New version OCCT stability test

[RPGWiZaRD]

Yea many thanks for this nice tool Tetedeiench, your long time efforts are very appriciated! By answering in this thread it's obvious to me it's not only your application that is good but I think you have a really great personality as a coder too.

While it's impossible to give proof of this application is better at finding errors than others, for me it's enough that it has detected errors in less than 30 mins when Prime ran a couple of hours. OCCT to me is clearly the #1 Stresstest I will continue using for years to come.

PS, yes I wouldn't say no to a small donation either from my side either, people like you deserve to get paid no matter they're interested in what they're doing or not.

[GAR]

This program makes explorer crash on vista 64, but program is still going and passing the test fine, same setup works fine on XP....so its not the overclock, the program doesnt work well on Vista 64

[eleeter]

...the program doesnt work well on Vista 64

Running now under Vista x64, no explorer crashes for me. Neat program running it now, never heard of it before.

[Nicksterr]

I can't get the program to put a load on my system (q9450), is it a gradual load or something? Using the 2.0.0a release.

[tranceaddict]

It is a good program for testing stability quickly but it is not as good as Prime/Orthos. On two different CPUs I found I can pass the 2h OCCT stability test, yet fail Prime between 10-20 minutes. I needed an additional 0.02V to make it 12+ hour Prime stable.

[CERO]

same applys for peeps yet they will be 2 hours prime stable and fail OCCT within 10-20min, therefore u should use both

[flexy]

It is a good program for testing stability quickly but it is not as good as Prime/Orthos. On two different CPUs I found I can pass the 2h OCCT stability test, yet fail Prime between 10-20 minutes. I needed an additional 0.02V to make it 12+ hour Prime stable.

what version do you use?

I use OCCT 2.0.0.b b1 - very senstive.

Use MEM for NB/memory timings...use "CPU" for CPU.

[Richard Dower]

Any new version of this program out?

[tranceaddict]

what version do you use?

I use OCCT 2.0.0.b b1 - very senstive.

Use MEM for NB/memory timings...use "CPU" for CPU.

Same version as you, the latest one.

[Nicksterr]

Anyone know why I can't get the program to stress? Is there somthing I can do to troubleshoot? prime95 25.6 works just fine...

[jonny_ftm]

This program makes explorer crash on vista 64, but program is still going and passing the test fine, same setup works fine on XP....so its not the overclock, the program doesnt work well on Vista 64

If explorer crashes, in X64 envirement, it means memory is getting corrupted despite the X64 protection architecture. This means your PC is unstable while stressing it with that tool. But, you can choose to ignore it and live with random lockups/errors

I can't get the program to put a load on my system (q9450), is it a gradual load or something? Using the 2.0.0a release.

It is only a display bug on latest versions appearing when you choose a setting with the 1mn idle monitor feature. Check load in taskmanager, it should be the right display of 99%

It is a good program for testing stability quickly but it is not as good as Prime/Orthos. On two different CPUs I found I can pass the 2h OCCT stability test, yet fail Prime between 10-20 minutes. I needed an additional 0.02V to make it 12+ hour Prime stable.

Compare what can be compared: Prime95 in small FFT mode to OCCT in CPU mode (they are the same, with OCCT better managing priorities in my expierience and from what the author said. This is proven by a same load temperatures on all cores for both programs, but achieved faster in OCCT) and Prime95 Blend to OCCT in RAM mode
If you use different modes, you can't compare

[XSAlliN]

E2160 + X cooler at 3000 Mhz (or more):

-what cooler do you use?
-how high those the temp go in *C at 98% CPU usage?

[Branjo]

Hello all, just like to start by saying that I find OCCT invaluable, I am using the 2.0a ATM and one thing I have noticed is that it reports the cores of my Q6600 differently than all other info apps, CoreTemp, RealTemp and Everest Ultimate. I set the cores as 0,1,2,3 meaning core 1,2,3,4 but core 4 always reads as core 3 in the other apps I mentioned. Doesn't amount to a hill of beans really but i have just wondered has anyone else noticed this or have this small issue. I am hoping that OCCT is correct as my core temps for most apps under load seem to give the high,low,low,high readings with the cores pinned as 1-4 but OCCT gives high,low,high,low with cores pinned as 0-3.
I have to say the inbuilt temp monitoring was a sgreat addition, and the graphs at the end of the test are good for determining PSU performance also, totally a 10/10 testing app.

Peace
Branjo

[winguy]

Hi,

I wish to undervolt my T7300 CPU on my laptop (other settings left default), which OCCT v2.0.0a test (CPU, RAM or Mix) should I run to check for stability?

Thanks.

[Speederlander]

Will this fully test an 8-core 5400/Harpertown system?

[Pisklink]

I would like to thank Tetedeiench very much for this nice program!!!! +1 credits

I just overclocked my comp. and I'am using this tool at the moment to test for stability. I like it most of all because it's just so much faster in finding errors instead of waiting for hours to pass. And eventually ending up that a voltage setting only needed one notch up.

I have only two questions tho. I'm using version 2.0.0a and is it correct that the temp. monitor reports 10 degrees higher than is actually is? And in addition is it normal that your comp. makes some weird noises when testing in mix mode?

[Infa]

That it reports 10c higher is becouse tjunction is set to 105c instead of 95c as it "should" be.

its not THAT weird that its coming some sounds from your components, probably its your psu, i dont know the word for the component in english that makes the sound, its those small things with copper wired around them, sometimes, when they are bad insulated, or are toughing something thats beside it, they make small vibrations during heavy load, this is very annoying and if you have this problem you should RMA your psu or whatever is making the sound, almost all hardware have this component. But its mostly common that its the psu or vga card, when you're making load the them.

[Pisklink]

I stress tested my comp. for 8 hours with OCCT but when I start Orthos it will fail in all modes within 10 min. While playing Call of Duty 4 for 1 hour isn't a problem. What could cause this problem? Wrong Orthos maybe? I have a beta version by some guy called Johnson.

[r4st4m4n]

I stress tested my comp. for 8 hours with OCCT but when I start Orthos it will fail in all modes within 10 min. While playing Call of Duty 4 for 1 hour isn't a problem. What could cause this problem? Wrong Orthos maybe? I have a beta version by some guy called Johnson.

I just had a similar problem. Orthos would pass the 1h Mix test but Orthos / Prime95 would fail within seconds. It was my RAM which was not stable. It seems Orthos even in Mix mode is not stressing the RAM a lot.

[stasio]

The latest beta version is OCCT 2.1.0 beta 02 dated 19.Jun 2008
DOWNLOAD

[WaterFlex]

Thanks. Works fine

[Pisklink]

The latest beta version is OCCT 2.1.0 beta 02 dated 19.Jun 2008
DOWNLOAD

Ok, I will give it a try but what if it is stable with OCCT and not with orthos? Could somebody provide a link to the correct Orthos version for testing dual cores? Thanks.

[stasio]

Could somebody provide a link to the correct Orthos version for testing dual cores?

Orthos version for testing dual cores dated 20.April 2006
SP2004 Orthos Edition

[Pisklink]

Thank you very much!!!! I will try both Orthos and OCCT as soon as I'm home from work.

[kiwi]

Ok, I will give it a try but what if it is stable with OCCT and not with orthos? Could somebody provide a link to the correct Orthos version for testing dual cores? Thanks.

It means it is not stable. But if it's not OCCT stable, orthos will fail for sure

Better run newer prime instead of orthos:
http://mersenne.org/gimps/p95v256.zip