Attention GPU Folders: Make sure you are set up for QRB units

Status
Not open for further replies.
Well I got my first 8057, on my laptop.

GPU 680m (essentially a lower-clocked GTX 670, GK104, 1344 shaders, 256-bit mem, 854/1800 core/mem)
TPF: 3m 45s
PPD: 60-61K

Do you know what the power draw is for this system? The PPD/W for high end laptops may be good.
 
HFM says that I have 4 WU finished and 4 WU that have failed. I don't get it... everything looks like it is fine:
Code:
23:29:27:WU01:FS00:0x15:Completed  50000000 out of 50000000 steps (100%).
23:29:27:WU02:FS00:Connecting to assign-GPU.stanford.edu:80
23:29:27:WU01:FS00:0x15:Finished fah_main status=0
23:29:27:WU01:FS00:0x15:Successful run
23:29:27:WU01:FS00:0x15:DynamicWrapper: Finished Work Unit: sleep=10000
23:29:27:WU02:FS00:News: Welcome to Folding@Home
23:29:27:WU02:FS00:Assigned to work server 171.67.108.144
23:29:27:WU02:FS00:Requesting new work unit for slot 00: RUNNING gpu:0:"GF110 [Geforce GTX 580]" from 171.67.108.144
23:29:27:WU02:FS00:Connecting to 171.67.108.144:8080
23:29:28:WU02:FS00:Downloading 56.77KiB
23:29:29:WU02:FS00:Download complete
23:29:29:WU02:FS00:Received Unit: id:02 state:DOWNLOAD error:OK project:8057 run:0 clone:502 gen:6 core:0x15 unit:0x000000076953ee30508b38c1e164ecf6
23:29:37:WU01:FS00:0x15:Reserved 326720 bytes for xtc file; Cosm status=0
23:29:37:WU01:FS00:0x15:Allocated 326720 bytes for xtc file
23:29:37:WU01:FS00:0x15:- Reading up to 326720 from "01/wudata_01.xtc": Read 326720
23:29:37:WU01:FS00:0x15:Read 326720 bytes from xtc file; available packet space=786103744
23:29:37:WU01:FS00:0x15:xtc file hash check passed.
23:29:37:WU01:FS00:0x15:Reserved 20400 20400 786103744 bytes for arc file=<01/wudata_01.trr> Cosm status=0
23:29:37:WU01:FS00:0x15:Allocated 20400 bytes for arc file
23:29:37:WU01:FS00:0x15:- Reading up to 20400 from "01/wudata_01.trr": Read 20400
23:29:37:WU01:FS00:0x15:Read 20400 bytes from arc file; available packet space=786083344
23:29:37:WU01:FS00:0x15:trr file hash check passed.
23:29:37:WU01:FS00:0x15:Allocated 544 bytes for edr file
23:29:37:WU01:FS00:0x15:Read bedfile
23:29:37:WU01:FS00:0x15:edr file hash check passed.
23:29:37:WU01:FS00:0x15:Allocated 36782 bytes for logfile
23:29:37:WU01:FS00:0x15:Read logfile
23:29:37:WU01:FS00:0x15:GuardedRun: success in DynamicWrapper
23:29:37:WU01:FS00:0x15:GuardedRun: done
23:29:37:WU01:FS00:0x15:Run: GuardedRun completed.
23:29:41:WU01:FS00:0x15:+ Opened results file
23:29:41:WU01:FS00:0x15:- Writing 384958 bytes of core data to disk...
23:29:41:WU01:FS00:0x15:Done: 384446 -> 353227 (compressed to 91.8 percent)
23:29:41:WU01:FS00:0x15:  ... Done.
23:29:41:WU01:FS00:0x15:DeleteFrameFiles: successfully deleted file=01/wudata_01.ckp
23:29:41:WU01:FS00:0x15:Shutting down core 
23:29:41:WU01:FS00:0x15:
23:29:41:WU01:FS00:0x15:Folding@home Core Shutdown: FINISHED_UNIT
23:29:41:WU01:FS00:FahCore returned: FINISHED_UNIT (100 = 0x64)
23:29:41:WU01:FS00:Sending unit results: id:01 state:SEND error:OK project:8057 run:0 clone:491 gen:7 core:0x15 unit:0x000000086953ee30508b38c070b328c3
23:29:41:WU01:FS00:Uploading 345.45KiB to 171.67.108.144
23:29:41:WU01:FS00:Connecting to 171.67.108.144:8080
23:29:41:WU02:FS00:Starting
23:29:41:WU02:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/kendrak/AppData/Roaming/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/beta/Core_15.fah/FahCore_15.exe -dir 02 -suffix 01 -version 701 -lifeline 4480 -checkpoint 15 -gpu 0
23:29:41:WU02:FS00:Started FahCore on PID 11756
23:29:41:WU02:FS00:Core PID:7408
23:29:41:WU02:FS00:FahCore 0x15 started
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:*------------------------------*
23:29:42:WU02:FS00:0x15:Folding@Home GPU Core
23:29:42:WU02:FS00:0x15:Version                2.25 (Wed May 9 17:03:01 EDT 2012)
23:29:42:WU02:FS00:0x15:Build host             AmoebaRemote
23:29:42:WU02:FS00:0x15:Board Type             NVIDIA/CUDA
23:29:42:WU02:FS00:0x15:Core                   15
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:Window's signal control handler registered.
23:29:42:WU02:FS00:0x15:Preparing to commence simulation
23:29:42:WU02:FS00:0x15:- Looking at optimizations...
23:29:42:WU02:FS00:0x15:DeleteFrameFiles: successfully deleted file=02/wudata_01.ckp
23:29:42:WU02:FS00:0x15:- Created dyn
23:29:42:WU02:FS00:0x15:- Files status OK
23:29:42:WU02:FS00:0x15:sizeof(CORE_PACKET_HDR) = 512 file=<>
23:29:42:WU02:FS00:0x15:- Expanded 57621 -> 261042 (decompressed 453.0 percent)
23:29:42:WU02:FS00:0x15:Called DecompressByteArray: compressed_data_size=57621 data_size=261042, decompressed_data_size=261042 diff=0
23:29:42:WU02:FS00:0x15:- Digital signature verified
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:Project: 8057 (Run 0, Clone 502, Gen 6)
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:Assembly optimizations on if available.
23:29:42:WU02:FS00:0x15:Entering M.D.
23:29:44:WU02:FS00:0x15:Tpr hash 02/wudata_01.tpr:  2252366912 2484876154 920962738 85761893 3389626746
23:29:44:WU02:FS00:0x15:GPU device id=0
23:29:44:WU02:FS00:0x15:Working on Giving Russians Opium May Alter Current Situation
23:29:44:WU02:FS00:0x15:Client config unavailable.
23:29:44:WU02:FS00:0x15:Starting GUI Server
23:29:47:WU01:FS00:Upload 74.11%
23:29:49:WU01:FS00:Upload complete
23:29:49:WU01:FS00:Server responded WORK_ACK (400)
23:29:49:WU01:FS00:Final credit estimate, 18229.00 points
23:29:49:WU01:FS00:Cleaning up
23:30:45:WU02:FS00:0x15:Setting checkpoint frequency: 500000
23:30:45:WU02:FS00:0x15:Completed         3 out of 50000000 steps (0%).
23:32:48:WU02:FS00:0x15:Completed    500000 out of 50000000 steps (1%).

Am I missing something or is HFM not like these new WU?
 
Well I got my first 8057, on my laptop.

GPU 680m (essentially a lower-clocked GTX 670, GK104, 1344 shaders, 256-bit mem, 854/1800 core/mem)
TPF: 3m 45s
PPD: 60-61K

Lol...is that an optimus enabled unit? I got the quadro k1000 in my laptop...
 
HFM says that I have 4 WU finished and 4 WU that have failed. I don't get it... everything looks like it is fine:
Code:
23:29:27:WU01:FS00:0x15:Completed  50000000 out of 50000000 steps (100%).
23:29:27:WU02:FS00:Connecting to assign-GPU.stanford.edu:80
23:29:27:WU01:FS00:0x15:Finished fah_main status=0
23:29:27:WU01:FS00:0x15:Successful run
23:29:27:WU01:FS00:0x15:DynamicWrapper: Finished Work Unit: sleep=10000
23:29:27:WU02:FS00:News: Welcome to Folding@Home
23:29:27:WU02:FS00:Assigned to work server 171.67.108.144
23:29:27:WU02:FS00:Requesting new work unit for slot 00: RUNNING gpu:0:"GF110 [Geforce GTX 580]" from 171.67.108.144
23:29:27:WU02:FS00:Connecting to 171.67.108.144:8080
23:29:28:WU02:FS00:Downloading 56.77KiB
23:29:29:WU02:FS00:Download complete
23:29:29:WU02:FS00:Received Unit: id:02 state:DOWNLOAD error:OK project:8057 run:0 clone:502 gen:6 core:0x15 unit:0x000000076953ee30508b38c1e164ecf6
23:29:37:WU01:FS00:0x15:Reserved 326720 bytes for xtc file; Cosm status=0
23:29:37:WU01:FS00:0x15:Allocated 326720 bytes for xtc file
23:29:37:WU01:FS00:0x15:- Reading up to 326720 from "01/wudata_01.xtc": Read 326720
23:29:37:WU01:FS00:0x15:Read 326720 bytes from xtc file; available packet space=786103744
23:29:37:WU01:FS00:0x15:xtc file hash check passed.
23:29:37:WU01:FS00:0x15:Reserved 20400 20400 786103744 bytes for arc file=<01/wudata_01.trr> Cosm status=0
23:29:37:WU01:FS00:0x15:Allocated 20400 bytes for arc file
23:29:37:WU01:FS00:0x15:- Reading up to 20400 from "01/wudata_01.trr": Read 20400
23:29:37:WU01:FS00:0x15:Read 20400 bytes from arc file; available packet space=786083344
23:29:37:WU01:FS00:0x15:trr file hash check passed.
23:29:37:WU01:FS00:0x15:Allocated 544 bytes for edr file
23:29:37:WU01:FS00:0x15:Read bedfile
23:29:37:WU01:FS00:0x15:edr file hash check passed.
23:29:37:WU01:FS00:0x15:Allocated 36782 bytes for logfile
23:29:37:WU01:FS00:0x15:Read logfile
23:29:37:WU01:FS00:0x15:GuardedRun: success in DynamicWrapper
23:29:37:WU01:FS00:0x15:GuardedRun: done
23:29:37:WU01:FS00:0x15:Run: GuardedRun completed.
23:29:41:WU01:FS00:0x15:+ Opened results file
23:29:41:WU01:FS00:0x15:- Writing 384958 bytes of core data to disk...
23:29:41:WU01:FS00:0x15:Done: 384446 -> 353227 (compressed to 91.8 percent)
23:29:41:WU01:FS00:0x15:  ... Done.
23:29:41:WU01:FS00:0x15:DeleteFrameFiles: successfully deleted file=01/wudata_01.ckp
23:29:41:WU01:FS00:0x15:Shutting down core 
23:29:41:WU01:FS00:0x15:
23:29:41:WU01:FS00:0x15:Folding@home Core Shutdown: FINISHED_UNIT
23:29:41:WU01:FS00:FahCore returned: FINISHED_UNIT (100 = 0x64)
23:29:41:WU01:FS00:Sending unit results: id:01 state:SEND error:OK project:8057 run:0 clone:491 gen:7 core:0x15 unit:0x000000086953ee30508b38c070b328c3
23:29:41:WU01:FS00:Uploading 345.45KiB to 171.67.108.144
23:29:41:WU01:FS00:Connecting to 171.67.108.144:8080
23:29:41:WU02:FS00:Starting
23:29:41:WU02:FS00:Running FahCore: "C:\Program Files (x86)\FAHClient/FAHCoreWrapper.exe" C:/Users/kendrak/AppData/Roaming/FAHClient/cores/www.stanford.edu/~pande/Win32/AMD64/NVIDIA/Fermi/beta/Core_15.fah/FahCore_15.exe -dir 02 -suffix 01 -version 701 -lifeline 4480 -checkpoint 15 -gpu 0
23:29:41:WU02:FS00:Started FahCore on PID 11756
23:29:41:WU02:FS00:Core PID:7408
23:29:41:WU02:FS00:FahCore 0x15 started
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:*------------------------------*
23:29:42:WU02:FS00:0x15:Folding@Home GPU Core
23:29:42:WU02:FS00:0x15:Version                2.25 (Wed May 9 17:03:01 EDT 2012)
23:29:42:WU02:FS00:0x15:Build host             AmoebaRemote
23:29:42:WU02:FS00:0x15:Board Type             NVIDIA/CUDA
23:29:42:WU02:FS00:0x15:Core                   15
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:Window's signal control handler registered.
23:29:42:WU02:FS00:0x15:Preparing to commence simulation
23:29:42:WU02:FS00:0x15:- Looking at optimizations...
23:29:42:WU02:FS00:0x15:DeleteFrameFiles: successfully deleted file=02/wudata_01.ckp
23:29:42:WU02:FS00:0x15:- Created dyn
23:29:42:WU02:FS00:0x15:- Files status OK
23:29:42:WU02:FS00:0x15:sizeof(CORE_PACKET_HDR) = 512 file=<>
23:29:42:WU02:FS00:0x15:- Expanded 57621 -> 261042 (decompressed 453.0 percent)
23:29:42:WU02:FS00:0x15:Called DecompressByteArray: compressed_data_size=57621 data_size=261042, decompressed_data_size=261042 diff=0
23:29:42:WU02:FS00:0x15:- Digital signature verified
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:Project: 8057 (Run 0, Clone 502, Gen 6)
23:29:42:WU02:FS00:0x15:
23:29:42:WU02:FS00:0x15:Assembly optimizations on if available.
23:29:42:WU02:FS00:0x15:Entering M.D.
23:29:44:WU02:FS00:0x15:Tpr hash 02/wudata_01.tpr:  2252366912 2484876154 920962738 85761893 3389626746
23:29:44:WU02:FS00:0x15:GPU device id=0
23:29:44:WU02:FS00:0x15:Working on Giving Russians Opium May Alter Current Situation
23:29:44:WU02:FS00:0x15:Client config unavailable.
23:29:44:WU02:FS00:0x15:Starting GUI Server
23:29:47:WU01:FS00:Upload 74.11%
23:29:49:WU01:FS00:Upload complete
23:29:49:WU01:FS00:Server responded WORK_ACK (400)
23:29:49:WU01:FS00:Final credit estimate, 18229.00 points
23:29:49:WU01:FS00:Cleaning up
23:30:45:WU02:FS00:0x15:Setting checkpoint frequency: 500000
23:30:45:WU02:FS00:0x15:Completed         3 out of 50000000 steps (0%).
23:32:48:WU02:FS00:0x15:Completed    500000 out of 50000000 steps (1%).

Am I missing something or is HFM not like these new WU?


Indeed, my hfm web interface says I got 20 failed on one box ATM.
 
Wow.... Laptops just showed up.

I am actually getting 2min 55seconds now, working on my second 8057.

Yeah it's pretty sweet to have roughly the power of a 660Ti in my laptop, plus a fast CPU and a big SSD makes this laptop the first time I actually own a laptop that is fast enough. With 32GB of ram on this sucker, I can pretty much do anything.

With an estimated 65k-70k PPD, this laptop can do pretty much the same PPD as my 4.2Ghz Gultown hexcore, at least with the current points system.

If this points system holds up I will be for sure buying a desktop Kepler card to fold full time.

Lol...is that an optimus enabled unit? I got the quadro k1000 in my laptop...

Yes, it is an optimus enabled machine, which also means I can fold on the GPU and have no UI lag as windows runs on the HD4000. The HD4000 is actually pretty decent, it can play Diablo 3 pretty decent and many other lither games.

Do you know what the power draw is for this system? The PPD/W for high end laptops may be good.

Not exactly, but I do have a kill-a-watt, I can find out. The crazy thing is you can get laptops with two of these GPU's in them...
 
Last edited:
HFM says that I have 4 WU finished and 4 WU that have failed. I don't get it... everything looks like it is fine:
Am I missing something or is HFM not like these new WU?

Note that HFM denotes a new core download as a failed WU. I've mentioned it to Harlam357 already.
 
any benefit to doing this with older GPU .. like the GTX260?

and to follow this up, how about a 9800gtx+?

I don't know if the new style WU will be ported to pre-Fermi. That being said, once the beta phase is over, the benchmark GPU at the moment is a stock GTX460, so it may be that you need to fold on that level of hardware or faster to benefit from GPU QRB. Who knows? Exciting times!
 
So it is failing?

Are you getting points? I think that the message is that HFM misreads these units as fails, but they are actually fine. I really need to get a video card to play with....
 
When i get a 8057 WU i get lots of points and my PC is usable (lags only a bit) when i fold on GPU+SMP 8 (i7 920).
But when i get a different WU unit (other than 8057) the points ofc drops, but my PC also becomes pretty much unusable (extreme lag).
Changing the settings back to non-beta settings doesn't remove the lag.

Any advice?
 
When i get a 8057 WU i get lots of points and my PC is usable (lags only a bit) when i fold on GPU+SMP 8 (i7 920).
But when i get a different WU unit (other than 8057) the points ofc drops, but my PC also becomes pretty much unusable (extreme lag).
Changing the settings back to non-beta settings doesn't remove the lag.

Any advice?

Tis the nature of CPU+GPU folding.

Your CPU is loaded up and busy. Most every web browser out there now leverages GPU acceleration...and your GPU is at 100% load. 8057s are easy WUs with short TPF IIRC, just about every other SMP unit out there is a long slow WU that is harder.
 
Tis the nature of CPU+GPU folding.

Your CPU is loaded up and busy. Most every web browser out there now leverages GPU acceleration...and your GPU is at 100% load. 8057s are easy WUs with short TPF IIRC, just about every other SMP unit out there is a long slow WU that is harder.
But before i made the settings for QRB units i could fold on CPU+GPU and still use the PC (it was a bit laggy but not extremely laggy) on all WUs.
 
But before i made the settings for QRB units i could fold on CPU+GPU and still use the PC (it was a bit laggy but not extremely laggy) on all WUs.

With the most recent batch of GPU WU? (outside of 8057?
 
With the most recent batch of GPU WU? (outside of 8057?
I don't know if they were the most recent batch..
I use 7.2.9

All i know is that after i made the QRB settings my PC lags alot more (not when i have a 8057 WU though)
 
Last edited:
I don't know if they were the most recent batch..
I use 7.2.9

All i know is that after i made the QRB settings my PC lags alot more (not when i have a 8057 WU though)

It is probably because of the new 2.25 FAHCore.

So, I ended up with 4 straight 8057's yesterday, and now am back to 8018's. That was a nice little point boost!
 
so it says one of my 680's will be making 174k ppd and the other 176k ppd. is this correct?? on 8057
 
I'm still shocked to see these numbers

your telling me i should of kept all my 460's i was folding on.

TPF's are 1min 41 seconds. and 1 min 42 sec.

cray thing is my temps are staying at 50c. on my old 580's i would be in the high 80's low 90s
 
I need to know what GTX 660 ti numbers are like ASAP
 
I just got a 8057. I'm on a 4GB GTX 680 overclocked 1175/3105 w/c'ed.

Getting 171k PPD.
 
I need to know what GTX 660 ti numbers are like ASAP

Check out the link to the stats spreadsheet that was posted by Grandpa_01 on the first page of this thread. I think it's post #3 or something. It's got numbers from others running 660Ti's.
 
Last edited:
could this possibly give eVGA back the "edge" to reclaim the top spot?
 
Dear GPU QRB:

I love you. You make my GPUs seriously sexy and have handed me my first 1M point day. I now have raging geek wood.

Love and hugs,
The Beta Team
 
could this possibly give eVGA back the "edge" to reclaim the top spot?

Definitely, which is why we need to jump on these units with everything we have. EVGA has closed the ppd gap to under 1.5M from our peak of ~7M.
 
I thought I was set to get these for the last week, and did not get one.

Yesterday morning, I added the next-unit-percentage tag which I did not already have, and still no love from stanford. Oh well, I'll get some soon enough.

 
I need to know what GTX 660 ti numbers are like ASAP

I can let you know tonight.


We need to start geting things ready. I know there a bunch of GPUs that have been idle (myself included).

Spin 'em up [H]! It is cold outside and the PPD is good!
 
I have had several 8057's on my GTX-460.
The last one I got had a lower ppd at the same TPF, hmmm.

8057WUc.jpg


8057WUlowppdc.jpg
 
I have had several 8057's on my GTX-460.
The last one I got had a lower ppd at the same TPF, hmmm.

8057WUc.jpg


8057WUlowppdc.jpg

They may be making adjustments to the QRB

Or if you had a pause in there it would affect the PPD because of a longer return time.
 
I don't pause anything, this is on a dedicated folder.
I don't even surf the web with it, it has no updates or even AV.
It gave me that low ppd right from the first frame and stayed there.
I'll see when it gets the next one, it is in its routine of giving me 8018's til I get another 8057.
 
The 660ti numbers we have are barely over 100k.

I just bought a GTX 580 for $200. I think I made the right move.
 
Loading up my 470 now.
-
In the v6 gui client, I added -betateam to the "additional client parameters" box in the config screen. Hope that's correct.
 
My 560Ti, at 960MHz core, is running about 1:40 TPF and the PPD is about 174K.
Still getting these only sporadically. Two or three then back to 8018 for one or two.
 
The 660ti numbers we have are barely over 100k.

I just bought a GTX 580 for $200. I think I made the right move.

I agree that for now that is the case. Going forward with keplar support in a new core (whenver that happenes) I think the 660ti cards will fly.
 
I can't believe i dumped my 460's less than two months ago for what felt like peanuts.

I'd love to get something but money is tight right now :(
 
My 560Ti, at 960MHz core, is running about 1:40 TPF and the PPD is about 174K.
Still getting these only sporadically. Two or three then back to 8018 for one or two.
the ppd reports we are hearing are all over the place for these new Wu.

we need more numbers to compare. At this point, from my perspective the numbers are pretty random.
 
Status
Not open for further replies.
Back
Top