Just to verify...

[hiigaran]

bloody hell, already up to three threads...hopefully i dont make a habit out of this...anyway, quick question:

how do i know i have the largest work units options set (the bigadv, or whatever its called)? i think i have it set up, but im not sure if i am missing anything.

[PantherX]

Since you haven't listed your system specifications in the post nor in your profile, that would be hard. The easiest method is to post the FAHlog here in the Forum within the Code Tags (in Full Editor mode) along with the hardware and we can help you.

[hiigaran]

GTX 590, i5 2500K @4.7 Ghz

Code: Select all

--- Opening Log file [December 4 11:19:57 UTC] 


# Windows GPU Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.41r2

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\FAH\GPU
Executable: C:\Program Files (x86)\FAH\GPU\Folding@home-Win32-gpu.exe
Arguments: -gpu 1 

[11:19:57] - Ask before connecting: No
[11:19:57] - User name: hiigaran (Team 212997)
[11:19:57] - User ID: 59CA1B3357073298
[11:19:57] - Machine ID: 2
[11:19:57] 
[11:19:57] Gpu type=3 species=20.
[11:19:57] Loaded queue successfully.
[11:19:57] 
[11:19:57] + Processing work unit
[11:19:57] Core required: FahCore_15.exe
[11:19:57] Core found.
[11:19:57] Working on queue slot 06 [December 4 11:19:57 UTC]
[11:19:57] + Working ...
[11:19:57] 
[11:19:57] *------------------------------*
[11:19:57] Folding@Home GPU Core
[11:19:57] Version                2.20 (Tue Aug 2 12:06:37 PDT 2011)
[11:19:57] Build host             SimbiosNvdWin7
[11:19:57] Board Type             NVIDIA/CUDA
[11:19:57] Core                   15
[11:19:57] 
[11:19:57] Window's signal control handler registered.
[11:19:57] Preparing to commence simulation
[11:19:57] - Ensuring status. Please wait.
[11:20:07] - Looking at optimizations...
[11:20:07] - Working with standard loops on this execution.
[11:20:07] - Previous termination of core was improper.
[11:20:07] - Going to use standard loops.
[11:20:07] - Files status OK
[11:20:07] sizeof(CORE_PACKET_HDR) = 512 file=<>
[11:20:07] - Expanded 124328 -> 501826 (decompressed 403.6 percent)
[11:20:07] Called DecompressByteArray: compressed_data_size=124328 data_size=501826, decompressed_data_size=501826 diff=0
[11:20:07] - Digital signature verified
[11:20:07] 
[11:20:07] Project: 7620 (Run 40, Clone 0, Gen 106)
[11:20:07] 
[11:20:07] Entering M.D.
[11:20:09] Will resume from checkpoint file work/wudata_06.ckp
[11:20:09] Tpr hash work/wudata_06.tpr:  1018710238 776515362 3795911217 1702810041 3231553119
[11:20:09] calling fah_main gpuDeviceId=0
[11:20:12] Working on Protein
[11:20:12] Client config found, loading data.
[11:20:12] Starting GUI Server
[11:21:15] Resuming from checkpoint
[11:21:15] fcCheckPointResume: retreived and current tpr file hash:
[11:21:15]    0   1018710238   1018710238
[11:21:15]    1    776515362    776515362
[11:21:15]    2   3795911217   3795911217
[11:21:15]    3   1702810041   1702810041
[11:21:15]    4   3231553119   3231553119
[11:21:15] fcCheckPointResume: file hashes same.
[11:21:15] fcCheckPointResume: state restored.
[11:21:15] fcCheckPointResume: name work/wudata_06.log Verified work/wudata_06.log
[11:21:15] fcCheckPointResume: name work/wudata_06.trr Verified work/wudata_06.trr
[11:21:15] fcCheckPointResume: name work/wudata_06.xtc Verified work/wudata_06.xtc
