GPU can only get to 60°C before crashing (No BSOD)

[arilmikhail]

Hey guys newbie here,
I'm having a crashing issue with my computer when my Sapphire Nitro+ RX570 gets to 60°C. The screen goes black, and if I was watching a video or playing a game, the audio would glitchy-ly  distort before the computer tries to reboot after approximately 3 minutes. Of note, my fans and RGB lights throughout the entire system still remain on and functional throughout this. This crashing temperature sorta varies around 54-60°C but the max I've ever gotten it to is that 60°C. Ambient temps are around 30-35°C.

Also, the synthetic stress test I've ran (AMD Wattman built-in stress test, Aida64, Heaven Benchmark) don't trigger this issue and I can get up to 80°C before getting some slight artifacting which goes away after the stress test are completed and closed. This crashing issue only occurs randomly when I leave the GPU fans on default settings (with Zero fan mode on) or when I'm playing games (despite setting a very aggressive fan curve)

Specs;
CPU: AMD Ryzen 5 3600

GPU: Sapphire Nitro+ RX 570 4GB
Mobo: Gigabyte B450 Aorus Elite 
RAM: 16GB (8x2) 3200Mhz Corsair Vengeance LPX DDR4 C16
PSU: Cooler Master MWE Gold V2 750W 80Plus Gold
Boot: Adata SU650 120GB
CPU Cooler: Aigo Darkflash Gale Dual tower cooler

Case: Aigo Darkflash Leo Atx with 6 DR12 Chassis fans

Additonal misc hardware: Comfast AX200 PCI-E Wifi/Bluetooth Card
 

At first I thought it was a simple GPU temperature issue as my RX 570 4GB was second hand. So I tried cleaning out the heatsink and fans and applied new thermal paste and thermal pads. Still didn't fix the issue. So I figured maybe it's just a faulty GPU then. But then I tried plugging it in 2 other computers, I don't get this crashing issue whatsoever and can get it to run up to 85°C with no issues.

I borrowed a Sapphire RX 560, Asus GTX 1060, and Leadtek GTX 650 to see if it's just an odd compatibility of my card with the rest of my hardware and yet they all suffer the same crashing issue. 
Then I thought maybe it's a PSU issue as I was using an Aigo 80Plus Bronze 750W, so I replaced it with a Cooler Master 80Plus Gold 750W. The issue still persist. I even tried lowering and increasing the power limits of my GPU through both Wattman Radeon Software and MSI Afterburner, but saw no changes to the issue. 
The only "overclocking" I've done is enabling XMP for my RAM kit to run at the rated 3200Mhz. Even disabling this didn't fix the issue. 

As of right now the only way my computer functions without crashing is by setting an aggressive fan curve, opening the chassis up, and cooling down my room with an AC unit to prevent that 60°C cap from being reached but even with all that, I will still reach those temperatures (such as when playing Halo Infinite) and result in crashing. 
Any help would be much appreciated

 

[da na]

Welcome to the forums!

It definitely seems like a stability issue with the GPU hardware. Tried undervolting and underclocking? 

[arilmikhail]

3 minutes ago, Mel0nMan said:

Welcome to the forums!

It definitely seems like a stability issue with the GPU hardware. Tried undervolting and underclocking? 

Unfortunately I have... The only "stable" underclocking for my gpu was keeping it at 300Mhz. With that underclock, the temps obviously don't rise and I never get a crashing issue... But that doesn't really bode well for my gaming performance. 

I'm doubtful (and seriously hopeful) that it isn't my RX 570 itself as when I insert it stock (no overclocking, underclocking, or messing around with fan curves) it works flawlessly in 2 other systems. 

I'm worried it might be my motherboard as that's a relatively expensive part in my build and I'm unsure if RMA-ing it is still possible. 

Is there anything else that I could possibly try to rule out? 

[Farm-Boy_7130]

1 hour ago, arilmikhail said:

Unfortunately I have... The only "stable" underclocking for my gpu was keeping it at 300Mhz. With that underclock, the temps obviously don't rise and I never get a crashing issue... But that doesn't really bode well for my gaming performance. 

I'm doubtful (and seriously hopeful) that it isn't my RX 570 itself as when I insert it stock (no overclocking, underclocking, or messing around with fan curves) it works flawlessly in 2 other systems. 

I'm worried it might be my motherboard as that's a relatively expensive part in my build and I'm unsure if RMA-ing it is still possible. 

Is there anything else that I could possibly try to rule out? 

Try PSU if you have a spare lying around. If this isn't it, Try Memory. Linus, and I quote said: "If you have a problem that is so unexplainable and nothing else can explain it, It's almost always either Memory and or RAM."

 

Best of luck bro!

[arilmikhail]

1 hour ago, Farm-Boy_7130 said:

Try PSU if you have a spare lying around. If this isn't it, Try Memory. Linus, and I quote said: "If you have a problem that is so unexplainable and nothing else can explain it, It's almost always either Memory and or RAM."

 

Best of luck bro!

Thanks for the suggestions man... I tried out different PSU's already and there was still no difference. Gave RAM a shot, different sticks, different slots... and still no difference... 

I tried stripping down the computer to it's bare essentials: one stick of ram (trying out different sticks), only boot drive, gpu. It still had the same crashing behaviour. At this point I'm pretty sure it's either a motherboard problem or a cpu one... 

[Farm-Boy_7130]

14 minutes ago, arilmikhail said:

Thanks for the suggestions man... I tried out different PSU's already and there was still no difference. Gave RAM a shot, different sticks, different slots... and still no difference... 

I tried stripping down the computer to it's bare essentials: one stick of ram (trying out different sticks), only boot drive, gpu. It still had the same crashing behaviour. At this point I'm pretty sure it's either a motherboard problem or a cpu one... 

Unfortunately that looks to be the case. Try a different CPU (Contact the CPU Manufacturer and ask about RMA) then a different Motherboard. Best for luck one again!

[arilmikhail]

Update on my issue, 

I got another motherboard into my system. An A320m this time. That didn't solve my problem, the crashing issue still persisted. 
After alot of depressive rage, I finally started fiddling around again. Lo and behold... I found the issue. 

My DP cable was to blame. 

 

I tried out a HDMI cable and a different DP cable and BOOM the crashing was nowhere to be seen again. 

I feel so dumb for not considering it or even testing it as a variable. But that's just one of life's painful lessons I guess. Thanks for all the support. I'll be back to gaming now.