Folding Forum

Posted: **Wed May 02, 2012 9:21 am**

Leonardo wrote:Yes, regardless of operating system or Folding platform, regardless of how many Folding clients you run, your original, single passkey is valid. You should never change that.

Right. Great then.

Leonardo wrote:If you plan on gaming, I wouldn't mess with dual booting Windows/Linux. Just run Folding in a VM and keep it going, even when gaming. I would imagine that most "pure" Linux Folding machines are dedicated Folding rigs, Folding exclusively or nearly so. That's the case with my Linux boxes.

But wouldn't folding and gaming at the same time really drop the folding performance?

Posted: **Wed May 02, 2012 12:12 pm**

If you game two hours per day and folding performance is reduced by 50%, you lose 1 hour per day of production. If you turn FAH off while you game you lose 2 hours per day of production.

Posted: **Wed May 02, 2012 2:22 pm**

ChasR wrote:If you game two hours per day and folding performance is reduced by 50%, you lose 1 hour per day of production. If you turn FAH off while you game you lose 2 hours per day of production.

Oh yes. Good point.

Although I don't know how the games will respond to that.

Posted: **Wed May 02, 2012 2:34 pm**

If folding in a VM, be sure to set its ungrabbed priority to idle in order to minimize the effect on games. Also set the VM memory allocation to the lowest amount possible and still avoid swapping, unless you have a lot of ram.

Posted: **Wed May 02, 2012 2:38 pm**

How much memory should I give the VM for SMP folding?

Posted: **Wed May 02, 2012 2:47 pm**

With a full Ubuntu install and the average WU, 800 MB is usually enough. On a machine with 6 or 8 GB of ram, I'd give the VM 1GB, so as not to get caught out by some future WU that uses a lot of ram.

Posted: **Wed May 02, 2012 2:53 pm**

Unfortunately I only have 4GB, but I'm planning to upgrade to 8 in the near future. I'm gonna give it 1GB then. 3GB is a bit tight but I think still manageable. Thanks ChasR

Posted: **Wed May 02, 2012 5:18 pm**

iceman1992 wrote:But wouldn't folding and gaming at the same time really drop the folding performance?

It depends on the game, WU that is currently running and the machine that it is running on, for my games i usually leave folding running - only when i i'm running dawn of war do i shut down f@h

Posted: **Wed May 02, 2012 7:33 pm**

Nathan_P wrote:My observations on identical Hardware:- dual L5640's, 12 gb ram, asus M/B. Windows machine max PPD 49,000 (daily driver with AV software etc), Ubuntu 10.10 machine with kraken (dedicated folder) max PPD 53k. Thats a lot less than 15% .

Three months ago I had Q9400 processor, which worked with VM + Ubuntu 11.04. After a conversation with Grandpa_01, I tried Ubuntu 10.10. With/without BFS Scheduler, PPD was less than 11.04.
With Q9400 the difference in PPD between Win7/Ubuntu 11.04 was about 9-10%.

After the upgrade (see signature last row), this difference was already Wn7/11.04 ~ 15%. My computer is dedicated (+3-4 hours surfing on the Internet).
If any significance - VM ware/Workstation/v8.0.1(1.0 GB RAM for Ubuntu).

Sorry for bad English

Posted: **Wed May 02, 2012 8:30 pm**

I've not seen a difference of more than 1% in the various Ubuntu flavors, when comparing frame times on the same WU, and I have had up to 20 Ubuntu instances of various versions going at one time. If you like the Unity desktop then 11.04 and later will work fine. If you don't like Unity (want a more windows like environment), use 10.10 or 10.04. I see about a 15% improvement with VM/Ubuntu over Win SMP, on the same WU, on any version of Ubuntu back to 9.04.

Posted: **Wed May 02, 2012 11:31 pm**

Here is the result with Win7/64bit.