[11:21:15] fcCheckPointResume: name work/wudata_06.edr Verified work/wudata_06.edr
[11:21:15] fcCheckPointResume: state restored 2
[11:21:15] Resumed from checkpoint
[11:21:15] Setting checkpoint frequency: 400000
[11:21:15] Completed  31600001 out of 40000000 steps (79%).
[11:26:05] Completed  32000000 out of 40000000 steps (80%).
[11:31:14] Completed  32400000 out of 40000000 steps (81%).
[11:36:32] Completed  32800000 out of 40000000 steps (82%).
[11:41:55] Completed  33200000 out of 40000000 steps (83%).
[11:47:23] Completed  33600000 out of 40000000 steps (84%).
[11:52:54] Completed  34000000 out of 40000000 steps (85%).
[11:58:26] Completed  34400000 out of 40000000 steps (86%).
[12:03:58] Completed  34800000 out of 40000000 steps (87%).
[12:09:32] Completed  35200000 out of 40000000 steps (88%).
[12:15:06] Completed  35600000 out of 40000000 steps (89%).
[12:20:40] Completed  36000000 out of 40000000 steps (90%).
[12:26:17] Completed  36400000 out of 40000000 steps (91%).
[12:31:52] Completed  36800000 out of 40000000 steps (92%).
[12:37:28] Completed  37200000 out of 40000000 steps (93%).
[12:43:04] Completed  37600000 out of 40000000 steps (94%).
[12:48:40] Completed  38000000 out of 40000000 steps (95%).
[12:54:17] Completed  38400000 out of 40000000 steps (96%).
[12:59:54] Completed  38800000 out of 40000000 steps (97%).
[13:05:31] Completed  39200000 out of 40000000 steps (98%).
[13:11:10] Completed  39600000 out of 40000000 steps (99%).
[13:16:49] Completed  40000000 out of 40000000 steps (100%).
[13:16:50] Finished fah_main status=0
[13:16:50] Successful run
[13:16:50] DynamicWrapper: Finished Work Unit: sleep=10000
[13:17:00] Reserved 759396 bytes for xtc file; Cosm status=0
[13:17:00] Allocated 759396 bytes for xtc file
[13:17:00] - Reading up to 759396 from "work/wudata_06.xtc": Read 759396
[13:17:00] Read 759396 bytes from xtc file; available packet space=785671068
[13:17:00] xtc file hash check passed.
[13:17:00] Reserved 47688 47688 785671068 bytes for arc file=<work/wudata_06.trr> Cosm status=0
[13:17:00] Allocated 47688 bytes for arc file
[13:17:00] - Reading up to 47688 from "work/wudata_06.trr": Read 47688
[13:17:00] Read 47688 bytes from arc file; available packet space=785623380
[13:17:00] trr file hash check passed.
[13:17:00] Allocated 544 bytes for edr file
[13:17:00] Read bedfile
[13:17:00] edr file hash check passed.
[13:17:00] Allocated 37087 bytes for logfile
[13:17:00] Read logfile
[13:17:00] GuardedRun: success in DynamicWrapper
[13:17:00] GuardedRun: done
[13:17:00] Run: GuardedRun completed.
[13:17:03] + Opened results file
[13:17:03] - Writing 845227 bytes of core data to disk...
[13:17:03] Done: 844715 -> 816491 (compressed to 96.6 percent)
[13:17:03]   ... Done.
[13:17:03] DeleteFrameFiles: successfully deleted file=work/wudata_06.ckp
[13:17:03] Shutting down core 
[13:17:03] 
[13:17:03] Folding@home Core Shutdown: FINISHED_UNIT
[13:17:06] CoreStatus = 64 (100)
[13:17:06] Sending work to server
[13:17:06] Project: 7620 (Run 40, Clone 0, Gen 106)
[13:17:06] - Read packet limit of 540015616... Set to 524286976.


[13:17:06] + Attempting to send results [December 4 13:17:06 UTC]
[13:17:06] Gpu type=3 species=20.
[13:17:09] + Results successfully sent
[13:17:09] Thank you for your contribution to Folding@Home.
[13:17:09] + Number of Units Completed: 20

[13:17:13] - Preparing to get new work unit...
[13:17:13] Cleaning up work directory
[13:17:13] + Attempting to get work packet
[13:17:13] Passkey found
[13:17:13] Gpu type=3 species=20.
[13:17:13] - Connecting to assignment server
[13:17:14] - Successful: assigned to (171.64.65.105).
[13:17:14] + News From Folding@Home: Welcome to Folding@Home
[13:17:14] Loaded queue successfully.
[13:17:14] Gpu type=3 species=20.
[13:17:16] + Closed connections
[13:17:16] 
[13:17:16] + Processing work unit
[13:17:16] Core required: FahCore_15.exe
[13:17:16] Core found.
[13:17:16] Working on queue slot 07 [December 4 13:17:16 UTC]
[13:17:16] + Working ...
[13:17:16] 
[13:17:16] *------------------------------*
[13:17:16] Folding@Home GPU Core
[13:17:16] Version                2.20 (Tue Aug 2 12:06:37 PDT 2011)
[13:17:16] Build host             SimbiosNvdWin7
[13:17:16] Board Type             NVIDIA/CUDA
[13:17:16] Core                   15
[13:17:16] 
[13:17:16] Window's signal control handler registered.
[13:17:16] Preparing to commence simulation
[13:17:16] - Looking at optimizations...
[13:17:16] DeleteFrameFiles: successfully deleted file=work/wudata_07.ckp
[13:17:16] - Created dyn
[13:17:16] - Files status OK
[13:17:16] sizeof(CORE_PACKET_HDR) = 512 file=<>
[13:17:16] - Expanded 124709 -> 501826 (decompressed 402.3 percent)
[13:17:16] Called DecompressByteArray: compressed_data_size=124709 data_size=501826, decompressed_data_size=501826 diff=0
[13:17:16] - Digital signature verified
[13:17:16] 
[13:17:16] Project: 7621 (Run 281, Clone 0, Gen 122)
[13:17:16] 
[13:17:16] Assembly optimizations on if available.
[13:17:16] Entering M.D.
[13:17:18] Tpr hash work/wudata_07.tpr:  640666086 310993874 676074779 1645945204 376350
[13:17:18] calling fah_main gpuDeviceId=0
[13:17:18] Working on Protein
[13:17:18] Client config found, loading data.
[13:17:19] Starting GUI Server
[13:18:20] Setting checkpoint frequency: 400000
[13:18:20] Completed         3 out of 40000000 steps (0%).
[13:23:55] Completed    400000 out of 40000000 steps (1%).
[13:29:31] Completed    800000 out of 40000000 steps (2%).
[13:35:07] Completed   1200000 out of 40000000 steps (3%).
[13:40:47] Completed   1600000 out of 40000000 steps (4%).
[13:46:23] Completed   2000000 out of 40000000 steps (5%).
[13:52:00] Completed   2400000 out of 40000000 steps (6%).
[13:57:35] Completed   2800000 out of 40000000 steps (7%).
[14:03:12] Completed   3200000 out of 40000000 steps (8%).
[14:08:48] Completed   3600000 out of 40000000 steps (9%).
[14:14:24] Completed   4000000 out of 40000000 steps (10%).
[14:20:01] Completed   4400000 out of 40000000 steps (11%).
[14:25:37] Completed   4800000 out of 40000000 steps (12%).
[14:31:13] Completed   5200000 out of 40000000 steps (13%).
[14:36:49] Completed   5600000 out of 40000000 steps (14%).
[14:42:25] Completed   6000000 out of 40000000 steps (15%).
[14:48:03] Completed   6400000 out of 40000000 steps (16%).
[14:53:39] Completed   6800000 out of 40000000 steps (17%).
[14:59:17] Completed   7200000 out of 40000000 steps (18%).
[15:04:52] Completed   7600000 out of 40000000 steps (19%).
[15:10:28] Completed   8000000 out of 40000000 steps (20%).
[15:16:03] Completed   8400000 out of 40000000 steps (21%).
[15:21:41] Completed   8800000 out of 40000000 steps (22%).
[15:27:19] Completed   9200000 out of 40000000 steps (23%).
[15:32:56] Completed   9600000 out of 40000000 steps (24%).
[15:38:35] Completed  10000000 out of 40000000 steps (25%).
