Question Power supply slowly killing components?

nishaburiy

Reputable
Mar 24, 2018
22
3
4,525
After 2 years of serving me well, right after the warranty period has ended, my RTX 2080 has bit the dust. It now has artifacts even with very low load, and Furmark crashes the system within a minute.
Knowing that two years is a very short lifespan for a card, I'm worried that my power supply was at fault here, and the next GPU I buy and all my other components will all die eventually as well due to the bad PSU. The 2080 was cleaned regularly and the temperature never exceeded 75c. I currently have a GTX 770 running in the same PC - hell of a downgrade, I know! - and everything seems to be working fine even when I put the system under as much load as possible (Cinebench and Furmark running at the same time).

Here's the relevant PC specs incase they're useful:
Ryzen 5 3600
Rtx 2080 Palit dual (RIP)
CoolerMaster V850

How can I verify that my power supply was not at fault here and the death of my 2080 was just an unfortunate occurrence.
 

nishaburiy

Reputable
Mar 24, 2018
22
3
4,525
Have you tried, changing or reflashing the bios on the rtx 2080. ?You have nothing to lose and it may bring it back from the dead. VGA Bios Collection: Palit RTX 2080 8 GB | TechPowerUp
I was considering re-flashing but I was worried that the GPU might crash halfway through and brick the card. I guess It really wouldn't make a difference at this point. I'll give it a try and will update on how it goes. Thanks for the idea.
 

nishaburiy

Reputable
Mar 24, 2018
22
3
4,525
Sorry to hear that:( When you test it with furmark, do the temps rise on gpu and memory, or does it just start artificating?<---not sure if that is a word.
A quick update on my situation would be that I RMAd my PSU and they actually confirmed that the unit was actually faulty and gave me a replacement unit. After changing it I'm still getting crashes on Furmark after the GPU has enough time to heat up, although the max temp the die reaches is 70. The card sadly doesn't have any VRM/VRAM temperature sensors so I have no idea if that's what is causing the problem. I'll send an email to Palit as a last resort to see if they agree to replace the card. If not I guess I'll just crack it open and reapply paste and clean the card to see if that fixes anything.
 

KyaraM

Admirable
You can use an app to monitor voltage, temps etc for the gpu. Install the HW Monitor Classic, HWMONITOR | Softwares | CPUID it is a very small app, and watch the values as you run the gpu on furmark. Maybe you can find an anomaly with memory heat or voltage.
Not every card has all sensors available to programs like that. Sounds like TCs card simply doesn't have the sensor available, just like with my old 1070.
 

KyaraM

Admirable
I never had a Rtx2080, but I do have a Rtx 3080, it shows gpu, voltage, power draw and hot spot and hopefully memory temps on this app. Can't hurt to try.
Amd my 3070Ti shows them, too. That doesn't say anything. You are making assumptions here. TC explicitly mentioned the card not having that sensor, and I remember from others that their RTX 2000 cards didn't, either. You really can't compare old cards with your new one. And just a quick google search tells me that only EVGA RTX 2000 cards had the feature:
View: https://www.reddit.com/r/nvidia/comments/bguqvz/which_rtx_2080_cards_have_temperature_sensors_for/
 
Amd my 3070Ti shows them, too. That doesn't say anything. You are making assumptions here. TC explicitly mentioned the card not having that sensor, and I remember from others that their RTX 2000 cards didn't, either. You really can't compare old cards with your new one. And just a quick google search tells me that only EVGA RTX 2000 cards had the feature:
View: https://www.reddit.com/r/nvidia/comments/bguqvz/which_rtx_2080_cards_have_temperature_sensors_for/
You are preaching to the choir, my main point of using the app was to glean any OTHER information. Let go of the vram lesson.