26 Jan
2012
26 Jan
'12
2:37 p.m.
On Thu, Jan 26, 2012 at 3:31 PM, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
If I had a 9000A-Z80 pod for my 9010A, I'd probably already have it
fixed (I probably use my Fluke for testing memory more than any other
single thing - even back in the day, a "DMA test" for a COMBOARD was
really just running a memory test on the quadrant of 68K memory space
that caused Qbus/Unibus DMA transactions... we could test local ROM,
local RAM, or remote RAM via DMA, all with the same buttons and the
same tool). Fortunately, I have a 6502 pod as well as a 68000 pod, so
the Fluke has plenty of targets at my house.
Yes... there are other ways to run address testing, bus testing,
etc... they are more work than popping the processor, plugging in a
vintage purpose-built tester and hitting well-understood buttons. The
lack of a Z80 pod pushes this project further down the stack,
especially since I have little software for the M3 that requires more
than 32K.
-ethan
The video monitor is the same circuit, so is the PSU.
So there shouldn't
ahve been any more or less ripple.
I more meant that one might have been in nicer shape than the other
one, not that one was designed better.
I should see
about tracing down the memory fault on my M3 - I've tried
memory tests and moving the 4116s around, but it seems to only see 32K
of the 48K that's installed. ?It's possible that a '157 or something
further upstream from the DRAM is faulty.
Obvious things would be address decoding or the RAS/CAS drivers.