F5F Stay Refreshed Power Users Overclocking Consider using LinX / IBT for stress testing.

Consider using LinX / IBT for stress testing.

Consider using LinX / IBT for stress testing.

B
brightzewl
Junior Member
33
01-20-2016, 11:35 AM
#1
I've been adjusting my 6700K recently. I realized that pushing it beyond 4.6GHz would require very high voltages to keep it stable. When I tried 4.6GHz with 1.385V for two hours using OCCT, it worked smoothly between 65-75c. But with IBT or LinX, it hits 95c in just two minutes and starts throttling. Gaming feels fine though. Should I think my overclocking is unstable or try other stress tests besides OCCT? P.S. Sorry for the unclear English.
B
brightzewl
01-20-2016, 11:35 AM #1

I've been adjusting my 6700K recently. I realized that pushing it beyond 4.6GHz would require very high voltages to keep it stable. When I tried 4.6GHz with 1.385V for two hours using OCCT, it worked smoothly between 65-75c. But with IBT or LinX, it hits 95c in just two minutes and starts throttling. Gaming feels fine though. Should I think my overclocking is unstable or try other stress tests besides OCCT? P.S. Sorry for the unclear English.

S
ScoooobyDoo
Junior Member
29
01-26-2016, 04:57 PM
#2
Speedster :
mcnumpty23 :
i wouldn't waste time on stress testing
if you're just playing games and using it normally, it's fine. Stress testing isn't really necessary beyond what regular use does. Thanks for clarifying!
It makes sense to check your OC by simulating everyday use. The issue is I don't always have time for long gaming sessions—stress tests are helpful for stability checks. I usually run stress tests for about 2-4 hours with different programs (like OCCT, Real Bench, x264), and if it works well then I test it overnight. If it passes both the stress test and my regular gaming, I think it's stable.
What's the point of LinX and IBT if they rarely reflect real-time usage? Not that those tests...
S
ScoooobyDoo
01-26-2016, 04:57 PM #2

Speedster :
mcnumpty23 :
i wouldn't waste time on stress testing
if you're just playing games and using it normally, it's fine. Stress testing isn't really necessary beyond what regular use does. Thanks for clarifying!
It makes sense to check your OC by simulating everyday use. The issue is I don't always have time for long gaming sessions—stress tests are helpful for stability checks. I usually run stress tests for about 2-4 hours with different programs (like OCCT, Real Bench, x264), and if it works well then I test it overnight. If it passes both the stress test and my regular gaming, I think it's stable.
What's the point of LinX and IBT if they rarely reflect real-time usage? Not that those tests...

C
CozyTea
Member
106
01-26-2016, 06:43 PM
#3
i wouldn't bother with stress testing
if in gaming and normal use it's fine, but it's not really required to push it beyond everyday demands
C
CozyTea
01-26-2016, 06:43 PM #3

i wouldn't bother with stress testing
if in gaming and normal use it's fine, but it's not really required to push it beyond everyday demands

P
PaigeOfTheBook
Senior Member
733
02-12-2016, 07:44 AM
#4
I wouldn't waste effort on stress testing unless in gaming or regular use—it's not a strict requirement beyond daily operation. It helps ensure stability when you're not using it intensively. The best check is through everyday use, which is what matters most. Stress tests are handy for long sessions with programs like OCCT, Real Bench, and x264; if they pass alongside gaming, it’s reliable. LinX and IBT are useful but don’t always reflect real-time performance, though they’re still decent. Any other stress test suggestions?
P
PaigeOfTheBook
02-12-2016, 07:44 AM #4

I wouldn't waste effort on stress testing unless in gaming or regular use—it's not a strict requirement beyond daily operation. It helps ensure stability when you're not using it intensively. The best check is through everyday use, which is what matters most. Stress tests are handy for long sessions with programs like OCCT, Real Bench, and x264; if they pass alongside gaming, it’s reliable. LinX and IBT are useful but don’t always reflect real-time performance, though they’re still decent. Any other stress test suggestions?

