Charity and Potential Ron Paul Publicity via ::Folding@home::

[JimInNY]

Your showing up now. Just took a bit. Welcome to folding on here.

600 points, 1 unit! Woooot! 19th place... with a bullet! lol

Glad to be here. Two more units going on this machine and one on the old one. Let's get this team into the top 100!

 

[JimInNY]

I have to say that I see no difference in my computers performance with two units running at once on here. And it's like like this thing is all that special. It's an AMD A4-3400 APU with Radeon HD Graphics 2.7o GHz.

Anyone shying away out of fear or something shouldn't really have to worry too much. If I can do this stuff anyone can.

Now join the team dammit!

 

[specialkornflake]

I have to say that I see no difference in my computers performance with two units running at once on here. And it's like like this thing is all that special. It's an AMD A4-3400 APU with Radeon HD Graphics 2.7o GHz.

Anyone shying away out of fear or something shouldn't really have to worry too much. If I can do this stuff anyone can.

Now join the team dammit!

By two units running at once, you mean one for the CPU/SMP and one for the GPU? You are running the Beta client? You are very correct, even 15 year old computers can contribute, the whole point of distributed computing is you don't need to be powerful! Yet, the GPUs are probably the most effective from what I've seen.

 

[gerryb]

Why is the GPU thread worth 1835 base pints, but my 4 CPU core thread is only worth 108 points?

 

[specialkornflake]

Why is the GPU thread worth 1835 base pints, but my 4 CPU core thread is only worth 108 points?

High-performance clients often require more computing resources. GPU clients typically run on dedicated systems, 24 hours a day, and use more processing power, more disk space, more network resources, more system memory, etc. Also, a major part of the scientific benefit is dependent on rapid turnaround of work units; hence we assign short deadlines for GPU work units. To reward those contributors for donating resources beyond the typical CPU client, for completing these work units very quickly within the short deadlines, and for contributing to the development of our next-generation capabilities, we currently set a benchmark value proportional to these demanding GPU work units. Without the GPU clients and your additional contributions, we would not be able to complete many important projects.

http://folding.stanford.edu/English/FAQ-NVIDIA-GPU3#ntoc22

 

[GunnyFreedom]

Why is the GPU thread worth 1835 base pints, but my 4 CPU core thread is only worth 108 points?

Different processes requiring different numbers of cycles to return work units. Some work units require 10x the processor cycles of other work units, project depending. The 'points' are more tied to processor cycles, while the work units are more ties to the given project that your client is processing.

You could be doing Alzheimer's one day, cancer the next. I'm doing Huntington's Disease on my monster machine right now, which apparently requires massive cycles because it's worth a lot of points. The client will examine each macine's capabilities and assign projects to suit.

The beta client does not seem to be using my GPU like it's using everyone else's. it recognizes my GPU, but I only have one folding slot instead of two. Maybe both my CPU and my GPU are on the same folding slot, or maybe the MacOS client just acts differently, no idea, but it seems that every processor I have is all on the one project, which is probably why my WU's have such ridiculously high scores.

 

[specialkornflake]

The beta client does not seem to be using my GPU like it's using everyone else's. it recognizes my GPU, but I only have one folding slot instead of two. Maybe both my CPU and my GPU are on the same folding slot, or maybe the MacOS client just acts differently, no idea, but it seems that every processor I have is all on the one project, which is probably why my WU's have such ridiculously high scores.

Consider updating your video card drivers. This resembles a problem I had, although if you are using a Mac I don't have any experience.

 

[GunnyFreedom]

Congrats! You have any more to add? I find the GPUs to be far more effective than my Cores.

Is there something special you have to do to enable GPU processing? I have a recognized graphics card, but my work que seems to be my cores only. it doesn't act like it's talking to my GPU even though to alleges to support it. MacOS. MacPro quad core dual, better yet, here:

 

[GunnyFreedom]

Consider updating your video card drivers. This resembles a problem I had, although if you are using a Mac I don't have any experience.

Allegedly my video card updates to the latest and greatest via automatic system updates, so if I don't already have the latest revision then something is wrong.

 

[specialkornflake]

Allegedly my video card updates to the latest and greatest via automatic system updates, so if I don't already have the latest revision then something is wrong.

My first thought was for you to go to Configure, Expert and add two name/value pairs- smp true and gpu true. That's how my computer running both is configured. After some research, I can't find any evidence that Folding@Home supports GPUs on Macs. I know very little about Macs, my only idea would be to try and run all of this through Bootcamp on your mac with a windows installation.

 

[JimInNY]

I am running the beta client, yes.

I'm not sure what I mean about 2 units running at once, I just know that on the beta client screen I see two working at one time.

 

[specialkornflake]

I am running the beta client, yes.

I'm not sure what I mean about 2 units running at once, I just know that on the beta client screen I see two working at one time.

