Dual GTX 970 System - Low PPD After Rebuild

If you think it might be a driver problem, see viewforum.php?f=79

Moderators: Site Moderators, FAHC Science Team

Post Reply
Unicorn
Posts: 7
Joined: Sat Apr 30, 2016 4:22 pm

Dual GTX 970 System - Low PPD After Rebuild

Post by Unicorn »

Yesterday I rebuilt one of my open air dual GTX 970 machines into a 4U rack case. Unfortunately I lost the BIOS profile due to a dead motherboard battery, but I remembered most if not all of the overclock settings and had it back up and running again in no time. I left it on a bench in the comms room folding overnight and came back 24 hours later to check on it and install it into a cabinet, but instead of the ~600k PPD it normally produces, it's now churning out a mere ~120K PPD. Both cards are folding on FahCore_21, one is on 48k PPD and the other at 65k PPD.

Does anyone know what might be going on here? This has crippled my efforts and I would like to get it running properly again.

I had attempted to install a third card in the machine, but I didn't have enough PSU cables for my old 1250W Enermax to spread the load across the rails and it refused to boot with all three installed, so I pulled one and put it back into another board for now until I get some connectors from Hong Kong and make up some more PCIE cables.

System specs as follows:

i7 920 @ 4.0 GHz
12GB Corsair Vengeance @ 1600MHz
EVGA X58 Classified
2x MSI GTX 970 G4 Gaming 4GB
Enermax Revolution 85+ 1250W
240GB Crucial BX200 SSD boot disk
Client version 7.4.4
Windows 8.1 x64
GeForce driver 364.72

Log files to follow.
Unicorn
Posts: 7
Joined: Sat Apr 30, 2016 4:22 pm

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by Unicorn »

