mlewis@unocss.UUCP (Marcus S. Lewis) (05/23/89)
I can't believe I am having these problems! I just upgraded my old-MB
Z using the Wideman method to 768K. This was to free up my UCI board
for another system, and I went from 704K to 768K in about 3 hours.
No problem. When I was using the UCI board, I had very occasional
parity/buss errors that shut the system down. Very annoying, but
completely eliminated by using a 200K Ramdisk. I loaded 200K of
stuff in there and never had a problem. This doesn't work any
more. Now I am getting parity errors at random intervals, but at a
fairly high frequency. Since my wife and kids try to use this thing,
this is totally unacceptable. What's going on. I have tried to
disable parity by outputting various values to port 0FCH, the memory
control latch, all of which so far have resulted in truly bizarre
behaviour and an immediate need to reboot. The memory system passes
multiple passes of both the Zenith diagnostics and the ROM Monitor
memory test. I had a single error on Sunday, not repeated, but that
whole bank got replaced anyway. The Zenith diagnostics won't touch
the parity stuff, since I don't have a Z-205. I get a parity error
immediately on that test for banks 3-6. HELP! The 64K chips I
pulled were 200ns parts and the 256K chips installed are 150ns, so
I don't think response time is the problem.
Marc
--
Na khuya mne podpis'? | Internet: cs057@zeus.unl.edu
| UUCP: uunet!btni!unocss!mlewis
Go for it! | Bitnet: CS057@UNOMA1
---------------------------------------------------------------------------usenet@cps3xx.UUCP (Usenet file owner) (05/24/89)
In article <808@unocss.UUCP> mlewis@unocss.UUCP (Marcus S. Lewis) writes: >I can't believe I am having these problems! I just upgraded my old-MB >Z using the Wideman method to 768K. This was to free up my UCI board >for another system, and I went from 704K to 768K in about 3 hours. >No problem. When I was using the UCI board, I had very occasional >parity/buss errors that shut the system down. Very annoying, but Do you have a UCI Eazy87 8087 kit installed? I had a similar problem when I put one in my 100. A call to UCI resulted in a suggestion to solder a small cap across two terminals of the Eazy87 board. That cleared it up and all's been fine ever since. Unfortunately, I don't remember which terminals and what cap size. If you need that info, I can pull the machine apart and look. Mark Urban-Lurain Computer Science Department Michigan State University urban@cpswh.cps.msu.edu
GUBBINS@TOPS20.RADC.AF.MIL (Gern) (05/24/89)
If I understand what you did: Converted old MB 192K to 768K by Wiedamann method and pulled all S-100 RAM boards. I would replace U153 with a 74S280 for a start. If things don't get better, I now have 2 spare New Type motherboards one 5MHz 64K RAM IC parts, one 8MHz 256K IC parts that I will let go to the desparate for $35 and $75 respectively. Neither contain any RAM. They are pull outs from machines I bought for parts and gaurenteed to work. Gern -------