My PC is new to Folding@home. help please.

[rs199208]

AMD Phenom II X2 555 Black Edition Callisto 3.2GHz 2 x 512KB L2 Cache 6MB L3 Cache Socket AM3
Unlocked to a:
AMD Phenom II X4 B55 Overclocked to 3900 Mhz on all 4 cores with a Noctua NH-D14 Heatsink.
MB- MSI 870-G45
4GB G.Skill RipJaws DDR3-2000 @1600 Mhz
Gigabyte 6850 1GB dual fans on heatsink, overclocked GPU @ 850 Mhz and Memory @ 1200Mhz.
HD- OCZ Vertex 2
Windows 7 64 ultimate
Thermaltake Pure Power 500 watt

I am running the GPU3 client with the r800 string.
Should I put any other strings in to use the CPU cores?
Right now the client is saying:

Donor
Core is not running now.

Progress:
108000/10000000

Thanks, Rich

[PantherX]

Welcome to the F@H Forum rs199208,

Can you please post the FAHlog (viewtopic.php?p=160884#p160884) so that we can understand what the F@H Client is doing and help you with troubleshooting?

FYI, the flag for the CPU Client to utilize all available CPUs is -smp but it would be nice if we viewed your FAHlog first.

[kiore]

Hi, I think he has this solved by running -smp 2 and the gpu at present, will ask him to pop back here with a "solved" or any follow up questions.
As he reports his cores running at 79% with the above I am unsure whether -smp 3 advisable without throttling the gpu, but some AMD/ATI gpu people may have better advice.

[rs199208]

kiore helped me solve most of it I think.
If you have any suggestions that would be nice.


F@H SMP 2 core and GPU3 client running. by rs199208, on Flickr


Gpu3 r800

Code: Select all

[19:54:44] + Attempting to send results [August 20 19:54:44 UTC]
[19:54:44] Gpu type=1 species=4.
[19:54:46] + Results successfully sent
[19:54:46] Thank you for your contribution to Folding@Home.
[19:54:46] + Number of Units Completed: 24

[19:54:50] - Preparing to get new work unit...
[19:54:50] Cleaning up work directory
[19:54:50] + Attempting to get work packet
[19:54:50] Passkey found
[19:54:50] Gpu type=1 species=4.
[19:54:50] - Connecting to assignment server
[19:54:50] - Successful: assigned to (171.64.65.102).
[19:54:50] + News From Folding@Home: Welcome to Folding@Home
[19:54:50] Loaded queue successfully.
[19:54:50] Gpu type=1 species=4.
[19:54:52] + Closed connections
[19:54:52] 
[19:54:52] + Processing work unit
[19:54:52] Core required: FahCore_11.exe
[19:54:52] Core found.
[19:54:52] Working on queue slot 06 [August 20 19:54:52 UTC]
[19:54:52] + Working ...
[19:54:52] 
[19:54:52] *------------------------------*
[19:54:52] Folding@Home GPU Core - Beta
[19:54:52] Version 1.24 (Mon Feb 9 11:00:12 PST 2009)
[19:54:52] 
[19:54:52] Compiler  : Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86 
[19:54:52] Build host: amoeba
[19:54:52] Board Type: AMD
[19:54:52] Core      : 
[19:54:52] Preparing to commence simulation
[19:54:52] - Looking at optimizations...
[19:54:52] - Created dyn
[19:54:52] - Files status OK
[19:54:52] - Expanded 98782 -> 492188 (decompressed 498.2 percent)
[19:54:52] Called DecompressByteArray: compressed_data_size=98782 data_size=492188, decompressed_data_size=492188 diff=0
[19:54:52] - Digital signature verified
[19:54:52] 
[19:54:52] Project: 5733 (Run 1, Clone 646, Gen 469)
[19:54:52] 
[19:54:52] Assembly optimizations on if available.
[19:54:52] Entering M.D.
[19:54:58] Tpr hash work/wudata_06.tpr:  667992579 804318383 3522811897 4027976731 2727189601
[19:54:58] Working on Protein
[19:54:58] Client config found, loading data.
[19:54:58] Starting GUI Server

SMP -smp 2

Code: Select all

[19:46:53] + Attempting to send results [August 20 19:46:53 UTC]
[19:47:03] + Results successfully sent
[19:47:03] Thank you for your contribution to Folding@Home.
[19:47:03] + Number of Units Completed: 4

[19:47:07] - Preparing to get new work unit...
[19:47:07] Cleaning up work directory
[19:47:07] + Attempting to get work packet
[19:47:07] Passkey found
[19:47:07] - Connecting to assignment server
[19:47:07] - Successful: assigned to (171.64.65.54).
[19:47:07] + News From Folding@Home: Welcome to Folding@Home
[19:47:08] Loaded queue successfully.
[19:47:11] + Closed connections
[19:47:11] 
[19:47:11] + Processing work unit
[19:47:11] Core required: FahCore_a3.exe
[19:47:11] Core found.
[19:47:11] Working on queue slot 05 [August 20 19:47:11 UTC]
[19:47:11] + Working ...
[19:47:11] 
[19:47:11] *------------------------------*
[19:47:11] Folding@Home Gromacs SMP Core
[19:47:11] Version 2.27 (Dec. 15, 2010)
[19:47:11] 
[19:47:11] Preparing to commence simulation
[19:47:11] - Looking at optimizations...
[19:47:11] - Created dyn
[19:47:11] - Files status OK
[19:47:12] - Expanded 1764940 -> 2253449 (decompressed 127.6 percent)
[19:47:12] Called DecompressByteArray: compressed_data_size=1764940 data_size=2253449, decompressed_data_size=2253449 diff=0
[19:47:12] - Digital signature verified
[19:47:12] 
[19:47:12] Project: 6073 (Run 0, Clone 13, Gen 475)
[19:47:12] 
[19:47:12] Assembly optimizations on if available.
[19:47:12] Entering M.D.
[19:47:18] Mapping NT from 2 to 2 
[19:47:18] Completed 0 out of 500000 steps  (0%)
[19:57:48] Completed 5000 out of 500000 steps  (1%)

Mod Edit: Changed Quote Tags To Code Tags - PantherX

[bruce]

Running FahCore_11 plus the drivers for ATI tend to use one full CPU to keep the GPU supplied with data. In V6, this can be mitigated somewhat by setting Environment Variables.

Beta support for the r800 and above is provided using a new OpenCL core and the V7 beta client. Optimizing that platform has yet to be done, so at this point it also tends to take one full CPU or maybe a little more.

On an X4, it may make sense to test it with either -smp 2 or -smp 3 and see which works best for you. The -smp 2 choice is probably more stable but it seems to depend a lot on which projects you're assigned whether -smp 3 is a good option or not.

[rs199208]

UPDATE: After installing Vmware and linuxfah's preconfigure Linux http://www.linuxforge.net/docs/crunching/fah-vmware.php
I am happy with the results on this little chip.

Unit 1 finished with 95 percent of time to deadline remaining.
[00:22:56] Updated performance fraction: 0.948482
[00:22:56] Project: 6063 (Run 0, Clone 172, Gen 294)
took 6.52 hr to complete!
Each frame consistent 4min 30sec
That is a huge improvement!
I am not going to hack this rig up for the i7 I'll just have to get the parts else where.
And this "was" a dual core that I bought last year for $90.