PC restarted during execution of prime95 (Z390 Aorus Master/9900k)
PC restarted during execution of prime95 (Z390 Aorus Master/9900k)
Whenever the system starts shutting down and temperatures reach 90-100°C or above, it’s very likely the issue is temperature-related. Even if the core or package temperatures aren’t the cause, other components like VRMs could be affected. Throttle protections offer limited help and often fail to stop damage. Following best practices isn’t always the right approach for someone who isn’t a competitive overclocker or an expert enthusiast. Pay attention to what Computronix advises—he conducts extensive testing with Intel processors, far more than most hobbyists do. He is the go-to expert for Intel CPUs and testing methods.
Thank you very much for the info. But just tell me one thing..what's the point of testing the CPU with non-avx workloads? even some games today and especially chess softrware (which I often use) incorporate avx instructions, and I optimally want to know that my overclock can handle those instructions.
I used the latest prime95 version 29.8. And why Gigabyte recommend version 27.9 in their overclocking guide if you say to use only version 26.6? Actually I care more about stability and error detection rather than just thermal performance. Like I said, in Aida64 (all tests combined), OCCT (large data set) I was able to pass 1 hour+ without any issues with temps well below 85. But prime95 (29.8) small ffts just overkill. I actually think that my overclock was stable under that load but because of the high temperature it is rebooting and preventing me from complete at least 30 minutes.
I guess I will try version 26.6 just to say I passed prime95 😁 and If I pass the rest of the tests (occt large, AIDA64, Realbench..) than I will just keep that overclock at 1.3v-1.31v.
Thank you all very much for all the help.
can you share the hwinfo64 sensor tab screenshot (expand to view all) for additional details? (before and during the stress test prior to crashing)
Truegenius, I don't plan to repeat the prime95 blend test again... I'm already concerned I may have harmed the CPU. I chose to keep Prime95 separate and focus on other stress testing tools. Computronix, I just came across your old post: Is version 26.6 of Prime95 more accurate? Hi everyone, I've been heavily overclocking my i5 6600k recently, and my system was built last week. I've started using the latest Prime95 version to check CPU stability, and once the temperature hit 82°C! People have mentioned that Prime95 version 26.6 is more realistic, so I'm proceeding... forums. I just completed an hour of Realbench stress test (it's putting stress on the whole system, including the GPU at full capacity!) and passed with temperatures around 85°C and below (89 was the maximum, but note... it's running at 9900k in air 😁). Including the Benchmark, I ran six tests. As I mentioned earlier, I've already cleared OCCT large set and Aida64. Prime95 turned out to be a mistake. I'm sorry I went through this—I damaged my CPU. 🙁 At least I managed something. EDIT: I tried again with prime95 (version 29.8), but this time I completely disabled AVX, so far it passed 40 minutes of the blend test. The temperature reached 90°C max, which is acceptable for stress testing, similar to other 9900k results (including Derbauer with NH-D15). Is this the correct approach when using Prime95? Would disabling AVX help mimic version 26.6?
Extreme overclockers often push systems beyond safe limits without concern for longevity. These individuals prioritize immediate performance gains over long-term reliability. For those planning daily use and expecting years of operation, such approaches should be avoided. The focus should remain on maintaining stable temperatures, especially when using high-performance applications that demand more heat. Disregarding thermal safety simply because it’s acceptable is not advisable. Adjusting settings to match recommended limits—like keeping temperatures below 80°C—is essential for sustained performance.
Thank you for your feedback. However, as mentioned earlier, I believe the temperatures are acceptable for the applications I use every day. Still, I think a maximum temperature of 80°C under load might be too low, especially for 9900k, and 85°C seems more appropriate. Once the CPU reaches above 85°C, I will definitely consider reverting to stock clocks. So far in gaming—where I mostly play—it stays below 80°C.
If you aim to discover your accurate baseline core temperatures, you must test your processor in the same way Intel does. This involves adhering to a process that matches Intel's datasheets. The best method is to closely imitate Intel's testing conditions and procedures. This includes normalizing, reducing, or removing the three main factors: environment, hardware, and software. In post #8 I mentioned: Please refer to Sections 11 & 12: Intel Temperature Guide - https://forums. Sincerely, your questions indicate you haven’t done this before. I hoped you had reviewed it by now, since the Guide details proper thermal testing methods. Keep in mind we’re here to assist. Darkbreeze and SgtScream have offered excellent guidance. That’s why we provide links for our members to explore further. PLEASE read it... carefully... after which I’ll be pleased to address your queries.
The CPU wasn't harmed; it has built-in safeguards that will automatically reduce clocks and voltages when temperatures rise above 80°C. If this doesn't work, the CPU will shut down immediately if it thinks it's entering a burn-out state. There won't be any blue screen or warnings—just a simple shutdown. This could occur once the temperature exceeds 90°C, but you should be safe.
Variety of AVX tasks is challenging performance, yet specific offsets can be utilized. (Intel's XTU supports AVX offset adjustments along with peak multipliers, and allows easy control of core voltage; in case of a crash while XTU is active, the system will revert to default settings.)
As previously mentioned, version 26.6 is the standard, while others resemble 110-120% overloads...(my tests show that newer P95 versions after v26.6 actually raise temperatures by 4-7°C compared to v26.6)
Most motherboards probably have TDP or power limits between 95-110 watts in their BIOS, which likely triggered a hard reset; typically the system would simply reduce performance near maximum temperatures. Running at 5 GHz usually means peak TDP exceeds 160 watts at 1.37V...
I’m sorry to hear the NH-D15 couldn’t cope... you can always opt for sub-5 GHz multipliers, since most users are GPU-bound already. I doubt your applications or games will be affected by only 4.8 GHz, still a 100 MHz boost over the typical core turbo settings, etc... That final 100-200 MHz segment is what demands higher voltage and probably increases temperatures by 5-8°C...
Most gaming CPU workloads are about 5°C lower than what’s observed even in Prime 95 blended mode, and 8-10°C less than the temperatures recorded with P95 or small FFTs.