Code: Select all

--- Opening Log file [May 2 20:52:51 UTC] 


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.34

                          http://folding.stanford.edu

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

Launch directory: D:\_SMP2_FAH
Executable: D:\_SMP2_FAH\FAH6.34-win32-SMP.exe
Arguments: -smp -verbosity 9 -advmethods 

[20:52:51] - Ask before connecting: No
[20:52:51] - User name: artoar_home (Team 32435)
[20:52:51] - User ID: 26FFA9FE11EB3C94
[20:52:51] - Machine ID: 1
[20:52:51] 
[20:52:51] Loaded queue successfully.
[20:52:51] - Preparing to get new work unit...
[20:52:51] Cleaning up work directory
[20:52:51] + Attempting to get work packet
[20:52:51] Passkey found
[20:52:51] - Will indicate memory of 1024 MB
[20:52:51] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 7
[20:52:51] - Connecting to assignment server
[20:52:51] Connecting to http://assign.stanford.edu:8080/
[20:52:51] - Autosending finished units... [May 2 20:52:51 UTC]
[20:52:51] Trying to send all finished work units
[20:52:51] + No unsent completed units remaining.
[20:52:51] - Autosend completed
[20:52:53] Posted data.
[20:52:53] Initial: 8F80; - Successful: assigned to (128.143.199.97).
[20:52:53] + News From Folding@Home: Welcome to Folding@Home
[20:52:53] Loaded queue successfully.
[20:52:53] Sent data
[20:52:53] Connecting to http://128.143.199.97:8080/
[20:52:54] Posted data.
[20:52:54] Initial: 0000; - Receiving payload (expected size: 1255124)
[20:52:57] - Downloaded at ~408 kB/s
[20:52:57] - Averaged speed for that direction ~474 kB/s
[20:52:57] + Received work.
[20:52:57] + Closed connections
[20:52:57] 
[20:52:57] + Processing work unit
[20:52:57] Core required: FahCore_a3.exe
[20:52:57] Core found.
[20:52:57] Working on queue slot 06 [May 2 20:52:57 UTC]
[20:52:57] + Working ...
[20:52:57] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 06 -np 4 -checkpoint 9 -verbose -lifeline 3432 -version 634'

[20:52:57] 
[20:52:57] *------------------------------*
[20:52:57] Folding@Home Gromacs SMP Core
[20:52:57] Version 2.27 (Dec. 15, 2010)
[20:52:57] 
[20:52:57] Preparing to commence simulation
[20:52:57] - Looking at optimizations...
[20:52:57] - Created dyn
[20:52:57] - Files status OK
[20:52:59] - Expanded 1254612 -> 2077012 (decompressed 165.5 percent)
[20:52:59] Called DecompressByteArray: compressed_data_size=1254612 data_size=2077012, decompressed_data_size=2077012 diff=0
[20:52:59] - Digital signature verified
[20:52:59] 
[20:52:59] Project: 7501 (Run 0, Clone 89, Gen 397)
[20:52:59] 
[20:53:01] Assembly optimizations on if available.
[20:53:01] Entering M.D.
[20:53:07] Mapping NT from 4 to 4 
[20:53:11] Completed 0 out of 500000 steps  (0%)
[20:56:09] Completed 5000 out of 500000 steps  (1%)
[20:59:00] Completed 10000 out of 500000 steps  (2%)
[21:01:51] Completed 15000 out of 500000 steps  (3%)
[21:04:43] Completed 20000 out of 500000 steps  (4%)
[21:07:34] Completed 25000 out of 500000 steps  (5%)
[21:10:26] Completed 30000 out of 500000 steps  (6%)
[21:13:17] Completed 35000 out of 500000 steps  (7%)
[21:16:09] Completed 40000 out of 500000 steps  (8%)
[21:19:00] Completed 45000 out of 500000 steps  (9%)
[21:21:52] Completed 50000 out of 500000 steps  (10%)
[21:22:57] Killing all core threads
[21:22:57] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown at user request.
[21:22:57] ***** Got a SIGTERM signal (2)
[21:22:57] Killing all core threads
[21:22:57] Could not get process id information.  Please kill core process manually

Folding@Home Client Shutdown.

Here is the result with VM / Ubuntu 11.04.

