XFX Issues

[Malique2]

I Have a XFX 9070 Swift. I don't OC. Tried latest drivers, rolled back to prior drivers and same exact issue. I'm stumped. My other rig with 6 GPUs 5000/6000 series has zero problems.

I have been haven't been able to run FOH it yet. It downloads and starts folding, I watch the PPDs start counting down, that goes on for a few minutes then it fails and restarts over, and the process goes on and on and on.I did capture this in the log, which shows the failure.

23:58:25:I1:WU73:********************************************************************************
23:58:25:I1:WU73:Project: 18260 (Run 21, Clone 2, Gen 261)
23:58:25:I1:WU73:Digital signatures verified
23:58:25:I1:WU73:Folding@home GPU Core27 Folding@home Core
23:58:25:I1:WU73:Version 8.2.1
23:58:25:I1:WU73: GPU info: Platform: OpenCL: AMD Accelerated Parallel Processing
23:58:25:I1:WU73: GPU info: PlatformIndex: 0
23:58:25:I1:WU73: GPU info: Device: gfx1201
23:58:25:I1:WU73: GPU info: DeviceIndex: 0
23:58:25:I1:WU73: GPU info: Vendor: 0x1002
23:58:25:I1:WU73: GPU info: PCI: 10:00:00
23:58:25:I1:WU73: GPU info: Compute: 2.0
23:58:25:I1:WU73: GPU info: Driver: 3661.0
23:58:25:I1:WU73: GPU info: GPU: true
23:58:25:I1:WU73: Checkpoint write interval: 12500 steps (5%) [20 total]
23:58:25:I1:WU73: JSON viewer frame write interval: 2500 steps (1%) [100 total]
23:58:25:I1:WU73: XTC frame write interval: 5000 steps (2%) [50 total]
23:58:25:I1:WU73: TRR frame write interval: disabled
23:58:25:I1:WU73: Global context and integrator variables write interval: disabled
23:58:25:I1:WU73:There are 3 platforms available.
23:58:25:I1:WU73:Platform 0: Reference
23:58:25:I1:WU73:Platform 1: CPU
23:58:25:I1:WU73:Platform 2: OpenCL
23:58:25:I1:WU73: opencl-device 0 specified
23:59:54:I1:WU73:Attempting to create OpenCL context:
23:59:54:I1:WU73: Configuring platform OpenCL
00:00:12:I1:WU73:ERROR:exception: Error invoking kernel reduceEnergy: clEnqueueNDRangeKernel (-54)
00:00:12:I1:WU73:Saving result file ..\logfile_01.txt
00:00:12:I1:WU73:Saving result file science.log
00:00:12:I1:WU73:Folding@home Core Shutdown: BAD_WORK_UNIT
00:00:14:E :WU73:Core returned BAD_WORK_UNIT (114)
00:00:14:I1:Default:Added new work unit: cpus:0 gpus:gpu:10:00:00
00:00:14:I1:WU73:Uploading WU results
00:00:14:I1:WU74:Requesting WU assignment for user Maliki2 team 64
00:00:14:I2:OUT21:> POST /api/results HTTP/1.1

[Joe_H]

Some things that might help are which OS you are using, which version of the F@h client, and how much RAM you have installed. I do know Project 18260 is particularly large and may use a larger amount of RAM during initialization.

Did you see WUs from other projects at all?

[Malique2]

Some things that might help are which OS you are using, which version of the F@h client, and how much RAM you have installed. I do know Project 18260 is particularly large and may use a larger amount of RAM during initialization.

Did you see WUs from other projects at all?

OS is Windows 11. Version is 8.4.9 but did it in previous versions. 32 GB ram. I'm only seeing WUs from my CPU. I've let the GPU run a bunch of times with same results of BAD WORK UNIT. I'm just stumped. No overclock. Brand new card. 6 GPU rig next to me with 5000/6000 series cards, folding away, but this turd just doesn't want to. The code below is what I just got trying to run it again.

