• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Best Software For Full Stress Test

P

Panda

n00b
Joined
Aug 21, 2008
Messages
22
I was wondering if there was any good full stress testing software available for the post-overclock to verify stability on the system. AT the moment, my tests are just :

Memtest 64 (linux live CD) && 3DMark Vantage.

Does anyone recommend any other good software packages to test overall system stability...preferably free.
 
Everest is good for a full package. Prime 95 to test your CPU overclock.
 
OCCT Tool in power supply test mode.

Nothing will hurt your whole system harder than this believe me.
 
I use Orthos or Prime95 for CPU Testing, I run Memtest for 20 passes (overnight) to make sure memory is solid)

For GPU testing, I tend to throw 3dMark or FutureMark (though this ends up being a good test for my cooling system too.)
 
How about both at once?

uber.jpg
 
I run memtest86+ and loop:

test 5: for timing and/or vdimm (and sometimes ioh) errors
test 7: for lack of vvt or vvt to vdimm related errors. The test will freeze if close to required minimum vvt

Those just assume passing memtest86+ and are just minimums before I boot into Windows. If your getting errors in 86+ you could end up chasing your tail when stressing in Windows.

Fwiw I learned some from testing but really learned it all from Eva2000's post on i4 forum. I have that thread bookmarked as it becomes very useful when trying to figure out memtest86+ errors.
 
Last edited:
Yea its a work in process. Just did that the other night and only the cpu is really overclocked so stressing the gpu to max was irrelevant at the time. Just trying to find an alternative to OCCT psu test.

I did try to allocate threads to P95 and Furmark before about a week ago but Windows bsod on me before I could launch P95. I had allocated 6 threads to P95 and left the other 2 for everything else so I'm not sure if that is a possibility or not as the overclock now is much more stable so will need to try again.
 
enginurd said:
# [RAM] Memtest* (bootable ISO), for at least 15 passes with ZERO errors.

* Memtest86+ v4.00 @ Memtest.org (2009-09-22)
* Memtest86 v3.5 @ Memtest86.com (2009-01-03)

# [CPU] Prime95* Small FFTs test or Intel Burn Test, for at least 12 hours with ZERO errors at acceptable temperatures. Please research the safe temps for your CPU.

* Prime95 v25.9 @ Mersenne.org (2009-03-15)
* IntelBurnTest v2.3 @ Guru3D (2009-07-05)

# [GPU] FurMark, ATI Tool, or 3DMark loop, for at least 6 hours with no visual defects (fragmenting/discoloration/pixilation/etc).

* FurMark v1.7.0 @ Geeks3D.com (2009-07-03)
* ATI Tool v0.26 @ TPU (2006-12-08)
* 3DMark®06 (Build 1.1.0) @ Futuremark

# [HDD] HD Tune Full Error Scan

* HD Tune v2.55 (2008-02-12)

# [HDD] Drive Fitness Test (or any other manufacturer's provided diagnostic test utility, available from their websites).

* Drive Fitness Test v4.16 @ Hitachi
* Data Lifeguard Diagnostic @ WD
* SeaTools @ Seagate/Maxtor
* Hutil @ Samsung

# [CPU/RAM] Prime95* Blend test, for at least 24 hours with ZERO errors.
# Monitor all temperatures during testing. These programs are recommended:

* Core Temp
* Real Temp
* ATI Tool
* Riva Tuner
* Speedfan

# Other programs that may be helpful:

* CPU-Z
* GPU-Z
* HD Tach
* ATTO
* Speedfan
* CrystalMark
* Everest Trialware
* Sandra Lite
* More? Tell us here!
Click to expand...

the above seems pretty thorough
 
I would just comment that so far I've found that if you can pass memtest86+ test 5 and 7 for 24 hours of looping without an error it is highly unlikely that you will bsod in Windows. One thing I've found recently is that with the 1366 socket all three of the RTL (round trip latencies) need to be at least 1 looser then the previous. So n,n+1,n+2 where n is the timing given the first channel. If not it can take a long time for the error to show up in memtest86+ and make you want to pull out your hair trying to do everything in Windows to get it stable.
 
Getting closer. Need to swap in my two 5870s to see if I get the same numbers. Using the DX10 path I was able to load up both gpus and keep them around 100% the entire time. Using OGL 3.0 did the same as last and would only load one gpu. I used Afterburner to kick the clocks up on both gpus other wise they would just sit there at 500/507.

Kombuster: 1440x900 DX10, stress + post + 8x AA
P95 custom: FFTs 2048 to 4096 using 3072MB ram not run in place
uberstress.jpg
 
You must log in or register to reply here.
Back
Top