home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!pipex!unipalm!uknet!brunel!concurrent.co.uk!nnw
- From: nnw@concurrent.co.uk (Neil Watson)
- Newsgroups: comp.sys.ibm.pc.hardware
- Subject: Re: 3 chip vs 9 chip SIMMS
- Keywords: SIMMS
- Message-ID: <1992Nov20.152110.24880@concurrent.co.uk>
- Date: 20 Nov 92 15:21:10 GMT
- References: <1992Nov19.110550.14222@concurrent.co.uk>
- Sender: usenet@concurrent.co.uk (NetNews System)
- Organization: Concurrent Computer Corporation, Slough, England
- Lines: 46
- Nntp-Posting-Host: bugs.concurrent.co.uk
-
- To follow up my own problem with what appears to be a solution...
-
- In article <1992Nov19.110550.14222@concurrent.co.uk> nnw@concurrent.co.uk (Neil Watson) writes:
- [para deleted] Thanks to Phil Lee for providing the interesting
- articles.
- >
- >I now have purchased a new 4Mb of memory for my (Companion - an
- >AMI/ETEQ chipset) 386DX-40, and they don't work. I either get parity
- >errors or just plain "won't boot" with the new memory installed.
- >The problem seems to worsen as the machine warms up.
- >
- >According to the documentation the motherboard will accept 256k, 1M or
- >4M SIMMS. I thought the fact that it took the latter implied that the
- >new 3 chip 1Mb SIMMS would be OK....
- >
- >I recall seeing a very informative posting on how the different refresh
- >cycles worked for 3 chip vs 9 chip SIMMS, but didn't save it (fool!).
- >However I'm not sure this is the problem - I may have a genuinely faulty
- >module (as it fails more often as it warms up?).
- >
- >The memory vendor has suggested "turn off the parity checking", but as
- >it's there for a purpose, I'm unwilling to do that. I think that that
- >will just mean that I don't get errors detected!
- >
-
- Anyway, bouquet of the day goes to Robert Tashjian (tashjian@netcom.com)
- who suggested checking if the BIOS supported Hidden Refresh or Slow
- Refresh. Sure enough the AMI BIOS I have (1990 vintage) has "Concurrent
- Refresh" and "AT type refresh" options. Setting the "AT type refresh"
- instead of the default makes it all seems to work. Aparently the BIOS is
- smart enough to figure out what it can and cannot do with the 4Mb simms,
- but didn't know about the style of refresh needed for the 1Mb 3 chip
- ones.
-
- Anyway, all seems well after an hour of running. The previous longest
- I'd managed with the new SIMMS was 5 mins!
-
- Thanks (publicly) gents, I hope this information may be of help to
- others out there in netland.
- Neil
-
- --
- Neil Watson, ESDG Product Support, Concurrent Computer Corp
- G0BLD 227 Bath Road, Slough, Berks, SL1 4AX, England
- Phone: (+44) 753 513360 FAX: (+44) 753 513303
- (nnw@slough.ccur.com, nnw@concurrent.co.uk or 100021,3041 @ Compuserve)
-