Code: Select all

[22:18:23] Trying to send all finished work units
[22:18:23] + No unsent completed units remaining.
[22:18:23] - Preparing to get new work unit...
[22:18:23] Cleaning up work directory
[22:18:24] + Attempting to get work packet
[22:18:24] Passkey found
[22:18:24] - Will indicate memory of 1024 MB
[22:18:24] - Connecting to assignment server
[22:18:24] Connecting to http://assign.stanford.edu:8080/
[22:18:26] Posted data.
[22:18:26] Initial: 8F80; - Successful: assigned to (128.143.199.97).
[22:18:26] + News From Folding@Home: Welcome to Folding@Home
[22:18:26] Loaded queue successfully.
[22:18:26] Sent data
[22:18:26] Connecting to http://128.143.199.97:8080/
[22:18:27] Posted data.
[22:18:27] Initial: 0000; - Receiving payload (expected size: 1253778)
[22:18:32] - Downloaded at ~244 kB/s
[22:18:32] - Averaged speed for that direction ~260 kB/s
[22:18:32] + Received work.
[22:18:32] Trying to send all finished work units
[22:18:32] + No unsent completed units remaining.
[22:18:32] + Closed connections
[22:18:32] 
[22:18:32] + Processing work unit
[22:18:32] Core required: FahCore_a3.exe
[22:18:32] Core found.
[22:18:32] Working on queue slot 07 [May 2 22:18:32 UTC]
[22:18:32] + Working ...
[22:18:32] - Calling './FahCore_a3.exe -dir work/ -nice 19 -suffix 07 -np 4 -checkpoint 6 -verbose -lifeline 1959 -version 634'

[22:18:32] 
[22:18:32] *------------------------------*
[22:18:32] Folding@Home Gromacs SMP Core
[22:18:32] Version 2.27 (Dec. 15, 2010)
[22:18:32] 
[22:18:32] Preparing to commence simulation
[22:18:32] - Looking at optimizations...
[22:18:32] - Created dyn
[22:18:32] - Files status OK
[22:18:32] - Expanded 1253266 -> 2077012 (decompressed 165.7 percent)
[22:18:32] Called DecompressByteArray: compressed_data_size=1253266 data_size=2077012, decompressed_data_size=2077012 diff=0
[22:18:32] - Digital signature verified
[22:18:32] 
[22:18:32] Project: 7501 (Run 0, Clone 316, Gen 217)
[22:18:32] 
[22:18:32] Assembly optimizations on if available.
[22:18:32] Entering M.D.
                         :-)  G  R  O  M  A  C  S  (-:

                   Groningen Machine for Chemical Simulation

                            :-)  VERSION 4.5.3  (-:

        Written by Emile Apol, Rossen Apostolov, Herman J.C. Berendsen,
      Aldert van Buuren, Pär Bjelkmar, Rudi van Drunen, Anton Feenstra, 
        Gerrit Groenhof, Peter Kasson, Per Larsson, Pieter Meulenhoff, 
           Teemu Murtola, Szilard Pall, Sander Pronk, Roland Schulz, 
                Michael Shirts, Alfons Sijbers, Peter Tieleman,

               Berk Hess, David van der Spoel, and Erik Lindahl.

       Copyright (c) 1991-2000, University of Groningen, The Netherlands.
            Copyright (c) 2001-2010, The GROMACS development team at
        Uppsala University & The Royal Institute of Technology, Sweden.
            check out http://www.gromacs.org for more information.


                               :-)  Gromacs  (-:

[22:18:38] Mapping NT from 4 to 4 
Reading file work/wudata_07.tpr, VERSION 4.5.3-dev-20101129-58a6b (single precision)
Starting 4 threads
Making 1D domain decomposition 4 x 1 x 1
starting mdrun 'BETA-LACTAMASE NDM-1 in water'
109000000 steps, 436000.0 ps (continuing from step 108500000, 434000.0 ps).

NOTE: Turning on dynamic load balancing

[22:18:38] Completed 0 out of 500000 steps  (0%)
[22:21:03] Completed 5000 out of 500000 steps  (1%)
[22:23:28] Completed 10000 out of 500000 steps  (2%)
[22:25:52] Completed 15000 out of 500000 steps  (3%)
[22:28:16] Completed 20000 out of 500000 steps  (4%)
[22:30:42] Completed 25000 out of 500000 steps  (5%)
[22:33:07] Completed 30000 out of 500000 steps  (6%)
[22:35:33] Completed 35000 out of 500000 steps  (7%)
[22:38:00] Completed 40000 out of 500000 steps  (8%)
[22:40:24] Completed 45000 out of 500000 steps  (9%)
[22:42:48] Completed 50000 out of 500000 steps  (10%)

The same project - 7501.

For Win7 average TPF - 172 sec.
For Ubuntu average TPF - 145 sec. (172/145 ~18%)

According to SMP bonus point calculator:

Win7 - TPF - 2:52 min - 18 429 PPD
Ubuntu-TPF- 2:25 min - 23 800 PPD (PPD percentage difference - 29%)

I have no explanation why it is so much difference to me.
In Win7 I only have MS Office and several small programs installed. AV is MS Security Essentials.

