table cannot be hard-coded because Keyboard Expander
rules to be followed.
you. Use the remaining 18 bytes for uninitialized data if you wish.
can move about in memory. You have to assign the start address of the
ZS$(CD,L)= S$+ LIST. +A$
table to the variable, ZV, in your profile program. When this program
•
is run, Keyboard Expander will copy the table into your program
will action the List command when you press C=I [1].
space.
If the complete string does not fit on the line, add in the remainder on
You will also need this table if you are initiating background routines
I
following lines. For example
or parallel programs from within a running ML program. Here you make use of the USER JUMP address at $311/$312. Store the address
ZS$(TD.B)=ZS$(TD,B) + REMAINDER.
for your table in low/high byte format. Then change location S310
•
from its normal S4c ("jmp" instruction), to any other value. Keyboard
will continue the string from the previous line. If you define more than 255 characters in a string, you'II get a
Expander will detect the change and copy the table into your program
space. You can be certain the table has been copied when $310
contains the jump instruction again (within one timer interrupt
BASIC
error. To get round this, you need to link the string to some other key.
period).
characters. You may link it to the key, Y, as follows:
In PAL source code, your table should look something like this:
Suppose that the string on the original key, C= - D is near or at 255
61
Nov.1916: � 7,lll!le 01
KEVECTOR
= • •= •+3 BRSTART BREND • = • +3 • = • +3 PPSTART PPEXCL • = • +3 PPSHARE • = • +3 •= • +3 PPEND NOTUSED • = • + 18
;START ADDRESS ;START BACKGROUND ;END BACKGROUND ;START PARALLEL ;RUN EXCLUSIVE
TABLE 1 - KE Y VARIABLES
;RUN SHARED
IREVSI � to IZl [Q) to � IEII�IE2J[ELJ IDE[! IRETORI'll ICORSORRIGRTI CORSoROOWI'JI msARROWI [EFTARROW! POONOI RO�El !SPACE! � Q 0 Ill [±]EJ El G 0 [J
;END PARALLEL ;NOT RELEVANT
Background Routines JSR BRSTART will initiate a background routine from your inter rupt routine or ML program, provided one is not already in existence. Pass the start address in the
.X/.Y registers in low/high byte format. A
zero status return means your routine has started. The routine is coded as a subroutine to run within the keyboard interrupt. It will be called on every clock tick until it disables itself with aJMPBREND. ParaUelPrograms A Parallel program has the same start/ end interface
as
the back
VARIABLES A toZ NO to N9 F1,F3,F5,F7 DE RE AI DO UA LA PO HO SP AT
SE
CM
SL
PL
Ml
EO
AS
PE
co
ground routine, only use PPSTART/PPEND instead. When first started, it will multitask with BASIC by getting scheduled after every alternate timer interrupt. This is all right for a memory bound program, but if you're doing 1/0 to the disk, printer, etc, BASIC will be
2 - ARRAY/STORAGE
TABLE
almost unusable. In this case and also if you need to lock out the
STORAGE
keyboard, you must run exclusively by doing a JSR PPEXCL. When you want to multitask again do a JSR PPSHARE. CommandLine The interrupt routine assigned to the ICTRLIIRETURNI function has
DESCRIPTION
ZP(3,62)
REPOSITIONED KEYS
ZR(3,62)
ROUTINES
ZS$(3,62)
STRINGS
zv zc
JUMP VECTOR TABLE COMMAND LINE
the privilege of receiving the command line if you assign the start address of a
40 byte data area to the variable, ZC, in the profile program. On this keypress, the line will be converted to ASCII and
TABLE 3 - SHIFT FACTOR
copied into the assigned area before your routine is entered. This will allow you to implement commands like renumber, delete, merge, etc, by starting a parallel program if necessary. lostalllog MLPrograms Procedures to do this have been discussed in the section on Keypo wer. Normally,
as
VAR
USE
VAR
USE
NO
ARRAY
--
----
so
ARRAY
SF
TARGET
SHIFT
CD
ARRAY
CF
TARGET
COMMODORE
TO
ARRAY
TF
TARGET
CONTROL
MODE NORMAL
for SCROLL.OBJ, you shouldn't need to run any
initialization code. But if you have to, the I NSTALL program is smart enough to enter your initialization routine provided you start your
TABLE 4 - VARIABLES USED IN DEFINING STRINGS
program as follows:
VAR
DESCRIPTION
;to in it routine
S$
Always the first in a new string
.ASC' 'INIT'' ;special string
L$
Always the first in a linked string
R$
To encode a RETURN function
N$
To encode a SHIFT RETURN function
0$
To encode a QUOTE character
JMP !NIT INIT (code)
;init routine
CONCLUSION
LN$ Normal factor when linking to another ke y
Keyboard Expander works intimately with the Kernel keyboard inter rupt to provide the interface through which you can manipulate any key to your advantage. It has the potential and flexibility to become
LS$
Shift factor when linking to another key
LC$
Commodore factor when linking to another key
LT$
Control factor when linking to another key
whatever you want it to be, even a total keyboard utility. So if you do develop any useful routines for it, don't be shy. Send them to the TRANSACTOR so that we all can enjoy them. I'll really look forward to that.
69
Nov. 1986: YoUne 7, Issue 03
Keyboard Expander: Generates PRG file • KE' on disk.
1620data 97,
LD
1630data
0,
0,
4, 97,
0
0,
1
72,
5 42
1
0,
KD
1660data
0,
•
42,
1 115,
4, 97,
1670data 82,
4, 97,
0,
•
1030 for j
NO
1040 if ch<>276146 then print· checksum e rro r' : end
GL
1680data
4,124,
CD
1050 rest o re
MM
1690data
5,
GN
1060 open 8,8,8, O:ke,p,w
OP
1700data 37.
MC
1710data 129, 76, 167,
Fl
1
0,
0,
AM
•
5
0,
1020 rem second loopwrites f il e
•
29,
4, 97,
GP
ch+ x : n ext
1
1640data248, 1650data 97.
=
0,
PI CH
: ch
0
0,
1010 rem first loop en sures checksum iscorrect x
97,
4, 97,
08
1 to 2745 : rea d
4,
1,188,
1000 rem program to create file • k e • ond isk
=
1. 135,
0,
GN
0, 97,
4, 72, 4,
0,
0,
1
0,
0,
1.
0,
0,
3, 97,
4, 21,
•
•
1
4, 97,169,191
Bl
1070 for j
IL
1080 print#8,chr$(x); : next
JH ·1720data 32,157,
CJ
1090 closeS : end
GJ
1730data212, 97,240,
IL
1100data
0,158, 50
FD
1740data133,212, 96, 96, 97, 169,
Gl
1110data 48, 54,
51, 58,162,
0
8C
1750data 133,216, 96, 96, 98, 238, 32,208
LK
1120data 32, 62,
9, 32, 75,
9, 32, 75
FH
1760data 96,
PH
1130data
6
NG
1770data 98,206,134,
AI
1140data169,
70, 32, 210, 255, 96,165, 25
FF
1780data 97, 162,
JD
1150data240, 39, 32, 75,
9, 133, 53, 32
FF
1790 data246, 97,160,
DP
1160data 75,
9, 133, 54,165, 55,133, 33
HL
1800data145, 245,
FM
1170data 56,229, 53,133, 55, 133, 27,133
PM
1810data230,246, 96,202, 97. 208,246,
IL
1180data 51,133, 53,165, 56,133, 34,229
JH
1820data 96,
NM
1190data 54, 133, 56,133, 28,133, 52, 133
PL
1830data232, 97,181,217, 97, 48,
27,133, 29,165, 28,133
JM
1840 data232, 97, 208,249, 97, 224, 25, 97
=
1 to 27 45 : rea dx
1.
8, 13,
9, 32, 75,
8, 10,
0,
0,
9, 201, 129,240,
96,
5, 97, 169, 64,129
5, 97,169,
0, 97, 133
7,
1
97,169,
•
97
1, 97
98,238, 33, 208, 96, 96 2, 98, 173,134,
2
4, 97,160,216, 97,132 0, 97,132,245,
97
96,200, 97, 208,251, 97 96
97, 166,214, 97. 134,245, 96 3, 96
10
1200data 54, 165,
K8
1210data 30, 169,
5,145, 29
LG
1850 data 176,
AL
1220 data200, 208, 251, 230, 30,202,208,246
JN
1860data 97,149,217, 98, 32, 255,233, 96
AO
1230data169,
3, 24,101, 27,133, 29, 169
HC
1870data232, 97,208,240, 97,166,245, 98
0,168, 162,
0,101, 28, 133, 30,160,
0, 169
El
12, 97,181,217, 97,
9,128
1880data 32. 240, 233, 97, 164, 213, 98,
32
IM
1240data
HC
1250 data 75, 145, 29, 200, 169, 69,145, 29
KM
1890 data 36,234, 97,169, 32, 97,145,209
PJ
1260data200,169, 88,145, 29. 200,169, 80
OK
1900data 98, 32,218,228, 96,136,
29,200,200,200,200,200,165
CK
1910data
28,145, 29
CN
1920 data 96, 97,164,211, 97, 196,213, 97
34
J8
1930data240, 81, 97,132,245, 97,166,214
27,133, 29,165, 28
MO
1940data129, 32,234,
9, 133
E8
1950data 32,174,
33,201,128,144
PN
1960data249, 97,208, 59,129, 32, 174,
9,165, 26, 41,127
MC
1970data 97. 176,
KM
1270 data145,
MC
1280data 27. 145, 29, 200. 165.
4, 97,196,211'
97, 48
97,176,242, 96
Nl
1290data200,165, 33, 145, 29,200, 165,
NF
1300data145, 29,165,
FL
1310data133, 30,160,
CD
1320data 26,201,
KE
1330data
MF
1340data201,
75,
9, 24
FJ
1980data245, 97,240, 236, 97,164,245,
LH
1350data1 01, 27,133, 29, 32, 75,
9,101
AA
1990data 16, 44, 96,200, 97,196,213, 97
PI
1360data 28,133, 30, 76, 171'
8, 76,103
88
2000data240, 55, 97, 176, 66, 97,192,
9
PC
2010data 97,208, 49, 96, 232, 97,208, 46
75
AF
2020data 97,164,211, 97, 240, 30, 97, 166
9, 76,171
18
0, 32, 75,
5,240.
3, 32, 42,
4,208, 21'
9, 201,
32,
1' 208, 20, 32, 75,
GO
1370data
KJ
1380data 24,101, 27, 32,
LM
1390data
9, 101,
DG
1400data
8,201,
IL
I
LJ
49,
28, 32, 49, 2, 208,
15,
9, 32, 32, 75,
9
8M
6,
4
97,208,249, 97,132 97 40
4, 97,176, 19
7. 129, 32,223,
4, 97
Nl
2050data176, 12, 97. 240,249,129, 32,223
3, 208, 17
KF
2060data
9, 24,101, 27, 32, 75
CF
2070data 32,174,
9, 76,171,
8,201,
4, 97, 176,
5, 97,208,249,129
4,
97, 134,214, 97,132
GL
1420data
AN
1430data 32.
NP
1440data
9,101, 28, 32, 49,
9, 76,171
OH
2080data211, 96 ,
8M
1450data
8, 201, 96, 48, 13, 41,159,170
KM
2090data 96,
HK
1460data232, 32, 42,
9,202, 208,250, 76
PL
2100data 97,208,
H8
1470data171.
