Nvidia RTX 2080 TI underperforming

[FUFU]

Hi guys,

I'm using an Nvidia RTX 2080 TI since today and I'm getting around ~ 1,3 million ppd. I read that more than 2 million ppd. would be normal for this graphics card.
Yes, I'm using my passkey and I'm running windows 10 64 bit on an AMD Threadripper 1950X.

What am I doing wrong?

Thanks in advance!
FUFU

Code: Select all

09:26:54:Saving configuration to config.xml
09:26:54:<config>
09:26:54:  <!-- Network -->
09:26:54:  <proxy v=':8080'/>
09:26:54:
09:26:54:  <!-- User Information -->
09:26:54:  <passkey v='********************************'/>
09:26:54:  <team v='43829'/>
09:26:54:  <user v='FUFU'/>
09:26:54:
09:26:54:  <!-- Folding Slots -->
09:26:54:  <slot id='1' type='GPU'>
09:26:54:    <paused v='true'/>
09:26:54:  </slot>
09:26:54:</config>
09:26:56:Saving configuration to config.xml
09:26:56:<config>
09:26:56:  <!-- Network -->
09:26:56:  <proxy v=':8080'/>
09:26:56:
09:26:56:  <!-- User Information -->
09:26:56:  <passkey v='********************************'/>
09:26:56:  <team v='43829'/>
09:26:56:  <user v='FUFU'/>
09:26:56:
09:26:56:  <!-- Folding Slots -->
09:26:56:  <slot id='1' type='GPU'>
09:26:56:    <paused v='true'/>
09:26:56:  </slot>
09:26:56:</config>
09:27:00:FS01:Unpaused
09:27:01:WU00:FS01:Connecting to 65.254.110.245:8080
09:27:01:WU00:FS01:Connecting to 65.254.110.245:8080
09:27:02:WU00:FS01:Assigned to work server 155.247.166.220
09:27:02:WU00:FS01:Requesting new work unit for slot 01: READY gpu:0:TU102 [GeForce RTX 2080 Ti Rev. A] M 13448 from 155.247.166.220
09:27:02:WU00:FS01:Connecting to 155.247.166.220:8080
09:27:04:WU00:FS01:Downloading 837.11KiB
09:27:07:WU00:FS01:Download complete
09:27:07:WU00:FS01:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:14166 run:19 clone:33 gen:81 core:0x21 unit:0x0000007f0002894c5c38c096d2eca26e
09:27:07:WU00:FS01:Downloading core from http://cores.foldingathome.org/Win32/AMD64/NVIDIA/Fermi/Core_21.fah
09:27:07:WU00:FS01:Connecting to cores.foldingathome.org:80
09:27:07:WU00:FS01:FahCore 21: Downloading 3.47MiB
09:27:08:WU00:FS01:FahCore 21: Download complete
09:27:09:WU00:FS01:Valid core signature
09:27:09:WU00:FS01:Unpacked 11.80MiB to cores/cores.foldingathome.org/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe
09:27:09:WU00:FS01:Starting
09:27:09:WU00:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:\Users\Stefan\AppData\Roaming\FAHClient\cores/cores.foldingathome.org/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 00 -suffix 01 -version 705 -lifeline 13816 -checkpoint 15 -gpu-vendor nvidia -opencl-platform 0 -opencl-device 0 -cuda-device 0 -gpu 0
09:27:09:WU00:FS01:Started FahCore on PID 14092
09:27:09:WU00:FS01:Core PID:3740
09:27:09:WU00:FS01:FahCore 0x21 started
09:27:09:WU00:FS01:0x21:*********************** Log Started 2019-04-06T09:27:09Z ***********************
09:27:09:WU00:FS01:0x21:Project: 14166 (Run 19, Clone 33, Gen 81)
09:27:09:WU00:FS01:0x21:Unit: 0x0000007f0002894c5c38c096d2eca26e
09:27:09:WU00:FS01:0x21:CPU: 0x00000000000000000000000000000000
09:27:09:WU00:FS01:0x21:Machine: 1
09:27:09:WU00:FS01:0x21:Reading tar file core.xml
09:27:09:WU00:FS01:0x21:Reading tar file integrator.xml
09:27:09:WU00:FS01:0x21:Reading tar file state.xml
09:27:10:WU00:FS01:0x21:Reading tar file system.xml
09:27:10:WU00:FS01:0x21:Digital signatures verified
09:27:10:WU00:FS01:0x21:Folding@home GPU Core21 Folding@home Core
09:27:10:WU00:FS01:0x21:Version 0.0.18
09:27:22:WU00:FS01:0x21:Completed 0 out of 12500000 steps (0%)
09:27:22:WU00:FS01:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
09:27:57:Saving configuration to config.xml
09:27:57:<config>
09:27:57:  <!-- Network -->
09:27:57:  <proxy v=':8080'/>
09:27:57:
09:27:57:  <!-- Slot Control -->
09:27:57:  <power v='full'/>
09:27:57:
09:27:57:  <!-- User Information -->
09:27:57:  <passkey v='********************************'/>
09:27:57:  <team v='43829'/>
09:27:57:  <user v='FUFU'/>
09:27:57:
09:27:57:  <!-- Folding Slots -->
09:27:57:  <slot id='1' type='GPU'/>
09:27:57:</config>
09:28:22:FS01:Finishing

[foldy]

Did you put it in pcie x16 slot (check with tool GPUz)?
Have you anything else running on the CPU, because it needs one CPU core to feed GPU, maybe try increase windows power options min CPU speed.
Maybe it just takes a days until PPD settles in. What is your TPF (time for 1%) for current work unit and project id?

[FUFU]

Yes, I put it into a pcie x16 slot.
I'm running BOINC additionally. I'll try to clear one core for FAH. Good point thanks!

For two work units it took around 1:10 minutes (project id 14166 and 14150) and for two others around 2:20 minutes to get from 0% to 1% (project id 14127 and 14126). Is this normal?

[FUFU]

Ok, I paused two cores now it says 2,5 million ppd.
Looks better! 1:23 hours for project 14167 sounds fast I think?

Edit:
Now I mean what you meant by TPF. It says now 50 seconds.

[foldy]

Looks good now!

[FUFU]

Hm, now I am back on 1,1 million (TPF 45 seconds / project 14163). Nothing else runs on the CPU. Is it normal, that the ppd is so fluctuating?

[foldy]

The project 14163 only has 6000 atoms and that may be too less to use all your GPUs shaders. While your previous project 14167 with 2.5M PPD had 152515 atoms. So other projects with more atoms will be faster again.
Maybe you can get more bigger projects by selecting a specific cause in FAHControl?
At this website you can check the project values and PPD http://www.linuxforge.net/bonuscalc2.php

[MeeLee]

Could be that BOINC is sharing the GPU, and resources are split between the two programs?

I know a lot of projects run great on mid-class graphics cards like the RTX 2060 to a 2080, sometimes even these cards don't see the full potential (especially with projects in the 13k that could eat almost 25% of PPD, compared to projects of 14k).

[FUFU]

Ok, cool thanks! I wasn't aware of the fact that this could have to do with the number of atoms. Is there a kind of overview, which shows what project number contributes to what cause (e.g. cancer)?

[MeeLee]

Ok, cool thanks! I wasn't aware of the fact that this could have to do with the number of atoms. Is there a kind of overview, which shows what project number contributes to what cause (e.g. cancer)?

https://apps.foldingathome.org/psummary