PPD Database - HFM exports needed

[Sarr]

Yeah, I only fold on linux, but I'd like to contribute to the database, it would be useful to have average performance not JUST on one OS too, folding on linux especially for dedicated 24/7 machines or folding farms isn't uncommon, I imagine.

[Kebast]

Yeah, I only fold on linux, but I'd like to contribute to the database, it would be useful to have average performance not JUST on one OS too, folding on linux especially for dedicated 24/7 machines or folding farms isn't uncommon, I imagine.

If you have a windows machine you can run HFM on that and just connect to all your linux clients. I fold on both windows and linux, and just run HFM on the windows PC but connect to the linux clients there.

[Darth_Peter_dualxeon]

I wrote a quick and dirty bash script that does this: https://github.com/patrick-tavares/foldingathome_tools. Not sure if it reports on everything you want but it may be a start

It does not report anything for me. I'm running fahclient in a terminal window. Somehow the installed service did not like my gpu previously, and I don't care to fix it right now.
but I do have the copy of the text in the terminal window.

I think these are the relevant lines in the log:

Code: Select all

$ cat log.txt | grep "Received Unit"
19:24:37:WU02:FS02:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:16423 run:2589 clone:0 gen:56 core:0xa7 unit:0x0000003c96880e6e5e8a8cf46f1ef611
19:24:37:WU01:FS00:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:14616 run:931 clone:3 gen:43 core:0xa7 unit:0x000000309bf7a4d65e8324c5d22d2ed8
21:38:50:WU03:FS02:Received Unit: id:03 state:DOWNLOAD error:NO_ERROR project:14375 run:1095 clone:1 gen:19 core:0xa7 unit:0x000000142879986c5e88706fbb40e5ac
21:53:46:WU02:FS00:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:14584 run:0 clone:1218 gen:73 core:0xa7 unit:0x000000510d5262775e7a6b36a714872b
22:42:29:WU00:FS01:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:11744 run:0 clone:6524 gen:52 core:0x22 unit:0x0000004b8ca304f15e6bc42e774cf05b
00:09:01:WU01:FS01:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:11759 run:0 clone:10982 gen:16 core:0x22 unit:0x0000001980fccb0a5e6eb55d1f701841
00:54:06:WU00:FS00:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:13850 run:0 clone:3115 gen:45 core:0xa7 unit:0x0000003a287234c95e725912fa137521
01:00:03:WU02:FS02:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:14627 run:575 clone:1 gen:42 core:0xa7 unit:0x0000002b0002894c5e878a236dae8513
02:34:54:WU03:FS01:Received Unit: id:03 state:DOWNLOAD error:NO_ERROR project:11741 run:0 clone:5507 gen:7 core:0x22 unit:0x000000178ca304f15e6bc5a092244c4a
04:29:19:WU01:FS02:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:14623 run:275 clone:2 gen:18 core:0xa7 unit:0x000000149bf7a4d65e893dc4286f856b
04:38:10:WU02:FS00:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:13861 run:0 clone:6602 gen:103 core:0xa7 unit:0x000000710d5a98395e730e9594f64877
04:43:03:WU00:FS01:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:11741 run:0 clone:1538 gen:52 core:0x22 unit:0x000000518ca304f15e67d8166ac24708
06:11:16:WU03:FS00:Received Unit: id:03 state:DOWNLOAD error:NO_ERROR project:13860 run:0 clone:23971 gen:28 core:0xa7 unit:0x0000001d0d5a98395e940a122b55a6bf
07:48:30:WU00:FS02:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:14616 run:459 clone:1 gen:48 core:0xa7 unit:0x000000329bf7a4d65e832558460009f4
08:07:19:WU02:FS01:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:14417 run:0 clone:1325 gen:16 core:0x22 unit:0x000000160d5262775e84c8a54ad37bab
08:09:26:WU01:FS00:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:16406 run:1472 clone:1 gen:68 core:0xa7 unit:0x0000004fa8f5c67d5e801e7ce797cbd8
10:24:09:WU03:FS00:Received Unit: id:03 state:DOWNLOAD error:NO_ERROR project:16411 run:3699 clone:3 gen:51 core:0xa7 unit:0x00000039a8f5c67d5e8771085945583e
11:05:30:WU04:FS02:Received Unit: id:04 state:DOWNLOAD error:NO_ERROR project:16401 run:683 clone:0 gen:66 core:0xa7 unit:0x0000004b96880e6e5e814c63268164d8
11:13:23:WU01:FS01:Received Unit: id:01 state:DOWNLOAD error:NO_ERROR project:14433 run:0 clone:676 gen:16 core:0x22 unit:0x000000140d5262775e8b4d54650c5a0e
12:55:58:WU00:FS00:Received Unit: id:00 state:DOWNLOAD error:NO_ERROR project:14413 run:0 clone:60 gen:46 core:0xa7 unit:0x000000340d5262775e7aeea8924ab0ae
13:55:26:WU02:FS02:Received Unit: id:02 state:DOWNLOAD error:NO_ERROR project:16427 run:0 clone:180 gen:29 core:0xa7 unit:0x0000001fa8f5c67d5e9248e24a1e6856
14:00:11:WU03:FS01:Received Unit: id:03 state:DOWNLOAD error:NO_ERROR project:11743 run:0 clone:1320 gen:7 core:0x22 unit:0x000000108ca304f15e67e11a8911675c


