Solved

7600k 5ghz error in prime65 but stable in all benchmarks and aida64 stresstest 30 mins

Hello awesome people that want to help, I have an Intel Core i5 7600k (yeah, with brand, model and all), OCed to 5ghz, all right until here, stable in aida 64 (10 mins at least), heaven an superposition without crashes, all apps and games running well, in fact, I didn't realize my problem until 1 month after OCing, so, what's the problem? that in prime65, when I try to run ANY test, it says "Hardware error, 0.5, expected less than 0.4, blah blah" in all workers (4 to be exact), so, my question is, if it's stable in ALL, even in the prime95 benchmark (Oprions>benchmark), why in all of the torture tests it shows the same result that an unstable overclock? (P.S: 1.275v in the vcore, i alredy tried to increase it a little bit, but I don't want to do it)
Reply to ivomola.im
11 answers Last reply Best Answer
More about 7600k 5ghz error prime65 stable benchmarks aida64 stresstest mins
  1. So to recap if I got it right - It's perfectly stable EVERYWHERE but in Prime95?
    Reply to TehPenguin
  2. TehPenguin said:
    So to recap if I got it right - It's perfectly stable EVERYWHERE but in Prime95?


    Yep, even in the prime95 benchmark, but in the "torture tests" (Small FTTs, and all that stuff the tests why people download prime95), it says that no, and my question is if that matters, because I don't need that the head of lara croft renders 0.00000000001 decimal points to the right, you know? With framerates higher than 60 and temperatures under control (I have liquid cooling, so I don't care), I'm just fine
    Reply to ivomola.im
  3. If prime95 is the only instance where your overclock shows any stability issues then I'd just ignore it, personally. Especially if you've used the overclocked CPU for over a month now and had no issues with it.

    I'll be honest I do not know what exact application prime95 simulates(if any) but it seems to be one that you do not care for too much.
    Reply to TehPenguin
  4. TehPenguin said:
    If prime95 is the only instance where your overclock shows any stability issues then I'd just ignore it, personally. Especially if you've used the overclocked CPU for over a month now and had no issues with it.

    I'll be honest I do not know what exact application prime95 simulates(if any) but it seems to be one that you do not care for too much.


    Is a very valid answer, but everyone says "if prime95 fails your oc is unstable, omg omg!!", and I even read somewhere in the internet in this same forum that if this exact same thing happens your cpu is corrupting data, so for that reason I want to make sure that nothing wrong is happening with my cpu, and for that reason too is why I want a "more detailed" answer, but if someone confirms me that nothing bad is happening to my pretty good OCer i5 7600k, then I'll consider my problem solved
    Reply to ivomola.im
  5. Best answer
    there's two sides of the fence

    the side that say if it's not stable for hours upon hours of stress testing, then it's not stable
    then the side that say if it's stable and not crashing for what you use it for then it's fine

    As long as your temps are good and you don't have any system stability issues for your use, then technically, it's fine

    All it means is that systems that can pass hours of torture testing have a much lower chance of issues.

    I thought my OC was fine, then along came GTAV and boom... loads of crashing. Had to up the voltage a little, but it had passed prime95 just fine
    Reply to blockhead78
  6. 10 mins "at least" doesn't mean "stable". Run OCCT / P95 for at least an hour for heavy load testing, and Realbench for an hour for light load testing.

    There is no one utility that's going to tell you if your overclock is perfectly stable, because they're all using different instruction sets. I can tell you, though....that you'll continue to have problems, if the limit of your testing to call it "stable".....is 10 mins long.

    TLDR: If you don't have the patience to stability test properly, it's your own fault....and your files you're screwing up. Enjoy.
    Reply to Vellinious
  7. Vellinious said:
    10 mins "at least" doesn't mean "stable". Run OCCT / P95 for at least an hour for heavy load testing, and Realbench for an hour for light load testing.