[15:44:14] Completed  10400000 out of 40000000 steps (26%).
[15:49:53] Completed  10800000 out of 40000000 steps (27%).
[15:55:32] Completed  11200000 out of 40000000 steps (28%).
[16:01:13] Completed  11600000 out of 40000000 steps (29%).
[16:06:42] Completed  12000000 out of 40000000 steps (30%).
[16:11:58] Completed  12400000 out of 40000000 steps (31%).
[16:17:37] Completed  12800000 out of 40000000 steps (32%).
[16:23:16] Completed  13200000 out of 40000000 steps (33%).
[16:28:56] Completed  13600000 out of 40000000 steps (34%).
[16:34:31] Completed  14000000 out of 40000000 steps (35%).
[16:40:05] Completed  14400000 out of 40000000 steps (36%).
[16:45:40] Completed  14800000 out of 40000000 steps (37%).
[16:51:15] Completed  15200000 out of 40000000 steps (38%).
[16:56:52] Completed  15600000 out of 40000000 steps (39%).
[17:02:31] Completed  16000000 out of 40000000 steps (40%).
[17:08:08] Completed  16400000 out of 40000000 steps (41%).
[17:13:44] Completed  16800000 out of 40000000 steps (42%).
[17:19:20] Completed  17200000 out of 40000000 steps (43%).
[17:25:00] Completed  17600000 out of 40000000 steps (44%).
[17:30:35] Completed  18000000 out of 40000000 steps (45%).
[17:36:11] Completed  18400000 out of 40000000 steps (46%).
[17:41:47] Completed  18800000 out of 40000000 steps (47%).
[17:47:27] Completed  19200000 out of 40000000 steps (48%).
[17:53:02] Completed  19600000 out of 40000000 steps (49%).
[17:58:37] Completed  20000000 out of 40000000 steps (50%).
[18:04:12] Completed  20400000 out of 40000000 steps (51%).
[18:09:46] Completed  20800000 out of 40000000 steps (52%).
[18:15:21] Completed  21200000 out of 40000000 steps (53%).
[18:20:56] Completed  21600000 out of 40000000 steps (54%).
[18:26:31] Completed  22000000 out of 40000000 steps (55%).
[18:32:06] Completed  22400000 out of 40000000 steps (56%).
[18:37:40] Completed  22800000 out of 40000000 steps (57%).
[18:43:12] Completed  23200000 out of 40000000 steps (58%).
[18:48:44] Completed  23600000 out of 40000000 steps (59%).
[18:54:19] Completed  24000000 out of 40000000 steps (60%).
[18:59:53] Completed  24400000 out of 40000000 steps (61%).
[19:05:28] Completed  24800000 out of 40000000 steps (62%).
[19:11:03] Completed  25200000 out of 40000000 steps (63%).
[19:16:39] Completed  25600000 out of 40000000 steps (64%).
[19:22:17] Completed  26000000 out of 40000000 steps (65%).
[19:27:52] Completed  26400000 out of 40000000 steps (66%).
[19:33:27] Completed  26800000 out of 40000000 steps (67%).
[19:39:02] Completed  27200000 out of 40000000 steps (68%).
[19:44:38] Completed  27600000 out of 40000000 steps (69%).
[19:50:14] Completed  28000000 out of 40000000 steps (70%).
[19:55:50] Completed  28400000 out of 40000000 steps (71%).
[20:01:26] Completed  28800000 out of 40000000 steps (72%).
[20:07:01] Completed  29200000 out of 40000000 steps (73%).
[20:12:36] Completed  29600000 out of 40000000 steps (74%).
[20:18:11] Completed  30000000 out of 40000000 steps (75%).
[20:23:48] Completed  30400000 out of 40000000 steps (76%).
[20:29:23] Completed  30800000 out of 40000000 steps (77%).
[20:35:00] Completed  31200000 out of 40000000 steps (78%).
[20:40:40] Completed  31600000 out of 40000000 steps (79%).
[20:46:14] Completed  32000000 out of 40000000 steps (80%).
[20:51:49] Completed  32400000 out of 40000000 steps (81%).
[20:57:24] Completed  32800000 out of 40000000 steps (82%).
[21:02:34] Completed  33200000 out of 40000000 steps (83%).
[21:07:46] Completed  33600000 out of 40000000 steps (84%).

