Rant:
Hey, after 5 days of debugging, I finally found out what was making my new system crash/BSOD. It was HwInfo64, latest version (got my new setup this monday, been testing from mon to friday non-stop). I fucking love this program but this time it almost killed my brain and made me return a brand new setup. I even bought a new motherboard as I thought the one I had was faulty.
How did I find out? At first I thought it was my PSU. So I started logging files and inspected them right after the crash to check for voltage spikes/drops. All voltages/currents/power/temperatures seemed stable. Rails stable, less than 1% diff. So I decided to increase the sample rate to 100ms instead of the default on hwinfo64 (1500ms or something) so I could get better graphs. What happened? Crashes after 10-20 seconds instead of random crashes after 1-2hours. Most of the crashes were actually JUST AFTER I closed a benchmark, which seemed odd as fuck. I thought psu couldn't handle voltage drops but it wasn't the case, it worked on my other computers.
I even made a post on reddit/tomshardware/etc and no one had a f* clue of what was going on. Here's the post at reddit:
At first I THOUGHT it was my motherboard or my PSU. After more than 40 hours of testing (other computers, benchmarks, etc) I found out they all my hardware parts all 100% fine!!!!
Context:
I built a 100% new system:
14700k + Maxsun z790 Icraft + 64gb DD5 Corsair 6000mhz + MSI MAG A850GL 850W + RX590 (old, need it to test games). Built it Monday night, installed latest hwinfo64 version.
I kept getting every random crash - idle, load, benchmarks, they seemed very random. At first I thought it was my old GPU, tried it on my old computer, then I tried my PSU on my old computer, then my old PSU on this new computer. I made every hardware combo possible (my old computer is 4770 + H81M+ 16GB ddr3 - 5000% stable with anything, including hwinfo64).
I did memtest, processor tests, disk texts, intel validation tests, EVERYTHING passed flawlessly.
Early conclusions: I isolated it the crashes 100% to the motherboard at first - I only had crashes with maxsun+14700k. No crashes with my PSU on ANY computer. No crashes with my 11y/o PSU on my 14700k computer. "Oh, im a genius, it's the motherboard's fault then, right?" WRONG. However, after literally buying a new mobo and literally going for the last test for the sake of being tilted, which was reinstalling my W11/W10 (tried on both) for the 5th time, I decided to not install anything and run benchmark for 6 fucking hours. 6 hours aida64 stress test (including disks) ---> STABLE, NO CRASHES.
I was like FUCKING WTF. "Ok, let's crash this, no way!!!!!!!!". Opened OCCT and AMD CPU stress test TOGETHER. Ran it for 5 hours. NO CRASHES. 3d mark 2 hours stress? No crashes. Then I decided to install HW64 to check the temps and BAM, crashes after 2 minutes. After this moment I knew it, when I was literally quitting and exhaust from testing non-stop for days: It was hwinfo for sure.
Then I googled "maxsun" + "hwinfo": V7.68 "Enhanced sensor monitoring on Maxsun mainboards." I had the latest version from monday on every fresh install I did (first thing I installed, then installed benchmarks) (7.72 or something)
Conclusion: MAXSUN Z790 ICRAFT + HWINFO64 -----> BSODS, CRASHES, INSTABILITY.
There's no way this is a coincidence - I've done days of every single test on this computer now - 10000% STABLE without HWINFO64 installed now after 5 days bashing my head against the wall. I lost most of my logs because I formatted this computer way to many times as I was tilted, tired and frustrated.
Question:
Why did this happen? Is it some specific thing that HWINFO64 pulls out of this motherboard or are the sensors unreliable and any other program reading the sensors would make it crash? I ask this because if any program reading this mobo sensors would make it crash, then it's actually better to get a new mobo.
100% fact: crashes seemed somewhat random, hard to isolate what exactly is making it crash. But making HWINFO64 get samples faster makes the system crash almost instantly (1-5 minutes max) without a BSOD - system freezes and I can't move my mouse (on w11) and on W10 it instantly reboots.
Sorry if this sounds a bit rude but I would NEVER think that something I've relied on for some many years (using it on every single computer since beta) would give me so much pain, including monetary loss (couldn't work for 4 days). HWINFO64 always felt like the FIRST program I install on any computer as it was literally my number #1 testing buddy.
I can try to get new logs if it helps the program in some sort of way but after literally 50+ crashes in a week I'm legit scared to open it and traumatized to go through crashes again.
Thanks, sorry for the ultra long text but I'm exhausted, I got a baby to feed and couldn't work for almost a week because of system instability. Looking for any sort of answer/explanation of what might have caused this(I know I should have posted some logs, I'm way too tired to find out how to do it and to reproduce more crashes). Should I still return my mobo? I think it's not a good indicator if reading its sensors makes it crash.
