Trex miner worker cruising along then suddenly drops from 232 watts to 182 then suddenly reports 304 watts?

I have 6 cards in a mining rig, some are 3080’s, in this case its a 3080. I have dual 1200 watt psus, 3 cards per psu balanced (riser + card on each cable x 3 on each psu).

Hours go by (12) suddenly my one miner on the 3080 craps out (I have the power limit exit option at 270), it suddenly quits at 303 watts (and just moments before reported only 190 watts)

Any ideas what would cause this (my other 5 workers kept going just fine)?

This particular card i had a harder time reaching higher hash rates, less stable. But i have also had it make it 3+ days just fine (at around 900 on memory and -165 on core).

example command

t-rex.exe -d 2 -a ethash -o stratum+tcp://addresshere -p x -i 22 –kernel 0 –dag-build-mode 2 –gpu-report-interval 5 -w 3080D2 –pl 68 –fan 79 –cclock -165 –mclock 900 –exit-on-cuda-error –exit-on-high-power 270

**Also, since adding the 6th card, here i had to set pl to 68 to get the wattage up to 232 or so (for some odd reason). It hashes out at around 96.5 max or so.

Any thoughts?

What do you think?

10 Points
Upvote Downvote

Leave a Reply

Your email address will not be published. Required fields are marked *

GIPHY App Key not set. Please check settings

One Comment

  1. I would guess that the driver for that card is crashing causing the causing the lower power draw, followed by recovery and clocks being reset to default as indicated by the high power draw.

BeeSafer ๐Ÿ Newly launched project. 500$ market cap. We’re aiming to become the bees best friend! Come join us on TG ๐Ÿ If you have any questions then feel free to ask !

Missing all attestation assignments on prater