28 Aug
2007
28 Aug
'07
4:12 p.m.
On 28 Aug 2007 at 20:40, Ensor wrote:
However, I tried the board in at least one '286
based AT clone which was
capable of being switched down to 4.77MHz and it still didn't want to know.
Not only that, but when installed in an AT class machine, the board simply
didn't appear in the computer's memory/IO map - verified by "prodding"
the
locations it was mapped to through "debug".
I dug mine out to see what might be causing the problem. It looks
pretty simple--the "bus" end is pretty much an 8255A connected
through a PAL (unlabeled, I suspect this is what it is) address
decoder. It shouldn't be that hard to suss out why it doesn't work
on a faster CPU.
A curious thing is that CMOS 4040's are used as some sort of address
counter (24 bit? Would mean that the design could potentially handle
some big EPROMs).
It's surprising how much I've forgotten since I've last had the box
out. It has 4 28 pin ZIFs on the EPROM module (gang programming, I
guess) and a note I made somewhen taped to it: "Note: National
27256Q is programmed as half of a 27512A". I don't even think I have
any 27256Qs anymore.
Cheers,
Chuck