Go Back   Hardware Canucks > HARDWARE CANUCKS COMMUNITY > HardwareCanucks F@H Team

    
Reply
 
LinkBack Thread Tools Display Modes
  #1 (permalink)  
Old June 20, 2010, 08:31 PM
LCB001's Avatar
Folding Captain
 
Join Date: Feb 2008
Location: Aylmer QC.
Posts: 1,774

My System Specs

Thumbs up SMP2 A3 Core Gets Speed Boost With v2.22

Stanford has released a new version (v2.22) of the SMP2 A3 core that some users are getting better ppd with, of course YMMV.

Fah-Addict News has a good explanation of the new cores improvements, the linux version was released first;

Quote:
The SMP2 core is already well-liked amongst most members of the folding community for its ease of use compared to the older a1 and a2 cores. However, it has not been showing its full potential. A new revision, v2.22, has been released for the Linux SMP2 client, which results in speed boosts of around 15%, and contains science fixes and improvements. In short, it works faster and better than its predecessors.

For the curious, we have a technical explanation of where the performance gains have been found. Historically, in the a1 and a2 cores, the unit's processing was separated into two different types of node, typically one thread or process per node (seen in the log file as "Making 1D domain decomposition 2 x 1 x 1"). PP (particle to particle) nodes undertake the calculations governing the interactions between the different atoms being modeled, whilst PME (Particle Mesh Ewald) nodes calculate the electrostatic field the atoms are suspended in, which is affected by every atom in the simulation.

The difficulty arises in deducing the correct division between PP and PME nodes. PP nodes make up the vast majority of the calculations, but the precise number of nodes required per unit depends on the number of atoms in the simulation, and of course the number of cores available for processing. An incorrect or suboptimal balance between PP and PME nodes will result in lower performance as some nodes (cores of your CPU) will not be sufficiently loaded to make full use of the available hardware.

In version 2.22 of the a3 core, this conundrum has been solved by making all nodes capable of performing either PP or PME calculations, automatically adapting as the unit is processed to give the correct number of PP and PME nodes at any given moment in the simulation. Normally this would only offer a small performance increase, but on non-dedicated machines this newfound flexibility should result in a large performance gain as there is much less danger of any one core/node getting bogged down by having to share CPU time with another task: the composition of the nodes will be dynamically updated to allow another core to process that part of the simulation.

To install this new version, delete the current core and the client will download the new version automatically. As always, it is not recommended to perform this update mid-unit, as it is probable that the new core will reject the checkpoint created by the previous core and restart the unit. If you are absolutely certain that your machine will not be switched off before the unit ends, it may be possible to delete the old core without stopping the client. The old core will be held in RAM processing the unit, and the client will then update to the new core version when it starts the next unit.

We welcome reports of actual figures for any performance gains.
Today it was released for windows;

Quote:
Windows users can rest comfortably in the knowledge that they have not been forgotten.

Version 2.22 of the SMP2 A3 core - recently released for Linux - is now also available on Windows. This core provides the same improvements and feature set of the Linux version, which was announced few days ago. Hopefully the performance gains seen under Linux will also accompany this Windows version.
For more info the Folding Forums have an interesting thread on the new core going ; Folding Forum • View topic - A3 windows core v2.22 released.

__________________
Folding For Team 54196

Reply With Quote
  #2 (permalink)  
Old June 20, 2010, 09:03 PM
Banned
F@H
 
Join Date: Sep 2009
Location: Montreal, QC
Posts: 5,415

My System Specs

Default

Very nice. Hopefully my clients will grab it automatically once they're done, because I'd rather not have to delete a unit.
Reply With Quote
  #3 (permalink)  
Old June 20, 2010, 09:07 PM
chrisk's Avatar
Folding Captain
 
Join Date: Jul 2008
Location: GTA, Ontario
Posts: 7,401

My System Specs

Default

I was still on 2.19. I added the oneunit flag to let the current WU finish properly, then I deleted the core, queue, and made sure the work folder was empty. Downloaded 2.22 automatically upon next start, and my PPd seems to be about 10% better. ~13.5K PPD on i7 920 3.6GHZ. (P6014)
__________________
Fold for team #54196
Reply With Quote
  #4 (permalink)  
Old June 20, 2010, 09:29 PM
LCB001's Avatar
Folding Captain
 
Join Date: Feb 2008
Location: Aylmer QC.
Posts: 1,774

My System Specs

Default

I'm hoping the new core will help improve the low PPD from the p2684 WU's, they are getting to be a pain in the ....
__________________
Folding For Team 54196

Reply With Quote
  #5 (permalink)  
Old June 21, 2010, 12:01 AM
Top Prospect
 
Join Date: Apr 2009
Posts: 125

My System Specs

Default

Now I'm going to have to shut mine down so I can add in the oneunit flag. Been getting some work units lately with rather large TPF's, had one 20 minutes and one 12. Too bad they aren't all like the 3 minutes and change unit I had a couple days ago.
Reply With Quote
  #6 (permalink)  
Old June 21, 2010, 07:43 PM
MVP
F@H
 
Join Date: Jan 2010
Posts: 262
Default

When I red that news yesterday I just stopped the windows clients, delete the core and restarted it (on 3 clients) and I did not failed a WU.

