more threads ... help please.

ozaudio

ozaudio

ozaudio

Soldato
Joined
23 Jan 2010
Posts
4,076
Location
St helens
im getting very annoyed, today ive been and bought a new case, psu and a gtx 560.

i feel its a wast of cash as my q6600 wont even fold!

i am sure im setting it up correctly , can some body make sure this is correct?



and when its running all it ever stays at is this

 
certainly looks fine as you can see it has a PPD status and is able to estimate the output just hasnt got as far as giving you and ETA to complete.

Although I think if your folding with a gpu you get better numbers by setting the CPU threads from -1 to 3 (in your case) and leaving a core to run the GFX offload.

I thought the q6600 had threads also? does it not have HT enabled?

edit: Although - dont try and run BONIC and Folding at the same time. I think you will be really asking too much of it then.
 
Couple of things:

1. Show us your task manager sorted by CPU
2. Have you setup a passkey? You won't get any SMP bonus points until you do.
3. Plus you have to complete the WU's within the time limits and not get more than 20% failures. If you do then you won't get the QRB (quick return bonus) for SMP WU's

A Q6600 at stock should get around 5K PPD all things being setup correctly.
 
Last edited:
ok so i changed system and stilll wont fold? another 3570k :D with a gtx 560, net problem?


22:21:27:Adding folding slot 00: READY gpu:0:"GF114 [GeForce GTX 560]"
22:21:27:WU00:FS00:Connecting to assign-GPU.stanford.edu:80
22:21:28:Saving configuration to config.xml
22:21:28:<config>
22:21:28: <!-- FahCore Control -->
22:21:28: <core-priority v='low'/>
22:21:28:
22:21:28: <!-- Folding Slot Configuration -->
22:21:28: <gpu v='true'/>
22:21:28:
22:21:28: <!-- Network -->
22:21:28: <proxy v=':8080'/>
22:21:28:
22:21:28: <!-- User Information -->
22:21:28: <passkey v='********************************'/>
22:21:28: <team v='10'/>
22:21:28: <user v='ozaudio'/>
22:21:28:
22:21:28: <!-- Folding Slots -->
22:21:28: <slot id='1' type='SMP'/>
22:21:28: <slot id='0' type='GPU'/>
22:21:28:</config>
22:21:28:WU00:FS00:News: Welcome to Folding@Home
22:21:28:WU00:FS00:Assigned to work server 171.67.108.141
22:21:28:WU00:FS00:Requesting new work unit for slot 00: READY gpu:0:"GF114 [GeForce GTX 560]" from 171.67.108.141
22:21:28:WU00:FS00:Connecting to 171.67.108.141:8080
22:21:29:WU00:FS00:Downloading 57.22KiB
22:21:29:WU00:FS00:Download complete
22:21:29:WU00:FS00:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:8044 run:0 clone:593 gen:21 core:0x15 unit:0x0000001b6953ee2d50626b1fc9e052cf
22:21:29:WU00:FS00:Starting
22:21:29:WU00:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/andy/AppData/Roaming/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/Core_15.fah/FahCore_15.exe -dir 00 -suffix 01 -version 702 -lifeline 2204 -checkpoint 15 -gpu 0
22:21:29:WU00:FS00:Started FahCore on PID 2456
22:21:29:WU00:FS00:Core PID:1192
22:21:29:WU00:FS00:FahCore 0x15 started
22:21:30:WU00:FS00:0x15:
22:21:30:WU00:FS00:0x15:*------------------------------*
22:21:30:WU00:FS00:0x15:Folding@Home GPU Core
22:21:30:WU00:FS00:0x15:Version 2.25 (Wed May 9 17:03:01 EDT 2012)
22:21:30:WU00:FS00:0x15:Build host AmoebaRemote
22:21:30:WU00:FS00:0x15:Board Type NVIDIA/CUDA
22:21:30:WU00:FS00:0x15:Core 15
22:21:30:WU00:FS00:0x15:
22:21:30:WU00:FS00:0x15:Window's signal control handler registered.
22:21:30:WU00:FS00:0x15:Preparing to commence simulation
22:21:30:WU00:FS00:0x15:- Looking at optimizations...
22:21:30:WU00:FS00:0x15:DeleteFrameFiles: successfully deleted file=00/wudata_01.ckp
22:21:30:WU00:FS00:0x15:- Created dyn
22:21:30:WU00:FS00:0x15:- Files status OK
22:21:30:WU00:FS00:0x15:sizeof(CORE_PACKET_HDR) = 512 file=<>
22:21:30:WU00:FS00:0x15:- Expanded 58080 -> 257038 (decompressed 442.5 percent)
22:21:30:WU00:FS00:0x15:Called DecompressByteArray: compressed_data_size=58080 data_size=257038, decompressed_data_size=257038 diff=0
22:21:30:WU00:FS00:0x15:- Digital signature verified
22:21:30:WU00:FS00:0x15:
22:21:30:WU00:FS00:0x15:Project: 8044 (Run 0, Clone 593, Gen 21)
22:21:30:WU00:FS00:0x15:
22:21:30:WU00:FS00:0x15:Assembly optimizations on if available.
22:21:30:WU00:FS00:0x15:Entering M.D.
22:21:32:WU00:FS00:0x15:Tpr hash 00/wudata_01.tpr: 2513867283 654228654 3194749316 1053838830 283322540
22:21:32:WU00:FS00:0x15:GPU device id=0
22:21:32:WU00:FS00:0x15:Working on Gallium Rubidium Oxygen Manganese Argon Carbon Silicon
22:21:32:WU00:FS00:0x15:Client config unavailable.
22:21:32:WU00:FS00:0x15:Starting GUI Server
22:22:35:WU00:FS00:0x15:Setting checkpoint frequency: 500000
22:22:35:WU00:FS00:0x15:Completed 3 out of 50000000 steps (0%).
 
