search

trexminer / T-Rex

T-Rex NVIDIA GPU miner with web control monitoring page
2.64k stars 439 forks source link

Error: Can't stop device (ID=1, GPU #1). cuda exception: CUDA_ERROR_UNKNOWN #1329

Open snxs48 opened 2 years ago

snxs48 commented 2 years ago

i get this error after a minute. Even without any OC. my rig: 3x 3060 v1, x79 motherboard. Already checked risers and cabels.

andriy2011 commented 2 years ago

I'm getting the same error on 3060ti

trexminer commented 2 years ago

@snxs48 what's your driver version? Make sure you're using 510.xx drivers if running Linux or 512.xx in case of Windows.

grenelt commented 2 years ago

Get the same error frequently on the last card after 5 minutes - and another failure preceeding this on changing cards.

Very strange and and i have to downclock them to ridiculous low values (3060TI LHR > 54 MH/s, 2070 > 39 MH/s, 3070 > 59 MH/s) to get some stability. Windows 10, 512.77

I'm very disappointed since the full unlock is out... It's very hard to find the right combinations of gpus and settings.

sign1818 commented 2 years ago

Since the full unlock I had to add more power to the LHR gpu's (3060 V2), Now all are fine.

grenelt commented 2 years ago

@sign1818 How do you add "more power"? Higher Power Limit? More power supplies? Higher lock-cv values? Higher PCH voltage? Or did you just yell "More Power" - like this guy: https://www.youtube.com/watch?v=YQwYNca4iog

sign1818 commented 2 years ago

Like in the video 🤣🤣🤣 I use a higher Power Limit per gpu, +/- 10Watt more then I used for 0.25.15 and no lock-cclock anymore.

cwkoomy commented 2 years ago

1of my 3060ti always crash, now every miner test also become dag has been damaged, check overclocking settings. lower mem clock from 2550mem to 0mem to -500 to -1000, still crash. is my card vram faulty already ? before unlock 75% never face this problem. lolmiber DAG repair also fail... and reboot again & again

alemmer commented 2 years ago

1of my 3060ti always crash, now every miner test also become dag has been damaged, check overclocking settings. lower mem clock from 2550mem to 0mem to -500 to -1000, still crash. is my card vram faulty already ? before unlock 75% never face this problem. lolmiber DAG repair also fail... and reboot again & again

I’m having the exact same issue with my 3060ti

federalarty commented 2 years ago

I have the same problem. Could it not be Windows? Logical memory ?

cwkoomy commented 2 years ago

I have the same problem. Could it not be Windows? Logical memory ?

i using hiveos

alemmer commented 2 years ago

Just sent my card back to Gigabyte for RMA, I’m assuming my cards vram went bad

cwkoomy commented 2 years ago

Just sent my card back to Gigabyte for RMA, I’m assuming my cards vram went bad

before 100% unlock, nothing problem, after this new 100% unlock, 1 of card dead.. so bad. other 4samsung1hynix still runing good. all just 4months card. temp around 45-55 everyday.

cwkoomy commented 2 years ago

Just sent my card back to Gigabyte for RMA, I’m assuming my cards vram went bad

before 100% unlock, nothing problem, after this new 100% unlock, 1 of card dead.. so bad. other 4samsung1hynix still runing good. all just 4months card. temp around 45-55 everyday.

lucky can claim warranty. after check is vram faulty. now my card run good after warranty.