DA
1480data
NC
1490data 29,208,
L8 JG
1510data
MC
1520data 45, 165,
EO
1530data
JD
1540da\a 32, 76,101,
9, 169,
J8
1550data 108, 27,
IE
8,
4, 97, 176, 15, 97,240
2030data214, 129, 32,223, 2040data 97,208,
9, 129
9, 32, 75
1410data 24,101, 27, 32, 49, 75,
4, 97,208,
0,
32, 75,
9, 32, 49
IJ
96, 96, 136,
97,208,
3
56, 97, 176, 18, 97,192, 39 1
•
96, 202, 97,177,209
2110 data 97,201, 58, 97,240,
6, 97,201
2, 97,201,
44, 96
0,145, 29,230
GP
2120 data 32, 97,240,
2, 230, 30, 40, 96,169
FL
2130data 24, 96,
1500data106,133, 31,169,
9, 133, 32, 169
CN
2140data214, 97,192, 40, 97, 144,
1' 133, 25, 96,
8, 165, 31,197
AG
2150data 165,211,
32,229, 46,176, 13,160
FK
2160data 96,232,129, 32,234,
31,230, 31, 208,
7,230
LF
2170data 32,
5,
40, 96
EA
2180data 97,176,
0, 16,129,
42, 10
FN
2190data247, 97,201, 32, 97, 240, 191, 97
1560 data 129, 76,188.
9, 99, 75, 69, 88
KA
8F
1570data 80,
0,
0, 97
PO
2200 data208, 186, 97, 166. 214, 97,164,211 2210data 97,192, 40, 97, 144, 1 . 96,202
KD
1580data
0,
0,
0
N8
2220 data 97. 160,
C8
1590data
1.
21,
4,
4, 68,
0, 97,
0
KO
2230data 32,208, 97,169, 16, 98,141, 32
IF
1600data
0,
1.
51,
4, 97,
0,
0,
1
EA
2240data208,
8F
1610 data 47,
0,
1.
55,
4
JP
2250data208,129, 32, 104,
1he Ttcw�Ntto.
9, 96,
0,177.
8,160,
0,
1
16,
0,
1,
•
4, 97,
26,
4, 24, 4,
0,
97,
7'0
96, 97,164,213, 97,166 7, 97
97,201, 40, 97, 176, 4,
97,201
97. 208,204, 129, 32,223,
98,
1 4
4, 97, 201, 32, 97,240
0, 97. 240, 176, 98, 172 77, 17,208, 98,141, 17 5, 97,224,223 Nov. 1986: 'Mime 7, ...... 03
I
CG
97,208, 249, 98, 140, 32,208, 97 2260 daa t
KK
2900 daa t
EM
2270data
16, 98,141, 17,208, 96, 96
JJ
2910data 96, 104, 129, 141,
AC
2280data129, 32, 165,
5, 98, 238, 32,208
OP
2920data129,141, 12,
KB
2290data129, 32,135,
5,129, 32,104,
5
FM
2930data 96, 72, 129,173,
AH
2300 data 96, 138, 97, 41,164, 97,201,164
BN
2940data129,173,
LL
2310data 97,240,246, 97,201,132, 97,208
KM
2950data
JA
2320 daa t
6, 98, 206, 32,208,129, 32,174
PN
MF
2330 daa t
5, 98, 76,126,234, 97,162,127
HO
2960data 96, 72,129,173, 3, 4, 96, 72 4, 129, 141, 8, 4 2970data129, 173, 12,
NG
2340daa t 98,142,
1' 220
JO
2980data129,173, 11 '
4,129, 141'
7,
4
JO
2350 daa t 96, 96, 96, 72, 97, 169, 32, 129
LD
2990data129, 173, 10,
4, 129,141'
6,
4
DO
2360data 45, 15,
4, 97,240, 12, 97,160
GO
3000data129,173,
9,
4,129,141,
5,
4
NG
2370data 15, 98, 140, 24,212, 96,168, 96
NM
3010data129,142,
4,
4,129, 140,
3,
4
LA
2380data136, 97,208,253, 98,140, 24,212
PC
3020data129, 108,
0,
4, 129, 108, 13,
4
GO
2390daa t 96, 104, 96, 96, 96, 104, 96, 24
oc
3030data129, 173, 14,
NK
2400data 97,105,
4, 96
FF
3040data142, 13,
KP
2410data 104, 97, 105,
4
KN
3050 daa t 169,
Bl
3060data141.
IB
[
9,
2420data 97,169,
0, 220, 98, 174,
1, 129,141,
0,
0,129, 141,
8, 129, 45, 15,
1'
4, 97
ED
2430daa t 208,
HO
2440data
NB
2450 data 97,169,247,129.
NA
2460 data141 . 15,
AB
2470 daa t
6, 131, 160. 80,
6, 98,142,
IL
2480daa t
3, 98,140, 21'
6,
4,129,173,
14,
4, 96, 96, 97, 169, 129 2, 97, 169,
4, 96, 96, 130,162, 80
CF
3100 daa t 169,
1 ' 129, 44,
20
JA
15, 129, 142, 3110 daa t
3, 97, 162, 49
FO
3120daa t
4,129
JA
3130data 129, 141,
4, 96, 96, 130, 162, 79
HC
3140 data 98,174, 141,
EA
2520daa t 130,162, 19,
MH
2530data 97. 208,
OF Bl
2540daa t 235, 98, 142, 143, 2, 98, 140, 144 2550data 2, 96, 96, 97, 165. 207, 97,240
FO
2560data 12, 97' 165,206, 98, 174,135,
Ll
9, 131, 160, 79,
9,
4, 97, 162, 2,
7, 4,
7
KL
3160daa t 129,173, 15,
72, 97, 160
Gl
t 169, 3170 daa
6, 97
4, 97,208
4, 129, 140, 96,138,
2, 97, 224,
3150daa t 240, 113, 97,224,
19,
2.
0, 129, 142,
Fl
7,131, 160,
4, 97, 162,
4,129, 76.249,
10
97, 208,
1, 129, 141
4, 96, 96, 97, 169, 65, 129 2,
2510 daa t
4
0, 129
3090daa t 141,
FG
4,
0, 96, 96, 97, 169,
Gl
2500 daa t 140,
72
4, 96, 72, 129, 173
4,129
KF
4, 96,
4
4, 97
3080 data
0,
7,
8,
4,129,140, 14,
HG
2490 daa t 97, 160, 234. 129, 142,
4, 96, 104
8. 129
8, 98, 76,126,234 45, 15,
11'
4
4, 97,208,
3070 daa t 97,208, 2,
96,104,129,141, 10,
4, 96, 72,129, 173,
NG
CB
1'
5,
4,
8, 129, 13, 15
7, 97,169.
4, 97, 208,
9,
7,
6,
8 4
96, 96 3, 97
97, 208, 72
0, 97, 240, 12, 97
0, 129, 141, 13,
0, 97,169
MG
3180 daa t 160, 129, 141, 14,
HN
97,165, 55,129, 205, 11' 3190daa t
2
NJ
t 208, 33, 3200daa
2570data 97,160.
0, 97, 132,207, 98, 32
PA
3210data
GH
2580data 19,234,
96, 96, 97' 165,203, 97
FP
3220data 97, 134, 55,
AH
2590data201, 63,
97,176, 75, 97, 160,
0
OF
14, 3230 daa t
AN
2600data 96, 10, 96, 10, 96, 10, 97,144
NA
3240daa t 97, 166, 45, 97, 164, 46, 97, 134
LM
2610data
2, 96, 200, 96,200, 96, 10, 97
JE
NC
2620daa t 144,
2, 96,200, 96, 24,130, 105
NN
3260daa t 132, 50,129, 32,207'
KH
2630data 16,
0, 97,133, 245, 96, 152,131
GP
3270data 49, 97,160,234, 98,142, 20,
NJ
2640daa t 1 05, 16,
0, 97,133,246, 97,169
FL
3280daa t 98,140, 21'
OL
2650data
0, 98,174,141,
2, 97,240, 13
AB
3290 daa t 165, 203, 97, 224,
FB
2660data 96, 24, 97,105.
4, 96,202, 97
JH
3300data 97,197,197.
DA
2670daa t 240,
7, 97,105,
4, 96,202, 97
LK
3310data 63, 97,240,
LB
2680data240,
2, 97, 105,
4,
96, 168, 97
BC
3320daa t 208, 22,
EP
2690data177,245,
97,240, 27, 96,170, 97
LK
3330 daa t
JK
2700data 41, 63, 97. 133,203, 96, 138, 97
NE
3340data129,141, 15,
4,
NO
2710daa t 162,
0, 97, 41,192, 97,240, 12
ON
3350data129, 32, 113,
KA
2720 daa t 96, 232, 97,201'
64, 97, 240,
7
FP
3360 data129, 44, 15,
MK
2730data 96,232, 97,201' 128, 97,240,
2
oc
3370 daa t 236,142,
PK
2740data 96, 232, 96,232, 98,142,141,
2
CN
3380data
ED
2750data 96, 96, 96,200, 97,177,245,129
FL
3390 daa 77,138, t
2 , 98,141, 138,
DC
2760data 141, 16,
4, 96, 200, 97,177,245
OB
3400 daa t 76, 133,
7, 97,224,
NB
2770data129,141, 17'
4, 96,200, 97,177
EK
3410data
AH
2780daa t 245, 129,141, 18,
4, 96, 96, 97
JL
3420daa t 32, 235,
GE
2790daa t 169, 76,
3, 97,240
PK
3430daa t 164, 203, 97,169,
EJ
2800data 23.
98, 173, 17
FB
3440data
CL
2810data
3
Ll
3450data208, 16,
LM
2820 data 97, 133, 246, 97, 160, 35, 129, 185
BE
2830 daa t 152,
FH
2840daa t 16,248, 129, 173, 14,
PE
2850 daa t
IH
2860 data129, 44,
EF
2870data 48,
GJ
2880daa t 129,141,
PN
2890daa t 96,104, 96, 170, 96, 104, 129, 141
.,.... liGWGctor
98,205, 16,
98,141,
16,
3,
3, 97,133,245, 98, 173, 18,
GO Cl
9, 97' 145,245, 96, 136, 97
0,
3250data 47,
4,
0, 97, 208, 14 0, 97
97, 165, 56, 129,205, 12
97,208,
26,129, 174, 13,
0
97, 134, 51,129,172
0, 97, 132,
56, 97,132, 52
97, 132, 48, 97. 134,
49, 97
5, 97,162 3
3, 97,208, 34, 97 4, 97,208, 34
97,240, 30, 97,201 9, 97, 201'
51,
97
97, 133. 197,129, 76.
31
97,169, 128, 129, 77, 15,
4
98, 142, 142,
2
5, 129,
76, 73,
8
4, 97,
48,248, 98
2, 97,240,
23, 97,224
5, 97, 208, 11.
