• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Yeesh - this is bad

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

jamespetts

Member
Joined
Aug 10, 2003
Location
UK
I ran Memtest86 (v. 3.0) overnight last night on my main (and not overclocked) system. It returned 158 memory errors in 9 hours. 158!

It's a really old system by modern standards, i440BX chipset, PIII-600 CPU, with 4x 64MB (256MB) RAM. Most of the modules are no-name things, although I think that the most recent module that I added (I upgraded them one at a time) was Crucial.

As I said, nothing's overlclocked - the CAS latency is already at 3.0; I think that I'll have to check whether there are any other timings that can be made slacker, although I have some doubts; timings were less adjustable back in 1998 when I bought the motherboard.

It's a good job that I'm just about to replace the whole system with a brand new P4 based computer (with good quality OCZ RAM), but I'm not throwing this computer away; it's going to be used as a backup server running Linux.

Is this number of errors normal for memory modules run at their stock speed (or, in the case of some of the modlues, which are rated at 133Mhz, less than their stock speed) for up to fice years? Or is something horribly wrong?

I did run a test on another, even older computer (AMD K6-2 233Mhz, my backup server) for just under an hour (that has some really old SIMMS, even som 16Mb SIMMS), and no errors were detected during that time.

This does concern me - is there anyone out there who can cast light on just how serious a problem that this is?
 
First start by testing each individual memory module in the board on it's own, and repeat until you find the dodgy stick - I bet it's just one crappy one.
 
I thought that it might come to that. Is that particular level of errors likely to generate a lot of problems (such as my mad disappearing fonts issue, outlined in "Microsoft Operating Systems")?
 
Hmm, by the way, is there a possibility that the problem could be caused by using PC133 RAM in place of PC100? I know that it'd be running under-spec and in theory should be more reliable, but is it possible that there is some small architectural incompatibility?

They don't sell PC100 RAM any more...
 
I don't think so, the RAM is exactly the same type but spec'd to run at a higher frequency. You would just be underlclocking it.
 
Back