Code: Select all

*********************** Log Started 2016-04-30T14:50:06Z ***********************
14:50:06:************************* Folding@home Client *************************
14:50:06:      Website: http://folding.stanford.edu/
14:50:06:    Copyright: (c) 2009-2014 Stanford University
14:50:06:       Author: Joseph Coffland <joseph@cauldrondevelopment.com>
14:50:06:         Args: 
14:50:06:       Config: C:/Users/Matt/AppData/Roaming/FAHClient/config.xml
14:50:06:******************************** Build ********************************
14:50:06:      Version: 7.4.4
14:50:06:         Date: Mar 4 2014
14:50:06:         Time: 20:26:54
14:50:06:      SVN Rev: 4130
14:50:06:       Branch: fah/trunk/client
14:50:06:     Compiler: Intel(R) C++ MSVC 1500 mode 1200
14:50:06:      Options: /TP /nologo /EHa /Qdiag-disable:4297,4103,1786,279 /Ox -arch:SSE
14:50:06:               /QaxSSE2,SSE3,SSSE3,SSE4.1,SSE4.2 /Qopenmp /Qrestrict /MT /Qmkl
14:50:06:     Platform: win32 XP
14:50:06:         Bits: 32
14:50:06:         Mode: Release
14:50:06:******************************* System ********************************
14:50:06:          CPU: Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz
14:50:06:       CPU ID: GenuineIntel Family 6 Model 26 Stepping 5
14:50:06:         CPUs: 8
14:50:06:       Memory: 3.99GiB
14:50:06:  Free Memory: 2.49GiB
14:50:06:      Threads: WINDOWS_THREADS
14:50:06:   OS Version: 6.2
14:50:06:  Has Battery: false
14:50:06:   On Battery: false
14:50:06:   UTC Offset: 1
14:50:06:          PID: 5140
14:50:06:          CWD: C:/Users/Matt/AppData/Roaming/FAHClient
14:50:06:           OS: Windows 8.1
14:50:06:      OS Arch: AMD64
14:50:06:         GPUs: 2
14:50:06:        GPU 0: NVIDIA:5 GM204 [GeForce GTX 970]
14:50:06:        GPU 1: NVIDIA:5 GM204 [GeForce GTX 970]
14:50:06:         CUDA: 5.2
14:50:06:  CUDA Driver: 8000
14:50:06:Win32 Service: false
14:50:06:***********************************************************************
14:50:06:<config>
14:50:06:  <!-- Network -->
14:50:06:  <proxy v='proxy:8080'/>
14:50:06:  <proxy-pass v='********'/>
14:50:06:  <proxy-user v='mnesbitt084'/>
14:50:06:
14:50:06:  <!-- Slot Control -->
14:50:06:  <power v='full'/>
14:50:06:
14:50:06:  <!-- User Information -->
14:50:06:  <passkey v='********************************'/>
14:50:06:  <team v='35947'/>
14:50:06:  <user v='Unicorn'/>
14:50:06:
14:50:06:  <!-- Folding Slots -->
14:50:06:  <slot id='0' type='CPU'>
14:50:06:    <cpus v='6'/>
14:50:06:  </slot>
14:50:06:  <slot id='1' type='GPU'/>
14:50:06:  <slot id='2' type='GPU'/>
14:50:06:</config>
14:50:06:Trying to access database...
14:50:06:Successfully acquired database lock
14:50:06:Enabled folding slot 00: READY cpu:6
14:50:06:Enabled folding slot 01: READY gpu:0:GM204 [GeForce GTX 970]
14:50:06:Enabled folding slot 02: READY gpu:1:GM204 [GeForce GTX 970]
14:50:06:WU03:FS02:Starting
14:50:06:WU03:FS02:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Matt/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 03 -suffix 01 -version 704 -lifeline 5140 -checkpoint 15 -gpu 1 -gpu-vendor nvidia
14:50:06:WU03:FS02:Started FahCore on PID 6112
14:50:07:WU03:FS02:Core PID:6132
