🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.sys.zenith
Article #190

Zenith 386/16 memory board question

#190
From: Shamus McBride
Date: Tue, 11 Aug 1992 21:39
29 lines
1490 bytes
While running Turbo C++ my machine kept getting a memory parity error.
Memory diagnostics didn't find the error. I set a trap on the NMI
interrupt handler entry, then when the error occured again I checked
ports F4H, F5H, F6H, and F7H for errors (technical manual 15-14). F7
showed 11110111 which implies an error in Bank 3 of the 4th card (of 4
1Meg cards).

So my question is how are the banks laid out on the
386/16 memory card?

    +----------------------------------------+
    | [][][][][][][][][]  [][][][][][][][][] |
    |                                        |
    | [][][][][][][][][]  [][][][][][][][][] |
    |                                        |
    |                            +----------++
    +-IIIIIIIIIIIIII-IIIIIIIIIII-+

Drawn from memory. 4 banks, TL, TR, BL, BR.


Any suggestions on how to better isolate the offending chip.
I usually start swapping chips around, but it often takes
2 to 3 hours to find the bad(?) chip ;-(

--
Shamus Mc Bride           | tecslm1@sdc.boeing.com      
Boeing Computer Services  | tecslm1%sdc@atc.boeing.com
(206) 865-5047            | uw-beaver!bcsaic!sdc!tecslm1
Technical Details
Message-ID: <6373@oval.BOEING.COM>
Path: arc1-ca.pugleaf.net!sparky!uunet!bcstec!bcsaic!sundry!sdc!tecslm1

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!bcstec!bcsaic!sundry!sdc!tecslm1
From: tecslm1@sdc.boeing.com (Shamus McBride)
Newsgroups: comp.sys.zenith
Subject: Zenith 386/16 memory board question
Message-ID: <6373@oval.BOEING.COM>
Date: 11 Aug 92 21:39:27 GMT
Distribution: na
Organization: Boeing Computer Services (ESP), Seattle, WA
Lines: 29