Hi guys, i have a Maximus VI Hero motherboard, and i currently overclocked it to 4.0 GHz. I am a little bit of a noob, when it comes to overclocking, but if i can boot into the OS and everything, is it considered stable? Or is there anything else i need to test? Also, I have heard of Prime95, but i heard you do not want to use that, while on adaptive mode. So, can anyone tell me how to know if my settings are stable? Thanks.
Knowing if my OC is stable.
- Thread starter xIce52
- Start date
Solution
I like to see at least 3 hours of Prime95 before I consider an OC stable, some will say much more. To each his own.
D/L and install the Intel Extreme Tuning Utility and then you can quickly and easily change between Adaptive mode and static mode without rebooting: http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-boards-software-extreme-tuning-utility.html
Yogi
fkr
Splendid
- May 24, 2009
- 2,646
- 0
- 21,460
so here is the story to the best of my knowledge and this only seems to be true for a haswell chip. the prime95 overwritews some of the voltage tables and messes with the testing when using adaptive.
I would use fixed voltages to find the overclock that works best for your system. I would also use prime 95 to test while using the fixed voltages. once you have tested using prime95 and watching your temps with a program like hwinfo then you can go back into your bios use the adaptive settings then test with aida and make sure when using aida that you are only testing the cpu and the cache and other things available. If you use the other test while in adaptive mode it will have the same result as using prime95 and it will potentially let your CPU spike up to a vcore that may not be safe
btw if you put up your full specs it is always appreciated so we can always verify and know that we are giving our best advice.
I would use fixed voltages to find the overclock that works best for your system. I would also use prime 95 to test while using the fixed voltages. once you have tested using prime95 and watching your temps with a program like hwinfo then you can go back into your bios use the adaptive settings then test with aida and make sure when using aida that you are only testing the cpu and the cache and other things available. If you use the other test while in adaptive mode it will have the same result as using prime95 and it will potentially let your CPU spike up to a vcore that may not be safe
btw if you put up your full specs it is always appreciated so we can always verify and know that we are giving our best advice.
RRDiei
Honorable
- Nov 10, 2013
- 6
- 0
- 10,510
I have been overclocking my new Haswell 4670k for a while, and done quite a bit of stress testing recently, so I hope I can help.
In my experience, I have been able to boot to the OS with different speeds and voltages and do some web browsing and word processing without any problems. However, once I started running a stress test, I would get BSODs even after only a couple of minutes into it. For an OC to be considered stable, you need to make sure that your CPU can cope even when under stress. In other words, you may be able to boot the OS and do some light work without problems, but playing a demanding game will make your computer freeze or give you BSODs.
As Y0GI said, there are lots of stress tests out there. Prime95 is very widely used, but I have read in several places that it is not optimised for Haswell, and does not test all of the processor's capabilities reliably. For Haswell, I have seen Intel Extreme Tuning Utility and Aida64 recommended for stress-testing purposes. Recommended times, again, vary depending on the source: from 6 hours of Aida64, to 8 hours on IXTU, to 24 hours on each. Once again, it depends on your preferences.
Myself, I have been using Aida64 for 8 hours to test my OCs, and I have used IXTU in the background to keep an eye on voltages and temperatures (especially the temperatures!). Then again, this is all preferences, but I have been able to find stability using this method (after a lot of trial and error).
Finally, I would recommend using manual voltages, rather than adaptive, when stress testing your OC. With manual voltages, you know at all times what your CPU is getting, and hence you can work out what the maximum voltage is to achieve stability. After you've worked this out, you can then go back to adaptive and set your maximum stable voltage as your upper limit.
I hope this helps. Good luck with your OC!
In my experience, I have been able to boot to the OS with different speeds and voltages and do some web browsing and word processing without any problems. However, once I started running a stress test, I would get BSODs even after only a couple of minutes into it. For an OC to be considered stable, you need to make sure that your CPU can cope even when under stress. In other words, you may be able to boot the OS and do some light work without problems, but playing a demanding game will make your computer freeze or give you BSODs.