97,169, 128, 98 2, 129
6, 97, 208
4, 97,169, 32, 97,208,208,129 5, 129, 32, 61'
4, 97,240,
6, 97
8, 129, 45, 15
28, 97, 196,197, 97
97, 192, 64, 97, 240, 12
3460data129, 173, 20,
4,
34 70 daa t 152, 129, 206, 20,
97,240, 4,
6, 96
97,240,
22
DB
3480data 96, 96,129, 32,165,
1
CB
3490 daa t
4, 97,240, 83, 97 81, 97, 80, 5, 97,169, 0
LE
3500data201, 63, 97, 176. 106, 97, 197, 197
EP
3510data 97,240,102, 97, 36,157, 97, 16
4, 96, 104, 96, 168
EB
3520data 98,129, 32, 113,
3, 129, 32, 211'
4, 97,240
6, 97, 169,
2,
2,
IF 71
0, 129, 141, 20,
5, 97,169
4, 96,152, 97
5, 97,166,212
3530 daa t 97,208, 86,129, 174, 18,
4, 97
Nov. 1916: Vlalume 7,
111 â&#x20AC;˘ 03
[
32,201'
5
4190data 129, 76, 146,
8, 129, 76, 73,
8
PL
4200 data 129, 173, 17'
4, 97,133,245,129
1
BB
4210data 173,
4, 98,205
BN
4220data 129,141, 16,
34, 129,173, 10
GK
4230data
14, 16,
0, 97, 240, 29, 97,133,246,129
DL
4240data
16,
0, 97' 133,245, 97,169
CG
4250data 129,174, 19,
39
NM
4260data 112,
9, 97
JK
OL
3640data 145, 245, 96,136, 97, 16,246, 97
GK
4280data 141, 19,
Gl
3650data 162, 141, 98, 32, 38, 235,129, 32
GC
4290data 129, 76,214,
6,129, 76,228,
6
Nl
3660data 155,
5, 97, 164
ED
4300data 129, 76,252,
6, 129, 76,234,
6
2, 98, 142, 142
EF
4310data 129, 76,238,
6,129, 76,244,
6
15
NE
4320data 129,
76, 218,
5,129, 76,135,
5
3, 98, 76, 72,235
HG
4330data 129, 76,165,
5,129, 76,174,
5
97,165, 212, 97
FE
4340data 129, 76, 201'
5, 129, 76,235,
5
97' 165, 203, 97,208, 10
NH
4350data
08
3540data 240, 81' 129, 174, 16,
Kl
3550data
12, 129, 32,102,
HH
3560data
PC
3570data 184,
CH
3580data
GO
3590data 141
AE
3600data
CL
3610data 173,
CK
3620data
CK
3630data
OM
4180data
0
CF
97,177,245, 97,240, 67,129, 76 97,208, 41' •
9, 97,160,
8, 97,133,197, 97,201' 97,169,
2, 97,208, 9,
0,129, 141, 19,
4, 97,160,
97,177 , 2 09,129, 32, 113,
5, 129, 32, 218,
3670data 203, 98,174,141, 3680data
2,129,108, 17'
KK
3690data
4, 97,112,
3700data 129, 32,165,
5,
4, 129, 44,
98,206, 32, 208,129,
4270data
4, 97' 133,246, 96,
18,
2, 97,
4, 97
9,128, 97, 144, 4, 97, 208,
5
4, 97
9, 64, 97,201'
8,129, 77, 19,
97,208,
96, 120,
4, 97, 41' 63, 129
4,129, 44, 16,
2, 97,
96
34
4,129
4, 97,169, 34, 96, 96
97,169,
0, 97,162, 20
GE
3710data 133, 245,
HG
3720data
98,174,141,
96
HL
4360 data 129, 157,
HB
3730data
96, 97, 162, 29, 97,208, 23, 97
NO
4370data 250,129, 32,174,
CH
4380data
Cl
4390data 133, 99,129, 173,
KL
4400 data
98, 97, 162,144, 96, 56, 98, 32
IL
4410data
73,188, 98, 32,221' 189, 97, 160
4, 97, 197, 197, 97
JH
4420data
2, 97, 160, 32, 129, 140, 20
MH
4430data
2,
97, 240,
1'
PL
3740data 166, 212, 97,208, 12, 97, 201'
JO
3750data
PO
97, 240, 224, 97,201'
3760data 220, 97, 201'
51'
7
3, 97,144
97,240,216, 97 .
0,
4, 96, 202, 97, 16 5, 129, 32,201
5, 96, 88,129, 173, 11' 12,
0, 97
0, 97,133
DE
3770data 162, 148, 98, 44,138,
BH
3780data
EJ
3790 data 240,
CN
3800data
32, 31'
4, 98, 32, 38
HM
4440data 245, 97,169, 13, 98, 32,210,255
PO
3810data 235, 129, 32, 135,
5, 97,165,245
AB
4450 data
97,162,
FE
3820data
97,133,212, 97, 230,197, 98, 32
JJ
4460data
32, 210,255, 96, 232, 97,224, 41
HC
3830data
72, 235,129, 32,174,
co
4470data
97, 144,245, 96, 96, 112, 75, 69
MC
3840data 148,
2
KB
4480data
89, 66,
CP
3850data
2, 96
DO
4490data
88, 80, 65, 78, 68, 69, 82, 96
MH
3860data 200, 97,177,245, 97,133,203,129
CE
4500data
13,118, 40, 67, 41'
6, 129
EB
4510data
56, 54, 32,
4, 97, 240, 87, 97, 162
AE
4520data
89, 32, 83, 84, 65, 78, 76, 69
EE
4530data
89
11'
97, 160,
4,129,
5,
97, 169,
IM
3870data
AK
3880data 206, 16,
ME
3890 data
KJ
32, 235, 1'
97,201'
5, 129, 76
3, 97,208,
0, 98,141,141, 5,129, 32, 61'
97,208,
3900data 142, 19,
2, 97, 16
2, 97, 162,
0,129
1'
98, 185,
0,
98, 32, 210,255,
1'
97,240,
6
96,200, 97,208
0, 129,189,
1
•
10, 98
79, 65,. 82, 68, 32, 69 32, 49, 57
65, 85, 66, 82, 69
4, 97' 162, 255, 98, 142 9, 97,208
OG
3910data
3,220,129, 32,102,
Nl
3920data
3, 129, 32, 135,
10
3930data
97,201, 10, 97,176,247,129,172
HA
3940data
19,
EL
3950data
97,201,
KD
3960data
15,
EN
3970 data
97, 240,
6, 129, 32, 135,
FO
3980data
76, 224,
IN
I
MF
LP LL
4, 97,240
Keyboard Expander: Generates PRG file "SCROLL.OBJ"
5, 97,165,198
EK
1000rem programto create file
5, 97,144,192, 129, 44
08
1010 rem first loop ensures checksum is correct
4, 97, 80, 26, 97,165, 198
GP
1020rem second loop
5,129
FA
1030 for j
8, 97,133,212, 97,162
PO
1 040 ifch<>86049then print·checksumerror· : end
32, 135
CD
1050restore
5, 97' 165, 198, 97, 208,249, 97
MJ
1060open 8,8,8, O:scroll.obj, p ,w
4, 97, 177
81
1070 for j
38,235,129
IL
1080printN8, chr$x ( ); : next
4, 97
CJ
1090closeS :end
5, 97,165
AK
1100data
EP
1110data 1120data
0,
0,
0,
0,
0,
1130data
0,
0,
0,
0,
4, 97, 177,245, 97,240, 65
3990data 148, 98, 32,
38, 235,129,
AN
4000data
PA
4010data 133, 216, 129, 172, 19,
DB
4020data 245, 96,170, 98, 32,
OF
4030data 238, 19,
JC
4040data 208, 191, 129, 32,135,
4,129,206, 16,
=
scroll.obj on disk
•
•
writes file
1to 741 : read x :ch
ch +x : next
=
•
•
=
1to 2745 : read x
0, 157, 76, 48,157, 76, 64, 157
BG
4050data 198, 97' 208,249, 98, 141,
PC
4060data 129, 76, 146,
Cl
4070data
97,240,238,129,238, 19,
4, 129
oc 10
HL
4080data
32; 195,
98, 140,140,
2, 98
CE
1140data
0,
0,
0,
NC
4090data 238, 32, 208,129, 32, 135,
5, 97
ME
1150data
0,
0,
0,
FK
41 00data 165, 198,
3
KA
1160data
0,
0, 32, 88,157, 32, 176,158
FC
4110data 220, 129, 32, 135,
32, 104
GL
1170data
JF
4120 data
5
OM
1180data 208,
OF
4130data
97, 16,251'
98,206, 32,208,129
NO
1190data 169, 24,133, 214, 32, 108,229, 32
AL
4140data
32,201,
5, 129, 32, 31'
4,129
GO
1200data 243, 157,
EO
4150data
76,189,
8, 98, 76, 52,234,129
FF
1210data
FH
4160data
32,235,
5, 97' 165,203, 97,201
KB
1220 data 169, 255,141, 41,157,141, 42,157
Fl
4170data
11' 96, 104, 96, 104
OH
1230data 104, 24,105,
lMllaleCICb
5,
3,220
8,129, 206, 16,
97,208, 249, 98,141, 5, 129,
5, 97, 48, 17,129, 32,104,
1'
97,208,
4
72
0,
0
0,
0,
0
0,
0,
0,
0
0,
0,
0,
0,
0
0,
0,
0,
0,
0
76, 43,159, 76, 145,159,
32, 114, 157, 32,225,158,240,245 2;1 '
32, 88,157, 32,176, 158 32,225,158,240,238, 76
27,157,140, 33,157,142, 34, 157 1, 170,104,105,
0
Nov. 1916: Ylllume 7, 1 11ue 03
I
KA
1240 data168, 76, 18,157,162, 255,232,224
KF
1880 data162.
JO
1250data 25,240, 11,181,217, 16,247, 32
HK
1890data202, 16,247,169,
80
1260data124,158, 144,242,176,
6, 169,255
PJ
1900data169, 56, 141, 12.208, 169, 127, 141
KA
1270 data133,
20, 133, 21, 32, 19, 166, 164
AA
1910data 29,208,141, 21,208,162, 25,142
PE
1280 data 95, 228, 44,208,
43,208
JL
1920data 46,157, 96,169.
80
1290 data
5,166
Nl
1930data 76, 15, 157, 36, 157, 16,244,
LD
1300 data 44, 164, 43, 136, 134,
64,132, 63
PK
1940data207, 159,166,214,236, 46,157,240
LP
1310 data160,
4,200,208
88
1950data233, 142, 46,157,181,217, 48,
PP
1320 data249,
63, 197, 95. 208
HG
1960data202, 138, 10, 10, 10, 24,1OS, 49
EK
1330data245, 200, 177, 63, 197, 96, 208,238
JM
1970data162, 12.157,
EL
1340data 136,152, 24,101, 63,133, 95,165
IC
1980data249, 169,
IG
1350data 64,105,
0,133, 96,162, 24,181
JN
1990 data
DG
1360data217.
5, 162, 23, 32,255,233
HE
2000data159,173,134,
MC
1370data162,
9
CO
2010data 11,141, 47,157,162,
LH
1380data128.133,217,169, 39, 133,213,162
KE
2020data208, 202, 16,250, 96
N8
1390data
5, 196,
1, 96,202,228, 44, 176, 0,177,
63, 240,
96,200, 177,
48,
0, 32,104,233,165,217,
1, 142, 146,
8,157,
0. 208,233, 48,202 8,141, 10,208
0,141, 21,208 32 1
1.208,202.202. 16
0, 162, 40,228, 213, 176
2,169, 127, 141, 23,208, 76, 144 2,205, 47,157,240 6,157, 39
2,202,134,214, 32
18
1400data240, 233, 76, 68, 158, 162, 25,202
KJ
1410data 48, 11, 181,217, 16,249, 32,124
FN
1420 data 158, 144,244, 176,
EG
1430data 20,133, 21, 72,208, 21,224, 24
CK
1440 data208,
4, 169, 13, 208, 245,224, 23
ON
10 rem -------------------------------
DA
1450data208,
5,232,181,217, 16,243,169
HC
12 rem
CD
1460data255,208,232,230, 20,208,
BM
14 rem
HO
1470data 21, 32, 19,166,165,218, 48,
CL
1480data
NO
1490data233,162,
lA
1500 data104, 48,
AJ
1510data177, 95,240, 37,141,
PN
1520data
9, 169,255, 133
9,128,133,218,162,
Keyboard Expander: Profile Program
2,230 9
EO IP
1, 32,255
keyboard expander profile prog
ke.pf 16 rem ------------------------------18 rem (prog a r m variables)
CH
20 zx = O: zy =O:zd =O:zi=O:zj =O : z b =O:zp= 0
1
ED
22 rem
35,157,169
EA
24 rem ( a f ctor for repositioned keys)
0. 145, 95, 132, 15,165, 95.133
Fl
MJ
1530data 63,165, 96,133, 64,169,255,133
KD
28 rem
JP
1540data 20. 133.