Posted: **Thu May 03, 2012 1:29 am**

Nathan_P wrote:It depends on the game, WU that is currently running and the machine that it is running on, for my games i usually leave folding running - only when i i'm running dawn of war do i shut down f@h

I see. OK then I'll try to leave it on, as long as my games don't lag I don't mind. I don't game that much anyway lately.

artoar_11 wrote:After the upgrade (see signature last row), this difference was already Wn7/11.04 ~ 15%. My computer is dedicated (+3-4 hours surfing on the Internet).
If any significance - VM ware/Workstation/v8.0.1(1.0 GB RAM for Ubuntu).

That sounds great, so 11.04 makes more than 10.10 in your case. Have you tried 11.10 or 12.04?

ChasR wrote:If you don't like Unity (want a more windows like environment), use 10.10 or 10.04. I see about a 15% improvement with VM/Ubuntu over Win SMP, on the same WU, on any version of Ubuntu back to 9.04.

Yes I'm a fan of unity

I already have ubuntu 12.04 running on my laptop (I don't use it for folding, at least for now), and seeing the improvements you guys reported (although some report less) I'm gonna install it on my desktop.

artoar_11 wrote:According to SMP bonus point calculator:

Win7 - TPF - 2:52 min - 18 429 PPD
Ubuntu-TPF- 2:25 min - 23 800 PPD (PPD percentage difference - 29%)

I have no explanation why it is so much difference to me.
In Win7 I only have MS Office and several small programs installed. AV is MS Security Essentials.

Wow awesome

. 29% is a huge difference. I can't wait to try it myself when I have time. Hopefully I'll see such a difference.

Thank you guys for taking the time to help me on this. Really really appreciate it.

Posted: **Thu May 03, 2012 4:26 am**

iceman1992 wrote:
Nathan_P wrote:It depends on the game, WU that is currently running and the machine that it is running on, for my games i usually leave folding running - only when i i'm running dawn of war do i shut down f@h
I see. OK then I'll try to leave it on, as long as my games don't lag I don't mind. I don't game that much anyway lately.

Some games may lag noticeably and others probably won't since FAH is going to be consuming only CPU resources which run at a very low priority. (Hopefully your VM will be set up to accommodate this properly.) If it lags enough to bother you, you can probably give FAH a smaller enough fraction of the CPUs that any lag becomes unnoticeable -- and that still may leave something for FAH.

Posted: **Thu May 03, 2012 6:26 pm**

bruce wrote: games may lag noticeably and others probably won't since FAH is going to be consuming only CPU resources which run at a very low priority. (Hopefully your VM will be set up to accommodate this properly.) If it lags enough to bother you, you can probably give FAH a smaller enough fraction of the CPUs that any lag becomes unnoticeable -- and that still may leave something for FAH.

I'll do that.

Right now I'm booting ubuntu and I can't even download the linux client. I go to folding.stanford.edu, and the "other platforms" link doesn't do anything.

And the main link isn't even there. Confused

Posted: **Thu May 03, 2012 6:36 pm**

iceman1992 wrote:
bruce wrote: games may lag noticeably and others probably won't since FAH is going to be consuming only CPU resources which run at a very low priority. (Hopefully your VM will be set up to accommodate this properly.) If it lags enough to bother you, you can probably give FAH a smaller enough fraction of the CPUs that any lag becomes unnoticeable -- and that still may leave something for FAH.
I'll do that. Right now I'm booting ubuntu and I can't even download the linux client. I go to folding.stanford.edu, and the "other platforms" link doesn't do anything.

And the main link isn't even there. Confused

Works on my computer. Google Chrome with AdBlocker running on Windows 7 64-bit SP1. I think you may have Javascript disabled or something like that. They use some fancy scripting for that which doesn't seem to show up on your machine for some reason but works for me. Hmm.

64-bit Debian / Mint / Ubuntu:
https://fah-web.stanford.edu/file-relea ... _amd64.deb
https://fah-web.stanford.edu/file-relea ... -1_all.deb
https://fah-web.stanford.edu/file-relea ... _amd64.deb

32-bit Debian / Mint / Ubuntu:
https://fah-web.stanford.edu/file-relea ... 2_i386.deb
https://fah-web.stanford.edu/file-relea ... -1_all.deb
https://fah-web.stanford.edu/file-relea ... 2_i386.deb

See also http://folding.stanford.edu/English/LinGuide

Folding Forum

SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?

Re: SMP on Win7 or Linux?