That looks good. Our GPUs seem to be frequently assigned 1835 credits so I can assume you are running your GPU and CPU effectively. Switch your view to Expert in the top right and you can see a lot more information. If you posted that picture we would know even more.

 

[GunnyFreedom]

Strange, I went into configure and all I had to do was select to use GPU and viola, there it is. Says ready, and download is at 0% with unknown WU's so we'll see if it takes.

And I would not curse my POSIX box with Windows to save my life! I especially would not use Boot Camp to operate entirely in Windows just to run F@H. As a networking and server professional, M$ gets plenty of my blood sweat and tears in the field, they don't get a dime of my personal money for any reason whatever.

It looks like the client recognizes the GPU as ready to fold, but the work cue is just stuck at download zero with no assigned project.

I guess I'll wait until tomorrow to see if it ever downloaded any work for the GPU.

ETA: 5 attempts, waiting on WS Assignment

So it looks like the GPU is ready to fold, but the work server doesn't know what to do with the request. How odd.

 

[JimInNY]

That looks good. Our GPUs seem to be frequently assigned 1835 credits so I can assume you are running your GPU and CPU effectively. Switch your view to Expert in the top right and you can see a lot more information. If you posted that picture we would know even more.

 

[GunnyFreedom]

Log text:

05:40:14:WU00:FS01:Connecting to assign-GPU.stanford.edu:8080
05:40:15:WU00:FS01:News: Welcome to Folding@Home
05:40:15:WARNING:WU00:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment
05:40:15:ERROR:WU00:FS01:Exception: Could not get an assignment
05:42:51:WU00:FS01:Connecting to assign-GPU.stanford.edu:80
05:42:52:WU00:FS01:News: Welcome to Folding@Home
05:42:52:WARNING:WU00:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment
05:42:52:WU00:FS01:Connecting to assign-GPU.stanford.edu:8080
05:42:52:WU00:FS01:News: Welcome to Folding@Home
05:42:52:WARNING:WU00:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment
05:42:52:ERROR:WU00:FS01:Exception: Could not get an assignment
05:44:27:WARNING:Caught signal SIGPIPE(13) on PID 51
05:44:28:WARNING:Caught signal SIGPIPE(13) on PID 51
05:44:28:Server connection id=7 ended
05:44:30:Server connection id=8 on 0.0.0.0:36330 from 127.0.0.1
05:47:06:WU00:FS01:Connecting to assign-GPU.stanford.edu:80
05:47:06:WU00:FS01:News: Welcome to Folding@Home
05:47:06:WARNING:WU00:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:80': Empty work server assignment
05:47:06:WU00:FS01:Connecting to assign-GPU.stanford.edu:8080
05:47:07:WU00:FS01:News: Welcome to Folding@Home
05:47:07:WARNING:WU00:FS01:Failed to get assignment from 'assign-GPU.stanford.edu:8080': Empty work server assignment
05:47:07:ERROR:WU00:FS01:Exception: Could not get an assignment

 

[JimInNY]

That looks good. Our GPUs seem to be frequently assigned 1835 credits so I can assume you are running your GPU and CPU effectively. Switch your view to Expert in the top right and you can see a lot more information. If you posted that picture we would know even more.

On my other computer, my older pentium 4 I am running the simple client. Would there be any advantage to using the beta client on that machine, and if so, do I have to wait for the current unit it is working on to finish?

 

[GunnyFreedom]

I'm guessing it has something to do with the GPU work server assignment IP 0.0.0.0

ETA -- no, that IP would be set by assign-GPU.stanford.edu:8080 depending on the project that was assigned. It's just really confused why I'm asking for GPU work it seems

 

[specialkornflake]

On my other computer, my older pentium 4 I am running the simple client. Would there be any advantage to using the beta client on that machine, and if so, do I have to wait for the current unit it is working on to finish?

I don't think there would be much benefit because the Beta Client is good for using your GPU and multi-core CPUs and a Pentium 4 is not multi-core I don't think. You should instruct your current WU to "Finish" or "Pause when finished" before switching.

 

[specialkornflake]

I'm guessing it has something to do with the GPU work server assignment IP 0.0.0.0

ETA -- no, that IP would be set by assign-GPU.stanford.edu:8080 depending on the project that was assigned. It's just really confused why I'm asking for GPU work it seems

This thread might apply: http://foldingforum.org/viewtopic.php?f=67&t=18295

 

[GunnyFreedom]

You were right the first time. GPU processing only works on Windows. Found the answer in the folding forum. People have tried enabling GPU on MacOS before but the client does not support it. I'll just have to wait until they enable GPU processing on POSIX compliant systems before I can GPU fold, because I am NOT buying and installing Windows to do this.

ETA - found it here: http://foldingforum.org/viewtopic.php?f=67&t=20459&start=45