A new member! And wait there is more!

Just wanted to report back. I could only hold out about 10 seconds before I had to put together the system. Everything went very smoothly with not one hiccup (though I haven't tried to get my DOS program to print yet). A whole boatload of thanks to everyone here for the help!

I was a bit bummed that I didn't have much time to play around with the new system once I got it up and running because of prior engagements. I just got the SMP client running this morning, and looking forward to some [H]ard Folding. Thanks again!
 
Don't be a stranger.

Let we come over to bbspot and infest it :D
 
ive got a good socket 775 heatsink i would be willing to throw in to the fray. and possibly either 2 x1gb or 2x2gb sticks of DDR2 if i you dont already have that- if i can find em in my parts pile.
 
ghost6303 said:
ive got a good socket 775 heatsink i would be willing to throw in to the fray. and possibly either 2 x1gb or 2x2gb sticks of DDR2 if i you dont already have that- if i can find em in my parts pile.
Click to expand...

He has his system up and running.

We however, are always looking for donated parts so we can cobble together systems and give them away to fellow folders.

If you want to donate parts to the [H]orde send me a PM.

This is what we where able to do with the last batch of donated parts, and we have another 2 systems in the works.

http://hardforum.com/showthread.php?t=1511675
 
I have the system up and running now, and had a little confusion as to which folding program I should run. I think I'm all set now though.

Right now I have the CPU client running, but it doesn't seem to be cranking out as many units as the GPU client was doing for me. For my set up what would be the most efficient use of folding clients?
 
A good GPU will crank out more points with the client than your standard CPU client. But a lot of us run both on our systems at the same time, many running an SMP client for the CPU and multiple GPU clients as cards in the system. I myself run the console versions of the Windows SMP client for the CPU and two console based GPU clients. That produces about 20-22,000 ppd. :)
 
Is there any specific settings I need to set up to have 2 running at the same time without bogging down my system? I turned the GPU client on and saw CPU usage go to 100%
 
bbspot said:
OK, I think I've got it figured out now.
Click to expand...

Excellent
thumbsup.gif
 
I've gone through the FAQs, and the SMP and the regular CPU folding program, but I think I'm doing something wrong.

I have a Core 2 Duo E8400 and according to HFM.net my PPD for the CPU is 221.6 while the PPD for my GPU is 1685.1. Are these numbers in line with what's expected, or does something seem fishy?

I'm running the 6.23 CPU client on Windows 7. I have 4GB of memory. I have both the GPU and the CPU clients running at the same time.

Thanks for any help.
 
bbspot said:
I've gone through the FAQs, and the SMP and the regular CPU folding program, but I think I'm doing something wrong.

I have a Core 2 Duo E8400 and according to HFM.net my PPD for the CPU is 221.6 while the PPD for my GPU is 1685.1. Are these numbers in line with what's expected, or does something seem fishy?

I'm running the 6.23 CPU client on Windows 7. I have 4GB of memory. I have both the GPU and the CPU clients running at the same time.

Thanks for any help.
Click to expand...

If you are using an ATI video card to fold, i've always heard that the client tends to be a bit more CPU intensive versus when folding on an nVidia card. Those numbers seem a tad low, but someone else will be able to help you better than I could. What video card are you using? On my laptop with the Turion 64 X2 Ultra at 2.5ghz, the SMP client gets about 400-500 ppd with client cpu utilization set at 55% to keep temps down.
 
bbspot said:
I've gone through the FAQs, and the SMP and the regular CPU folding program, but I think I'm doing something wrong.

I have a Core 2 Duo E8400 and according to HFM.net my PPD for the CPU is 221.6 while the PPD for my GPU is 1685.1. Are these numbers in line with what's expected, or does something seem fishy?

I'm running the 6.23 CPU client on Windows 7. I have 4GB of memory. I have both the GPU and the CPU clients running at the same time.

Thanks for any help.
Click to expand...

ahh you have the wrong one.. you have the old standard client..

http://www.stanford.edu/group/pandegroup/folding/release/FAH6.29beta-win32-SMP-mpich.exe you want that one.. use zero82z's windows SMP guide it will take you about 3-4 minutes to set it up.. and has all the links you will need..

http://hardforum.com/showthread.php?t=1511766

use that guide..
 
I had the SMP client installed, but that was even lower on the PPD scale. It looked like I would get 1 unit done in 3-4 days (if that). From what I read, it seemed like you needed at least a quad-core to get the SMP units done in time. Am I reading things wrong?
 
I'm not a seasoned folder by any means, but I'm using a E8400 running at 4.1ghz using the MPICH client. While getting A3 projects I'm getting between 2600-3200 ppd, while running gpu clients.

Overclock that cpu, to around 3.6. That will allow you to complete the work on time.
 
bbspot said:
I had the SMP client installed, but that was even lower on the PPD scale. It looked like I would get 1 unit done in 3-4 days (if that). From what I read, it seemed like you needed at least a quad-core to get the SMP units done in time. Am I reading things wrong?
Click to expand...

