• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

F@H GPU Client 2 - It's here!

F

Foxery

Golden Member
Announced 4/10/08

If you have a 2xxx or 3xxx card, please feel free to check out the new GPU client. You can download it from here

and a FAQ is here

We will next put this on our main download page, assuming that this stage of beta testing looks good.

Please keep in mind that this is beta software. The client/core may be unstable (although it looks ok so far), the points benchmarking may change due to software or policy changes, etc. With that in mind, we'd appreciate any comments you may have -- please post them in this forum section. Thanks!
 
You beat me to it here LOL! My 2900XT is already folding away. Don't forget to force 3D clocks with ATITool 0.27 Beta 4. The PPD is decent and this should pair very nicely with an overclocked quad core. I'm running both the GPU and SMP clients simultaneously on my old Opteron 185 to see how my system handles it.
 
Answers to the ever-crucial question, how many points? 🙂

"We benchmark it on a dedicated computer with an ATI Radeon 3850 GPU, running in a Dell Inspiron 531, with a 2.16 GHz dual core AMD 64 X2 4000+.
Points = 1000 * (DaysPerWU)"

The FAQ also says they still expect it to need a dedicated CPU core to feed it data.

I can't wait to get home tonight!
 
I believe it relies on hardware features that were new in the 2000 series.

Mine's running now... interesting how much of a CPU hog it still is. I dedicated a full core of my 3GHz Core2, and it is only keeping the GPU client at 80% activity. PPD is fantastic, though, since the benchmark system is an older AthlonX2 system; their reference card must be starving for data!

(Use RivaTuner to check the GPU Usage and temperature.)
 
Holy crap the WU's are small, 97 points! Looks like it plays well with the SMP client, I have 2 GPU clients running and 1 SMP client. Each GPU is using 100% of a core, and the SMP client is using the other 2 cores. Looks like each 3870 is earning 920PPD. This gives me about the exact same PPD as 2 SMP clients. (2 x 1800, now its 1 x 1800 + 2 x 920).

Also, anyone running Vista and looking for the .dll's, they are here: C:\Users\USERNAME\AppData\Roaming\Folding@home-gpu
 
What's your rig? My 3850 is at 1500 PPD right now. 1xSMP running next to it only dropped from 3000 to 2500, so my total is now 4000! A little more overclocking this weekend and I'm sure I can get 4700 out of a single box.

All of my fans are making more noise now, of course 😉 I'll live.
 
Originally posted by: Foxery
What's your rig? My 3850 is at 1500 PPD right now. 1xSMP running next to it only dropped from 3000 to 2500, so my total is now 4000! A little more overclocking this weekend and I'm sure I can get 4700 out of a single box.

All of my fans are making more noise now, of course 😉 I'll live.
Click to expand...

I tested further. The -gpu X tag isn't working on my comp, so I was running 2 clients on a single GPU.

To get the most PPD, I found that running 1 SMP client and 1 GPU2 client works best. In this situation, the GPU2 client earns 1600PPD and the SMP client earns ~2300PPD.

Previously I was earning ~3400 PPD with 2 SMP clients.

My rig is:
Q6600 @ 3.0ghz
2 x HD3870
 
Okay so I'm new to this Folding on the PC- I use it on my PS3 but it's so simple that way so I am a bit stuck here. I have a 3850 Crossfire rig in sig....I downloaded this Beta, read the FAQ....installed it and ran it. ATM I am only running 52% GPU activity on one GPU in the ATI CCC....the FAQ mentions that in order for the client to work with two GPU's I have to copy the Folding@home-gpu folder from AppData to gpu-2 directory...yet I cannot ind such a directory. Any help appreciated on getting this setup 🙂.
 
Hi Sylvanas,
I'm glad to see another folder. In order to get 100% GPU activity, update the graphics drivers to the most recent version. I don't have 2 GPUs, so I can only guess how it works: go to the appdata folder, copy the folding folder to a new folder in the addata directory. Now go to the original folding directory, create a shortcut to the application and add -gpu 0 as an argument, go to the new directory and do the same thing, but add -gpu 1 as an argument. Also make sure crossfire is disabled. Good luck

This thread may help you: http://foldingforum.org/viewtopic.php?p=17967
 
Originally posted by: Philippart
Hi Sylvanas,
I'm glad to see another folder. In order to get 100% GPU activity, update the graphics drivers to the most recent version. I don't have 2 GPUs, so I can only guess how it works: go to the appdata folder, copy the folding folder to a new folder in the addata directory. Now go to the original folding directory, create a shortcut to the application and add -gpu 0 as an argument, go to the new directory and do the same thing, but add -gpu 1 as an argument. Also make sure crossfire is disabled. Good luck

This thread may help you: http://foldingforum.org/viewtopic.php?p=17967
Click to expand...

Thanks for the info Philippart - will give it a go when i get home.
 
Originally posted by: LOUISSSSS
so why is it again that nvidia gpu's can run FAH? i miss my x1900xt kind of
Click to expand...

I assume you meant 'can't run'?

IIRC ,basically Nvidia don't work closely with F@H dev team whilst ATI/AMD do, & they have a vested interest to because they are sponsors too.Their are some problems running F@H on Nvidia cards which needs Nvidias help, their isn't much so no GPU client for Nvidia cards 🙁, again IIRC 😉
 
Actually, running on two video cards is currently broken. But yes, definitely update to Catalyst 8.3 for best performance on the one...

RE: nVidia, basically their drivers are broken and they don't care. ATI cares, and their developers work with Stanford on the F@H client.
 
There are more comments about the new GPU2 client here.
I guess I would like to stress this:
It's also an important time to note the contributions of the GPU1 client: we've been able to learn a great deal of running GPU's "in the wild," which is very different than in the lab, and what we've learned has gone into GPU2. GPU2 is much easier to run, more reliable scientifically, and has more advanced algorithms. Due to these differences, GPU1 has become obsolete. However, we are working to write a paper with what we've learned from GPU1 and acquiring data for that paper will need some more GPU1 cycles.
Click to expand...
That the GPU1 client will become obsolete fairly soon with no WUs to crunch 🙁
This seems somehow not right to me. If I had bought a card so I could use the GPU1 client I would be bumming a bit now. Although it now makes sense why they did not "push" the old GPU client at the folding forum. It's like they knew it was going to go by the wayside.

On a positive note: it sounds like the GPU1 client laid the foundation for a much better GPU2, so if you contributed on a GPU1, all future results will be part yours also!

So learning from this, with how fast vid cards change technology, the same thing could happen to this new GPU2 client.
 
I don't know why they can't maintain some level of backward compatibilty with older GPUs with the new client.
I doubt many people would of bought an XT19xx solely for F@H, but still it is a shame that they won't be able to crunch again in the near future, including my X1950 Pro 🙁, think I'll give it a last run before they kill the client.

Any idea when the GPU1 gets shut down?
 
Originally posted by: Assimilator1
I don't know why they can't maintain some level of backward compatibilty with older GPUs with the new client.
Click to expand...

The new client relies on hardware features that were new in the 2000 series. 1x00 series cards are incapable of processing the instructions.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top