Coleco ADAM
Aaron T Dingus
megahertz?
Also, can anyone make a *hopefully educated* guess of the machine's speed
in MIPS?
Sorry if this is a FAQ (sarcasm).
--
Aaron T. Dingus | The Ohio State Universty
Professional athelete | adi...@magnus.acs.ohio-state.edu
Richard F. Drushel
From actual Coleco schematics for the ADAM:
The Z80 clock crystal is 7.15908 mHz, divided by 2 to
give 3.57954 mHz.
The 6801 microprocessors for each ADAMnet device use 4.0
mHz crystals.
As far as MIPS are concerned, it depends greatly upon
whether the code uses device I/O or not. The ADAM does *NOT*
utilize the Z80 interrupt structure for tapes, disks, keyboard,
etc. (The video system, however, uses the NMI to move sprites.)
Instead, each device has its own 6801 microprocessor, which
feeds device data to a master 6801, which then leaves data and
status information in special memory locations mapped to the
top page (255) of Z80 address space. While the EOS operating
system has routines to initiate and terminate I/O requests while
going off to do something else, almost all existing software has
used higher level EOS calls which poll the device endlessly until
the desired result is achieved or until there is an error. So in
practical terms, code which has device I/O through the EOS
interface spends lots of time in a polling loop. Straight
computational code, however, will "fly" along...:)
I suppose you could pick an average Z80 instruction,
figure out how many T and M cycles it would require, then figure
out how many times it would execute at phi=3.57954 mHz. My Z80
bible from Zilog gives each instruction time in microseconds
assuming a 4 mHz clock, so that would give a slightly fast estimate
for the ADAM.
--
Rich Drushel ** r...@po.CWRU.edu *** Biology Ph.D. Student ** Cleveland FreeNet
Co-Sysop, Coleco ADAM Forum --- Assistant Sysop, Science Fiction & Fantasy SIG
"Solda pung apfashat ro des-marno, / Marn ladir o armag noth yeni arno. / Hell
miryat it, / Jambo iat it, / Os lasse wei ticip kati baldo." / Old Ennish poem
John Howard Osborn
>Rich Drushel ** r...@po.CWRU.edu *** Biology Ph.D. Student ** Cleveland FreeNet
>Co-Sysop, Coleco ADAM Forum --- Assistant Sysop, Science Fiction & Fantasy SIG
>"Solda pung apfashat ro des-marno, / Marn ladir o armag noth yeni arno. / Hell
>miryat it, / Jambo iat it, / Os lasse wei ticip kati baldo." / Old Ennish poem
You're probably as good a person to ask as any about ADAMs...
Can you still buy "stringy-floppy" media?
Did the daisy wheel on the printer really melt under heavy use?
I heard the Coleco designed a floppy disk drive and CP/M system for
the ADAM but that it was never released. True?
I seem to remember that you could either buy a Colecovision and later buy
the rest of an Adam or buy an Adam and run Colecovision cartridges? AM I
just confused? (I *do* remember the flap over the Atari 2600 emulation.)
Just curious...
-
-John H. Osborn
-osb...@cs.utexas.edu
Rob McCool
: In article <1992May12.2...@usenet.ins.cwru.edu> r...@po.CWRU.Edu (Richard F. Drushel) writes:
: just confused? (I *do* remember the flap over the Atari 2600 emulation.)
:
Out of curiosity, does anyone know what sort of main microprocessor the
2600 used? I heard it was a 6502 but opened it and found no such beast...
or at least, no such labeled beast. Was it a Z-80? Something 4-bit? Does
anyone really care?
--
------------------------------------------------------
Rob McCool: NCSA STG System Administrator
stga...@ncsa.uiuc.edu ro...@ncsa.uiuc.edu r-mc...@uiuc.edu rmc...@imsa.edu
It was working ten minutes ago, I swear...