As Y0GI said, there are lots of stress tests out there. Prime95 is very widely used, but I have read in several places that it is not optimised for Haswell, and does not test all of the processor's capabilities reliably. For Haswell, I have seen Intel Extreme Tuning Utility and Aida64 recommended for stress-testing purposes. Recommended times, again, vary depending on the source: from 6 hours of Aida64, to 8 hours on IXTU, to 24 hours on each. Once again, it depends on your preferences.
Myself, I have been using Aida64 for 8 hours to test my OCs, and I have used IXTU in the background to keep an eye on voltages and temperatures (especially the temperatures!). Then again, this is all preferences, but I have been able to find stability using this method (after a lot of trial and error).
Finally, I would recommend using manual voltages, rather than adaptive, when stress testing your OC. With manual voltages, you know at all times what your CPU is getting, and hence you can work out what the maximum voltage is to achieve stability. After you've worked this out, you can then go back to adaptive and set your maximum stable voltage as your upper limit.
I hope this helps. Good luck with your OC!
fkr
Splendid
- May 24, 2009
- 2,646
- 0
- 21,460
read through these as there is so much info that reiterating here and coming from multiple sources may get difficult to parse through
http://www.overclock.net/t/1411077/haswell-overclocking-thread-with-statistics
http://www.overclock.net/t/225371/intel-cpu-essentials-collected-guides
http://www.overclock.net/t/1411077/haswell-overclocking-thread-with-statistics
http://www.overclock.net/t/225371/intel-cpu-essentials-collected-guides
mundungus7
Honorable
- Nov 17, 2013
- 13
- 0
- 10,520
I never call an OC stable until Ive ran Prime for a couple hours and the temps stay below 75 or so. (celcius) Tools of the trade for me: CPUZ-to monitor clocks and voltages CORETEMP- for temperatures and PRIME95-to stress test
fkr
Splendid
- May 24, 2009
- 2,646
- 0
- 21,460
hwinfo 64 is my favorite monitoring tool as it seems to have the most categories to see what is going on.
this is my computer, this gives you low and highs as well as the average. if you click on a value it will bring up a bar graph
http://prntscr.com/24w8rj
this is my computer, this gives you low and highs as well as the average. if you click on a value it will bring up a bar graph
http://prntscr.com/24w8rj
fkr
Splendid
- May 24, 2009
- 2,646
- 0
- 21,460
If you read through those guides they have a breakdown that describes exactly what you are looking for. In the end what you are looking for is at what point do you have to increase you voltage by larger amounts to achieve a higher overclock.
for example on my i5 2500k:
at 4.3 I am still at pretty much stock voltages. stock being 1.2 volts
to get to 4.5-4.6 I need to make a pretty good jump in voltage 1.4 volts
to get to 5.09 I have to go above 1.5 volts and that generates a ton of heat. right at 100C in prime 95
So i run 4.4 at 1.3 volts
I am only using a hyper 212 so that is the limit of this cooler.
the above chart is more or less what you are trying to discover for your chip.
I would never run a chip at or above 80C for extended use. I do not think that there is any real threat of running it that high but it is a risk. I have still yet to hear about peoples chips really degrading over time and mine has not even after two years of being overclocked, and everything in my system is overclocked.ram GPU's and CPU.
for example on my i5 2500k:
at 4.3 I am still at pretty much stock voltages. stock being 1.2 volts
to get to 4.5-4.6 I need to make a pretty good jump in voltage 1.4 volts
to get to 5.09 I have to go above 1.5 volts and that generates a ton of heat. right at 100C in prime 95
So i run 4.4 at 1.3 volts
I am only using a hyper 212 so that is the limit of this cooler.
the above chart is more or less what you are trying to discover for your chip.
I would never run a chip at or above 80C for extended use. I do not think that there is any real threat of running it that high but it is a risk. I have still yet to hear about peoples chips really degrading over time and mine has not even after two years of being overclocked, and everything in my system is overclocked.ram GPU's and CPU.
TRENDING THREADS
-
-
News Firefox user loses 7,470 opened tabs saved over two years after they can’t restore browsing session- Started by Admin
- Replies: 58
-
Question Looking to upgrade my pc for more video and photo editing with light gaming.
- Started by fuzmaster
- Replies: 3
-
News Apple debuts M4 processor in new iPad Pros with 38 TOPS on neural engine
- Started by Admin
- Replies: 80
Facebook
Twitter
Instagram