Really nice improvement on my 980x in smp2... was arounf 24000 ppd before and now 35000 ppd (still ain't 54000ppd I was having with A2 bigadv but great boost at least)
__________________
Reply With Quote
  #7 (permalink)  
Old June 21, 2010, 09:00 PM
sswilson's Avatar
Moderator
F@H
 
Join Date: Dec 2006
Location: Moncton NB
Posts: 14,496

My System Specs

Default

Quote:
Originally Posted by Charloz24 View Post
When I red that news yesterday I just stopped the windows clients, delete the core and restarted it (on 3 clients) and I did not failed a WU.

Really nice improvement on my 980x in smp2... was arounf 24000 ppd before and now 35000 ppd (still ain't 54000ppd I was having with A2 bigadv but great boost at least)

Same here. Seems like it picked up the new core no problem. I'll have to see what kind of performance I get out of it though.

Update: Still working on the same WUs and my PPD has actually gone down on both 6701 WUs. I'll have to see if maybe the "efficiencies" are applied when the Core first starts, but as it stands I'm not impressed. (Might even go so far as to get the old core out of the recycle bin and re-install it).
__________________
MSI Z87I Gaming AC / i5 4670K / 2X 4G Gskill 1866 DDR3 / XFX XTR 750 / EVGA GTX 680 SC+ 2GB / Intel DC S3700 200G / random 160G Sata HDD
Inwin 904 / Swiftech MCP655-b / Alphacool NexXxos XT45 120 Rad / 2X Scythe GT AP-15 / EK Supreme HF / Dell UltraSharp U2412M

Asrock AM1H-ITX / AM1 Athlon 5350 / 2X4G Gskill PC3-14900 / Intel 6235 Wi-Fi / 90W Targus Power Brick / 320G Seagate Momentus / Mini-Box M350 / 1X 22" Dell IPS / 1X 22" HP

Last edited by sswilson; June 22, 2010 at 04:43 AM.
Reply With Quote
  #8 (permalink)  
Old June 22, 2010, 11:36 AM
bojangles's Avatar
Hall Of Fame
F@H
 
Join Date: Jan 2008
Location: Oakville, ON
Posts: 2,683

My System Specs

Default

LOL why was charli943 banned after one post? Did I miss something here?

Anyway, nice to see some performance get popped out of FAH. I wish I was still doing it...money has to go where it belongs now that I live on my own :(.
Reply With Quote
  #9 (permalink)  
Old June 22, 2010, 01:23 PM
sswilson's Avatar
Moderator
F@H
 
Join Date: Dec 2006
Location: Moncton NB
Posts: 14,496

My System Specs

Default

Quote:
Originally Posted by bojangles View Post
LOL why was charli943 banned after one post? Did I miss something here?

Anyway, nice to see some performance get popped out of FAH. I wish I was still doing it...money has to go where it belongs now that I live on my own :(.
Sig Spammer. If you do a google search for the name you'll see 20+ links all showing it signing up for different forums over the last few days.... ;)
__________________
MSI Z87I Gaming AC / i5 4670K / 2X 4G Gskill 1866 DDR3 / XFX XTR 750 / EVGA GTX 680 SC+ 2GB / Intel DC S3700 200G / random 160G Sata HDD
Inwin 904 / Swiftech MCP655-b / Alphacool NexXxos XT45 120 Rad / 2X Scythe GT AP-15 / EK Supreme HF / Dell UltraSharp U2412M

Asrock AM1H-ITX / AM1 Athlon 5350 / 2X4G Gskill PC3-14900 / Intel 6235 Wi-Fi / 90W Targus Power Brick / 320G Seagate Momentus / Mini-Box M350 / 1X 22" Dell IPS / 1X 22" HP
Reply With Quote
  #10 (permalink)  
Old June 28, 2010, 06:20 AM
cozmosland's Avatar
Top Prospect
 
Join Date: Dec 2009
Posts: 232
Default

hey how can i tell if i have the new core using fahspy?
fahmon doesnt work for me anymore.
is there a tell tale sign since in that program it just says grogpu2 under core
Reply With Quote
Reply


Tags
f@h , folding@home , new core , smp , smp2 , stanford

Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
SMP2 (A3) Core Updated To v2.17 LCB001 HardwareCanucks F@H Team 5 March 22, 2010 10:58 AM
SMP2 client folding A1 core Charloz24 HardwareCanucks F@H Team 18 March 1, 2010 05:34 PM
AMD Implementing "Dynamic Speed Boost" on Upcoming Thuban CPUs FiXT Press Releases & Tech News 23 February 22, 2010 03:44 PM
SMP2 Core Progress Update - Some Will Have Bonus Plan LCB001 HardwareCanucks F@H Team 85 December 15, 2009 01:33 PM
FSB boost or GHZ boost? J2TheROC Overclocking, Tweaking and Benchmarking 14 August 13, 2009 09:33 PM