• 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.

Prime 95 CPU temps

C

cyclone3d

Fully [H]
2FA
Joined
Aug 16, 2004
Messages
18,192
What is the standard method of testing with Prime 95? Is it with Blend?

I am only running 1.2v at 4.5Ghz and am getting 79-80c on a couple cores when running small FFTs.

Blend is about 10c or so lower than that.

System specs are in sig.
 
Actually theres not a "proper way" or standard method yo use prime 95. Generally People want to be at least 8 hours of p95 small fft test. However thats not an indicative of fully stability as much People can pass 24 hours of prime and crash at the first minute of any heavy CPU game.

Blend test its more a generalized system test RAM, caché, CPU, etc. Its useful when RAM its overclocked or when the overclock its achieved via bclk or with the help of the bclk and again it Will not be any indicative of full stability. If you want to test stability then just play some heavy games i found crysis 3 and far cry 3 as the best stability test ever. I also use dragón age origins wich its fraking intensive not matter What hardware you have it Will use as much resources as it can.. if you can play couple of hours crysis and far cry without any crash, hardlock, BSOD then you are fine.
 
I personally think that Prime 95 is a waste of time, and I prefer Linx or IntelBurnTest. I can only imagine how hot they would get that i7 though, considering it's already pretty hot with Prime 95.
 
the current version of prime 95 is actually a greater overall burden on the system than ibt, especially if you go into the custom setting and specify the amount of memory you want the tests to absorb. prime is best used over the span of a number of hours as each calculative iteration is generally shorter than that of an ibt iteration depending on the amount of memory used. i've stopped using prime as i really can't afford the power bill of a system running full blast for half a day, especially when i'm building multiple systems in a month. ibt for the most part is a pretty accurate metric with a run time at most of a few hours.
 
cyclone3d said:
What is the standard method of testing with Prime 95? Is it with Blend?

I am only running 1.2v at 4.5Ghz and am getting 79-80c on a couple cores when running small FFTs.

Blend is about 10c or so lower than that.

System specs are in sig.
Click to expand...

Use blend with 85%-90% of your RAM. Certain FFT sizes create more stress and heat than others. Off the top of my head, 1344k, 1792k, and 2688k are the ones that are most stressful for Sandy/Ivy Bridge. Not sure about Haswell though.
 
Thanks for the info.

I generally don't use any type of stability testing program when I am overclocking except for when I first get a new CPU.

I am pretty sure I didn't use anything at all with my previous CPU.

Back in the day, I used to run all kinds of stability tests when overclocking and then decided it was a huge waste of time. As long as it is stable in games and all other application I use, it is good enough.

I'm kind of wondering if my NH-D14 is on it's way out as I am pretty sure that this CPU should not be running this hot with the small FFT test at only 1.2v.

Even the ASUS ROG article recommends up to 1.3v on air. I don't see how that is valid with doing stress testing as it would definitely go over the 95c temp.
 
You must log in or register to reply here.
Back
Top