Steve VanDevender
Out of curiosity, does anyone know what sort of main microprocessor the
2600 used? I heard it was a 6502 but opened it and found no such beast...
or at least, no such labeled beast. Was it a Z-80? Something 4-bit? Does
anyone really care?
It was a 6502 derivative (6510?). Specs for the 2600 were posted
in alt.sources a long time ago, and I was amazed. The 2600 had
128 bytes of RAM. Yup. And that was mapped into both the zero
page and the stack page so it was used for both purposes. 2600
programs basically had to control screen displays scan line by
scan line by modifying registers that controlled the
player/missile/background graphics. I now have an incredible
amount of respect for the Activision programmers who did all my
favorite games on that machine, because they earned their money
like few other programmers, shoehorning really great games into
2K and 4K ROM cartridges. I'd love to know how they did the
pseudorandom terrain generation in _River Raid_.
--
Steve VanDevender ste...@greylady.uoregon.edu
"Bipedalism--an unrecognized disease affecting over 99% of the population.
Symptoms include lack of traffic sense, slow rate of travel, and the
classic, easily recognized behavior known as walking."
Richard F. Drushel
In a previous article, osb...@cs.utexas.edu (John Howard Osborn) says:
>>Rich Drushel ** r...@po.CWRU.edu *** Biology Ph.D. Student ** Cleveland Free
>
>Can you still buy "stringy-floppy" media?
answers is *YES*. I don't think there are any new ones being made by Loran,
but there are still pretty good supplies of both original blanks and surplus
software tapes (e.g. SmartBASIC) available from ADAM vendors and places like
American Design Components. (The latter can be erased and re-initialized.)
It is also possible to make "new" tapes using standard audio cassette
tape (or even better, the special computer tapes made by Radio Shack for some
of their Color Computers). There are 2 methods:
(1) ADAM ddps have holes in them that standard audio cassette tapes
lack, while lacking holes audio tapes have. If you are careful, unscrew the
plastic shells from the ddp and the target audio tape, drill the appropriate
holes in each, re-assemble, then use a dual-deck cassette recorder to
copy the ddp to the audio tape. Set the levels pretty high, and copy both
sides. The reason for drilling holes is that a ddp won't fit into a tape
player, nor an audio tape into an ADAM tape drive.
(2) A device developed by Syd Carter called the MegaCopy Tape Formatter
allows an ADAM with 2 tape drives to do the job of a dual-deck cassette
recorder. It is unfortunately out of production and somewhat difficult to
find. It sold for about $50 and plugged into the motherboard headers for
tape drive 2.
I have tried in vain to find a way to invoke format routines for the
digital data drives. I believe that they are present in the ROM firmware for
the drive controllers, but haven't figured out how to call them up. Coleco
(and Loran) had to format their blank tapes *SOMEHOW*. All the EOS lets you
do is read/write blocks to an already-formatted tape. The reason for my
belief that the tape drives may have a format utility in them somewhere is
that the Coleco disk drives *DO*. It is a typical cute programmer hack:
** fill a 1K block of memory with FFh
** LD HL,#BASE_OF_BLOCK
** LD BC,#FFh ;hiword of block #
** LD DE,#CEh ;loword of block #
** LD A,#DEVICE_NUMBER ;drive to format (4=disk 1, 5=disk 2)
** CALL EOS_WRITE_BLOCK ;magically invokes disk format
Note that the block number spells FACE (haha). Substituting tape 1
(device=8) or tape 2 (device=24) does reformat the tape--instead it tries 3
times to find block FACEh (taking about 5 minutes to wind and rewind the
tape), then errors out (tapes have blocks 00h to FFh).
