My Gigabyte AORUS GeForce RTX 2080 Xtreme WATERFORCE 8G Graphics Card started overheating a bunch a month ago. I noticed that the radiator was cold and the coolant tubes were not vibrating at all. I’m pretty sure the pump is broken or for whatever reason it is not functioning. I took a giant air purifier fan I had in my house, removed the front of my computer case to blow air into it and I haven’t had it overheat in a month. I talked to live support and they said I have to post this in this forum to be eligible for warranty replacement. Could I get some assistance in verifying that this is a hardware issue to replace this card?
thanks
It’s quite odd gigabyte support sent you here since nvidia doesn’t build water-pumps and those also don’t need drivers. Also, the gpu seems to work, just the cooling is broken. I guess the support person just heard “linux” and instantly directed you here without even noticing “broken water-pump”. You can of course post the output of nvidia-smi without the make-shift cooler so I can tell you that the card is overheating (which you already know).
1 Like
Yeah, I mean its pretty obvious what’s happening, the customer support person said I just need someone from this forum to verify it. Is there some mod that needs to do this and give me a RMA or something?
Heres the smi
Sun Jan 2 10:18:26 2022
±----------------------------------------------------------------------------+
| NVIDIA-SMI 460.91.03 Driver Version: 460.91.03 CUDA Version: 11.2 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 GeForce RTX 2080 Off | 00000000:01:00.0 On | N/A |
| 37% 34C P0 51W / 275W | 2163MiB / 7979MiB | 13% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+
±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 1365 G /usr/lib/xorg/Xorg 1717MiB |
| 0 N/A N/A 2744 G /usr/lib/rstudio/bin/rstudio 25MiB |
| 0 N/A N/A 4096 G cinnamon 189MiB |
| 0 N/A N/A 21603 G …e/user/Telegram/Telegram 2MiB |
| 0 N/A N/A 22171 G …token=44147324119251083 34MiB |
| 0 N/A N/A 22453 G …oken=144848539331503283 11MiB |
| 0 N/A N/A 25801 G …oken=139353393324692967 86MiB |
| 0 N/A N/A 30573 G …oken=145377431817202039 33MiB |
| 0 N/A N/A 32248 G …n-US/Browser/firefox.real 12MiB |
±----------------------------------------------------------------------------+
Just file an RMA and DO NOT mention linux ;-)
No nvidia official here will ever give you a verification/RMA neither for their own cards and most definitely not for some other vendor, due to liability reasons.
Like Mart said, “do not mention the war, err, linux”. Otherwise you’ll get something back like “We do not support linux” or “You voided your warranty” (which of course is BS).
A bit more on topic: the nvidia-smi output doesn’t really show overheating, 34°C while rinning in P0 at 13% load isn’t unusual. Likely you need some time to heat it up? e.g. to 60°C.