^GPU client

not sure if they have them for the GPU client. i think i heard somewhere that they dont. but heres my SMP log as well:

EDIT: log too big, and cant post so soon. also, log too big on its own.

all i need to know though, is what flags should be set, and what options i have to have, and that should be it.

[patonb]

You can't do bigadv smp as you have only 4 threads/cores.

Gpus arent there yet, so its just a run and forget ABOUT it.

[hiigaran]

well that sucks. it takes a little more than 10 hours for me to complete a work unit. why would bigadv be locked out for me?

...and im assuming theres no way to force it?

[k1wi]

Because your CPU doesn't have 8 threads or 8 physical cores, it doesn't qualify. Plus, while your machine takes 10 hours to fold a regular -smp work unit, it doesn't guarantee that it'd complete a bigadv with it's deadlines... Come January the requirements increase again (to 16 cores/threads I believe), so your machine will be even less qualified for bigadv... (It's a great gaming chip, but F@H utilises all the threads/CPUs you can throw at it).

BUT there is NOTHING wrong with 'only' folding regular SMP work units... Spitting them out in 10 hours (it'll vary with the work unit) simply means you are helping the project advance fast and the 'premium' of bigadv over regular smp is not particularly large.

[GreyWhiskers]

Quick note. With the Stanford servers not making any bigadv work units available for the last few weeks, my I7 2600K system, which usually completes a p6900 bigadv work unit in 2.1 DAYS, for about 31K ppd, is now folding "regular" SMP work units. There is quite a variety, some of which take over a full day to complete on this machine. There is also quite a variability on the ppd across the set of SMP projects on each individual computer. The ranges I have completed include:

Project ID: 10125
Min. Time / Frame : 00:06:57 - 18,501.6 PPD
Avg. Time / Frame : 00:07:07 - 17,855.5 PPD [~11hours :40 min per WU]

Project ID: 10127
Min. Time / Frame : 00:18:27 - 19,513.9 PPD
Avg. Time / Frame : 00:19:04 - 18,574.9 PPD [~ 1 day 7 hours per WU]

Project ID: 10128
Min. Time / Frame : 00:07:02 - 18,521.1 PPD
Avg. Time / Frame : 00:07:10 - 18,006.6 PPD [~11hours :40 min per WU]

Project ID: 10130
Min. Time / Frame : 00:08:09 - 19,221.8 PPD [~ 14 hours per WU]
Avg. Time / Frame : 00:08:24 - 18,370.1 PPD

Project ID: 10134
Min. Time / Frame : 00:07:28 - 18,282.5 PPD [~ 12 hours 30 min per WU]
Avg. Time / Frame : 00:07:36 - 17,803.5 PPD

Project ID: 7138
Min. Time / Frame : 00:02:31 - 28,780.1 PPD [~4 hours 15 min per WU]
Avg. Time / Frame : 00:02:35 - 27,673.3 PPD

Project ID: 6990
Min. Time / Frame : 00:02:40 - 25,280.3 PPD [~ 4 hours 26 min per WU]
Avg. Time / Frame : 00:02:44 - 24,361.1 PPD]

What's the message - your CPU, configuration, operating system, competing tasks, memory, internet speed, etc. etc. all determine what you can do. When you go into SMP folding, even "regular" folding, you might have some WUs that take over a day to perform, and others that complete in several hours.

Or, Your Mileage May Vary

GW.

[hiigaran]

i see no reason why i couldnt complete such a work unit with my chip barring the lack of threads. even if the hyperthreading doubled the performance (which it clearly doesnt), then it should take 4 days of continuous running to complete...im assuming your processor is not overclocked, so tell me if it is otherwise. in any case, the fact that my CPU is clocked 43% higher than stock should allow me to complete the units considerably faster. granted, not 43% faster, but last i checked, i had 125 GFLOPS, according to BurnTest, which, from what ive seen on other processors, is a massive amount. surely the deadlines arent that short?

Because your CPU doesn't have 8 threads or 8 physical cores, it doesn't qualify. Plus, while your machine takes 10 hours to fold a regular -smp work unit, it doesn't guarantee that it'd complete a bigadv with it's deadlines... Come January the requirements increase again (to 16 cores/threads I believe), so your machine will be even less qualified for bigadv... (It's a great gaming chip, but F@H utilises all the threads/CPUs you can throw at it).