$ cat log.txt | grep credit
21:40:12:WU02:FS02:Final credit estimate, 2239.00 points
21:55:15:WU01:FS00:Final credit estimate, 4370.00 points
00:08:04:WU00:FS01:Final credit estimate, 77196.00 points
00:56:33:WU02:FS00:Final credit estimate, 4497.00 points
01:02:20:WU03:FS02:Final credit estimate, 3579.00 points
02:19:57:WU01:FS01:Final credit estimate, 125380.00 points
04:28:26:WU02:FS02:Final credit estimate, 3562.00 points
04:40:27:WU00:FS00:Final credit estimate, 5779.00 points
04:44:53:WU03:FS01:Final credit estimate, 126797.00 points
06:12:21:WU02:FS00:Final credit estimate, 2176.00 points
06:53:23:WU00:FS01:Final credit estimate, 126547.00 points
07:50:32:WU01:FS02:Final credit estimate, 3625.00 points
07:54:02:WU03:FS00:Final credit estimate, 2452.00 points
10:22:51:WU01:FS00:Final credit estimate, 3158.00 points
11:06:25:WU02:FS01:Final credit estimate, 196793.00 points
11:07:40:WU00:FS02:Final credit estimate, 3800.00 points
12:49:57:WU03:FS00:Final credit estimate, 3606.00 points
13:57:14:WU04:FS02:Final credit estimate, 2899.00 points
14:02:54:WU01:FS01:Final credit estimate, 180240.00 points
15:39:58:WU03:FS01:Final credit estimate, 88506.00 points 

[NoMoreQuarantine]

I wrote a quick and dirty bash script that does this: https://github.com/patrick-tavares/foldingathome_tools. Not sure if it reports on everything you want but it may be a start

Awesome, that might be just what I need. I'll take a look at it later when I'm off work.

[foldinghomealone]

I'll just continue to upload new results every day. I finally got all my cards clocked the same now, so the only difference is windows vs Ubuntu.

You provided already a bit data so I could fill the database. We have a good indication now.
You could also extend uploading periods to weekly.

[foldinghomealone]

7871 results of Core22 from the beginning of the beta phase through to February

That's some awesome folding you have done...

I updated the db and introduces a column active/inactive projects

[Kebast]

I'll just continue to upload new results every day. I finally got all my cards clocked the same now, so the only difference is windows vs Ubuntu.

