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

SuperPi questions

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

Zenoth

Member
Joined
Nov 26, 2004
Location
France, Lyon
I'd like to know if the 32M test in SuperPi (I'm using SuperPi Mod 1.4) is considered a good benchmark for Memory stability ? I mean stability indeed, not really performance. If it passes the 32M bench, then does it mean that my Memory is stable, or is it irrelevant as far as stability is concerned ? Or is it more of a CPU-stability/performance benchmark ?

Also, do I need to run two separate instances ? (for my X2)
 
IMHO, its not that tough of a stability test. I'd suggest memtest86. Superpi doesn't put a continuous load, nor will it fully test the memory. If you do want to use it though, you'd certainly want two instances at the least.
 
Also try Prime95 using the "blend" test to determine memory stability. I would let it run for 24 hours.
 
I think it's a good starting point, but it's definitely not considered stable in my book. I ussually start with running Memtest86+ for a few hours. I don't spend too much time running it because I ussually find what's stable in Windows is quite a few MHz below what's stable in Metest86+ anyway. In Windows I then run Windows Memtest overnight. I also use SuperPi, SP2004 (or Prime95), 3DMark, etc.
 
Thank you guys for your advices.

I also just read that Prime95's "Blend" test is supposed to be a good stability tester for Memory, is it true ? I thought that Prime95 only benched/tested CPU's. Is it the case ?
 
Zenoth said:
Thank you guys for your advices.

I also just read that Prime95's "Blend" test is supposed to be a good stability tester for Memory, is it true ? I thought that Prime95 only benched/tested CPU's. Is it the case ?

Here's what I do:

Run Memtest86 from a boot disk, maybe 4 - 5 full passes. If it passes here, run SuperPi 32 mb. If it passes that then run Prime95 or Orthos at least 10 -12 hours, more if you can, 24 hours preferably.

I've seen numerous occasion of passing Memtest86 and failing SuperPi or passing Memtest86 and SuperPi then failing Prime95 or Orthos.

If you get to Prime95 or Orthos and it runs for a while and fails, you're getting close but not prefection yet. If you fail Memtest86, you have a more severe problem and should fix it without accessing your operating system.

It's a progressive process taken one step at a time. Other testing methods will vary.
 
Back