>Did the daisy wheel on the printer really melt under heavy use?
buggy, but the hardware is still going strong after 8 years (10 for
ColecoVision video game units). More likely you would go deaf from the
klak-klak-klak of the noisy printer head solenoid :)
>I heard the Coleco designed a floppy disk drive and CP/M system for
>the ADAM but that it was never released. True?
sectors per block) and CP/M 2.2 were released. The drives are IBM-type
hardware, half-height, with obnoxious drive doors and woefully maldesigned
power supplies. (An external transformer supplies 9 VAC, some fat electro-
lytic capacitors on the DC side of the bridge boost it to a wobbly +13 VDC,
and then some overworked, under-heatsinked 78xx series regulators give +5
and +12 VDC to run both the controller and the drive mechanics. In
particular, the 7805 regulator will boil spit during normal operation. Since
the regulator section is part of the controller board, in the drive box
itself, these babies run *HOT*.) The disk drive is connected via 6-wire
phone cable to one of the ADAMnet ports (a 2nd disk drive is daisy-chained
from the 1st disk drive). Unlike the other ADAMnet devices (printer, keyboard
and tape drives), the disk drive controller uses a 6800 microprocessor with
an external firmware ROM, instead of the 6801 with internal (and hence
inaccessible) mask ROM. Consequently, the disk controller ROM has been
disassembled, and modified ROMs permit the use of IBM 360K (5.25") and 720K
(3.5") drive mechanics.
The CP/M 2.2 for the ADAM is a standard Digital Research CP/M. It
supports the use of a 64K memory expander as a RAMdisk (M:), and includes
2 special utility programs to copy programs between CP/M and EOS disks.
There are BIOS extensions to support the TMS 9928 video display processor
(set foreground/background/border colors) as well as a prototype (never
released) serial/parallel card which would have run off the ADAMnet and
been accessed as RDR:/PUN: and LST: devices. (Schematics of this prototype
have surfaced, but nobody knows how to write the firmware for the 6801 it
uses, much less burn it into a 6801.) The problem with CP/M 2.2 is that it
doesn't support any of the popular 3rd-party hardware available for the ADAM,
like larger memory expanders (up to 1 MB), the 320K and 720K disk drives,
hard drives (up to 60 MB), serial and parallel cards, etc. Tony Morehen
essentially disassembled, then rewrote, the ADAM BIOS to utilize all this
hardware; the result is T-DOS (Tony's DOS). T-DOS is completely CP/M 2.2
compatible, but the CCP is more like MS-DOS. For all practical purposes,
anyone with an ADAM who runs CP/M programs uses T-DOS instead of CP/M 2.2.
And TDOS is *FREE*!!
>I seem to remember that you could either buy a Colecovision and later buy
>the rest of an Adam or buy an Adam and run Colecovision cartridges? AM I
Expansion Module #3 to convert it into an ADAM computer. This expansion
unit was a cream-colored box about half the size of the regular ADAM, which
plugged into the expansion port at the back end of the ColecoVision. You
used a regular ADAM printer/power supply instead of the little black wall
box for the ColecoVision. There was space for 2 tape drives, 3 internal
expansion slots, and a sideport expansion edgecard, just like the stand-alone
ADAM. Expansion Module #3 contained the ADAM boot ROMs, 64K of RAM, and the
6801s for the ADAMnet devices; it utilized the Z80, VDP and sound chip of the
ColecoVision. The ADAM itself is a 2-board computer: an upper ColecoVision-
like board with the Z80 et al. and a lower board with ADAMnet and 64K RAM.
Thus, the ADAM is able to play ColecoVision game cartridges. There are 2
reset switches, one which accesses ADAM boot ROMs, and another which tries to
boot a game cartridge.
Expansion Module #1 allowed the ColecoVision to play Atari 2600 games.
I am not exactly certain how this worked. Some of the Atari games now exist
as disk/tape versions which can be run on the ADAM, while the actual Atari
cartridges have a very different edgecard connector compared to the Coleco
games. This in itself would lead me to suspect that, edgecard interface
aside, the guts of the Atari 2600 were identical to the ColecoVision (same
microprocessor, same video and sound chips, addressed identically, and the
same game OS). But (though I don't own one) I've seen the EM #1, and it
has a fair amount of circuitry in it--more, it seems, than just rerouting
one edgecard to another. In the ADAM, there is a fair amount of circuitry
devoted to allowing external video and audio signals to be gated into the
video modulator, superseding the normal ones. If the Atari 2600 used a Z80
but had different video and sound systems, maybe the EM #1 has that
circuitry in it, and just gates the final audio and NTSC video onto the
ColecoVision bus. Does anybody who has/had an Atari 2600 know anything
about its microprocessor, game OS, etc?