FA
30 rem (ar raydim e n sions sou r ce keys)
PC
1550data240, 13,173, 35,157,160,
1,145
GH
32 nd =O:sd =1:c d =2:td=3:rem •norm,s htf ,c= ,ct r l
OF
1560data 63, 169, 13, 32,210,255, 96,104
AE
34 rem
GO
1570data104,104,104, 76,111,158, 32,240
DL
36 rem (used in s t rings o f r keys)
KE
1580 data233, 160,255,200,177,209, 192,
39
EE
38 rem
LJ
1590 data240.
48
EN
40 s$ = ··:I$= chr$(0):i$= chr$(0):rem new,link and ia ' ctive
AF
1600data144, 30,201, 58,176, 26,152, 24
HG
42 q$ = chr$(34):r$ = chr$(13):n$ = chr$(141 ):rem q uote,
MC
1610 data101,209, 133, 122, 165,210, 105,
LF
1620data133,123, 32, 121,
8N
1630data 32,107,169,174,
ON
1640data 24, 96,120,173,
LJ
1650data 157,173,
Gl
1660data100,141,
8K
1670data
HL
1680data165,214,141, 37,157,165,211.141
HN
1690data 38,157, 169,
HN
1700 data 21, 157, 96, 120. 169,
0, 141, 146
LA
1710data
0,
AH
1720data 40,157,141,
GE
1730data204,174, 37,157,134,214,173, 38
MO
1740data157,133,211, 56,233, 40,144.
24, 134,214, 32,240,233 3, 32, 210,255, 160,
21. 76,215, 166,224, 11
38,201. 32.240, 245, 201.
26 sf=64:cf= 128:tf = 192:rem s htf ,c=.ctrl
e r t,shf t +re t
o
0,142, 36,157
LH
36,157, 56, 36
44 In$ = chr$(3):s 1 $ = chr$(1 ):l c$ = chr$(2):1t $ = chr$(4) :rem norm,s hft,c=,ct r l
3,141, 39
ME
46 rem
1,
3,141, 40,157.169
BH
f r keys) 48 rem (variables o
0,
3, 169,158,141,
ON
50 a=10:b=28:c=20:d=18:e=14:f =21:g=26:h=29
0,
1
:i=33:j=34:k =37:1 = 42:m =36
3, 32, 30,157,169,255,133,204 PM
0,133,211, 88, 32
2, 173, 39,157,141, 1,
3,169,
52 n =39:o=38:p=41 :q=62: r =17:s= 13:t =22:u=30 : v=31:w=9 :x=23:y=25:z= 12
GJ
3,173
54 nO=35:n1= 56:n2 =59:n3 =8:n4=11 :n5=16:n6=19 :n7 = 24:n8 =27: n 9=32
0,133
EG
56 f1 =4:f3=5:f5=6:f7 =3:de=O:re=1:ho=51:ri=2 :do=7: ua=54:1a=57
2
MH
58 p= i 40:mi= 43:eq = 53:po=48:at =46:as =49:sl =55
El
1750data133, 211, 32,108,229, 88,172, 33
JA
60 pe = 44:co=45:se= 50:cm =47:sp= 60
8A
1760data 157,174, 34,157, 196.197,208,
3
MF
62 rem
EO
1770data236,142,
2, 96,165,207,240, 12
LN
64 rem Ump ve ctor and command line)
JP
1780data 165,206, 17 4, 135,
0. 132
IG
66 zv=O:zc=O:rem add resses, 0= none
PF
1790data207, 32, 19,234, 96,162,153,160
CG
68 rem
GO
1800data159, 32, 12,157,208, 92,169,
0
MM
t on ar ray s torage) 70 rem (keydefini i
HM
1810data141, 46,157,141, 47,157,162, 63
EN
72dim zs$(3,62).zp(3,62).z r(3,62)
PE
1820data 157, 64,
2
IG
74 rem
JG
1830 data 169,255,157, 64,
3, 202, 16,250
DP
AE
1840data162,
2,157, 91,
3,202, 16,250
0 1
JD
1850data162,
6,169, 13,157,248,
2,160.
3,202, 16,250,162,
76 rem-----------------------------78 remdefine y o u r keyboard p rofile
7,202
KH
80 rem82 rem
Ll
1860 data 16,250, 173, 134,
2, 32,215, 159
ED
FA
1870data169, 96,141, 16,208, 56,169,216
LP n
betwee n lines 100 and 799
•• dont de f any new variables 84 rem-----------------------------••
Nov. 1986: Yolume 1, ...,. 03
I
EH
86rem
00
800 rem 802 rem
AE
CE Ll
(relocate tables)
OM
804 rem 806 zi=fre(O):zp=peek(55)+256•peek(56):rem prog start 808 zb = peek(zp+ 7)+256•peek(zp+8):rem dscr base 810 ifpeek(zp+3)=75andpeek(zp+4)=69and peek(zp + 5) = 88andpeek(zp +6) =80then814 812 print· error: cannot locate keyboard expander·:end
MJ
814 zx= (peek(47) +256•peek(48)) + 9:rem array defs
MM
816 forzi = Oto62:forzj Oto3 818 zd =zb +(zi•16)+(zj•4) 820 ifzp(zj,zi)thenpokezd.zp(zj,zi) 822 zy zr(zj,zi):ifzythenpokezd+2,zy-int(zy/256)•256
GB AP
LL OK HB NB GC KA 10 OF
HC CD JE
IE JM PF
DP
130 zs$(td,r)=s$+ "load"+q$+ ·o:·+i$+q$ + • ,8 • +r$+ ·run·+ r$:rem load-run
PL
132 zs$(td,i)=s$+ "open15,8,15,"+q$+"iO"+q$ + • :close15 +r$:rem init disk •
MN
•
HJ MO
CK
=
ON
FM AH EC AG BM DB PH OB GE
:new·+r$:rem test part 140 zs$(td,mi)= s$ + poke45,peek(43):poke46,peek(44): • 142 zs$(td,mi) =zs$(td,mi) + 'poke43, 1:poke44,8:clr "+r$ •
·
OK
=
:pokezd+3,int(zy/256) 824 iflen(zs$(zj,zi))=Othen830 826 zy=zx +(zi•12) +(zj•3)
FJ BL HF
828 pokezd+1,peek(zy):pokezd +2,peek(zy +1) :pokezd+ 3,peek(zy+2) 830 nextzj:nextzi 832 rem 834 ifzvthenpoke785,zv- int(zv/256)•256
DO
:poke786,int(zv/256):poke784,0 836 ifzcthenpokezb-7,zc-int(zc/256)•256 :po kezb- 6, int(zc/256) 838 zy =(peek(51)+256•peek(52))-9:rem new start 840 forzi=Oto8:pokezy+ zi,peek(zp +zi):next: pokezb -1,255 842 pokezb-5,zy - int(zy/256)•256:pokezb-4,int(zy/256) 844 poke55,peek(zb- 5):poke56,peek(zb- 4):clr
EG
:rem back to norm 144 fem 146 rem (basic progs on strings) 147 rem 148 rem read error channel 150 zs$(td,e) =s$ + • 8000open15,8,15:inputN15,a,b$,c,d :printa, b$,c,d 152 zs$(td,e)=zs$(td,e) + • :close15 +r$+ run8000 +r$+ "8ooo·+r$ •
AK GP
PL FP
BH
:rem new top mem MG
134 zs$(td,v)=s$+ "open15,8,15, • +q$+·va· +q$ + • :close15 +r$:rem validate 136 zs$(td,g)= s$+ "loac spcs],8 • +r$ :rem load from dir 138 zs$(td,pl)=s$+ • clr:poke43,peek(45):poke44,peek(46)
•
JE OH NA EK
•
•
858 input·filename for profile' ;I$ 860 sys57812f$,8 862 poke174,0:poke175, 160:poke193,peek(55) :poke194, peek(56):sys62954
Bl NN OE
864 clr:end
MJ
+q$+ '$0"+q$+ ':getN15,a$,b$"+r$ 158 zs$(td,d)=zs$(td,d)+ • 8001getN15,a$,b$:ifst<>O then8005 • + r$ 160 zs$(td,d) zs$(td,d)+ • 8002getN15,a$,b$ :printasc(a$+ n$) + asc(b$ +n$)•256; • +r$ =
162 zs$(td,d) zs$(td,d) + • 8003getN15,a$ :ita$= + q$+q $+ thenprint:goto8006 +r$ 164 zs$(td,d) =zs$(td,d)+In$+ chr$(a) 166 zs$(nd,a)=I$+·8004printa$;:goto8003 +r$ 168 zs$(nd,a)=zs$(nd,a) + 8005close15:poke198,0 :end"+r$ 170 zs$(nd,a)=zs$(nd,a) + • 8006k peek(203) :ilk= 64then8001 • +r$ 172 zs$(nd,a) zs$(nd,a) + • 8007ifk =62then8005 +r$ 17 4 zs$(nd,a)=zs$(nd,a}+ 8008goto8006 • +r$ + • run8000 • +r$ =
•
Keyboard Expander: Profile Definitions
10 OK LK 00 Kl PO 01
IH CJ OF GJ
IB
FF
100rem------------------------------1 02 rem keypower profile definitions kpower.defs 104 rem 106rem ---------- --------------- ------
•
=
=-
•
•
176 rem delete basic lines = s$+" 8000"+r$+" 178 +r$ 30012 +r$ +
GN KJ JN Kl
112 rem 114 zp(nd,ua)=do+ sf:zp(nd,la) =ri +sf 116 rem
GO EP
124 zs$(td.p)=s$+ "open15,8,15, +q$+ "so:· +i$ + q$ + :close15 • +r$:rem purge
•
• +r$+" 1 • +r$
74
(some basic keywords)
188 zs$(cd,a)= s$+ asc( • :zs$(cd,c) =s$+ • chr$( • :zs$(cd,d) =s$+·data· 190 zs$(cd,f)=s$ +·for" :zs$(cd,g) =s$+ • goto • :zs$(cd.�= s$+ ·input· 192 zs$(cd,l) =s$+ "list· :zs$(cd,m) =s$+·mid$(· •
LD
208 rem (interrupt routines)
10
•
3009"
206 rem
DG
126 zs$(td,n) =s$ + • open15,8, 15, • + q$+ • rO: • + i$ + • = +i$+q$+":close15" +r$:rem rename 128 zs$(td,l)=s$+ "load·+ q$+ • 0: • +i$ + q$+ ,8 +r$:rem load
•
MO
EO
•
184 rem 186 rem 187 rem
+r$+ '+r$
•
Cl
•
OK
•
LD
•
•
PJ
= zs$(td,z)+ • +r$+ +r$+ • 182 zs$(td, = zs$(td,z)+ • +r$+ o· +r$+
:zs$(cd,n)=s$+ "next' 194 zs$(cd,o) =s$+ ·open ·:zs$(cd,p)= s$+ ·print· :zs$(cd, r) =s$+ • right$( 196 zs$(cd,s)= s$+ • str$( 198 rem 200 rem (imp table and command line) 202 rem 204 zv=40204:rem zc =0 no command line•
JF
•
FF
•
•
•
108 rem 110 rem (reposition shift cursor keys)
118 rem (commands on strings) 120 rem 122 zs$(td,s) = s$+ ·save" +q$ + • 0: ' +i$+q$ + ',8 +r$:rem save
•
•
•
GF
•
154 tem list dir to screen 156 zs$(td,d) = s$+ • 8000n$=chr$(0):open15,8,0,'
•
846 rem 848 rem (store absolute version) 850 rem 852 input' create an absolute version - y/n' ;1$ 854 iff$<> "y andf$<> • n" then852 856 iff$ = 'n then864
•
•
Nov. 1M6: � 7, ..... 03
AP
210 rem
EM
212 zr(nd,fS)= 40192:zr(nd,f7)
=
40195
:rem list scroll down/up
IB
214 z r(sd,fS) = 40198:zr(cd,f5)
40201
=
:rem line rule enable/disable
GP
216 rem
Nl