Hey, after 5 days of debugging, I finally found out what was making my new system crash/BSOD. It was HwInfo64, latest version (got my new setup this monday, been testing from mon to friday non-stop). I fucking love this program but this time it almost killed my brain and made me return a brand new setup. I even bought a new motherboard as I thought the one I had was faulty.
How did I find out? At first I thought it was my PSU. So I started logging files and inspected them right after the crash to check for voltage spikes/drops. All voltages/currents/power/temperatures seemed stable. Rails stable, less than 1% diff. So I decided to increase the sample rate to 100ms instead of the default on hwinfo64 (1500ms or something) so I could get better graphs. What happened? Crashes after 10-20 seconds instead of random crashes after 1-2hours. Most of the crashes were actually JUST AFTER I closed a benchmark, which seemed odd as fuck. I thought psu couldn't handle voltage drops but it wasn't the case, it worked on my other computers.
I even made a post on reddit/tomshardware/etc and no one had a f* clue of what was going on. Here's the post at reddit:
At first I THOUGHT it was my motherboard or my PSU. After more than 40 hours of testing (other computers, benchmarks, etc) I found out they all my hardware parts all 100% fine!!!!
Context:
I built a 100% new system:
14700k + Maxsun z790 Icraft + 64gb DD5 Corsair 6000mhz + MSI MAG A850GL 850W + RX590 (old, need it to test games). Built it Monday night, installed latest hwinfo64 version.
I kept getting every random crash - idle, load, benchmarks, they seemed very random. At first I thought it was my old GPU, tried it on my old computer, then I tried my PSU on my old computer, then my old PSU on this new computer. I made every hardware combo possible (my old computer is 4770 + H81M+ 16GB ddr3 - 5000% stable with anything, including hwinfo64).
I did memtest, processor tests, disk texts, intel validation tests, EVERYTHING passed flawlessly.
Early conclusions: I isolated it the crashes 100% to the motherboard at first - I only had crashes with maxsun+14700k. No crashes with my PSU on ANY computer. No crashes with my 11y/o PSU on my 14700k computer. "Oh, im a genius, it's the motherboard's fault then, right?" WRONG. However, after literally buying a new mobo and literally going for the last test for the sake of being tilted, which was reinstalling my W11/W10 (tried on both) for the 5th time, I decided to not install anything and run benchmark for 6 fucking hours. 6 hours aida64 stress test (including disks) ---> STABLE, NO CRASHES.
I was like FUCKING WTF. "Ok, let's crash this, no way!!!!!!!!". Opened OCCT and AMD CPU stress test TOGETHER. Ran it for 5 hours. NO CRASHES. 3d mark 2 hours stress? No crashes. Then I decided to install HW64 to check the temps and BAM, crashes after 2 minutes. After this moment I knew it, when I was literally quitting and exhaust from testing non-stop for days: It was hwinfo for sure.
Then I googled "maxsun" + "hwinfo": V7.68 "Enhanced sensor monitoring on Maxsun mainboards." I had the latest version from monday on every fresh install I did (first thing I installed, then installed benchmarks) (7.72 or something)
Conclusion: MAXSUN Z790 ICRAFT + HWINFO64 -----> BSODS, CRASHES, INSTABILITY.
There's no way this is a coincidence - I've done days of every single test on this computer now - 10000% STABLE without HWINFO64 installed now after 5 days bashing my head against the wall. I lost most of my logs because I formatted this computer way to many times as I was tilted, tired and frustrated.
Question:
Why did this happen? Is it some specific thing that HWINFO64 pulls out of this motherboard or are the sensors unreliable and any other program reading the sensors would make it crash? I ask this because if any program reading this mobo sensors would make it crash, then it's actually better to get a new mobo.
100% fact: crashes seemed somewhat random, hard to isolate what exactly is making it crash. But making HWINFO64 get samples faster makes the system crash almost instantly (1-5 minutes max) without a BSOD - system freezes and I can't move my mouse (on w11) and on W10 it instantly reboots.
Sorry if this sounds a bit rude but I would NEVER think that something I've relied on for some many years (using it on every single computer since beta) would give me so much pain, including monetary loss (couldn't work for 4 days). HWINFO64 always felt like the FIRST program I install on any computer as it was literally my number #1 testing buddy.
I can try to get new logs if it helps the program in some sort of way but after literally 50+ crashes in a week I'm legit scared to open it and traumatized to go through crashes again.
Thanks, sorry for the ultra long text but I'm exhausted, I got a baby to feed and couldn't work for almost a week because of system instability. Looking for any sort of answer/explanation of what might have caused this(I know I should have posted some logs, I'm way too tired to find out how to do it and to reproduce more crashes). Should I still return my mobo? I think it's not a good indicator if reading its sensors makes it crash.
Last edited:
