Overclock testing parameter?

miftamail.huda

Prominent
BANNED
Sep 19, 2017
148
0
690
Does this below test is enough to ensure my cpu and memory stability ?
1. Google Stress app : 1 Hour
2. HCI Memtest : 500%
3. Prime 95, Small FFT : 5 hours
4. Prime 95, Blend test 14500 MB Memory (my memory is 16GB) : 12 hours

Thank you :bounce:
 
Solution
You don't need any of those other tests. Prime95 version 26.6, and ONLY version 26.6, for 15-24 hours (24 hrs is better as I have seen many systems throw errors between 15 and 24 hrs but rarely after 24 hrs) on Small FFT.

http://windows-downloads-center.blogspot.com/2011/04/prime95-266.html

15 minutes on Small FFT for thermal compliance verification. For memory, run Memtest86+ for 7 passes. That will take a LONG time to run, so be prepared. If you have not overclocked the memory past what it's XMP profile outlines, then you really don't NEED to test the memory, but it doesn't hurt either because there is always the slim chance of having a faulty stick.

If the Memtest is good for 3 passes, it's 90% likely that it's fine. 7 passes...
You don't need any of those other tests. Prime95 version 26.6, and ONLY version 26.6, for 15-24 hours (24 hrs is better as I have seen many systems throw errors between 15 and 24 hrs but rarely after 24 hrs) on Small FFT.

http://windows-downloads-center.blogspot.com/2011/04/prime95-266.html

15 minutes on Small FFT for thermal compliance verification. For memory, run Memtest86+ for 7 passes. That will take a LONG time to run, so be prepared. If you have not overclocked the memory past what it's XMP profile outlines, then you really don't NEED to test the memory, but it doesn't hurt either because there is always the slim chance of having a faulty stick.

If the Memtest is good for 3 passes, it's 90% likely that it's fine. 7 passes generally assures 99.9% reliability.
 
Solution

miftamail.huda

Prominent
BANNED
Sep 19, 2017
148
0
690


Im using latest Prime95, i think it is 29.4. Newer should be better right?

And i forgot to mention, i did Memtest86 through DOS as well, 12 passes, i leave the test running when i go to work
I tried all recommended app tester from all forum,

But, i have yet run small FFT for 15+ hours , will try this next time

Offcourse , i wont bother with stress test unless i do overclock

Ryzen 5 1600 3,8 GHz 1,25V
Corsair CMK16GX4M2B3000C15R 2933MHz 1,35V SOC 1,1V (16-19-19-36)
Strix B-350F
 
Newer is NOT better. Versions newer than 26.6 use AVX instruction sets which are not steady state and should not be used as a unilateral metric for testing thermal compliance or stability. If you plan to run applications that rely heavily on AVX instruction sets then it's understandable to use a version that incorporates those, but if not, it simply limits your thermal ceiling and creates unrealistic temperatures.




Further, as explained to me and many others on this forum by Computronix, who has far more experience with CPU architectures and testing procedures than 95% of the people you will ever meet, speak to or read about, there is this below. He is also the author of the Intel temperature guide, found here:

http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html



This pretty well sums things up and is equally relevant whether working with an Intel or an AMD system.



I can think of several reasons why x264 encoding or AVX / AVX2 / FMA3 apps won't work as a unilateral metric for thermal testing.

(1) A steady-state workload gives steady-state temperatures; encoding does not.

(2) Simplicity in methodology; most users would find encoding apps unfamiliar and cumbersome to accomplish a simple task.

(3) Most users such as gamers never run any apps which use AVX / FMA, so adaptive or manual voltage aside, it makes no sense to downgrade your overclock to accommodate those loads and temps.

(4) Standardization; Prime95 has been around since 1996; many users are familiar with it.

For the minority of users who routinely run AVX / FMA apps, then P95 v28.5 can be useful tweaking BIOS for thermal and stability testing.


regardless of architecture. P95 v26.6 works equally well across all platforms. Steady-state is the key. How can anyone extrapolate accurate Core temperatures from workloads that fluctuate like a bad day on the Stock Market?

I'm aware of 5 utilities with steady-state workloads. In order of load level they are:

(1) P95 v26.6 - Small FFT's
(2) HeavyLoad - Stress CPU
(3) FurMark - CPU Burner
(4) Intel Processor Diagnostic Tool - CPU Load
(5) AIDA64 - Tools - System Stability Test - Stress CPU

AIDA64's Stress CPU fails to load any overclocked / overvolted CPU to get anywhere TDP, and is therefore useless, except for giving naive users a sense of false security because their temps are so low.

HeavyLoad is the closest alternative. Temps and watts are within 3% of Small FFT's.
 
100% correct. That will get you as close as anybody can ever be to fully stable. Nothing is ever 100% stable as there can be a number of things aside from the CPU and memory that affect stability, but at least if you pass both of those you will have, as much as possible, minimized any potential for silent data corruption due to the configuration.
 

miftamail.huda

Prominent
BANNED
Sep 19, 2017
148
0
690
let me add one thing about prime95 (i'm just curious)

if newer version is using AVX (then limiting the thermal ceiling by 20C)
so is better if i can keep my CPU under 75C (both 26.6 or newer) ?
 
Anything under 80C at full load for 15 minutes or more is thermally compliant. Obviously, lower is better, but under 80°C is compliant with recommendations from Intel. I don't like seeing over 70°C during testing myself, but that's for reasons related to longevity more than anything else.