218 rem end of defs
EO OK GL
install 14 rem 16rem----------------------20 poke55,0:poke56,32:clr
MP MO PO EO JJ
26 inputN15,a,b$,c,d
•
•
=
Othen34
30 printa,b$,c,d:end 34 gosub44:a
1 :loadf$,8,1 •
36 if a$<>· kexp anda$<> in it· •
then40
FO
38 sysad
AJ IE
40 poke55,1b:poke56,hb:new 42rem •
44 open1,8,2,1$ +
NC KP
48 lb = asc(x$) 50 getN 1 ,x$:ifx$=
co
52 hb"' asc(x$):ad= lb+ 256•hb
46 get# 1,x$:ifx$=
•
,p,r" •
• •
thenx$= chr$(0) thenx$ = chr$(0)
FF
54 fori .. 1to3:getN1,x$:next
JA
56 a$=
• •
:fori= 1to4:get#1 ,x$
:a$= a$+ x$ 58 next:close1 :return
Keyboard Expander: U.t Seroll Source FN NL
OP
IN
MA IL PN
BC AO AN
LK
NN ON
00 FM
GJ AJ BA
HJ
GP HO GO
MM
LL
IJ KK
1 r9C'Tl -------------------·---------2 r em list scroll and line rule scroll.s 3 r em
4 (8fT\--------------------·-·---------10 open 2.8. 1, O:scroll.obj 12 rem open 4,4 •
14 sys 700 16.opto2 18·--20msgflg •
221s1x
24 blnsw
26gdbln
28 blnon 30 tblx 321dtb1
34 IXcadr
36gdcol 381stshf
40;---
-
-
-
-
-
42.- $9d00 44 ;:np
46
•
157 t97
204 206
207
214 217
;run mode ;last keypress :blink sw
;linell ;screen line links
646
;col for text
654
;last shift
647
listup listdn
;:np jmp lrule jmp lrkill
48
;install address
:scroll up :scroll down ;line ruler
;kill line ruler 50 52 . NO 54 ;vector tab lor background rtns EG 56 :set up by profile program 00 58 brstart • - . +3 ;start int rtn - . +3 ;end int rtn OF 60brend PM 62 ppstart • - • +3 ;start parallel rtn ;run excluSIVe CJ 64 ppexd • -.+3 JF 66 ppshare • • •+3 ;run shared OM 68ppend ;end parallel rtn • - • +3 ;reset cursor 70 rescur • - • +3 IH •
•
I
lhelluiNdllt
LB AD
LG
MP KN
OM MP
KH
ME MO NM
92savec 94 savev
-
= =
-
=
KD NN NA FK EO
skey+l sshf+1 savel+1 savex+1 savec+2 savev+2 firstI + 1
LL
PI 00
HM
96 firstI 98 first2 100: 102 ;---entry from interrupt • 1041ist u p :scroll up 106 Iiston jsr I up jsr 108 list in 110 lstup JSr listou1 112 jsr 114 beq lup 116 bne lexit 118; =
GB AH
120 ;---entryfrom interrupt
OE Gl IM KB
KJ
FP AO
KJ
FO HF PJ
DP
AC
GG
• 1221is1dn :scroll down JO Iiston js r OA 124 FG listin LC 1261dn jsr BF LA 128 Ida 1124 KF GH 130 sta $d6 DC jsr BM 132 $e56c NG ls td n HK 134 DO jsr lis t o ut jsr FO 136 PO PN 138 HF beq ldn GO 140 lexit AB jmp ppend . E A 142 EJ IL 144 :---initiate paraUel program GL Cl 1461iston :common point for upldn BE • LM GE 148 :save entry key sty skey SIX sshf 150 AD EK ; . ..and shift value Ida 1$11 152 HF FF A B 154 sta first1 NB Sta first2 EB 156 DB AN 158 ;ret addr • start of parallel prg NK pia GB 160 clc DC 162 Bl adc 111 NH PI tax 164 MC :low byte pia oc 1 66 MM 168 Fl adc 110 BH :high byte HK 170 LB tay AM DE 172 jmp ppstart :start parallel program EC 174; co DP 176 :---list up PA ML O E 1781stup ldx 1$11 AB NM 180 NB 1821u04 inx FB Nl 184 125 ;all lines cpx JB lu12 CK 186 FF beq 01 188 Ida $d9,x IC PO 190 bpi :linked lu04 KE gnum :getnum OL 192 HD jsr PA bee lu04 :none NF 194 bcs lu16 OL 196 BJ MD 198: HC Ida l$ff Jl LN 2001u12 ;high line n um sta $14 CJ 202 BO HJ 204 sta $15 JC FJ 2061u16 $8613 ;fond line GB JSr ME 208 BM ;lo by1 ldy $51 ;hi byt MK NO 210 cpx $2c b ne lu20 LE CK 212 =
•
BF GP
LL
88 savel 90savex
KC
28 close 1 :close15:ifa =
BC
PO
24 open15,8,15:open1,8,2,t$+ ,p,r
•
86 sshf
FN NM
18 ita= 1 then36 22 input· filename· ;f$
;remaining jmps
76; -----------------------78 ;the list scroll routines
NJ
LL
general install program
LO PE
• - • + 15
80: JH 82 :run time variables bgwofk GC 84skey
AE
10 rem ----------------------12 rem
GN OK GM
72 bgwork 74 .
=
Keyboard Expander: ·INSTALL· Program ON LJ JH
ML AM
=
JE GK
IN CF JC AK
HN
NH MP OK LJ
OE JE MJ
MA JG EO
•
214
cpy $2b
,With start
218 220;
rts
:no prev lines
dex
;lower
216
2221u20
224 226 228
230
232 2341u24
236
238 240 lu28
242
244lu32 246
bne lu20
cpx
bcs
ldx ldy dey siX sty ldy
Ida
beq
ony
$2c lu24
$2c $2b
;..than start
:correct
$31 10
:search lor
lu36
;end of line
bne lu28
71
BN
NO
IL 00 NP KB
OL FK
OA IG
$40
($3Q,y
LJ
IJ LO LL BJ
GH
BJ
248 250:
2521u36 254 256 258
rts
;no more
iny
;chk
bne lu32
;no match
Ida ($3Q,y anp $60
·chk ;link hi
Ida ($3Q,y anp $51 iny
260 262
264 266
bne lu32
268
dey
270 272 274 276 278 280 282
a de $31
sta Ida adc sta
ldx I da
288
bmi
290
ldx
292 2941u40 296
jsr ldx jsr
302 304
ora sta Ida sta
298 300 306 308 310 312
Ida
ldx
stx
dex
314
SIX
316 318
j r s jmp
320:
322 :---list down 3241s1dn 326 ldx 3281d04 dex 330 bmi Ida 332 334 bpi 336 jsr 338 340
342: 3441 d08 346
348 350 ld12
352 3541 d16 356 3581d18 360 3621d20 364 366
368 370
3721d24 374
3761d30 378
380
3821d32
384 386 388 390
$60 1124
$d9,x lu40 123
$e9ff
;chk last line :not linked ;2nd last ;erase
10
;to scrl dwn
$d9 1$80
;fix up link
$e968
$d9 1139
;and length
111
;set insrt
$d6
·horne
$d5
$292
$e9IO list
•
;position
•
125
;all lines
gnum
bee ld04
bcs
ld16
;linked ;get line num ;none
Ida
l$ff
;set
sta
sta
pha
$d9, x ld04
$14 $1 5
bne ld30 cpx 124 bne ld20
Ida
l$0d
bne ld12 cpx 123 bne inx
Ida bpi Ida bne inc bne inc
JSr Ida
ldx
jsr ldx
SIX
jsr
pia bmi
408: 4 10 :--4121ist
420 422
110
ld08
sta
3961d36 398
416 4 18
$51 $40
ld24
jsr
-
ldy
;..for
; .. 1stline :save a byte ·last •
;ret char ;2nd last
$d9,x ld18
ll$ff
ld12 $14 ld32 $15
$8613
$da
bmi ld36 ora 1$80
392 394
414
•
;makematch
tya clc
284 286
400 4 02 404 406
;link lo
$da 11
;Inurn ;find line ;2nd line ;not finked
;unlink it
1124
;erase line ;last line
$e910
;set curs
list $11d2
;printret
$e9ff $d6
:curs row
•
11 ($5Q,y
Ida beq ls24 sta savel Ida 10
Ncw.l916:�7.11-03
I
IM PC EH AL 10 EH HO OH 01 AH CD
_428 430 432
434 436 438 440 442 444 ;
JO
446;--4481serv
JK
450
M1
EE
LD FB NB KM HG HP
pp
482 484 gn10
470 472
CD KF LL BG
494 496
AJ JK EN
498 500
508 510 512
514
CA NE
516 5 18
OE IN EB PI
520 522 524
OP Kl
526 528 gn20 530 532;
EO LG OL
534;--5361istin
CJ JG CG lA KG OM
540:
EO
552 554 556
KK
EO
sta j mp
a
OM OL
;to max
EA ON KM GO KN PH
$14 $15
538 542 serv 544 546 548 550
-
jsr ldy iny Ida cpy beq
NF FP
626 628
;restore
JA
630
10
sta
;ret
FG IM
632 634
636 rcu10
jsr
;end
EP IH
638; 640 getkey
AB IG DO AL FE AH JO GA FG PB
642 644 646 648 650 652 654 gk40 656; 658;---not 660 rcurs 662 664 666 668 670 672
$ffd2
;clr rtss
ls24
;get line num ;set line start addr
($d1),y H39 gn20+ 1 ;end of line
cmp ll$3a bcs gn20+ 1 tya clc ade $d1 sta $7a Ida $d2 a de HO sta $7b jsr $79 savex stx jsr $a96b savex ldx sec bit $18
IE KF KL GG JO lA ED BP
bee gn20+ 1
;chrgot
AD MP
00
•
Kl KH OK NA Jl
savev+ 1
BC
Ida
ll<lserv
FJ
sta Ida
$300 ll>lserv $301
706 708 710 : 712 714 716 718 720 lr1 0
jsr rts
Seat3
;reset blink
� = � = =
832
;sprite data area
;sprite register 53248 ;sprite variables 2040 bgwork+ 13 bgwork + 14 •
ldx H<ruler ldy H>ruler JSr brstart bne lr30
;entry from interrupt ;start addr of bg rtn
;column
HK MN FB
752 754 756
EN
;run exclusively
IK NC
758 760 762
CM MB 01
582 584 586 588: 590;--5921istout
LO PO OH
594 596 598
PJ
•
CJ NH IF HF
HO $292
Fl PK NL
sta ldx
spr + 21 :enable 1125
796 7981r30
stx
lrspyl
800
.
11$71 spr+29 ;exp x
rts
'
802 ;---kill line rule • 804 lrkill -
sta
AA
810
jmp
spr + 21 ;dsabl brend ;finish
CK MP CL EH BL MK
812; 814·--816 ruler
AC OF FB JC
pp 01 10 KD MD IM EO FO
JG BB DB OF
'
=
•
818
bit
820 822; 8241r40
bpi
msgflg Irkill
826 828 830
jsr ldx cp x
832 834 836 838 840 8421r42 844 846 848 850 852 854 8561r44 858 860
LE BE
862 864 866
IE OG IP
868 870 872
AH EO
;start background rtn
NF
;resource is used!