BUT there is NOTHING wrong with 'only' folding regular SMP work units... Spitting them out in 10 hours (it'll vary with the work unit) simply means you are helping the project advance fast and the 'premium' of bigadv over regular smp is not particularly large.

im not fussed about the points, but arent larger work units more efficiently completed?

[Joe_H]

Yes, the deadlines for the bigadv WU's are that short. One has a preferred deadline of 4 days, others just 5 days. WU's not completed within that are reassigned, and you would only get base credit if turned in by the final deadline after passing the preferred deadline. And early next year the deadlines will be reduced even more, that was announced just about a month ago. They also announced a 16 core minimum, though the way the folding client tests for that means 8 cores with HT will get work too.

Basically people have been gently telling you that your machine while very fast, is not configured for doing bigadv folding. These WU's need many threads, not just brute force calculations, to be processed efficiently. The experimental bigadv folding was aimed towards server class machines with multiple CPU's and thus capable of doing many threads simultaneously. It is just that recent 4 and 6 core i7's with HT were "just good enough" to do them within the deadlines when OC'd enough.

In any case this is moot. At this time there are few if any bigadv WU's to process. There was a problem with the servers last month and they are still working on clearing that up.

[hiigaran]

alrighty, well no worries then. thanks.

[GreyWhiskers]

Another of my points (maybe too subtle) was that when we all start folding, we need to just let our systems loose on whatever WUs that Stanford throws at us. THEN, we look at some stats to see what our particular setup is getting.

BTW, my i7 2600K that ran the projects summarized above, is clocked at about 4.1 GHz, with its 8 GB of DDR-3 1600 memory. For the SMP projects, the i5's can get quite decent ppd counts - but you need to see how your system performs. the i5 has four floating point units like the i7, but doesn't have the HT threads.

Since you are running v6 rather than v7, you can use HFM.net (see the tools link at the top of the page) to give you a couple of summary status looks at your folding history. You set it up, and it in the background captures stats from each completed and "Early Unit End" work unit run. Over time, it will compile a quite useful history of the work you've done.

I used the "Benchmark Viewer" to give averaged TPF and ppd for up to the last three runs of a given project. Then, I use the Work Unit History viewer. It keeps a record of each completed WU on your local machine, and on networked machines you link in. one of the installations I have has over 2000 line items. I can capture the data through the DatAdmin Sqlite viewer, to export to Excel for more detailed analysis.

MtM is working on a very useful v7 version which is in testing now which will have many of the same features, along with some of its own.

So - for me, an important part of folding is keeping on top of how well your machines work across the wide variety of projects that Stanford assigns you. Any given set of numbers are representative of how your system is performing - that may lead you to looking at interferences, clocking, etc.

[hiigaran]

ahh, i use FAHMon. its basic, but its really all that i need. although, i dont know why, but it reports my SMP units with one amount of point values, and then in another area, the same unit has a considerably less value. for instance, the one im currently working on is reported as 1636 points, but then on the side, it shows it is 334 points. which one is it? meh, no biggie.

[Amaruk]

...for instance, the one im currently working on is reported as 1636 points, but then on the side, it shows it is 334 points. which one is it? meh, no biggie.

Both

It looks like you're working on Project 11060. It's worth 334 points (Credit) and 1,636 is your points per day (PPD) as calculated by FahMon. My guess is that PPD does not including the bonus.

[GreyWhiskers]

Folding@home Bonus Point Calculator is useful for you to see how the bonus is calculated. Drop down list selects the project - then you enter an average TPF and an estimated upload time (for some large work units, that can amount to several minutes, even on a fast internet connection). Then the calculated bonus points are shown.

GW

[hiigaran]

...for instance, the one im currently working on is reported as 1636 points, but then on the side, it shows it is 334 points. which one is it? meh, no biggie.

Both

It looks like you're working on Project 11060. It's worth 334 points (Credit) and 1,636 is your points per day (PPD) as calculated by FahMon. My guess is that PPD does not including the bonus.

actually, the PPD shows a tad less than 10000, for the SMP client.

so if ive understood this correctly, i get 1636 points for completing the work unit, plus a 334 point bonus? that might explain why i had a much larger value on it before...