14:50:07:WU03:FS02:FahCore 0x21 started
14:50:07:WU00:FS00:Starting
14:50:07:WU00:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Matt/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/Core_a4.fah/FahCore_a4.exe -dir 00 -suffix 01 -version 704 -lifeline 5140 -checkpoint 15 -np 6
14:50:07:WU00:FS00:Started FahCore on PID 5172
14:50:07:WU00:FS00:Core PID:5260
14:50:07:WU00:FS00:FahCore 0xa4 started
14:50:07:WU02:FS01:Starting
14:50:07:WU02:FS01:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Matt/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_21.fah/FahCore_21.exe -dir 02 -suffix 01 -version 704 -lifeline 5140 -checkpoint 15 -gpu 0 -gpu-vendor nvidia
14:50:07:WU02:FS01:Started FahCore on PID 5376
14:50:07:WU03:FS02:0x21:*********************** Log Started 2016-04-30T14:50:07Z ***********************
14:50:07:WU03:FS02:0x21:Project: 13104 (Run 1, Clone 52, Gen 12)
14:50:07:WU03:FS02:0x21:Unit: 0x00000008ab436c9056b4f80677272cb2
14:50:07:WU03:FS02:0x21:CPU: 0x00000000000000000000000000000000
14:50:07:WU03:FS02:0x21:Machine: 2
14:50:07:WU03:FS02:0x21:Digital signatures verified
14:50:07:WU03:FS02:0x21:Folding@home GPU Core21 Folding@home Core
14:50:07:WU03:FS02:0x21:Version 0.0.17
14:50:07:WU02:FS01:Core PID:5552
14:50:07:WU02:FS01:FahCore 0x21 started
14:50:08:WU00:FS00:0xa4:
14:50:08:WU00:FS00:0xa4:*------------------------------*
14:50:08:WU00:FS00:0xa4:Folding@Home Gromacs GB Core
14:50:08:WU00:FS00:0xa4:Version 2.27 (Dec. 15, 2010)
14:50:08:WU00:FS00:0xa4:
14:50:08:WU00:FS00:0xa4:Preparing to commence simulation
14:50:08:WU00:FS00:0xa4:- Looking at optimizations...
14:50:08:WU00:FS00:0xa4:- Files status OK
14:50:08:WU00:FS00:0xa4:- Expanded 118523 -> 268860 (decompressed 226.8 percent)
14:50:08:WU00:FS00:0xa4:Called DecompressByteArray: compressed_data_size=118523 data_size=268860, decompressed_data_size=268860 diff=0
14:50:08:WU00:FS00:0xa4:- Digital signature verified
14:50:08:WU00:FS00:0xa4:
14:50:08:WU00:FS00:0xa4:Project: 6397 (Run 56, Clone 19, Gen 277)
14:50:08:WU00:FS00:0xa4:
14:50:08:WU00:FS00:0xa4:Assembly optimizations on if available.
14:50:08:WU00:FS00:0xa4:Entering M.D.
14:50:08:WU02:FS01:0x21:*********************** Log Started 2016-04-30T14:50:07Z ***********************
14:50:08:WU02:FS01:0x21:Project: 9441 (Run 5, Clone 2, Gen 239)
14:50:08:WU02:FS01:0x21:Unit: 0x00000110ab436c9d56af1b281f22f8e7
14:50:08:WU02:FS01:0x21:CPU: 0x00000000000000000000000000000000
14:50:08:WU02:FS01:0x21:Machine: 1
14:50:08:WU02:FS01:0x21:Digital signatures verified
14:50:08:WU02:FS01:0x21:Folding@home GPU Core21 Folding@home Core
14:50:08:WU02:FS01:0x21:Version 0.0.17
14:50:10:WU02:FS01:0x21:  Found a checkpoint file
14:50:10:WU03:FS02:0x21:  Found a checkpoint file
14:50:13:WU00:FS00:0xa4:Using Gromacs checkpoints
14:50:13:WU00:FS00:0xa4:Mapping NT from 6 to 6 
14:50:14:WU00:FS00:0xa4:Resuming from checkpoint
14:50:14:WU00:FS00:0xa4:Verified 00/wudata_01.log
14:50:14:WU00:FS00:0xa4:Verified 00/wudata_01.trr
14:50:14:WU00:FS00:0xa4:Verified 00/wudata_01.xtc
14:50:14:WU00:FS00:0xa4:Verified 00/wudata_01.edr
14:50:14:WU00:FS00:0xa4:Completed 4584300 out of 5000000 steps  (91%)