FJ
Ida sta
HO lrspyl
sta
ldx
lrspyc H$31
sta
sprt,x
LC OH KL PM AG HO
lr1 0 112
NH KA MP
ldx Ida
NN OF
792 794
;s6 X
808
00
:make sprite
786 788 790
spr+10 1156 spr+12
110
PJ =
746
savec+1 110
1M Ttaii6CCtur
;clear blink
dex bpi
NO CH
;clr scroll
110 blnon
;s5 X
sta Ida sta Ida sta
Ida
oc
ldy sty
lr24 118
806
EP MB
used: same as rescur � • ;reset cursor blnon ;check blink Ida beq rc10 gdbln Ida ;orig char ; orig colr ldx gdcol
dex bpi Ida
BB PH
GB
740 742 744.
$d6 savee $d3
sei Ida sta
skey sshf lstx gk40 lstshf
LJ
Ida
�
•
736 7381r14
570 572 574 57 6 578 580
GO
;fix curs
LC CA
AP MP LO CM NC NB
ppexcl
;not on 2nd
JB OJ NM
rescur 11255 blnsw
EP
rcu10 $d3 $e56c
Ida 11$11 sta sprt,x dex bpi lr12 112 ldx Sta sprt+27,x
jsr Ida sta
$d3
1140
728 730 lr12 732 734
564 566 568
sta Ida sta eli jsr rts
HB
dex bpi ldx
sta
sta Ida
cpy bne cpx rts
DL EA 10
FO
eli ldy ldx
00
EJ GJ
$d6 savec+ 1 $d3
· '
558 560; 562 scur
;reset cursor ;set blink off
�
110 blnsw sa vee
722 724 726
LL
;save cursor
bee
;rst cursor ;set blink on
•
'
CL GK PK FM IF FE
•
782 784
686 688 sprt 690 spr 692 scr 694 lrspyl 6961rspyc 698 700 lrule 702 704
Ida sta Ida sta
�
IL HK BM
EC PN CP GL BN
•
$300 savev $301
$300 savev+ 1 $301
682; 684 ;basic line ruler routine
DC AE FN MD
:set err vee!
778 780
AC
OK
sei
savev
EA HA
;rst err vect
•
678; 680;·························
ML ;eva! num to binary
sta sec sbc
674 676 rc10
776
MB CH
NL
rts
=
Ida sta ldx stx Ida
;err from ;.. evaI fxd pt
622 624
PL
EK
a
:err vee entry
cmp H$20 beq gn10 cmp 11$30
=
608 610 612; 614 rcur 616 618 620
Ida sta Ida sta
KE OL EL
•
$e9f0 H$11
606
-
;into list
$a6d7
•
600; 602 rerv 604
NO HH
pla:pla pla:pla jmp
502 504 506
PN
NF
Ida sta Ida sta
;to fool list ;list qote fig
'
486 488 490 492
JP KK
$31 $60 $40 H$11
jsr rts
474. 476;--478 gnum 48 0
FJ
sta
462 464
AF KK LK IH
PK AH
($5f),y $1 $51
456 458 460 ls24
452 454
466; 4681s32
GP
sta sty Ida
cpx H$0b beq ls32 save! Ida ldy 1/1 ($3f),y sta Ida ll$0d
IE NK CN LO
AD KN
I
4 24 426
EG
�
;position
lrcolr tblx lrspyl beq lr30 stx lrspy1 Ida ldtb1 ,X bmi lr42 dex
txa asl asl
a a
asl a etc adc 1149 ldx sta
111 2 spr+1,x ;y pos
dex dex bpi Ida ldx cpx bcs Ida
8741r46 876 878·
•
;run
sta
jmp
lr44
110 1140 $d5 lr46 11$71
;ma x
spr+23 ;y exp lr30
'
880;--8821rcolr 884 886 888
-
•
890 lr50 892 894 1 r52
Ida txcadr cmp lrspyc beq lr58 sta lrspyc l dx H6 sta spr + 39,x;color
896 898 9001r58
dex bpi rts
lr52
902: 904. end
l r1 4
'
748 750 lr20
764 766 768
. 770 lr24 772 774
sta
dex bpi Ida jsr Ida sta sec Ida ldx sta sbc dex
116 1113 scr,x
;ptrs
lr20
txcadr lr50 ;color 11$60 spr+16 ;msb x 11216 118 spr,x 1148
76
;xpos
Nov. 1916: Wume 7, tnue 03
I
NewsBRK Submitting NEWS BRK Press Releases
Allow 2 to 6 Weeks
If you have a press release which you would like to submit for the NEWS BRK column, make sure that the computer or device for which the product is intended is prominently noted. We receive hundreds of press releases for each issue, and ones whose intended readership is not clear must unfortunately go
When a mail order arrives, it's processed and prepared within 5 business days. But we find it takes about 1-2 weeks for the post office to deliver it. Coupled with the fact that it takes i-2 weeks for the order to get from you to us means it will probably take somewhere between 2 and 6 weeks for an item to arrive from the day you mail the order. Please bear with us.
straight to the trash bin. It should also be mentioned here that we only print product releases which are in some way applicable to Commodore equipment. News of events such as computer shows should be received at least 6 months in advance.
Transactor Mall Order News
Our mail-order department is expanding nicely, but our mail-order card isn't. Seems we just can't find any more room to put more text without making it so
Transactor News
small that you can't read it. So, if you're using the card to order, we suggest you Transactor Writer's Guide Finally Finished
pull it out and cross-reference with the list below for more details.
That's right! After 3 years of collecting, compiling, re-arranging, and generally
• Transactor T-Shirts, $13.95 and $17.95
ensuring completeness, The T. Writer's Guide is done. We kept all those
As mentioned earlier, they come in Small, Medium, Large, Extra Large, and Jumbo. They're 13.95 each, $17.95 for the Jumbo. The Jumbo makes a good
requests in a file and have sent out about 200 so far. If you would like one, they're free for the asking. Call or write the office in Milton, Ontario.
night-shirt/beach-top - it's BIG. I'm 6 foot tall, and weigh in at a slim ISO pound s - the Small fits me tight, but that's how I like them. If you don't, we
Subscription Timing
suggest you order them I size over what you usally buy. The design is screened using a "super-opaquing" process so they should wear much longer than your
Remember, if we receive your subscription order within two weeks of the release of an issue, your subscription won't start until the next issue. The mail
ordinary screens and iron-ons.
list is sorted and printed about two weeks before each release date. All the mail is processed, sometimes the "morning of", the day we print up the list. If your subscription order comes in the day after, it won't appear on the "big list", which means it won't be sorted into our mailing at second class postage rates.
• The Transactor Book of Bits and Pieces "I, SI 4.95
Not counting the Table of Contents, the Index, and title pages it's 246 pages of Bits and Pieces from issues of The Transactor, Volumes 4 through 6. Even if you ,
have all those issues, it makes a handy reference - no more flipping through magazines for that one bit that you just know is somewhere ... Also, each item
We can't send out mags at first class postage rates, so if you sent your sub 2-3 weeks before a release date. and your Transactor doesn't show within 2-3
is forward/reverse referenced. Occassionally the items in the Bits column appeared as updates to previous bits. Bits that were similar in nature are also
weeks after the release date, you should probably buy one more off the news stand until your subscription gets "engaged".
cross-referenced. And the index makes it even easier to find those quick facts that eliminate a lot of wheel re-inventing.
Free Transactor rs with Mag+ Disk Subscription • The Tr@nS@Ctor 154I ROM Upgrades, $49.95
For a limited time only, subscribe or renew to a combination magazine and disk
You can burn your own using the ROM dump file on Transactor Disk 113, or
subscription, and we'll send you a free Transactor T-Shirt! You save 29% off the magazines, 16% off the disks, and get a Transactor T worth $I 3.95 ($I 7.95 if
you can get a set from us. This is an introductory offer to see how much response we get - we only have 50 sets available. But if they become popular, we'll certainly get more in, and the price may go up.There are 2 ROMs per set, and they fix not only the SAVE@ bug, but a number of other bugs too (as
you order the jumbo size!) The T-Shirts come in 5 sizes (red only), with a 3-color screen featuring Duke, our mascot, dressed in a snappy white tux, standing behind the Transactor logo done in yellow with black "3-D" borders. The
described in P.A. Slaymaker's article, Vol 7, Issue 02). Remember, if SAVE@ is
screen was done using a special "super-opaquing" process that cost us quite a bit more than those decals that crack and fade.Mine has been through the wash
about to fail on you, then Scratch and Save may just clobber you too. This hasn't been proven I 00%, but these ROMs will eliminate any possibilities short
three times and so far shows n o signs of "machine punishment" like my other Ts have.
of deliberately causing them (ie. allocating or opening direct access buffers before the Save).
Transactor Disk Price Increase
• The Micro Sleuth: C64/154 I Test Cartridge, $79.95 US., $99.95 Cdn.
This cartridge, designed by Brian Steele (a service technician for several schools in southern Ontario), will test the RAM of a C64 even if the machine is
A subscription to 6 Transactor Disks remains at $45.00. However, the price of single order Transactor Disks has been increased from $7.95 to $8.95 each -
too sick to run a program! The cartridge takes complete control of the machine.
another good reason to take advantage of the above offer!
It tests all RAM in one mode, all ROM in another mode, and puts up a menu with the following choices:
Refund Polley
I ) Check drive speed Should any product you order be defective on receipt, return it and we'll send you another for no additional charge. Recently we've had a few items returned
2) Check drive alignment 3) I 541 Serial test 4) C64 serial test 5) Joystick port I test
because "it's not quite what I wanted". We will credit your account (less shipping and handling) for purchases of other Transactor products, but we ask that you please be sure you need things like G-Links or RAM boards since we can't refund your money. While we're on the subject, although we've never had
6) Joystick port 2 test 7) Cassette port test 8) User port test
a subscriber ask for one, there are no refunds on subscriptions. 77
No¥.1916: Vc*lrne 7, ..... 03
A second board, that plugs onto the User Port, contains 8 LEOs that lets you zero in on the faulty chip. Complete with manual. Note: This is an introductory offer - prices may go up by next issue.
• Inner Space Anthology $14.95 This is our ever popular Complete Commodore Inner Space Anthology. Even after a year and a half, we still get inquiries about its contents. Briefly, The
GUNK is "transparent" - that means it won't interfere with programs, except those that rely on the serial routines which it replaces (ie. programs with built· in "lastloaders" lor the 1541 won't fike the presence of the GUNK). It has no manual (aside from one page of installation instructions) because it alters nothing and leaves everything unchanged! An on-board switch allows you to select Serial or IEEE. GUNK works with both the C64 and the Cl28 in 64 mode, but not on the VIC 20.
Anthology is a reference book- it has no "reading" material (ie. "paragraphs"). In 122 compact pages, there are memory maps for 5 CBM computers, 3 Disk Drives, and maps of COMAL; summaries of BASIC commands, Assembler and MLM commands, and Wordprocessor and Spreadsheet commands. Machine Language codes and modes are summarized, as well as entry points to ROM routines. There are sections on Music, Graphics, Network and BBS phone numbers, Computer Clubs, Hardware, unit-to-unit conversions, plus much
• The TransBASIC Disk $9.95 This is the complete collection of every TransBASIC module ever published up to Volume 7,1ssue01. There are over 120 commands at your disposal. You pick the ones you want to use, and in any combination! It's so simple that a summary of instructions fits right on the disk label. The manual describes each of the commands, plus how to write your own commands.
more.. . about 2.5 million characters total!