Code: Select all

03:03:24:I1:WU92:Version 8.2.1
03:03:24:I1:WU92: GPU info: Platform: OpenCL: AMD Accelerated Parallel Processing
03:03:24:I1:WU92: GPU info: PlatformIndex: 0
03:03:24:I1:WU92: GPU info: Device: gfx1201
03:03:24:I1:WU92: GPU info: DeviceIndex: 0
03:03:24:I1:WU92: GPU info: Vendor: 0x1002
03:03:24:I1:WU92: GPU info: PCI: 10:00:00
03:03:24:I1:WU92: GPU info: Compute: 2.0
03:03:24:I1:WU92: GPU info: Driver: 3661.0
03:03:24:I1:WU92: GPU info: GPU: true
03:03:24:I1:WU92: Checkpoint write interval: 100000 steps (5%) [20 total]
03:03:24:I1:WU92: JSON viewer frame write interval: 20000 steps (1%) [100 total]
03:03:24:I1:WU92: XTC frame write interval: 25000 steps (1.2%) [80 total]
03:03:24:I1:WU92: TRR frame write interval: disabled
03:03:24:I1:WU92: Global context and integrator variables write interval: disabled
03:03:24:I1:WU92:There are 3 platforms available.
03:03:24:I1:WU92:Platform 0: Reference
03:03:24:I1:WU92:Platform 1: CPU
03:03:24:I1:WU92:Platform 2: OpenCL
03:03:24:I1:WU92: opencl-device 0 specified
03:03:44:I1:WU92:Attempting to create OpenCL context:
03:03:44:I1:WU92: Configuring platform OpenCL
03:04:02:I1:WU92:ERROR:exception: Error invoking kernel reduceEnergy: clEnqueueNDRangeKernel (-54)
03:04:02:I1:WU92:Saving result file ..\logfile_01.txt
03:04:02:I1:WU92:Saving result file science.log
03:04:02:I1:WU92:Folding@home Core Shutdown: BAD_WORK_UNIT
03:04:02:E :WU92:Core returned BAD_WORK_UNIT (114)
03:04:02:I1:WU92:Uploading WU results
03:04:04:I1:WU92:Credited

[muziqaz]

Windows drivers hijacked your AMD drivers. Please reinstall AMD adrenalin software from AMD website

[Malique2]

Windows drivers hijacked your AMD drivers. Please reinstall AMD adrenalin software from AMD website

Reinstalled. Folding started with no percentage update. It was folding for awhile too! Rebooted and back to failing again.

[muziqaz]

Windows drivers hijacked your AMD drivers. Please reinstall AMD adrenalin software from AMD website

Reinstalled. Folding started with no percentage update. It was folding for awhile too! Rebooted and back to failing again.

Post the logs when there is no percentage update. I think you are also posting on Discord? No?
I'm out of ideas except blaming your system stability (not GPU). RAM/CPU. Also DDU your drivers, and install fresh Adrenalin. I see other 9070 users folding ok. Mine is on Linux, so cannot test

[DarkFoss]

Try lowering the clock of your gpu through the control panel I don't know the default for a 9070 but with my 7900xtx I have to lower it from 3020 Ghz to 2620. You could also try lowering the number of cores used by the cpu by 1 from what ever the default setting currently is in the fahclient.

[Malique2]

After using my head a little I just pulled it and slapped it into my multi mining/folding rig. And away it went. Now I have to figure out why this machine won't. The other rig has 1 5500xt, and 2 6600xt and 2 6600 and a 6700xt.

[Malique2]

So just had a complete failure N%N. So not sure what that means. I'm about to toss it in the third computer. Obvisouly it is running, I just can't figure out why this computer is having such a hard time folding. Ugh.

[muziqaz]

So just had a complete failure N%N. So not sure what that means. I'm about to toss it in the third computer. Obvisouly it is running, I just can't figure out why this computer is having such a hard time folding. Ugh.

NaN error is not your computer, it is the project

[Malique2]

Tried it in my third computer and it starting mining away. Now I need to figure out why it won't in my newest.

[toTOW]

Did you check the PSU of this machine ?

[Malique2]

Did you check the PSU of this machine ?

I work tomorrow but have Tuesday and Wednesday off. Im going to throw one of my 6600xts in and see if it folds. If it doesnt I'll sap the psu. This is a fairly new psu with not much use. I have a few extras I can try. I'll let you know.