    There is no one utility that's going to tell you if your overclock is perfectly stable, because they're all using different instruction sets. I can tell you, though....that you'll continue to have problems, if the limit of your testing to call it "stable".....is 10 mins long.

    TLDR: If you don't have the patience to stability test properly, it's your own fault....and your files you're screwing up. Enjoy.


    With 10 minutes at least I mean that the shortest stress test I ever ran is 10 minutes long, normally I test what people recommend, 30 minutes
    Reply to ivomola.im
  8. ivomola.im said:
    Vellinious said:
    10 mins "at least" doesn't mean "stable". Run OCCT / P95 for at least an hour for heavy load testing, and Realbench for an hour for light load testing.

    There is no one utility that's going to tell you if your overclock is perfectly stable, because they're all using different instruction sets. I can tell you, though....that you'll continue to have problems, if the limit of your testing to call it "stable".....is 10 mins long.

    TLDR: If you don't have the patience to stability test properly, it's your own fault....and your files you're screwing up. Enjoy.


    With 10 minutes at least I mean that the shortest stress test I ever ran is 10 minutes long, normally I test what people recommend, 30 minutes


    Your files
    Reply to Vellinious
  9. Vellinious said:
    ivomola.im said:
    Vellinious said:
    10 mins "at least" doesn't mean "stable". Run OCCT / P95 for at least an hour for heavy load testing, and Realbench for an hour for light load testing.

    There is no one utility that's going to tell you if your overclock is perfectly stable, because they're all using different instruction sets. I can tell you, though....that you'll continue to have problems, if the limit of your testing to call it "stable".....is 10 mins long.

    TLDR: If you don't have the patience to stability test properly, it's your own fault....and your files you're screwing up. Enjoy.


    With 10 minutes at least I mean that the shortest stress test I ever ran is 10 minutes long, normally I test what people recommend, 30 minutes


    Your files


    Look, I don't want to discuss, my cpu passed all the tests (even the game test with superposition benchmark, but my first "real" game arrives in 1 week roughly), and I also alredy choosen a best answer (btw, thanks blockhead78 and TehPenguin, but specially blockhead78), so, to recap, I are not having problems with any program except with prime95, and my question was if that can cause problems, the answer I received from 2 people says that unless I notice any "real" problems (notice the " sign), I shouldn't be worried about it, so, if that's the answer I receive with details and examples (again, thanks blockhead78), that is the answer I'll accept as a solution to my problem (if i can call it "problem"), so, again, thanks TehPenguin, blockhead78 and everyone that wanted to help, that definetly answered my question ;)
    Reply to ivomola.im
  10. Vellinious said:
    10 mins "at least" doesn't mean "stable". Run OCCT / P95 for at least an hour for heavy load testing, and Realbench for an hour for light load testing.

    There is no one utility that's going to tell you if your overclock is perfectly stable, because they're all using different instruction sets. I can tell you, though....that you'll continue to have problems, if the limit of your testing to call it "stable".....is 10 mins long.

    TLDR: If you don't have the patience to stability test properly, it's your own fault....and your files you're screwing up. Enjoy.


    Why so negative and edgy? D:
    Reply to TehPenguin
  11. TehPenguin said:
    Vellinious said:
    10 mins "at least" doesn't mean "stable". Run OCCT / P95 for at least an hour for heavy load testing, and Realbench for an hour for light load testing.

    There is no one utility that's going to tell you if your overclock is perfectly stable, because they're all using different instruction sets. I can tell you, though....that you'll continue to have problems, if the limit of your testing to call it "stable".....is 10 mins long.

    TLDR: If you don't have the patience to stability test properly, it's your own fault....and your files you're screwing up. Enjoy.


    Why so negative and edgy? D:


    I think the exact same thing now that you say it
    Reply to ivomola.im
Ask a new question Answer

Read More

Overclocking Intel Intel i5 7600k Prime95 Stress Test 5ghz