Re: Hardware emulation of 6509 using 6502?

Den Wed, 7 Mar 2018 08:59:32 +0100 skrev Anders Carlsson
<anders.carlsson@sfks.se>:
> By the way, was the MAX Machine really sold in early 1982 or is that 
> popular belief that it was sold in early 1982?
> 
> We're quite certain it was announced in early 1982, but when I 
> previously browsed through several Japanese computer and games
> magazines from 1982-83, the earliest mention of the MAX Machine was
> found in Nov-Dec 1982, at the same time as the Japanese C64 and it
> wasn't properly tested until a bit into 1983. If it had been on the
> market for nearly a year, that was unusually late for a games
> magazine to do a write-up spread over several pages.

I think we need some input on this from someone who actually understand
the language and preferable had been around in Japan at the time in the
early 80's.

> It has been speculated that the MAX was planned to be sold in early 
> 1982, but for whichever reason - hardware issues or lack of software
> - it got severly delayed. Somewhere there should be documents or
> books about this, but I don't know where to look.

A good question to ask is at which point did MOS Technologies diverge
from being a Commodore owned general IC manufacturer to Commodores
in-house "ASIC" manufacturer? It's well known that the VIC-1 were made
by MOS in hope of selling it to other companies than Commodore. A
qualified guess is that things started to change with the success of
the VIC 20.

Were VIC-II development initiated by MOS or by Commodore itself?

If it were initiated by MOS, were it even intended for a home computer
or rather arcade games or a game console (like the MAX, except without
the useless keyboard)?

If it were initiated by Commodore itself, were it supposed to be an
upgrade to the VIC-20, or for a new machine?

In this timeline we have the shift from PET 2001 to PET 2001N/3000
where they shifted from SRAM to DRAM without being tied up by how some
custom ICs work. That change could had taken place at any point in time
regardless of what chips MOS did manufacture (as the 74xx logic were
sourced from other compaies (except some 74LS245:s in atleast VIC-20
which actually in some cases are 65245 made by MOS (!) ). It actually
took place in 1979 (afaik) and that would had been around the turning
point where it started to be a good idea to use DRAMs for computers
with atleast 16k. (I assume that the 8k 2001N/3008 were only made in
really small numbers, probably to be able to continue some kind of 8k
model).

Another thing in the timeline is when 64kbit DRAMs were reasonable to
use in a home computer. The older 16kbit DRAMs did require three
different voltages and for a VIC-20 class computer that might have had
a noticeable impact on production cost (although the VIC-II and SID
ended up requiring 12V DC making it necessary to have two voltages in a
C64, so using 16k DRAMs would "only" had required one more voltage but
atleast a bunch of capacitors to generate a negative voltage from the
9V AC line). (Btw Atari 600XL did use 16k*4 bit DRAMs which also only
required a single 5V DC rail, so the introduction of Atari 600XL is
also an important step in the time line).

The MAX might had been a test bench for various production and market
stuff, and not really considered to technically be the best machine. At
the time, Atari had for a long time made the membrane keyboard 400
model and in Europe Sinclair made the ZX80 and ZX81 computers with
membrane keyboards (even worse than the Atari 400 afaik), so Commodore
might had wanted to gain some experience in manufacturing hardware with
membrane keyboards.

For a pure game machine with some limited programming cartridge it made
no sense to have exactly the same keyboard layout as the VIC 20. We can
probably assume that the (afaik) same keyboard layout and matris on MAX
as on VIC-20 were so they in the future could produce a budget computer.

I thing that at the time, Commodore management had many options and
they had to choose wisely. In a hypothetical other version of the
from-then future they might had made a machine with VIC-II but with
VIC-20 memory map and maybe 8k SRAM built in. They might had made two
versions, one with a full keyboard and one with a membrane keyboard and
maybe without some of the ports. (Cartridge, game and probably cassette
ports would had obviously been available).

Maybe they even had prototypes of that kind made, but then were taken
by surprise by the drop in RAM prices?

Another thing in the time line are other computers where 64k RAM made
any sense. The first PC used 16kbit DRAMs but it could be populated
with 64k and expanded to over a halv megabyte. Before that afaik it
were only CP/M computers (and maybe the small number of FLEX 9
6809 computers) that could make any use of 64k ram as most other
computers had loads of ROM that made bank switching necessary. A CP/M
computer usually had 2-4k ROM and 60-62k RAM available (loosing 2-4k of
the 64K ram), while most other computers had much more ROM. (An
exception is the Sharp MZ 700 and it's relatives which loads the basic
interpreter from cassette (!) and thus has about 60k ram available
without bank switching).

Btw judging by what happened with the TED/264 line it's obvious that
Commodore didn't really know what they were doing at some point in
time :)

> Another aspect that should also be considered is which were the first 
> C64 games to be available. Generally it is thought the first
> cartridge games were ports of the MAX versions, but those who were
> around in 1982 have different opinions on whether there were any
> cartridges to buy when the C64 was launched or if those games
> appeared several months later, in time for Christmas 1982. While it
> probably can be deducted which games were the first to be written, it
> isn't the same as the first games to be sold. Also HESware holds a
> claim for being the first 3rd party publisher of cartridge games,
> possibly even beating Commodore themselves.

If there were any limits on supply of the cartridges, Commodore would
had been fools by not directing them to Japan where the Max were sold.
Max could only use cartridges while the C64 could at least load software
from tape from day one. 

I'm not sure when the first 1541's did appear - that might also be an
indicator as there would be absolutely no reason for the 1541 to appear
before the C64, but the C64 might had been sold for a while before the
1540 were updated to 1541. It seems unlikely that though that there
were any extended time in which C64 were available in the shops but no
1541 or upgrade roms for 1540. ROMs for 1541 might had been produced in
a batch a while ago before the 1541 started to sell though, as they
probably could finish the small changes early while the C64 were in
prototype stage, making drives available for the prototype C64
development and still being able to use the rest of that rom batch
when C64 finally started to ship.

> If the MAX was already on the market in Japan, those cartridge 
> conversions to run on the C64 should have been done, manufactured and 
> ready to be sold by the day the C64 hit the market, no doubt about
> it. Again this is probably found in a book or two, though I don't
> know which one to look in and how accurate the sources would be.

Well, Afaik, there is no need to convert any cartridge from Max to C64.
Unless they try to do something really weird they should run on a C64
as is.

-- 
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.

       Message was sent through the cbm-hackers mailing list