[SOLVED] Freezes Troubleshooting PRO help needed (not a noob question)

[Goran Petric]

Ok', im in a bit of a pickle here. I build a pc for my friend from used components i bought, tested it and all works. He gets it, and it freezes - hard locks in games, screen goes gray or the image goes still and audio sound is like '''EERRRRRRRRRRRRRRR'', than you have to hard-restart it. I take it home, and i can barely make it crash, and it is tottaly random! I have been trying to reproduce it crashing so i can start troubleshooting but to no avail, in the 48h i have been tourchering it has crashed only 3 times. In bf4 twice, one time in singleplayer while just standing still and starring at screen, other time in multiplayer match in a helicopter, and once in GTA5 durring first tutoriall, i have been trying to make it do again but it just wont crash! I opened multiple games and benchmarks at the same time, on 2 monitors even, and nothing! I am burning it with prime95 and Fur mark to make it hot, and it is hot now, but nothing, it wont crash!
I was thinking maybe its PSU bit faulty, but since it holds up to burning stress testing.....
Specs:
Intel pentium g4600 (stock cooler)
Sapphire nitro RX 580 4G
Gigabyte ga-h110-d3a
Single stick 8gb ddr4 Kingston 2400 (in hw info its Clock is constantly fluctuating around 1066,1-1067,1, is that normal?)
Antec VP 500w
patriot burst 120g ssd+ wd 250gb hdd
4 cheap antec molex fans from antec case, i all put them on single molex power and stack it, was thinking that was doing the problems, but i can't reproduce the crash reliably to figure that out. It only randomly crashes sometimes in games, not in benchmarks stress testing (tried all unigine benchmarks, aida64 and furmark)

I allready did reinstall windows from scratch few months ago, test it it was fine, gave it to friend and than it crashes. Anyone have any idea what to do next except reinstall everything from scratch ? It looks like a software issue since hardware is holding up to stress testing. Temps are fine fine, gpu wont go above 62 and cpu above 60-70 durring haaard pushing with multiple games, only with prime95 and furmark i can get gpu to go up to 73 and cpu 95-100, with both prime and furmark running, cpu throttles to 3500/3600 from usuall 3700 when hitting 100.

Also the clocks of GPU seems a little low for 580, and it is awfully COOL (my personal 580 8gb is much hotter and louder than this one) and there is bios switch on gpu its on right position performance mode and getting only 1150 core and 2050 memory.

I checked the sticker it is real 580.

 

[cin19]

I think the problem may relate to cpu overheat when you play the games, do you try to use the aftermarket cpu cooler, because the temp ( 95-100 C ) is in the higher end. Try an aftermarket cpu cooler if you have one, make sure the cooler has the 120mm size fan.

Also when I look at the 2nd screen-shot picture, it shows you had the i3 6100. And that why the RAM runs at the 2133 mhz (1066x2), keep in mind only the 7th gen cpu, the RAM will run at 2400mhz, like the pentium g4600.

Also like you said, the PSU may have problem to power the rx580, because the Antec VP 500w is not the good PSU. You may try to lower the power limit to 80%-90% with software, like the MSI afterburner. If the PC can run well when you lower the power limit, and you know it is time to get other PSU.

 

[Goran Petric]

Yeah i forgot i put i3 6100 in it (same thing as g4560-4600 i had before). But its not the temps, you didn't read it carefully, it goes high temp only when stress testing it with prime95 and furrmark at the same time, in regular gaming it stays around 60.

I put another psu in (i have one more antec sitting around), rearanged molex fan supply, reinstalled all the drivers and trying to make it crash again. Im pretty sure it wont crash, ill give it to my friend and hell say its crashed -_-

Also, Antec VP is a good psu, i have 3 of them in systems right now, one is in my own personal ryzen 1800x & rx 580, never had any issues, antec vp is perfectly fine tier3 psu for that graphic card, there is no need to lower power of components, as i runned all the benchmarks i have at the same time and put the temps thro the roof, and it was fine, it wont crash like that on maximum power consumption.

 

[Goran Petric]

Ok, so i put another but same PSU, clean installed new drivers and all, testing it for 2 days and no crash so far, and also the gpu clocks are now different..... core 1340 / memory 1750, before it was 1150/2050. Could it be just bad amd drivers ? -_-

 

[cin19]

so i put another but same PSU

You said you had a few of them so that your mean it is same brand / model. Right?

Could it be just bad amd drivers ?

About you try to use the first PSU that caused the problem, and if you will not get the problem, then you know the problem is from the driver, otherwise the problem is from the first PSU.

For reference only: https://linustechtips.com/main/topic/1116640-psu-tier-list-40/