Quote from: dedizonesv2 on Today at 05:06:29 PM

Quote from: DrunkenBrahman on Today at 10:58:20 AM

Quote from: trexminer on Today at 07:48:06 AM

Quote from: FomoMiner on April 16, 2022, 01:12:06 AM

great results BUT it will lock and lock again & again non stop
you will have to lower your oc until it becomes the same hash as previous versions of trex.
nice try but failed (unfortunatly)

Please upload the full log file, I'll have a look. Thanks.

I set the settings of the cards on which the error appeared in the stock 0-0-0-0, but the error still occurs at the moment between "DAG generated" and "using kernel #4"
V 20220417 13:52:41 GPU #4: DAG generated [crc: 50f76ac8, time: 17775 ms], memory left: 2.83 GB
V 20220417 13:55:54 TREX: Can't find nonce with device [ID=3, GPU #3], cuda exception: CUDA_ERROR_LAUNCH_FAILED, try to reduce overclock to stabilize GPU state
X 20220417 13:52:41 GPU #4: using kernel #4

This is first variant, and the second is wnen card just skip "using kernel" step and show no hashrate, 0.00

You OC ? and Driver Nvidia vers ?

HiveOS 0.6-215@220415 N470.103.01

I fixed problem by adding
"dag-build-mode": "2,2,2,2,2,2"
"kernel": "4,4,4,2,4,4"