Now Macro's site says this about Object RAM:
This is the sprite and colour ram. To write to this the CPU has to go through the first of the buffer chips (8F and 8H or 9B *). To read the contents back, it also uses additional buffer chips at 4L, 4M and 5M. This occupies memory from 5800 to 58FF. A failure in this will normally be down to the ram chips (4F and 5F) or the afore mentioned buffer chips.
Mmm, the Fluke says it can't WRITE to the RAM. I am actually NOT suspecting the RAM itself, it came from the working Bootleg board so I'm pretty sure they are good.
Now 4L,4M and 5M buffer ICs are only used when reading, OK let's do some reads only from some memory addresses in this range, get same data for all adresses (AF, sometimes other values) but no error message.
When I write a single address there's no error but when I read the same address still AF so it's not working clearly.
So I put the RAM test in LOOP (keeps repeating) and start probing around. Soon enough I find that the Write line of the RAMs stays high all the time. That can't be right, it should co low before and during a write operation. So I backtrack the signal to chip 8M pin 12 and sure enough a constant high there.
I suspected that chip 74LS138 but.....I've jumpered a trace before between the pin 20 of both RAMs and pin 12 of 8M, which is where the schematics say the signal comes from. There was no conductance between these pins so I suspect there was a cut made (remember, those two RAMs were not on the board when it was still a Tazz-Mania.
I wondered if there might be something else going on, so I desoldered the jumper wire and sure enough the output of 8M, pin 12 was happily toggling when running the RAM test.....mmmmm.
Could it be the RAMs after all then ? Well I simply removed them and tested again.....NOPE, still the Write signal staying high on the sockets....
I wonder if they somehow made some new connection anywhere for the conversion...gotta dig some more....yippie !
