[Coco] Buyer beware

Mark McDougall msmcdoug at iinet.net.au
Mon Feb 23 17:57:25 EST 2009


Bill wrote:

> Screenshot at http://www.sq1bbs.com/3002coco.jpg

Is this image consistent each time, or somewhat random?

Years ago I got an old Juno First arcade PCB that had partly corrupt 
sprites. I suspected RAM but the trouble is the PCB had a sea of 4116's - 
all NON-socketed - and I had no schematics. So I had an idea...

I modified the MAME driver to simulate a single _faulty_ RAM chip in the 
sprite circuit. Then I ran the emulator, and checked what it showed against 
what I was seeing on my arcade cabinet. After switching bit positions a few 
times, I got exactly the same corruption I was seeing on my cab!

Back to the PCB, there was (IIRC) enough indication of which bank of memory 
was for sprites. I located what appeared to be the chip for that bit 
position, unsoldered it and replaced it with a socket and new RAM chip. 
Fired it up and - lo and behold - no more corruption!!!

Anyway, the point of this ramble is that perhaps doing the same in the Coco 
MESS driver might be an option - at least it might _prove_ it's a RAM 
problem if you can reproduce the same garbled graphics???

I'd give it a go myself but I'm about to head off to work... maybe later if 
you're still having problems???

BTW when you checked for loose chips, did you press down _quite firmly_ on 
each socketed chip? "Loose" doesn't mean "I can wiggle it when I touch it" 
in this case.

Regards,

-- 
|              Mark McDougall                | "Electrical Engineers do it
|  <http://members.iinet.net.au/~msmcdoug>   |   with less resistance!"



More information about the Coco mailing list