F5F Stay Refreshed Power Users Overclocking AIDA 64, IntelBurn Test, Prime 95, OCCT

AIDA 64, IntelBurn Test, Prime 95, OCCT

AIDA 64, IntelBurn Test, Prime 95, OCCT

Y
yarabi87
Member
174
11-14-2017, 05:03 AM
#1
Hello there,
Today I adjusted my 7600k to 4.4 GHz (planning to go to 4.5 GHz next). I changed the multiplier to 44, cache to 39, BCLK remained at 100.0 MHz, LLC set to level 1, and the voltage was 1.255. After booting into Windows, I proceeded with the tests.

First, I ran AIDA 64 for 10 minutes, stressing CPU and FPU, and it stayed stable. Here are the results:
Next, I performed IntelBurn Test v2, also maintaining stability after 10 tests:
Then, I tested two versions of Prime 95 (29.4 and 26.6), both remaining stable for about 10 minutes:
26.6: 29.4:

Finally, I conducted an OCCT CPU test in both 32 and 64-bit modes using a large dataset. However, the 64-bit mode failed after a few seconds, while the 32-bit mode completed within 10 minutes.
*UPDATE* I also ran Intel's XTU stress test for 10 minutes, and it passed:

I’m unsure what’s causing the issue. All other programs function normally, except for the 64-bit mode.
Is there something incorrect with my OC, or would you think it’s stable?
Thanks,
salms2s
Y
yarabi87
11-14-2017, 05:03 AM #1

Hello there,
Today I adjusted my 7600k to 4.4 GHz (planning to go to 4.5 GHz next). I changed the multiplier to 44, cache to 39, BCLK remained at 100.0 MHz, LLC set to level 1, and the voltage was 1.255. After booting into Windows, I proceeded with the tests.

First, I ran AIDA 64 for 10 minutes, stressing CPU and FPU, and it stayed stable. Here are the results:
Next, I performed IntelBurn Test v2, also maintaining stability after 10 tests:
Then, I tested two versions of Prime 95 (29.4 and 26.6), both remaining stable for about 10 minutes:
26.6: 29.4:

Finally, I conducted an OCCT CPU test in both 32 and 64-bit modes using a large dataset. However, the 64-bit mode failed after a few seconds, while the 32-bit mode completed within 10 minutes.
*UPDATE* I also ran Intel's XTU stress test for 10 minutes, and it passed:

I’m unsure what’s causing the issue. All other programs function normally, except for the 64-bit mode.
Is there something incorrect with my OC, or would you think it’s stable?
Thanks,
salms2s

A
ALargeBoulder
Member
58
11-14-2017, 12:37 PM
#2
Of course it's not stable. That's a lot of testing. OCCT works well since it quickly shows up bad overclocks, and the small set is ideal for that because it focuses on the CPU. But honestly, just 10 minutes of testing isn't enough. OCCT should be run for at least 8 hours or a full hour with the small set.
A
ALargeBoulder
11-14-2017, 12:37 PM #2

Of course it's not stable. That's a lot of testing. OCCT works well since it quickly shows up bad overclocks, and the small set is ideal for that because it focuses on the CPU. But honestly, just 10 minutes of testing isn't enough. OCCT should be run for at least 8 hours or a full hour with the small set.

W
Wapher
Junior Member
16
11-16-2017, 01:05 AM
#3
Of course it's not stable. lel
Ten minutes of tests isn't enough. OCCT really works because it quickly shows bad overclocks. The OCCT small set is best for that since it focuses on the CPU.
But honestly, running Prime for 10 minutes and thinking it's stable doesn't make sense. At least 8 hours or more and an hour of OCCT small set would be better.
Isn’t small set meant for thermals instead of testing? And why would the 64-bit version stop working after a few seconds when the 32-bit one did? Which Prime version should I use?
W
Wapher
11-16-2017, 01:05 AM #3

Of course it's not stable. lel
Ten minutes of tests isn't enough. OCCT really works because it quickly shows bad overclocks. The OCCT small set is best for that since it focuses on the CPU.
But honestly, running Prime for 10 minutes and thinking it's stable doesn't make sense. At least 8 hours or more and an hour of OCCT small set would be better.
Isn’t small set meant for thermals instead of testing? And why would the 64-bit version stop working after a few seconds when the 32-bit one did? Which Prime version should I use?

C
Char1ie_XD
Senior Member
578
11-16-2017, 01:12 AM
#4
Prime 26.6 is a CPU stress test. I successfully completed one hour of it to confirm my OC, along with 21 hours of Prime and 4 hours of Realbench.
The reason for failure isn't clear—should be an error message indicating instability. If it appears, the issue lies in the OC settings.
Could also be a thermal limit warning; adjusting OCCT parameters might resolve it.
Unless you overclocked RAM, consider dropping that setting.
C
Char1ie_XD
11-16-2017, 01:12 AM #4

Prime 26.6 is a CPU stress test. I successfully completed one hour of it to confirm my OC, along with 21 hours of Prime and 4 hours of Realbench.
The reason for failure isn't clear—should be an error message indicating instability. If it appears, the issue lies in the OC settings.
Could also be a thermal limit warning; adjusting OCCT parameters might resolve it.
Unless you overclocked RAM, consider dropping that setting.

I
50
11-16-2017, 06:51 AM
#5
Gaidax :
Prime 26.6, small test was a CPU stress test. I completed one hour of it to confirm my OC, together with 21 hours of Prime and 4 hours of Realbench.
What causes failure? How can we be sure? It should display an error message, but the key point is – if it fails then your OC isn’t stable. No need to circle around.
Regarding the issue, the error could be a thermal limit warning, which might be resolved by adjusting settings in OCCT. Apart from that, did you overclock RAM? If yes, remove it.
It simply says "Stopped. Error Detected on Core #1-4 (varies). I also ran Real Bench for 1 hour and passed without problem. And no, I never overclocked RAM. It’s still the same."
I
itsWammyKablam
11-16-2017, 06:51 AM #5

Gaidax :
Prime 26.6, small test was a CPU stress test. I completed one hour of it to confirm my OC, together with 21 hours of Prime and 4 hours of Realbench.
What causes failure? How can we be sure? It should display an error message, but the key point is – if it fails then your OC isn’t stable. No need to circle around.
Regarding the issue, the error could be a thermal limit warning, which might be resolved by adjusting settings in OCCT. Apart from that, did you overclock RAM? If yes, remove it.
It simply says "Stopped. Error Detected on Core #1-4 (varies). I also ran Real Bench for 1 hour and passed without problem. And no, I never overclocked RAM. It’s still the same."

S
Sapient
Junior Member
34
11-16-2017, 11:12 AM
#6
Does it report the specific error in the log? Alternatively, remove OC and check if the issue persists.
S
Sapient
11-16-2017, 11:12 AM #6

Does it report the specific error in the log? Alternatively, remove OC and check if the issue persists.

S
simonlh06
Member
50
11-16-2017, 07:16 PM
#7
I made a few adjustments, setting it to 4.5 GHz, 1.310 v, with AVX offset at 1, and it functions properly! No issues were found in the initial seconds. I ran OCCT 64-bit on a large dataset for 10 minutes without any errors detected.
S
simonlh06
11-16-2017, 07:16 PM #7

I made a few adjustments, setting it to 4.5 GHz, 1.310 v, with AVX offset at 1, and it functions properly! No issues were found in the initial seconds. I ran OCCT 64-bit on a large dataset for 10 minutes without any errors detected.