cancel
Showing results for 
Search instead for 
Did you mean: 

Best way to stability check a CPU OC

Rokha88
Level 8
Hi everyone,

Since a few days, I am trying to get a slight OC (4,7Ghz@1.225v BIOS) on my I7-7700K, but as it is a hot chip, I am quite confused by the very disparate results I get from a stress test to another, especially about temperatures.

What software would you recommend to stress-test the stability and for how long ? I have read that Realbench is a good test as it is quite realistic. Some recommend to run the benchmark for a few hours, some recommend the Stress-test for a few hours ? What about it ?

Another concern is that on IntelBurnTest, Prime95 (most recent version) & OCCT Linpack, I get INSTANTLY crazy temperatures (90 C°/194F° average), how can it be so hot instantly ? I have read that it may be due to the AVX instructions, as I don't get these temps on other software, or on older versions (example V266 of prime95 it seems to work great). The avg is then 73C° (163F°) Should I bother with it ?

For information, I have Maximus VIII Ranger, I7-7700K (4,7GHZ@1.225v bios), Scythe Fuma REV.B (X2 fans) CPU cooler, and Zalman Z11+ Case (1 front intake / 1 top exhaust, 1 rear exhaust).


Thanks for your answers :-).

Bye.
4,964 Views
13 REPLIES 13

Silent_Scone
Super Moderator
For the platform you are using I would still recommend Realbench. Prime95 induces an incredible amount of current based on a totally unrealistic AVX routine. For 9/10 users a video encode will be the most taxing AVX load put on the system, therefore Realbench is ideal for emulating a real work use. It depends on what one intends to use the system for.
13900KS / 8000 CAS36 / ROG APEX Z790 / ROG TUF RTX 4090

Silent Scone@ASUS wrote:
For the platform you are using I would still recommend Realbench. Prime95 induces an incredible amount of current based on a totally unrealistic AVX routine. For 9/10 users a video encode will be the most taxing AVX load put on the system, therefore Realbench is ideal for emulating a real work use. It depends on what one intends to use the system for.



Thank you. I have ran Realbench for 8 hours straight and set 16 RAM (which is my maximum) and it seems it has ran with no trouble @ 1.248v (hwmonitor).

The problem is that not I am trying to put it in adaptive mode but it seems it wants to go above the required 1.248v. How can I solve this as it gets to 1.264v, which I don't think is necessary?

Is it 100% sure that the adaptative will pull enough voltage, even at lower CPU charges?

Thank you.

Rokha88 wrote:
Thank you. I have ran Realbench for 8 hours straight and set 16 RAM (which is my maximum) and it seems it has ran with no trouble @ 1.248v (hwmonitor).

The problem is that not I am trying to put it in adaptive mode but it seems it wants to go above the required 1.248v. How can I solve this as it gets to 1.264v, which I don't think is necessary?

Is it 100% sure that the adaptative will pull enough voltage, even at lower CPU charges?

Thank you.


Adaptive uses Intel's VID table, so it is not possible to go below the minimum VID for said frequency. You would need to run a negative offset, however be aware this is applied to the entire VID stack, not just the turbo multipliers.
13900KS / 8000 CAS36 / ROG APEX Z790 / ROG TUF RTX 4090

Rokha88
Level 8
I had a look at it and the VID mentions 1.245v, the highest.

That's quite complicated to find the perfect balance between sufficent voltage at middle-high charge without going sky high, and not getting too low on idle loads..

I have heard of "SVID Behaviour" option but I can't find it in my bios, which is up to date still.

Rokha88 wrote:
I had a look at it and the VID mentions 1.245v, the highest.

That's quite complicated to find the perfect balance between sufficent voltage at middle-high charge without going sky high, and not getting too low on idle loads..

I have heard of "SVID Behaviour" option but I can't find it in my bios, which is up to date still.


As I said, you can apply either a negative or positive offset, however, this isn't true adaptive. The minimum VID is definined by the individual CPU. Not something within our control.
13900KS / 8000 CAS36 / ROG APEX Z790 / ROG TUF RTX 4090

Falkentyne
Level 12
One way to VERY quickly determine if you are unstable is to use java minecraft, if you own it.
Run HWinfo64 in sensors only mode and go to WHEA area.

Then load minecraft to main menu about 20 times in a row
MC puts a 100% load (almost) on all cores when loading the game.
If you are completely unstable, you will get CPU Cache L0 error.

Easy test.

I would not use a game to stress test a system, personally I would use Aida64 stress test. if it detects a system flaw it immediately shuts down the test and if not after about 30 minutes/heat soak of the equipment, you know you have stable system.

RedSector73 wrote:
I would not use a game to stress test a system, personally I would use Aida64 stress test. if it detects a system flaw it immediately shuts down the test and if not after about 30 minutes/heat soak of the equipment, you know you have stable system.


I agree with you. However there are some cases where you are temperature limited and can't run AIDA64 at all, because AIDA64 "Stress FPU" (the test you want to run normally), uses AVX instructions, and runs as hot as small FFT Prime95 with AVX disabled.

In that case, you need a less stressful stress test. My minecraft java (i do not know about the windows version) test is reliable because it puts a 100% all core burst gaming load on the CPU. This is different from a "Synthetic" load because of something called "transientse"--a burst gaming load will have worse transients, which helps "balance" the much higher current draw of the synethetic load dropping voltage more.

Give it a try.

Aida64 - System Stability Test
Can stress
CPU
FPU
Cache
System memory
Local disk(s)
GPU

As a whole or individually

I'd call that comprehensive stress test, besides once got stable in it never found anything that prove otherwise.