Z
79
02-12-2016, 06:36 PM
#5
Speedster :
mcnumpty23 :
I wouldn’t waste time on stress testing unless it’s for gaming or regular use—it’s not necessary to push it beyond normal daily demands. Thanks for clarifying!
It makes sense that evaluating your OC through everyday usage is the best approach. Since long gaming sessions aren’t always possible, stress tests become really helpful for checking stability. I usually run stress tests for about 2 to 4 hours across different programs (such as OCCT, Real Bench, and x264), and once I reach a stable result, I test it overnight. If it holds up alongside my gaming time, I treat it as reliable.
What’s the value of LinX and IBT if they rarely reflect real-time performance? Not exactly, but they’re still decent tools.
Do you have any suggestions for another stress test?
Test, test, test—those who skip testing their overclocking often end up with crashes or frequent Windows reformats. They don’t realize that an unstable CPU overclock can cause problems with memory controllers, leading to bad data reads and writes, errors, and corrupted files on your drives.
I usually start with quick checks before longer ones, and I’m only satisfied when everything passes.
A quick Cinebench run gives me a good initial idea of stability. If it fails, it’s unlikely to pass anything else. Running 50 tests with standard IBT settings takes about 10 minutes and accurately reflects stress and heat. Doing 50 with Max settings for RAM is time-consuming but comparable to running 16 hours of Prime95.
Realbench is great for everyday checks.
For overnight tests, I begin with AIDA64’s stress test—it’s quite straightforward to pass. If you fail, you’re unlikely to succeed with anything else; instability is a sign that you need more power or reduced speed.
I usually run a 8-hour stress test on Prime95 overnight.
Overnight, I limit realbench to 8–9 hours.
Here are my current results, but I started overclocking yesterday. I’m using a Ryzen R7 1700 at 3.9GHz, on an MSI B350m motherboard, with 16GB (2x8GB) G.Skill Trident Z RGB at full speed.
http://imgur.com/j2rGryt
IBT standard
http://imgur.com/a/DylSq
Realbench
http://imgur.com/a/Jubrb
Cinebench
http://imgur.com/a/qqmUr
9 and a half hours of AIDA64
Also, the motherboard temperature readings in AIDA64 are from the MOSFETs.
Z
zbartholomew18
02-12-2016, 06:36 PM #5

Speedster :
mcnumpty23 :
I wouldn’t waste time on stress testing unless it’s for gaming or regular use—it’s not necessary to push it beyond normal daily demands. Thanks for clarifying!
It makes sense that evaluating your OC through everyday usage is the best approach. Since long gaming sessions aren’t always possible, stress tests become really helpful for checking stability. I usually run stress tests for about 2 to 4 hours across different programs (such as OCCT, Real Bench, and x264), and once I reach a stable result, I test it overnight. If it holds up alongside my gaming time, I treat it as reliable.
What’s the value of LinX and IBT if they rarely reflect real-time performance? Not exactly, but they’re still decent tools.
Do you have any suggestions for another stress test?
Test, test, test—those who skip testing their overclocking often end up with crashes or frequent Windows reformats. They don’t realize that an unstable CPU overclock can cause problems with memory controllers, leading to bad data reads and writes, errors, and corrupted files on your drives.
I usually start with quick checks before longer ones, and I’m only satisfied when everything passes.
A quick Cinebench run gives me a good initial idea of stability. If it fails, it’s unlikely to pass anything else. Running 50 tests with standard IBT settings takes about 10 minutes and accurately reflects stress and heat. Doing 50 with Max settings for RAM is time-consuming but comparable to running 16 hours of Prime95.
Realbench is great for everyday checks.
For overnight tests, I begin with AIDA64’s stress test—it’s quite straightforward to pass. If you fail, you’re unlikely to succeed with anything else; instability is a sign that you need more power or reduced speed.
I usually run a 8-hour stress test on Prime95 overnight.
Overnight, I limit realbench to 8–9 hours.
Here are my current results, but I started overclocking yesterday. I’m using a Ryzen R7 1700 at 3.9GHz, on an MSI B350m motherboard, with 16GB (2x8GB) G.Skill Trident Z RGB at full speed.
http://imgur.com/j2rGryt
IBT standard
http://imgur.com/a/DylSq
Realbench
http://imgur.com/a/Jubrb
Cinebench
http://imgur.com/a/qqmUr
9 and a half hours of AIDA64
Also, the motherboard temperature readings in AIDA64 are from the MOSFETs.

G
guntaj800
Member
112
02-14-2016, 05:27 PM
#6
Speedster :
mcnumpty23 wouldnt bother with stress testing
if it's just for gaming and normal use – it's fine, but not really required beyond everyday demands.
Thanks for the reply!
It makes sense to verify your OC through regular playtime. Stress tests are handy when I don't have time for long sessions, helping check stability. I usually run stress tests for 2-4 hours with different programs (like OCCT, Real Bench, x264), and after a good result I test it overnight. If it works alongside my gaming, I think it's stable.
What's the value of LinX and IBT if they rarely reflect real-time use? Not exactly, but they're still decent.
Do you have any suggestions for another stress test?
I tend to prefer Asus Realbench because it simulates real-life usage better.
G
guntaj800
02-14-2016, 05:27 PM #6

Speedster :
mcnumpty23 wouldnt bother with stress testing
if it's just for gaming and normal use – it's fine, but not really required beyond everyday demands.
Thanks for the reply!
It makes sense to verify your OC through regular playtime. Stress tests are handy when I don't have time for long sessions, helping check stability. I usually run stress tests for 2-4 hours with different programs (like OCCT, Real Bench, x264), and after a good result I test it overnight. If it works alongside my gaming, I think it's stable.
What's the value of LinX and IBT if they rarely reflect real-time use? Not exactly, but they're still decent.
Do you have any suggestions for another stress test?
I tend to prefer Asus Realbench because it simulates real-life usage better.