• Super Kit 1541 $29.95 US, S39.95 Cdn • The Toolbox (PAL and POWER) $79.95
Super Kit is, quite simply, the best disk file utility there is. No more losing those valuable copy-protected originals (like what's happened to me twice too many
PAL and POWER from Pro-Line are two of the most popular programs for the Commodore 64. PAL is an easy-to-use assembler (most assembler listings in The Transactor are in PAL format), and POWER is a programmer's aid package
times). So far we've shipped over 300 Super Kits and orders continue to pour in.
that adds editing features and useful commands to the programming environ
• Gnome Speed Compiler S59.95 US, $69.95 Cdn
ment. They come with two nice manuals, and our price is $50 less than
This compiler is for BASIC 7.0 on the Commodore 128.
suggested retail!
• Gnome Kit Utility $39.95 US, $49.95 Cdn • A X IOOO Amiga I MEG RAM Box $729.00 ( +SIOO S&H) U.S., •
Gnome Kit is a Commodore 128 utility with enhancements for the BASIC editor (like Trace, Find, Renumber, Delete, Auto, etc.) as well as enhanced monitor commands, and floppy disk monitor functions.
$1035.00 ( +S25 S&H) Cdn AX2000 Amiga 2 MEG RAM Box $899.00 (+$100 S&H) U.S., S 1276.00 ( + S25 S&H) Cdn
The AX2000 adds 2 Megabytes of "fast" RAM to the Amiga, allowing more tasks
Transactor Dt.ka, Transactor Back wues, and Mlcroftcbe
to run in the system at once, or for use as a fast RAM-drive. The unit plugs into the expansion connector on the side of the Amiga and duplicates the connector
All issues of The Transactor from Volume 4 Issue 0 I forward are now available on microfiche. According to Computrex, our fiche manufacturer, the strips are
for other devices to plug into. Up to two RAM boards may be plugged in together Oimited by the Amiga'a power supply), adding 4 Megabytes. The box has "auto-config",so with Kickstart 1.2 the RAM will automatically be added to the system when it is booted. II you are using Kickstart 1.0 or 1.1 (no auto config), you can use the program included with the AX2000 to add the memory to the system, and change your startup-sequence to automatically add the memory on power-up. Standard expansion bus architecture was used in the design of the AX2000, ensuring compatability with all peripherals and operat· ing system releases. The unobtrusive steel box is the same height and colour as the Amiga, and snugs up to the side without taking up much extra space. The unit is built tough and comes with a I year manufacturer warranty.
the "popular 98 page size", so they should be compatible with every fiche reader. Some issue are ONLY available on microfiche- these are marked "MF only". The other issues are available in both paper and fiche. Don't check both boxes for these unless you want both the paper version AND the microfiche slice for the same issue. To keep things simple, the price of Transactor Microfiche is the same as magazines, with one exception. A single back issue will be S4.50 and subscrip tions are $15.00. The exception? A complete set of 18 (Volumes 4, 5, and 6) will cost just $39.95!
This seems to be the most highly-recommended Amiga RAM board, and the first one to actually be available, so we're selling it here at The Transactor. You
This list also shows the "themes" of until Volume 5 , Issue 01.
each issue. "Theme issues" didn't start
can order the AX2000 or the 1-Meg AXI 000 from the subscription form in this issue. Shipping and Handling to the U.S.A. is via courrier and includes all customs clearance, or you can opt to clear shipments yourself and have it shipped "collect".
• Pocket Writer C64 • Poc.ket Planner C64 • Pocket filer C64 • Pocket Writer Cl28 • Pocket Planner Cl28 Cl28 • Pocket Filer • Pocket Dictionary
• Vol. 4,lssue OJ (• Disk I) • Vol. 4 , Issue 04- MF only (• Disk I) • Vol. 4, Issue 02 (• Disk I) • Vol. 4 , Issue 05- MF only (• Disk I) • Vol. 4,1ssue 03 (• Disk I) • Vol. 4 , Issue 06- MF only (• Disk I) • Vol. 5.Issue 01 -Sound and Graphics . (• Disk 2) • Vol. 5 ,Jssue 02- Transition to Machine Language (• Disk 2) • Vol. 5 , Issue 03 - Piracy and Protection- MF only (• Disk 2) • Vol. 5, Issue 04 -Business & Education- MF only (• Disk 3) • Vol. 5, Issue 0 5- Hardware & Peripherals (.Disk 4) • Vol. 5, Issue 06- Aids & Utilities (. Disk 5) • Vol. 6, Issue 01 -More Aids & Utilities (.Disk 6) • Vol. 6 , Issue 02- Networking & Communications (.Disk 7) • Vol. 6 , Issue 03- The Languages (.Disk 8) • Vol. 6 , Issue 04- Implementing The Sciences (.Disk 9) • Vol. 6 , Issue 05- Hardware &Software Interfacing (.Disk 10) • Vol. 6, Issue 06- Real Life Applications (.Disk II) • Vol. 7, Issue OJ - ROM I Kernel Routines (• Disk 12) • Vol. 7, Issue 02- Games From The Inside Out (.Disk 13) • Vol. 7, Issue 03 - Programming The Chips (• Disk 14)
S39.95 US, S49.95 Cdn S39.95 US, $49.95 Cdn S39.95 $49.95 $49.95 $49.95
US, S49.95 Cdn US, $69.95 Cdn US, $69.95 Cdn US, $69.95 Cdn
$14.95 US, $19.95 Cdn
In our opinion, the Pocket packages from DigitalSolutions are the best you can get on their own - the fact that they work with each other makes them even better. Planner and filer data can be loaded into the Writer, Writer text can be sent to the Filer, and etcetera. The Dictionary spell checker works with both versions or the Writer.
• The GUNK C64 to IEEE Interface $49.95 The GUNK plugs into the cartridge port, but doesn't extend the port for more cartridges (for that you'll need a "motherboard" or some kind).The other side of the GUNK is an IEEE card-edge suitable lor a PET-IEEE cable. From there, any
Notes: The Transactor Disk "I contains all program from Volume 4, and Disk
"2 contains all programs from Volume 5, Issues 1-3. Afterwards there is a separate disk for each issue. Disk 8 from The Languages Issue contains COMAL
IEEE device can be accessed including disk drives, modems, printers, etc. The 71
ue 03 Nov. 1916: w..ne 1, 111
0.14, a soft-loaded, slightly scaled down version of the COMAL 2.0 cartridge. And Volume 6, Issue OS published the directories for Transactor Disks I to 9.
Sending Cheques For Transactor Products
Once connected to CIS, your iNet clock stops ticking. When you exit from CIS, iNet starts the clock again until you connect to another service. The iNet package comes with a lovely manual with reference cards. I haven't tried mine out yet, but next issue a have a complete run�own of my results. If you can't wait until then. call:
If you wish to send a cheque with your subscription/order form, or you wish to conceal your credit card number, you can use an envelope and tape it to the
Sylvia Abretti Telecom Canada
back of the subscription card. The post office has threatened to charge us extra
410 Laurier Avenue W. Room 240
for sloppy business reply mail so please try to use an envelope that is smaller than the card. Can't find one? Just trim the end oH the envelope and tape along that edge when fixing it to the card.
Ottawa, Ontario, KIP 6HS 1-800 267 7400 High-Powered FREE Terminal Program for the C64
The Transactor Communications Disk The "Transactor Communications Disk" is proving to be a bigger project than
CBterm/C64 is an aii-ML terminal program for the C64.1t is set far above all the rest of the terminal programs because of a number of special features. First and
we expected. Collating a suitable manual is turning into a task almost as large as putting together a Transactor magazine. But it's not ready yet so don't send
foremost, the price is right; CBterm is FREE.
any orders. More next issue.
As for the rest of the features, CBterm supports the following:
CompuServe and Quantum l.lnk
•
40 column display and a split-screen 80 column display. (No extra hardware needed).
By the time you read this, The Transactor should have sections up and running
•
on both CompuServe AND Quantum Link. •
On CompuServe use "GO CNV" for CommOdore News and Views. There you'll find a menu of popular Commodore related magazines. You'll be able to read articles that have appeared in previous issues, as well as download programs
XMODEM and Punter file transfer protocols available. The XMODEM protot col will even correctly download Compuserve .IMG files. 22.5K RAM buffer with read/write and save/load to disk. Disaply RAM to screen or transmit RAM to modem.
item later in News BRK). Questions concerning magazine related topics will
ASCII/PETSCH file conversion built in. • Works with any and all modems. 1600, 1650, 1660, 1670, HES, Mighty-Mo. Hayes, RS-232 adapters. • 300 or 1200 baud.
also be answered on a regular basis.
•
Auto dial and redial. (Even on a 1660 that does not have carrier detect.)
•
VIDTEX Cursor control and positioning mode for use with Compuserve.
we've published (software for downloading from CIS is available free- see the
On Quantum Link you'll find us in the "Meet the Press" section of the Commodore Information Service. You'll need the special software from Quan tum Link before you can access the articles and software. The Transactor will be featuring a regular section on our online activities for both services. Next issue we'll be publishing a "Getting Started" summary with complete step-by-step instructions for joining us online.
Industry News
•
Includes Low Res Graphics. •
•
Full DISK control - send commands or list the directory.
•
User-programmable FUNCTION KEYS to help automate your log-ins.
•
Terminal parameter controls to provide compatibility with all BBS/computer
systems. • Built-in HELP screen. • Super extra: Direct display of the COMPUSERVE RLE encoded High Resolu tion Pictures. This includes the NWS Radar Weather Maps, CB Users Pictures, FBI I 0 Most Wanted List pictures, etc.
The Commodore Show, September 20 and 21, 1986 •
The West Coast Commodore Association is having another show at the Los Angeles Airport Hilton, September 20th and 21st. Judging from their February show in San Francisco, this one should be just as big a success. See the ad in this issue.
Full printer support.Send RAM data or control codes to printer. Even alter the printer's secondary address while still on line.
The high resolution screen (Text or pictures) may be dumped to any STAR or EPSON dot matrix printer. Option overlay programs provide printing of these pictures in a low resolution mode to ANY STANDARD printer. And they can be saved to disk in KOALA picture format and can then be altered with KOALA software.
Twin Citlea 128: The Commodore 128 Journal
CBterm/C64 is available for downloading from Data Library 2 of the CB Interest Group SIG on the Compuserve network. This Data Library contains the
Last issue we published the address of The Twin Cities 128 Club. We copied the
CBterm/C64 program along with all supporting documentation and overlay programs. They may be downloaded with any XMODEM or Compuserve
address correctly from an incorrect copy. Here's the correct address:
executive program (Vidtex). Twin Cities 128 1607 Hewitt Avenue, Suite 4 PO Box4625 Saint Paul, MN 55104
For those people who cannot download, or who are not on the Compuserve network, CBterm/C64 is made available free by sending a blank disk, along with a stamped sell-addressed disk mailer to:
lNet2000 This is the service that the rest of the country has been waiting for! Unlike metropolitan areas, calling your favourite data base from a remote area means calling long distance to the closest public dial port of a national networking service. And that can cost more than the service itself! iNet 2000 is a service operated by Telecom Canada, and (I'm told) is also working on setting up services in the U.S. There is a $50 subscriber fee, and hourly charges. Then, from a menu, you can connect to over I000 other online services, (including CompuServe) through iNet. The best part is ... for just $3.00 per month you can call iNet using a 1-800 number! No more long distance charges!
Chris Dunn 5848 N. Whipple St. Chicago, II. 60659 You must include postage and the return mailer. Any questions can be directed to the SYSOP of the CB Interest Group SIG (CBIG) on Compuserve (GO CBIG) or via EZplex to (76703,717). CBterm, originally written for use on Compuserve and its CB simulator, has grown into the powerful program it is today and incorporates requests by users as to what they Nov.1Ue: Wurne 1,111 ue os
would like to see in a terminal program. New features are being added and are
multitasking and multiwindowing. Aegis Draw is only the beginning. We are
released in DL2 of the CBIG SIG.
planning 'expanded CAD' products in the future.
