Is Memtest86 still a valid memory test?

C

CitizenInsomniac

Limp Gawd
Joined
May 26, 2005
Messages
134
I recently ran into an OC issue where MemTest86+ 1.70 was passing all tests with flying colors, yet Orhots Large FFT tests failed repeatedly for the same configuration (see http://www.hardforum.com/showthread.php?t=1266410 for detail).

I did a lot of searching on these forums and googling around for info, and it turned out I wasn't the only experiencing such issues: MemTest86 passing tests, but Orthos/Prime95 failing Large FFTs or Blend tests.

It would seem to me that MemTest86 is no longer a valid test of RAM stability, at least on the new Intel dual-core chipsets (I can't speak for other chipsets).

Agree? Disagree?
 
I thought memtest was only for testing for actual bad RAM modules. The Prime95 tests are for testing stable OCs. This is what I've always been lead to believe.
 
You could probably find people with the opposite, as well. You should always use several tools to ensure the most scenarios are tested. Everything stresses the system in different ways.
 
keetha3 said:
I thought memtest was only for testing for actual bad RAM modules. The Prime95 tests are for testing stable OCs. This is what I've always been lead to believe.
Click to expand...
But that's not its only usage. If you overclock your RAM too far, you can easily make MemTest86 find errors and fail too. The problem seems to be that MemTest86 is generally recommended (I mean, just read the OC software sticky in this forum) as a test of OC memory stability - and yet clearly it's not sufficient on its own to fully diagnose OC memory issues. So is there even a point in continuing to use it and recommend it? (Aside from testing for real RAM hardware errors, I suppose.)
 
keetha3 said:
I thought memtest was only for testing for actual bad RAM modules. The Prime95 tests are for testing stable OCs. This is what I've always been lead to believe.
Click to expand...
But that's not its only usage. If you overclock your RAM too far, you can easily make MemTest86 find errors and fail too. The problem seems to be that MemTest86 is generally recommended (I mean, just read the OC software sticky in this forum) as a test of OC memory stability - and yet clearly it's not sufficient on its own to fully diagnose OC memory issues. So is there even a point in continuing to use it and recommend it? (Aside from testing for real RAM hardware errors, I suppose.)
 
I just double checked that link, and it seems to confirm my view. It's under "memory testing." As in to test for good/bad RAM. It's not under the stability or OC test categories.
 
You must log in or register to reply here.
Back
Top