[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y ] [Search | Free Show | Home]

How do you guys stress test your CPU overclocks? I've been

This is a blue board which means that it's for everybody (Safe For Work content only). If you see any adult content, please report it.

Thread replies: 25
Thread images: 3

File: 1434055698963.jpg (34KB, 480x802px) Image search: [Google]
1434055698963.jpg
34KB, 480x802px
How do you guys stress test your CPU overclocks?

I've been hearing talk here and there that using stress tests with AVX is not a good idea because it achieves temperatures you can never achieve in any other situation. But if you don't stress test the AVX instructions, how can you be sure the AVX component is stable? Is there a way to test AVX moderately? So that it is still used, but at lower temps?

Do you guys not stress test with AVX at all? What's your workflow?
>>
yui best girl
>>
>>61130948
i see yui, i like
>>
>>61130948
prime or cpuz for quick stabilty tests
asus real bench and occt for overnight stressing

i skip avx, just causes unnecessary heat
>>
>>61131018
>i skip avx
Then how do you know your CPU is stable if you're not testing that section?

Many programs, and even some games, use AVX instructions.
>>
>>61131032
it never gets hit hard enough in normal loads, there's a reason why avx stress will send your cpu to housefire levels and yet games never reach half of it.
>>
>>61131056
Yes, correct. But if you entirely skip testing AVX. How do you know your CPU will be stable when a program that uses AVX is running? I'm not talking about temps now.
>>
>>61130948
OCCT uses modified avx tests that doesn't produce much heat but will lock your computer up.
>>
>>61131080
That's interesting. Gonna check it out. Thanks.
>>
Just go for broke. If your system powers through the AVX barbecue and other edge cases just fine, it'll likely be rock solid stable for everything else as well. Holding back on tests because they might be unrealistically stressful is missing the point.
>>
>>61131072
because it's never crashed since??

i mean if you are really worried about that.. then just stress it. Personally i dont bother and ive never had issues, infact a lot of "stress tests" dont use avx so i have a feeling it doesnt contribute to system instability that much on its own.
>>
stresstesting is a meme you mong
>>
>>61131137
>because it's never crashed since??
That does not mean it's stable. The side effects caused by instability can go from a simple bit switch in a file (which might corrupt an OS file), to a whole disk partition corruption.

Also you might not be using an AVX app today, and you'll be using one tomorrow without you even knowing it.
This thread is about how to test AVX without it trying to generate as much heat as possible, which limits maximum overclocks.
>>
File: Saber_is_a_prostitute.gif (170KB, 850x900px) Image search: [Google]
Saber_is_a_prostitute.gif
170KB, 850x900px
>>61131007
>>61131015
>Yuifags
>>
>>61131168
that goes for any overclock anon, you can never know your system is 100% stable. A cosmic ray can equally cause a bit to flip and corrupt your data.

Trust me anon, avx stability is a meme, it only generates excessive heat which affects your otherwise "stable" OC. but w/e if your autism is ticking about avx stress using occt small fft until your heart is content. you'll just end up with a lower OC
>>
>>61131317
>avx stress using occt small fft
Is this the moderate AVX test that >>61131080 was talking about?
>>
>>61130948
Y U I
U
I

That said, most people here are morons who say Prime95 AVX is "too severe" or "unrealistic". The whole point of a stress test is to work the CPU as hard as it can possibly be worked. If the CPU survives that it's sure to survive any vidya games, kernel compilations, photo rendering, or anything else you want to throw at it. Do P95 blend - latest version, with AVX - for 24 hours straight to validate your final overclock.
>>
All you're doing is hammering your CPU with floating point shit that literally no other application does to that extent. If you want to burn test your CPU with "that" version of prime95 and use small FFT mode, you're just proving that your CPU can handle running prime95 in small FFT mode.

You can compile, render 3D art, encode video, and play 3 games simultaneously and your CPU will still never get hammered with floats the same way prime95 will.
>>
>>61130948
OCCT and Prime95 Small FFT for quick testing
Prime95 Blend for 12 hours or so for bulk stability and thermals testing
Throw in some IBT Very High for memory and IMC testing
Round it all off by running some distributed computing applications (SETI, Einstein, Milkyway, Collatz, and Asteroids) for up to a week and see if any workunits come back as invalid.

If all of the above passes cleanly, then i consider it stable.
>>
File: 1497848119828.png (30KB, 284x339px) Image search: [Google]
1497848119828.png
30KB, 284x339px
>>61130948
>how do you guys stress test your cpu overclocks

I can stress test my entire aged thinkpad by trying to run skyrim at anything above low settings.
>>
>>61131709
this. you are just killing your cpu.
>>
>>61131709
Duh. That's the point of a stress test, to be more difficult than any real application you'll ever run.
>>
>>61131738
This guys knows what's up, maybe you can stress test in some shitty games if you run your PC for a few hours and turn it off, but with constant sleep and wake cycles and month long uptimes, that's not gonna pass.
>>
>>61131738
But that means you're severely limiting your maximum OC to keep the Prime95 Small FFT temps in check.
>>
>>61130948
Use Furmark.
Thread posts: 25
Thread images: 3


[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y] [Search | Top | Home]

I'm aware that Imgur.com will stop allowing adult images since 15th of May. I'm taking actions to backup as much data as possible.
Read more on this topic here - https://archived.moe/talk/thread/1694/


If you need a post removed click on it's [Report] button and follow the instruction.
DMCA Content Takedown via dmca.com
All images are hosted on imgur.com.
If you like this website please support us by donating with Bitcoins at 16mKtbZiwW52BLkibtCr8jUg2KVUMTxVQ5
All trademarks and copyrights on this page are owned by their respective parties.
Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.
This is a 4chan archive - all of the content originated from that site.
This means that RandomArchive shows their content, archived.
If you need information for a Poster - contact them.