14:50:23:WU02:FS01:0x21:Completed 350000 out of 2500000 steps (14%)
14:50:23:WU02:FS01:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
14:50:31:WU03:FS02:0x21:Completed 560000 out of 2000000 steps (28%)
14:50:31:WU03:FS02:0x21:Temperature control disabled. Requirements: single Nvidia GPU, tmax must be < 110 and twait >= 900
14:51:18:WU00:FS00:0xa4:Completed 4600000 out of 5000000 steps  (92%)
14:54:00:WU00:FS00:0xa4:Completed 4650000 out of 5000000 steps  (93%)
14:56:13:WU02:FS01:0x21:Completed 375000 out of 2500000 steps (15%)
14:56:41:WU00:FS00:0xa4:Completed 4700000 out of 5000000 steps  (94%)
14:57:56:WU03:FS02:0x21:Completed 580000 out of 2000000 steps (29%)
14:59:25:WU00:FS00:0xa4:Completed 4750000 out of 5000000 steps  (95%)
15:02:03:WU02:FS01:0x21:Completed 400000 out of 2500000 steps (16%)
15:02:06:WU00:FS00:0xa4:Completed 4800000 out of 5000000 steps  (96%)
15:04:46:WU00:FS00:0xa4:Completed 4850000 out of 5000000 steps  (97%)
15:05:18:WU03:FS02:0x21:Completed 600000 out of 2000000 steps (30%)
15:07:27:WU00:FS00:0xa4:Completed 4900000 out of 5000000 steps  (98%)
15:07:55:WU02:FS01:0x21:Completed 425000 out of 2500000 steps (17%)
15:10:08:WU00:FS00:0xa4:Completed 4950000 out of 5000000 steps  (99%)
15:10:09:WU01:FS00:Connecting to 171.67.108.45:8080
15:10:11:WU01:FS00:Assigned to work server 155.247.166.219
15:10:11:WU01:FS00:Requesting new work unit for slot 00: RUNNING cpu:6 from 155.247.166.219
15:10:11:WU01:FS00:Connecting to 155.247.166.219:8080
15:10:12:WU01:FS00:Downloading 117.26KiB
15:10:12:WU01:FS00:Download complete
15:10:12:WU01:FS00:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:6396 run:66 clone:34 gen:103 core:0xa4 unit:0x000000810002894b5462c8a9a005369f
15:12:45:WU03:FS02:0x21:Completed 620000 out of 2000000 steps (31%)
15:12:49:WU00:FS00:0xa4:Completed 5000000 out of 5000000 steps  (100%)
15:12:49:WU00:FS00:0xa4:DynamicWrapper: Finished Work Unit: sleep=10000
15:12:59:WU00:FS00:0xa4:
15:12:59:WU00:FS00:0xa4:Finished Work Unit:
15:12:59:WU00:FS00:0xa4:- Reading up to 1254528 from "00/wudata_01.trr": Read 1254528
15:12:59:WU00:FS00:0xa4:trr file hash check passed.
15:12:59:WU00:FS00:0xa4:- Reading up to 109840 from "00/wudata_01.xtc": Read 109840
15:12:59:WU00:FS00:0xa4:xtc file hash check passed.
15:12:59:WU00:FS00:0xa4:edr file hash check passed.
15:12:59:WU00:FS00:0xa4:logfile size: 88161
15:12:59:WU00:FS00:0xa4:Leaving Run
15:13:02:WU00:FS00:0xa4:- Writing 1524429 bytes of core data to disk...
15:13:02:WU00:FS00:0xa4:Done: 1523917 -> 1289201 (compressed to 84.5 percent)
15:13:02:WU00:FS00:0xa4:  ... Done.
15:13:02:WU00:FS00:FahCore returned: FINISHED_UNIT (100 = 0x64)
15:13:02:WU00:FS00:Sending unit results: id:00 state:SEND error:NO_ERROR project:6397 run:56 clone:19 gen:277 core:0xa4 unit:0x000001560002894b5462c9cd7066f0de
15:13:02:WU00:FS00:Uploading 1.23MiB to 155.247.166.219
15:13:02:WU00:FS00:Connecting to 155.247.166.219:8080
15:13:02:WU01:FS00:Starting
15:13:02:WU01:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/Matt/AppData/Roaming/FAHClient/cores/web.stanford.edu/~pande/Win32/AMD64/Core_a4.fah/FahCore_a4.exe -dir 01 -suffix 01 -version 704 -lifeline 5140 -checkpoint 15 -np 6
15:13:02:WU01:F

