Question what is the best/safest overclocking benchmark ???

[indianajune1]

hey guys , a serious thanks for any advice. I'm about to do my first over clock on my i7700k and id be lying if I wasn't nervous. I have the asus apex ix mother board that has a profile to OC to 5ghz for you so I figured id try that but I'm wondering what the best benchmark is to (as safe as possible) test for stability ? I hear prime 95 thrown around a lot but I heard that's REALLY hard on your cpu. was thinking about OCCT instead. not looking to break any records here just gaming and everyday use. I own 3d mark but I heard that's more for gpu but if it will work I know that well. also how long would you realistically run it for 24 hours seems extreme and unsafe. thanks again !

 

[Dark Lord of Tech]

I always preferred OCCT.

 

[Lutfij]

Prime 95 would be a good start. FYI, there's nothing safe about overclocking. There is always a risk or failure.

 

[vistagamer6969]

Aida64

 

[indianajune1]

Prime 95 would be a good start. FYI, there's nothing safe about overclocking. There is always a risk or failure.

I understand this. That’s why I’m trying to be as safe as possible

 

[drea.drechsler]

hey guys , a serious thanks for any advice. I'm about to do my first over clock on my i7700k and id be lying if I wasn't nervous. I have the asus apex ix mother board that has a profile to OC to 5ghz for you so I figured id try that but I'm wondering what the best benchmark is to (as safe as possible) test for stability ? I hear prime 95 thrown around a lot but I heard that's REALLY hard on your cpu. was thinking about OCCT instead. not looking to break any records here just gaming and everyday use. I own 3d mark but I heard that's more for gpu but if it will work I know that well. also how long would you realistically run it for 24 hours seems extreme and unsafe. thanks again !

You're confusing stability test with benchmarks. They are two different things.

A benchmark is used to determine if your machine is performing as it should. A good CPU benchmark is Cinebench 20 as it puts a heavy, real-world rendering load on the CPU. It's also both fairly repeatable and scaleable. And it's popular too, so you should be able to find scores to compare to.

3dMark is really a GPU benchmark. It does present a CPU score but that is still dependent on GPU performance.

A stress test is used to de-rate a system to an assured level of stability. How stable you want is usually determined by how you plan on using it. For instance, look at a graphics production machine where it might render images that take several hours for a studio frequently under time pressure. Where time is money you'll probably want a very stable machine as just one crash that wastes several hours of work will easily outweigh the few minutes of render time potentially gained. In that case, if it can't pass a Prime95 run for 8 hours at least it shouldn't go in use there.

But if all you're using it for is to play games, then just run OCCT large data set for about 15-20 min's and be done with it. Games don't even put that heavy of a load on a CPU.

A decent stress test for a GPU is to run Valley continuously for about two hours. Be sure VSync, FreeSync, GSync and any other frame rate limiters are turned off. BTW: two hours because that's twice what you should be sitting at the display playing a game anyway. Think safe, be safe, live long and prosper.

 

[william p]

Prime 95 is a stress test for stability testing. It's not a benchmark. HWBOT has overclocking benchmarks, and CPUZ ranks CPU clock speed without regards to stability. For gaming benchmarks Firestrike is popular. Unigine Valley is another video intense benchmark. usrbenchmark.com gives rankings compared to other similar systems. Cinebench has separate tests for CPU and GPU rendering.
You should get baselines on most of these befor you overclock so you will be able to see what improvement you've made.

 

[indianajune1]

You're confusing stability test with benchmarks. They are two different things.

A benchmark is used to determine if your machine is performing as it should. A good CPU benchmark is Cinebench 20 as it puts a heavy, real-world rendering load on the CPU. It's also both fairly repeatable and scaleable. And it's popular too, so you should be able to find scores to compare to.

3dMark is really a GPU benchmark. It does present a CPU score but that is still dependent on GPU performance.

A stress test is used to de-rate a system to an assured level of stability. How stable you want is usually determined by how you plan on using it. For instance, look at a graphics production machine where it might render images that take several hours for a studio frequently under time pressure. Where time is money you'll probably want a very stable machine as just one crash that wastes several hours of work will easily outweigh the few minutes of render time potentially gained. In that case, if it can't pass a Prime95 run for 8 hours at least it shouldn't go in use there.

But if all you're using it for is to play games, then just run OCCT large data set for about 15-20 min's and be done with it. Games don't even put that heavy of a load on a CPU.

A decent stress test for a GPU is to run Valley continuously for about two hours. Be sure VSync, FreeSync, GSync and any other frame rate limiters are turned off. BTW: two hours because that's twice what you should be sitting at the display playing a game anyway. Think safe, be safe, live long and prosper.

Thanks . I’ll prob try OCCT. I’m just worried about doing any permanent damage so I was looking for something middle of the road while I figure it out.

 

[drea.drechsler]

Thanks . I’ll prob try OCCT. I’m just worried about doing any permanent damage so I was looking for something middle of the road while I figure it out.

I really wouldn't worry as you'd have to configure your PC extremely badly for either a benchmark or a stress test to do any damage, something like very high core voltage and/or poorly mounted heatsink. Even then you'd have to run one almost exclusively for several weeks on end to do it because modern systems have internal protections to prevent damage, e.g. they'll self-throttle when reaching a Tjmax condition or shut the system down when reaching a Tcritical condition.

And besides... the purpose for running the program is to identify a problem and take immediate steps to remedy it. Right?