More
referral
Increase your income with Hive. Invite your friends and earn real cryptocurrency!

12x RTX 3070, issue with one of them

Hy dear community,
I had 2 workers with RTX 3070 for a total of 12 GPU. ALL gpu are working perfectly in this setup of 2 workers.

I decided to move all the gpu to only one worker.
Now:

  • 11/12 gpu are recognised. It is allways the same card with issue (again it was working previously).
  • I did some gpu swap testing: same issue, same card. Issue do not care about pci slot, allways the same card with issue.
  • I changed all parts of hardware (new riser kit changed 2 times for all swap testing, power cables changed 2 times. So no issue possible with hardware.
  • I used B250 mining expert. I have couple rigs with 12 amd gpu, same motherboards and psu(1200w for each section + 400w for atx) and everything working fine.
  • It is my first 12 nvidia gpu with Asus B250

If you guys have some suggestions, it will be very appreciate.
Best regards,

Have you tested it isolated on its own to rule out issues with the card itself? If it can hash and operate alone, then at you can start narrowing down the problem back to the rig. Yes, it was operating ok in the previous setup, but it could have been affected “somehow” during the transition to the new setup.

Yes, solved.
It was a driver corruption on one of the two new gpu installed in this rig.
Thank you,

How did you update your corrupted driver. I have 3070 not detected aswell. unable to enum cuda gpus.no cuda capable device is detected. opencl driver version 20.40-1147287