S00:Started FahCore on PID 7724
15:13:03:WU01:FS00:Core PID:7092
15:13:03:WU01:FS00:FahCore 0xa4 started
15:13:03:WU01:FS00:0xa4:
15:13:03:WU01:FS00:0xa4:*------------------------------*
15:13:03:WU01:FS00:0xa4:Folding@Home Gromacs GB Core
15:13:03:WU01:FS00:0xa4:Version 2.27 (Dec. 15, 2010)
15:13:03:WU01:FS00:0xa4:
15:13:03:WU01:FS00:0xa4:Preparing to commence simulation
15:13:03:WU01:FS00:0xa4:- Looking at optimizations...
15:13:03:WU01:FS00:0xa4:- Created dyn
15:13:03:WU01:FS00:0xa4:- Files status OK
15:13:03:WU01:FS00:0xa4:- Expanded 119558 -> 271752 (decompressed 227.2 percent)
15:13:03:WU01:FS00:0xa4:Called DecompressByteArray: compressed_data_size=119558 data_size=271752, decompressed_data_size=271752 diff=0
15:13:03:WU01:FS00:0xa4:- Digital signature verified
15:13:03:WU01:FS00:0xa4:
15:13:03:WU01:FS00:0xa4:Project: 6396 (Run 66, Clone 34, Gen 103)
15:13:03:WU01:FS00:0xa4:
15:13:03:WU01:FS00:0xa4:Assembly optimizations on if available.
15:13:03:WU01:FS00:0xa4:Entering M.D.
15:13:05:WU00:FS00:Upload complete
15:13:05:WU00:FS00:Server responded WORK_ACK (400)
15:13:05:WU00:FS00:Final credit estimate, 1817.00 points
15:13:05:WU00:FS00:Cleaning up
15:13:09:WU01:FS00:0xa4:Mapping NT from 6 to 6 
15:13:09:WU01:FS00:0xa4:Completed 0 out of 5000000 steps  (0%)
15:13:44:WU02:FS01:0x21:Completed 450000 out of 2500000 steps (18%)
15:15:49:WU01:FS00:0xa4:Completed 50000 out of 5000000 steps  (1%)
15:18:30:WU01:FS00:0xa4:Completed 100000 out of 5000000 steps  (2%)
15:19:36:WU02:FS01:0x21:Completed 475000 out of 2500000 steps (19%)
15:20:06:WU03:FS02:0x21:Completed 640000 out of 2000000 steps (32%)
15:21:11:WU01:FS00:0xa4:Completed 150000 out of 5000000 steps  (3%)
15:23:55:WU01:FS00:0xa4:Completed 200000 out of 5000000 steps  (4%)
15:25:28:WU02:FS01:0x21:Completed 500000 out of 2500000 steps (20%)
15:26:36:WU01:FS00:0xa4:Completed 250000 out of 5000000 steps  (5%)
15:27:31:WU03:FS02:0x21:Completed 660000 out of 2000000 steps (33%)
15:29:16:WU01:FS00:0xa4:Completed 300000 out of 5000000 steps  (6%)
15:31:22:WU02:FS01:0x21:Completed 525000 out of 2500000 steps (21%)
15:31:57:WU01:FS00:0xa4:Completed 350000 out of 5000000 steps  (7%)
15:34:37:WU01:FS00:0xa4:Completed 400000 out of 5000000 steps  (8%)
15:34:54:WU03:FS02:0x21:Completed 680000 out of 2000000 steps (34%)
15:37:10:WU02:FS01:0x21:Completed 550000 out of 2500000 steps (22%)
15:37:18:WU01:FS00:0xa4:Completed 450000 out of 5000000 steps  (9%)
15:39:58:WU01:FS00:0xa4:Completed 500000 out of 5000000 steps  (10%)
15:42:20:WU03:FS02:0x21:Completed 700000 out of 2000000 steps (35%)
15:42:38:WU01:FS00:0xa4:Completed 550000 out of 5000000 steps  (11%)
15:43:03:WU02:FS01:0x21:Completed 575000 out of 2500000 steps (23%)
15:45:19:WU01:FS00:0xa4:Completed 600000 out of 5000000 steps  (12%)
15:48:17:WU01:FS00:0xa4:Completed 650000 out of 5000000 steps  (13%)
15:48:52:WU02:FS01:0x21:Completed 600000 out of 2500000 steps (24%)
15:49:41:WU03:FS02:0x21:Completed 720000 out of 2000000 steps (36%)
15:51:26:WU01:FS00:0xa4:Completed 700000 out of 5000000 steps  (14%)