I think what you where seeing was the credit without the bonus.

Do you have HFM.net installed to show you what ppd you will be getting witht he bonus?

Also, you need to fold 10 SMP WU in order to start getting the bonus while using a passkey.

The SMP client will crush the single core client even with a higher end dual core, but a quad does much better.
 
HFM.net says right now that the PPD is 107. I went through the advanced config and there was a setting for "disable CPU affinity lock" which defaults to "no." From my reading of it, it seems that it's better to have that set to "yes" so that's what I did. Is that right?

Also, I haven't overclocked this system yet. Waiting a little while to see how stable it runs at standard speed (I've had it shut down a couple times, which I think has to do with overheating, but haven't verified yet.)

Thanks for the help guys.
 
Thanks for the coretemp program, got that running now.

That's the guide I followed to install it. Here's my log (or at least the last section of it):

Code: 
# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.29

                          http://folding.stanford.edu

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

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\[email protected]
Arguments: -smp -advmethods 

[12:04:23] - Ask before connecting: No
[12:04:23] - User name: Brian_Briggs (Team 33)
[12:04:23] - User ID: 6A6448925B6630AF
[12:04:23] - Machine ID: 1
[12:04:23] 
[12:04:23] Loaded queue successfully.
[12:04:23] 
[12:04:23] + Processing work unit
[12:04:23] Core required: FahCore_a3.exe
[12:04:23] Core found.
[12:04:23] Working on queue slot 01 [June 1 12:04:23 UTC]
[12:04:23] + Working ...
[12:04:23] 
[12:04:23] *------------------------------*
[12:04:23] Folding@Home Gromacs SMP Core
[12:04:23] Version 2.19 (Mar 12, 2010)
[12:04:23] 
[12:04:23] Preparing to commence simulation
[12:04:23] - Ensuring status. Please wait.
[12:04:33] - Looking at optimizations...
[12:04:33] - Working with standard loops on this execution.
[12:04:33] - Previous termination of core was improper.
[12:04:33] - Going to use standard loops.
[12:04:33] - Files status OK
[12:04:33] - Expanded 1765536 -> 2251033 (decompressed 127.4 percent)
[12:04:33] Called DecompressByteArray: compressed_data_size=1765536 data_size=2251033, decompressed_data_size=2251033 diff=0
[12:04:33] - Digital signature verified
[12:04:33] 
[12:04:33] Project: 6067 (Run 0, Clone 147, Gen 23)
[12:04:33] 
[12:04:33] Entering M.D.
[12:04:39] Using Gromacs checkpoints
[12:04:40] Resuming from checkpoint
[12:04:40] Verified work/wudata_01.log
[12:04:40] Verified work/wudata_01.trr
[12:04:40] Verified work/wudata_01.edr
[12:04:40] Completed 194019 out of 500000 steps  (38%)
[12:09:23] Completed 195000 out of 500000 steps  (39%)
[12:33:57] Completed 200000 out of 500000 steps  (40%)
[12:57:15] Completed 205000 out of 500000 steps  (41%)
 
Your time per frames are way way off.... Ohhhh The ATI client is probably eating a whole core...

Try running this without the ATI card folding and see what your TPF goes to.

You can type your TPF in here and get an idea of what you should be getting: http://www.linuxforge.net/bonuscalc2.php
 
OK, I'll shut down the GPU to see what happens. I'll let you know how it goes.
 
Yes, I have my passkey installed. It appears I'm getting 374 PPD with the GPU off. Anyway to have both run to get the benefits of that, or should I just run the SMP?
 
Just for comparison, I had a stock e8400 folding over the weekend with nothing else running on the machine and I was getting about 10 minutes per frame. P6014, P6055 and P6071 all hovered around the 10 minute mark. HFM.net was reporting around 2800ppd for that machine.

What does cpu-z show for the clock speed?
 
With the GPU folder turned off it's taking 11-12 minutes per step. This is while I'm using the computer for work. 2-3 times faster than when the GPU folder was on. CPU-Z says clock speed of 3002.9 MHz.
 
That seems like it's inline with what you should be getting from the CPU. It sounds like HFM isn't reporting the bonus correctly for some reason though.

Using the calculator that Kendrak linked, should give you the approximate PPD you should be getting.
 
I think the HFM is taking into effect the slower times at the start, any shut down time and troubble shooting.

When he start a new WU the better PPD with just the SMP running should show though.
 
Yes, my PPD has been improving as it moves along. There was plenty of stopping and starting in the middle there. Without the GPU going it's running at an acceptable level. I have my laptop folding now also, but it's only a Core Solo processor.
 
If you ever have the chance to grab a cheap Q6600, do ett.
 
You must log in or register to reply here.
Back
Top