Quote from: JoepMeloen on February 26, 2018, 04:03:19 PM

Quote from: MarioMBM on February 26, 2018, 11:11:28 AM

Quote from: gsrcrxsi314 on February 25, 2018, 06:52:24 AM

Quote from: flatounet on February 25, 2018, 06:37:56 AM

Quote from: gsrcrxsi314 on February 25, 2018, 12:34:52 AM

Phoenix/OP:

i have 2 systems running just fine on Phoenix2.6

but now on a 3rd system with a brand new fresh install of win10 x64. and a single MSI GTX 1080ti im hitting the following error:

Code:

17586:19:01:11.605: GPU1 GPU1: DAG generated in 5.9 s (403.2 MB/s)
17586:19:01:13.312: main Eth speed: 0.000 MH/s, shares: 0/0/0, time: 0:00
17586:19:01:14.641: GPU1 CUDART error in CudaProgram.cu:422 : unspecified launch failure (4)
17586:19:01:14.641: GPU1 GPU1 search error: unspecified launch failure
17586:19:01:15.117: wdog Thread(s) not responding. Restarting.
17586:19:01:17.460: GPU1 GPU1: Starting up...
17586:19:01:17.497: GPU1 CUDART error in CudaProgram.cu:230 : all CUDA-capable devices are busy or unavailable (46)
17586:19:01:17.497: GPU1 GPU1 initMiner error: all CUDA-capable devices are busy or unavailable

I have DDU'd my driver, and reinstalled to no avail. nvidia driver 390.77, same driver as my other nvidia based rig thats working just fine

i got this error one time , it's about OC for me ....
reset OC ,and re-test or check/change riser after .....

There was no overclock applied. It was completely stock.

But I dont know what it was but after updating to the most recent version of Windows 10, the problem seems to have gone away.

Old version: 1511
New version: 1709

Works fine now.

Same as you i have 2 righs completly the same 1070 Dual OCs , they are basicaly mirrors.
1 works fine the other has the error you had, i have the latest windows 10 1709 and i am still getting that error after 10/12 hours of running.
Drives me crazy.

turn off all the nvidia audiodevices in devicemanager. for me it helped to solve some weird (driver i think) related stuff.

Thanks , i`ve just disabled all the audio related devices. and under clocked 1 gpu to the lowest possible and seems stable for 2/3 hours now.
I think i might have a bad riser that stops relaying info to the miner or something.