15:54:10:WU01:FS00:0xa4:Completed 750000 out of 5000000 steps  (15%)
15:54:48:WU02:FS01:0x21:Completed 625000 out of 2500000 steps (25%)
15:57:09:WU01:FS00:0xa4:Completed 800000 out of 5000000 steps  (16%)
15:57:11:WU03:FS02:0x21:Completed 740000 out of 2000000 steps (37%)
16:00:21:WU01:FS00:0xa4:Completed 850000 out of 5000000 steps  (17%)
16:00:43:WU02:FS01:0x21:Completed 650000 out of 2500000 steps (26%)
16:03:31:WU01:FS00:0xa4:Completed 900000 out of 5000000 steps  (18%)
16:04:27:WU03:FS02:0x21:Completed 760000 out of 2000000 steps (38%)
16:06:46:WU02:FS01:0x21:Completed 675000 out of 2500000 steps (27%)
16:06:53:WU01:FS00:0xa4:Completed 950000 out of 5000000 steps  (19%)
16:10:08:WU01:FS00:0xa4:Completed 1000000 out of 5000000 steps  (20%)
16:11:59:WU03:FS02:0x21:Completed 780000 out of 2000000 steps (39%)
16:12:47:WU02:FS01:0x21:Completed 700000 out of 2500000 steps (28%)
16:13:16:WU01:FS00:0xa4:Completed 1050000 out of 5000000 steps  (21%)
16:16:11:WU01:FS00:0xa4:Completed 1100000 out of 5000000 steps  (22%)
16:18:36:WU02:FS01:0x21:Completed 725000 out of 2500000 steps (29%)
16:19:04:WU01:FS00:0xa4:Completed 1150000 out of 5000000 steps  (23%)
16:19:11:WU03:FS02:0x21:Completed 800000 out of 2000000 steps (40%)
16:22:03:WU01:FS00:0xa4:Completed 1200000 out of 5000000 steps  (24%)
16:24:28:WU02:FS01:0x21:Completed 750000 out of 2500000 steps (30%)
16:25:25:WU01:FS00:0xa4:Completed 1250000 out of 5000000 steps  (25%)
16:26:32:WU03:FS02:0x21:Completed 820000 out of 2000000 steps (41%)
16:28:10:FS00:Paused
16:28:10:FS01:Paused
16:28:10:FS02:Paused
16:28:10:FS00:Shutting core down
16:28:10:FS01:Shutting core down
16:28:10:FS02:Shutting core down
16:28:10:WU03:FS02:0x21:WARNING:Console control signal 1 on PID 6132
16:28:10:WU03:FS02:0x21:Exiting, please wait. . .
16:28:10:WU02:FS01:0x21:WARNING:Console control signal 1 on PID 5552
16:28:10:WU02:FS01:0x21:Exiting, please wait. . .
16:28:11:WU03:FS02:0x21:Folding@home Core Shutdown: INTERRUPTED
16:28:11:WU03:FS02:FahCore returned: INTERRUPTED (102 = 0x66)
16:28:11:WU02:FS01:FahCore returned: INTERRUPTED (102 = 0x66)
16:28:12:WU01:FS00:FahCore returned: INTERRUPTED (102 = 0x66)
16:28:43:Removing old file 'configs/config-20151124-133838.xml'
16:28:43:Saving configuration to config.xml
16:28:43:<config>
16:28:43:  <!-- Network -->
16:28:43:  <proxy v='proxy:8080'/>
16:28:43:  <proxy-pass v='********'/>
16:28:43:  <proxy-user v='mnesbitt084'/>
16:28:43:
16:28:43:  <!-- Slot Control -->
16:28:43:  <power v='full'/>
16:28:43:
16:28:43:  <!-- User Information -->
16:28:43:  <passkey v='********************************'/>
16:28:43:  <team v='35947'/>
16:28:43:  <user v='Unicorn'/>
16:28:43:
16:28:43:  <!-- Folding Slots -->
16:28:43:  <slot id='0' type='CPU'>
16:28:43:    <cpus v='6'/>
16:28:43:    <paused v='true'/>
16:28:43:  </slot>
16:28:43:  <slot id='1' type='GPU'>
16:28:43:    <paused v='true'/>
16:28:43:  </slot>
16:28:43:  <slot id='2' type='GPU'>
16:28:43:    <paused v='true'/>
16:28:43:  </slot>
16:28:43:</config>
Unicorn
Posts: 7
Joined: Sat Apr 30, 2016 4:22 pm