CBterm is distributed free, for non-commercial or private use so that all C64 users can enjoy the world of telecommunicating. CBterm/C64 and all related
Aegis Draw requires a basic Amigacomputer, 512K or RAM, and one disk drive (although two drives and/ or a hard drive is highly recommended, as well as
files are (q 1985 by Chrisdos. They may NOT
purpose. All rights reserved. VIDTEX, EXplex and CB simulator are TM.
additional memory). Allowing use of the multitasking capabilities of the Amiga, Draw will give the user two windows to work i n - either two separate drawings,
Compuserve Inc. Chris Dunn (Chrisdos) is an information provider/SYSOP and
or two windows of the same drawing (more windows are allowed when more
is not an employee of Compuserve Inc.
memory is available). Some of Draw's significant features include:
The mail-in disk copy offer may be terminated at any time.
• Infinite levels of Zoom
be used for any commercial
• 16 customizable colours Hardware Interconnection Products
Master Software has announced four new interconnection products for Com
• Cloning of any part of a drawing • Full editing controls
• Lines, arcs, circles, rectangles, and freehand drawing tools • Independent window display features • Grid and Data snap (on/off toggle) • Save as IFF or ASCII files • Numeric display
modore computers: "Modem Master", "Modem Master Plus", "Y-NOT?", and the "80 Mono Cable".
• Background plotting/printing • Customizable plotter driver • Undo feature allows last operation to be undone • Parts library for each drawing • Full rotation of any object/part
"Modem Master" is a four-foot extender for the user port (modem port) of the VIC 20, Commodore 64, SX-64, Plus/4, and Commodore 128, allowing devices
• lnterchangable parts between drawings• Resizing of any object/part • Automatic dimensioning of any part or piece of drawing
to be placed in a location more convenient than behind the computer. The extender uses tangle-proof ribbon cable and connectors which are keyed to
• Customizable scale and grid size per window
prevent incorrect installation. List price of "Modem Master" is $29.95.
• Horizontal or vertical printout • Works with digitizing tablets, plotters, or printers supported by the Amiga.
"Modem Master Plus" is a "Modem Master" with a reset switch. The reset swithc is buffered to prevent electrical damage to thecomputer, and programs
• Numerical input for precise positioning of objects/drawings
to recover a BASIC program that was in memory at the time of reset are
Aegis Draw also takes advantage of keyboard input with corresponding keystrokes for the most popular tasks. This program can be used by novices as
included. Price is $34.95.
well as experienced CAD engineers for simple to complex design. Aegis Draw is available at authorized Amiga dealers and software stores. The suggested retail price isS199.95 (U.S.). For more information,contact:
The question of whether two printers can be used simultaneously with Commodore computers is answered: Y-NOT? "Y-Nor· is a six-foot long "Y" cable for the six-pin serial port of all Commodore computers. "Y-Nor· contains one male six-pin plug and two female six-pin jacks, and can be used
AEGIS DEVELOPMENT, INC. 2210 Wilshire Blvd. "277
to operate two printers or to separate the disk drive and printer to opposite sides of the computer set-up for added system flexibility. "Y-NOT?" is priced at $15.00.
Santa Monica, CA 90403
The "80 Mono Cable" will produce an
WCCA
(213) 306-0735
PRESENTS
80 column monochrome display from
the Commodore 128 in 80 column mode on any compositte colour monitor or monochrome monitor. It is six feet long and plugs into the RGBI port of the computer and into the video input jack of the monitor. The "80 Mono Cable", priced at $9.00, is ideal for word processing and spreadsheet applications. or any time a colour display is not needed. All prices include shipping to USA and Canada destinations. Discounts are
SEPTEMBER 20 & 21 1986 SHOW TIMES 10AM- 6PM P.S.T. LOS ANGELES AIRPORT HILTON CALL 213-410-4000 for hotel reservations
available for dealers, distributors, and computer clubs. Add 3.00 for C.O.D. (Maryland residents must include 5% sales tax.) Master Software 6 Hillery Court Randallstown, MD 21133
•
(301) 922-2962
• •
Aegis Ships CAD Program for the Amlga
•
EXHIBITS, EVENTS, AND DOOR PRIZES NATIONAL COMMODORE SPEAKERS SHOW SPECIALS & DISCOUNTS SEE THE LATEST INNOVATIONS IN HARDWARE/SOFTWARE TECHNOLOGY FOR THE COMMODORE MARKET
Santa Monica, CA - Aegis Development, Inc. has begun shipments of Aegis Draw, I 2-dimensional CAD program for the Commodore-Amiga computer.
The only West Coast exhibition and conference focusing exclusively on the AMIGA, Commodore 128, and C64 marketplace
This is the program that many owners and potential owners of Amigas have been waiting for. and Aegis Draw is being touted as the program the machine was created for.
REGISTRATION FEES: ONE DAY $10.00TWODAYS $15.00
"Based on our intitial orders and end user requests, there are thousands of people out there who have been waiting for Aegis Draw before buying their Amiga. Now that the product is shipping, you can expect to see Amiga sales
FOR MORE INFORMATION AND DETAILS CONTACT: WEST COAST COMMODORE ASSOCIATION, INC. P.O. BOX 210638 SAN FRANCISCO, CALIFORNIA 94121 (415)982-1 040 BETWEEN 8AM-5PM PST
increase significantly", Company president David Barrett stated. He went on to say that, "This program was created specifically for the Amiga, and is not simply a port from another computer. Therefore, we take advantage of many of the Amiga's characteristics which make this program so powerful, such as 10
New. 1916: Yl*lme 1,111ue 03
Computoons
TV
__..)
f•·""·
p ,... . .. ,d c"'""'
I
' j
�\ R
1• ··� 0 •·' ' "
DINNERS
VDT DINNERS
7 I vvv
��
�-
/��.1 �·
H£L(M YOU'QE APPL£ A�D l'N\lON\NODORE IT WON'T WORK
��;..._,. .
t-�-"" .Y • •
. • • •
I'VE SUN rr MAPP�N BEFOF?E! IIlEY START OIIT /)£/..11/J:R/N� TilE MAl/.. AND.., 13EFORE YOV KNOW IT tHeY'VE G-OT YOVI? c/IJB!
D m
..
.
..
0
0
0
o:-
0
•
•
•
•
0 0
.
.
.
.
0 •
..
. .
.
"I have an older model, but It's very intelligent...
.
.
'
.. "
81
Nov. 1986: Yolurne1, .._ oo
I
A must for software
developers Allows more programs to run simultaneo usly and faster Can be used to increase system RAM and/or as a FAST RAM DRIVE Uses standard memory bus architecture to allow for future compatibility Allows full use of memory expansion port for additional peripherals AX2000 2 MEG RAM Board $899.00 U.S. ($1276.00 CDN)
AX1000 1 MEG RAM Board $729.00 U.S. ($1035.00 CDN) Complete in case, nothing else to buy! 1 year manufacturer warranty!
DEALER INQUIRIES INVITED Comspec Communications Inc.
153 Bridgeland Avenue, Unit 5 Toronto, Ontario, Canada M6A 2Y6
(416) 787-0617
Mail order through The Transactor (see order card and News BRK) Shipping via courrier: within Canada add $25.00. To U.S.A. add $100.00 U.S.- includes customs clearance AMIGA is a registered trademark of Commodore Business Machine.
-路 -
Please send me 6 consecutive Taa:•Cidals starting with the next issue!
0 I'm s u bscribing to
I
·'
I I I
• .�
:\l
�
I
OU.S.A. $15. OForeign $2l.us. OAir Mail Overseas only) $40.us.
the magazine AND the disk, so please send m e my FREE T-Shlrt in the size indica ted below.
(
includes $4 . 1S
posa• 1»Cf • aue
0 Renewal (please include your Subscription Number from mailing label) 0 New Subscription (0 Please send microfiche instead of magazines) New address? 0
____ _
Name & Address�-
------------------
-:-:-(ptease include your postal/z ip code):
0 Transacto r T's OS. OM. OL. OXL, $13.95*. OJ umbo, $1 7.95* 0 Pocket Writer C64 $39.95* 0 Transactor B o o k of Bits and Pieces *I $14.95• 0 Pocket Planner C64 $39.95• 0 Transactor Microfiche Set (Vol4+Vol5+Vol6) 39.95• 0 Pocket Filer C64 $39.95* 0 AXIOOO I MEG Amiga board $729.00* +S&H (see below) 0 Pocket Writer Cl28 $49.95• 0 Pocket Planner C128 $49.95* 0 AX2000 2 MEG Amiga board $899.00* +S&H (see below) 0 Pocket Filer 0 $1oo• S&H encl. for each AX board, or 0 Ship collect Cl28 $49.95* 0 The Toolbox (PAL and POWER) $79.95* 0 Inner Space Anthology $14.95• Note: See News BRK for details on above items The Transactor Disk (154 I I 4040/MSD format) 0 Please send 6 co nsecutive disks t o c o rrespond with my magazine subscription: $45.00. • 0 Disk 9 . Send these diSks at $7.95* each. 0 Disk 10 0 Disk 1 0 Disk 5 0 Disk 1 1 0 Disk 2 0 Disk 12 0 Disk 6 0 Disk 3 0 Disk 7 0 Disk 13 0 Disk 4 0 Disk 8 0 Disk 1 4
SAVE 29%0FF Cover Price
0 Pocket Dictionary $14.95* 0 The TransBASIC Disk $9.95• 0 The G-Link Interface $49.95• 0 The Micro Sleuth $79.95• 0 1541 ROM Upgrade $49.95• 0 Super Kit 1541 $29.95* 0 Gn ome Speed Compiler $59.95* 0 Gnome Kit Utili ty $39.95•
Transactor Back wues: $4.50• each, MF $4.50* each (MF=Microftcbe) 0 Vol.6, lss.03 0 MF 0 Vo1.5, lss.02 0 MF 0 Vol.4,1ss.OI 0 MF 0 Vo1.6, lss.04 0 MF 0 Vo1.5,lss.03 MF only 0 Vol.4,lss.02 0 MF 0 Vo1.6, lss.05 0 MF 0 Voi.S, lss.04 MF only 0 Vol.4,lss.03 0 MF 0 Vol.4,lss.04 MF only 0 Vo1.6,lss.06 0 MF 0 Voi.S, Iss.05 0 MF 0 Vol.4,1ss.05 MF only 0 Vol.7,1ss.OI 0 MF 0 Vol.5,1ss.06 0 MF 0 Vol.4,1ss.06 MF only 0 Vol.6,lss.OI 0 MF 0 Vol.7,1ss.02 0 MF 0 Vol.5,1ss.OI 0 MF 0 Vol.6,1ss.02 0 MF 0 Vol.7,1ss.03 0 MF •
•
•
·
•
• Prices are in U.S. Dollara
NOTE: Prepaymentrequired. Purchase orders wiD be accepted ONLY it accompanied bypayment. 0 Cheque/MO. enclosed
0
Visa
0
MasterCard
Cheque•
0 Please send dealer information for The Transactor. -
Amoun
I
Acct.*
I use the following Commodore equipment: OC64 OC1 28 OVIC20 Disk Uni t: OSFD1001 01540/41
.._.,-
Dated
Ex[!ires OAmiga
04040
08128 08050
- - - - --- -- --------- -
-
016/ +4 01571
08032/96 08250
- -- ----
I
08/86
-
-
----- -
NO POSTAGE NECESSARY IF MAILED IN THE UNITED STATES
BUSINESS REPLY MAIL FIRST CLASS PERMIT NO 390
BUFFALO. NY
POSTAGE WIU BE PAID BY AOORESSEE
277 Linwood Avenue Buffalo, NY, 14209-9990
t . .ll.t t t.llt... t.t t.t 1.1 1.1 11 1... 11 .
••
••
••
••
••
••
•••