For completeness, Expansion Module #2 was an arcade-type driving
setup, with steering wheel and foot pedal. The steering wheel (called a
"spinner" in Coleco tech manuals) used the maskable interrupt of the Z80 in
Interrupt Mode 1 (automatic jump to 38h). I don't know if it can be used
with the ADAM or not. No ADAM software I know of uses any interrupts except
the NMI issued at 60 Hz from the VDP, and outside of games, even that is just
for a system clock.
>Just curious...
Glad for the opportunity to expound on my hobby computer :)
--
D John McKenna
> 2600 used? I heard it was a 6502 but opened it and found no such beast...
> or at least, no such labeled beast. Was it a Z-80? Something 4-bit? Does
> anyone really care?
>It was a 6502 derivative (6510?). Specs for the 2600 were posted
>in alt.sources a long time ago, and I was amazed.
So was I. It wasn't a 6510 - that was a CBM hack for the C64 (they added
an almost 8 bit I/O port with data direction register to locations 0 and
1). The 2600 used a 6507 - a REALLY cut down 6502. Only 13 address lines
(8K total - the machine allocated 4K to the cartrige and 4K to the
rest). Also no IRQ line. Very interesting. All of the timing had to be
done in software - timing loops, checking HSYNC and VSYNC flags on the
video chip. Blech. I have MUCH respect for the people who wrote such
triffic games on them.
John West
--
gu...@uniwa.uwa.edu.au For the humour impaired: insert a :-) every 3 words
Gene Dolgner
>Also no IRQ line. Very interesting. All of the timing had to be
>done in software - timing loops, checking HSYNC and VSYNC flags on the
>video chip. Blech. I have MUCH respect for the people who wrote such
>triffic games on them.
My best friend, for many years, used to write video game cartidges at Coleco.
After some discussions with him, the term, MUCH respect, describes it all.
Every byte was precious. The 2K/4K cartidges not only contained timing loops
and motion software, but also all displays and sounds/tunes. He said sometimes
changing the tune in a cartridge was all it took to keep it from fitting. It
is an absolute art to program this way. I wish I had half that knowledge.
Gene Dolgner, contractor at large (I may reappear at any time!)
gdol...@chelsea.Prime.COM
"Life goes slower when you graze." - Northern Exposure
Anthony J Stieber
>
> In a previous article, osb...@cs.utexas.edu (John Howard Osborn) says:
>>Can you still buy "stringy-floppy" media?
>
> If you mean "digital data packs" (ADAM's high-speed cassette tapes), the
>answers is *YES*. I don't think there are any new ones being made by Loran,
The Adam ddps are modified audio cassettes, while stringy floppies (aka
wafer tapes) are something else. I've only seen them once or twice,
but they are about the size of a credit card, and have a very thin
continuous loop of magnetic tape in them. The drive runs the tape in
one direction over the head, there is no rewind. To get to previous
data you just wait until it passes over the head again. Tapes were
available in various lengths, I think up to 100KB, maybe more.
I remember Percom selling them for the home computers of the time
(Apple ][, Atari 800, Pet, TRS-80) as much cheaper alternatives to
floppy drives. I later saw them for a TI computer (99/2?) that was
about the size of the Sinclair ZX-8x machines. The Sinclair QL also
had something similar called a microdrive, it may even have been the
same thing.
Anyone know more about these devices? I always wanted to get a closer
look at them.
--
<-:(= Anthony Stieber ant...@csd4.csd.uwm.edu uwm!uwmcsd4!anthony