You provided already a bit data so I could fill the database. We have a good indication now.
You could also extend uploading periods to weekly.

Cool, I'll just upload weekly now then.

[HaloJones]

7871 results of Core22 from the beginning of the beta phase through to February

That's some awesome folding you have done...

I updated the db and introduces a column active/inactive projects

The whole dataset before removing what you didn't want was over 50000 units. You get big datasets when you get to #152 in the total user stats

[foldinghomealone]

The whole dataset before removing what you didn't want was over 50000 units.

It's not really about not wanting it's more about that it doesn't matter if we have 2.000 WUs for the same GPU or 50.000. As the sample size is already very high it won't change the average much.
And it just would make me crazy handling 50.000 entries in google sheets

[Juggy]

The whole dataset before removing what you didn't want was over 50000 units.

It's not really about not wanting it's more about that it doesn't matter if we have 2.000 WUs for the same GPU or 50.000. As the sample size is already very high it won't change the average much.
And it just would make me crazy handling 50.000 entries in google sheets

How many samples would you say concludes a proper dataset?

Also, should we not be collecting device specific data for GPU's including boost clock speeds?

As we all know there are some GPU providers that have better board manufacturing than others as well as better OC firmwares and capabilities.

[HaloJones]

I think to try to have all the possible variables accounted for is too much complication. OS, motherboard, cpu, gpu type, card maker, cooling spec, overclock settings, cooling, usage of machine while folding, Project, Run, Clone, Gen.

All this will do (which is still a lot) is give a range for a card that a potential buyer/user could expect.

[Juggy]

I think to try to have all the possible variables accounted for is too much complication. OS, motherboard, cpu, gpu type, card maker, cooling spec, overclock settings, cooling, usage of machine while folding, Project, Run, Clone, Gen.

All this will do (which is still a lot) is give a range for a card that a potential buyer/user could expect.

Hmmm, for instance an OC card from XXX manufacturer may be $400 but from YYY may be $370.

There are lower priced manufacturers like Galax that still produce good kit.

Card manufacturer (Gigabyte)
Model Number (RTX 2080 Super)
Sub Model Number (Windfrorce OC)
Factory Clock Speed (1830MHz)

Surely that is not too much info?

[Endgame124]

Is there a, uh, more hands off approach? I would like to participate, but at the moment I would prefer to install something and let it do everything else hands off.

[NormalDiffusion]

Is there a, uh, more hands off approach? I would like to participate, but at the moment I would prefer to install something and let it do everything else hands off.

You just need to install and set up hfm.net (5 min) and let it run in the background. Once per week you export the database, rename your slots in the csv file (1 min with excel) and upload the csv. Not much work...

[Kebast]

I think to try to have all the possible variables accounted for is too much complication. OS, motherboard, cpu, gpu type, card maker, cooling spec, overclock settings, cooling, usage of machine while folding, Project, Run, Clone, Gen.

All this will do (which is still a lot) is give a range for a card that a potential buyer/user could expect.

Hmmm, for instance an OC card from XXX manufacturer may be $400 but from YYY may be $370.

There are lower priced manufacturers like Galax that still produce good kit.

Card manufacturer (Gigabyte)
Model Number (RTX 2080 Super)
Sub Model Number (Windfrorce OC)
Factory Clock Speed (1830MHz)

Surely that is not too much info?

Hopefully just given the average and standard deviation, people can conclude performance given various clocks. My cards are a good example. I've got them all running at a modest ~10% overclock, and they perform considerably better than the other GTX970 samples submitted. Even operating system makes a different. The cards in my Ubuntu machine give about 10% more PPD than my Windows machine.
It should all balance out given enough samples. Also, in the hot summer months I'll have to downclock my cards, as they sit in a hot room during the June - August months. So you'll get that under performance too eventually.
I originally had the clock speed in my HFM submittal, but figured that was too much to sort through.