• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Standard Client Folding prog ?'s

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Norcalsteve

Member
Joined
Sep 19, 2009
Location
Crestview, FL
Hey fellow Folders... a New guy here, (posted in New guy sticky already... somewhat) and just in time for 3rd place Folding!!!! Maybe the 1 WU i finished last night when i joined bumped us!! LOL!

So working with the Standard Client is really straight forward and simple... I have just a few clarification questions that i didn't pick up in the reading. When i right click on the F@H icon in the sys tray, i see a link called "Display" when i click it, it shows the spinning protein and i notice my work load increase (what i want) but when i dont touch the mouse or do anything on the comp for 15-30 seconds, it stalls to a crawl... until i move the mouse again (yes screen saver is turned off, and priorities are even high on the FAH). also, the Viewer in the F@H folder displays a different protein than what is on the "right click" selection of the F@H system tray icon. and that Viewer stalls too after a short period...

and last question, does Deino MPI/MPICH2 Clients work your processor almost like Prime95? if so, i want that one cause i want to fold faster and use more of my processor... this standard client is good, but i feel like i can do more with this rig of mine. I would also use my GPU's but i game on this computer too, and what i gather from the install guide, i would have to un link my cards and do a set up like that... unless one of you tell me otherwise.

Thanks all, and... WTF for 3rd!!!:bday:

EDIT: Ok, Downloaded the GPU tray client, now i dont have that issue with the display... also... is it ok to run the CPU and the GPU client together? Cause i am, and i see no impact from one to the other.
 
Last edited:
If you are talking about your signature rig and assuming you have 6 GB of ram, you want to finish the uniprocessor cllients you have running and install VMware Player 3.0.0 and run Linux (Ubuntu 9.10) in it and fold -bigadv WUs in it. At 4.0 GHz, you'll make almost 30,000 ppd compared to the uniprocessor client's several hundred ppd and the Win SMP client's several thousand ppd. I have to head to work, so hopefully someone will be around to help you.
 
Wow, i may just try that... yes i am folding on my Sig Rig ;-) i will read around on how to set up the way you suggested... I have Ubuntu 9.10 already (had a Grub issue with windows so i reformatted and made my RAID0) so now i will d/l VMware.

Since this post, i have installed the SMP client and the GPU client... figured how to make my GPU scream (a little) with 700-800 ns/day... i guess thats ok? NEED MOAR POWER!

Thanks for that input to help me fold faster!

any more input from you all would be much appreciated. I just need directions and How-To's! ;-P
 
Don't use the gpu viewer, it kills performance. Use HFM to monitor your progress. Generally all performance is referred to in terms of Points Per Day (ppd). HFM will give everything you need to know about the WUs you're folding and your statistics. Use EOC to monitor your overall statistics.
If you enter your user name and EOC ID number (495060) into HFM, it will report your EOC stats.
 
Don't use the gpu viewer, it kills performance. Use HFM to monitor your progress. Generally all performance is referred to in terms of Points Per Day (ppd). HFM will give everything you need to know about the WUs you're folding and your statistics. Use EOC to monitor your overall statistics.
If you enter your user name and EOC ID number (495060) into HFM, it will report your EOC stats.
As I'm reading through various thread where you make the same recommendation, I started wondering: Shouldn't you just create a macro that will enter the above message into the text box for you? :D
 
DELETED

Figuered how to make Sharing work for Fahmon in windows to see my Ubuntu SMP folding client in VMWare

Thanks for your help and higher PPD's...

is 15100 ppd with one GPU client in Windows and 1 CPU client in Ubuntu/VMware good?
 
Last edited:
After you have folded 10 SMP WU's, you'll be eligible for the extra large WU's, and their substantial bonus.

If you want to fold that intensely.

With the bonus points factored in, you should make nearly 25k per day, just from the extra large WU's, alone.

Of course, you can still fold with your gpu also.
 
How, just let it run for 10 WU's and in about two days they just start giving me extra large? or is there a flag i have to put after the starting command?

Also i found this:
Change "8" to however many cores you plan to run the client on. For example you can run the -smp 4 flag for quad core cpus and -smp 2 for duals.

Warning: If you had a work unit running with x amount of threads and after rebooting the vm you change it, it will lose the wu. Thanks for the heads up MADMAX22.

does this only pertain to -bigadv? or all types of SMP clients? i think i put -smp 4 in when i started and i want 7... should i just wait till the WU is done?
 
Last edited:
If you want to finish the WUs as fast as possible, stop the client. restart it with the -oneunit flag. When it finishes the WU reconfigure it for -smp 8. Using all cores you should be running 3:00/frame.

Whoever posted use -smp 2 for duals got it wrong. THe minimum number of threads the Linux SMP client runs is 4. If you enter -smp, -smp 2, or -smp 4, the client spawns 4 FahCore_a2 processes. THe maximum is something like -smp 64. THe Windows SMP Client will only spawn 4 processes, period. THis willl change with threads based SMP2.
 
sweet, i will do that... also, what flags are recommended for the GPU client in windows for faster/efficent folding? or shall i just overclock the cards? (Dunno though, that would be more money to throw at my Water Cooling loop) BTW, i am only folding on one GPU, i dont want to unbridge from my SLI, since i pause in the evening and shoot up some MW2!!
 
THe only flag I've used is -verbosity 9 and it doesn't seem to make the log any more verbose. None of the flags affect the GPU client production on a WU. Selecting -advmethods either in the advanced configuration or by command line may change the WUs you are assigned. Sometimes you'll get more high producers and sometimes just the opposite. Turning it on or off depending on the prevailing assignment can net you more points.
 
WARNING! newbe question ahead!

what is, or how do you, calculate "Frame time?"
Been trying to google it and forum comb it, but no dice...

As of now, i am running the VM/Ubuntu/SMP -bigadv -verbose 9 -smp 8 with all 8 core engaged successfully thanks to your help... but what type of performance am i looking at? I see 11169 PPD and 25.4K total for just the CPU WU. But, the ETA is in 2 days! is that normal? It took 32min from 1% to 2% and 33min from 2% to 3%

Is each % a frame? and the time about 32 min?

atleast my PPD is 20k
 
WARNING! newbe question ahead!

No problem! :thup:

what is, or how do you, calculate "Frame time?"
Been trying to google it and forum comb it, but no dice...

As of now, i am running the VM/Ubuntu/SMP -bigadv -verbose 9 -smp 8 with all 8 core engaged successfully thanks to your help... but what type of performance am i looking at? I see 11169 PPD and 25.4K total for just the CPU WU. But, the ETA is in 2 days! is that normal?

That time is fine for a -bigadv WU

It took 32min from 1% to 2% and 33min from 2% to 3%

Is each % a frame? and the time about 32 min?

Yes, and Yes.

atleast my PPD is 20k
 
Back