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by Unicorn »

MSI Afterburner reports that the GPU usage on both cards is hovering between 60-70% - not sure if that makes a difference or not as I never paid any attention to that value whilst it was performing well. I'll check it on some of my other rigs to find out what they are showing for GPU usage.
mmonnin
Posts: 324
Joined: Wed Dec 05, 2007 1:27 am

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by mmonnin »

If the GPU usage is not 95%+ then I would guess CPU or PCI-E 2.0 is holding you back.
Unicorn
Posts: 7
Joined: Sat Apr 30, 2016 4:22 pm

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by Unicorn »

What's really confusing and frustrating about this is that nothing whatsoever about this system except the case has changed, and it's now producing ~1/3rd of the points that it was for the past 6 months, on an open air test bench in an air conditioned room.

I have uninstalled and re-installed the client as well as the Nvidia drivers to absolutely no avail. Something has crippled this machine and I have no idea what.
toTOW
Site Moderator
Posts: 6307
Joined: Sun Dec 02, 2007 10:38 am
Location: Bordeaux, France
Contact:

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by toTOW »

Do the GPUs go back to full speed if you pause the CPU slot ?

Since you moved you hardware, make sure that everything is correctly seated in the connectors (PCIe, RAMs, CPU heatsink, ...), and that you don't have thermal paste issues on the CPU that could make it throttle ...
Image

Folding@Home beta tester since 2002. Folding Forum moderator since July 2008.
Unicorn
Posts: 7
Joined: Sat Apr 30, 2016 4:22 pm

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by Unicorn »

I've removed and re-seated everything. Pausing/removing the CPU slot from the machine does not make any difference. I tried that first before pulling and re-seating components. I am extremely confused by this. The cards both fold absolutely fine in other systems, but other cards will no longer achieve full GPU utilization in this machine.
Rel25917
Posts: 303
Joined: Wed Aug 15, 2012 2:31 am

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by Rel25917 »

Take a good look at all the bios settings as you say that did change(lost them all). Also what are the cards temps at? I wouldn't expect that big a loss from going from open air to a case but maybe an airflow problem?
Unicorn
Posts: 7
Joined: Sat Apr 30, 2016 4:22 pm

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by Unicorn »

Temps were not that bad; I was a little disappointed with the cooling performance of the 4u case but it needed to go in a rack, I can modify it to improve upon that and hopefully keep the fan noise levels at an acceptable level. I am having to run the card fans at about 75% duty cycle to keep the GPU temps between 65-70 degrees Celsius, but the Twin Frozr V coolers on these cards are very quiet, even at 100% they're almost inaudible over the 8x chassis fans which are my real noise concern - something I plan on managing with a GRID+ fan controller.

Anyway, after hours of pulling my hair out I solved this issue very late last night/early this morning. I remembered that when my old EVGA X58 Classified 3 based gaming system at home was running one of these 970s, it had a similar problem with extremely low GPU performance right after a CPU upgrade. I'd borrowed the i7 980X from it to test another board, and replaced it with an i7 930. When I swapped the 980X back into the board again I was experiencing almost unplayable frame rates in games like BF3 and GTAV, which the GTX 970 was previously not having any trouble with. After hours of troubleshooting that issue, the solution was another complete BIOS reset, which solved the problem.

The same thing happened last night with the dedicated folding machine. I loaded defaults, set the CPU overclock values from memory, booted into Windows and the system immediately began utilizing 95% of both GPUs. Something in EVGA's X58 BIOS restricts the performance of the PCIE lanes on the board. I went through every BIOS menu several times with a fine toothed comb and couldn't find anything even remotely related to PCIE performance or anything else that would bottleneck a GPU. All I know is that as soon as I remembered my saga with the low FPS on my gaming rig, I was almost certain this was the same thing. It's quite a common problem too, if you Google "GTX 970 low FPS" or "GTX 970 low GPU usage" you'll see plenty of other people having the same issues in games. Something that gets changed on the board during a CMOS reset/update bottlenecks the PCIE lanes and the only solution I have found with my X58 Classified boards is to load the defaults and go from there.

Thanks for your help, I wrote a fairly detailed conclusion to this in the hope that someone else who may be experiencing this problem will benefit from it in the future.
bruce
Posts: 20910
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: Dual GTX 970 System - Low PPD After Rebuild

Post by bruce »

Thanks for your help.

(... I only wish that somebody could figure out which BIOS setting was the culprit.)
Post Reply