So the GPU is crunching away. The SMP is almost stalled.

Elsewhere I seem to recall your saying you have set the SMP cores to 3? If so that could be an issue. SMP I understand needs an even number of cores, with the default -1 meaning the f@h client works on what is available. But there may also be issues with setting anything other than maxiumum cores. Certainly I saw an error in my logs when I was trying to change stuff around and had my 8 core i7 2600 allocated at only 6 cores, and indeed a WU froze. Try reverting to -1 slots?

But these days the SMP units will go slowly on a Q6600 CPU. By modern standards that CPU, which was once good, is now well outdated by power. The predicted 52 minutes per frame (1%) of the SMP WU indicates your problem. The real power in that machine is now in the GTX560 and it's there you need to maximise resources. You might be better off deleting the SMP core, and instead allocating three separate uniprocessor cores, leaving the final CPU core available to the GPU (GPU does need part of a CPU)?
 
this is a totaly rebuilt pc , 3570k set to -1 smp

the q6600 is on the auction site and will help pay for a PSU for 3 hardcore gpus :D

cpu folding up to 7273 now... its rising!!!!

waiting on funds for a good cooler to overclock this thing
 
Last edited:
As I was told in a recent other thread, don't rely on the f@h 7 client to give accurate PPD. HFM.net may be more reliable.

Remember different WUS have different points and some crunch a lot better than others evidenced by my own f@h 7 predicted PPD ranging from feeble mid 20k and currently suggesting over 50k.
 
its dropped back down to 6200 ppd for cpu now so ill wait for a new work unit and see what i get then. ive have a 3570k before and was decent ppd!

ill try HFM
 
If its on the same machine then it should work fine, just add a v7 client as localhost.

if its a remote machine then its a bit trickier, you have to have a password and add the hfm host's ip or open up the restrictions on the v7 client you are monitoring.
 
Biffa said:
If its on the same machine then it should work fine, just add a v7 client as localhost.

if its a remote machine then its a bit trickier, you have to have a password and add the hfm host's ip or open up the restrictions on the v7 client you are monitoring.
Click to expand...

Figured it out, was using an old installer that I had on my backup drive rather than downloading the latest.

My old Phenom II seems to be giving better PPD now @ stock than it used to at 4ghz
 
You must log in or register to reply here.
Back
Top Bottom