Folding@Home status update from Vijay

Wow, he must be learning that we don't apprieciate not knowing whats coming. Kudos to giving us a heads up!!
 
Holy Crap. Is he serious? I for one am looking forward to this :D

 
Is the client written differently or is it still the same SMP client?
To me this just looks like SMP2 is really just an updated Core
If so wont we have the same shitty issues on Windows?
 
So the WinSMP client is going to be more on par with LinSMP and as a Linux user I won't have any more reason to be smug? Blasphemy!
 
nitrobass24 said:
Is the client written differently or is it still the same SMP client?
To me this just looks like SMP2 is really just an updated Core
If so wont we have the same shitty issues on Windows?
Click to expand...

1) SMP2: Gromacs and Desmond. Much effort has gone into our "SMP2" project, the codename for the second generation SMP client. The main goal here was to make it MUCH easier to use. In order to do that, it meant getting rid of our use of MPI. We have had two approaches to this. Both ditch MPI by using threads instead. One was to switch to a new piece of software for the core. This has led to the "Desmond" core, based on software from DE Shaw Research. The second approach was to communicate the MPI issues with the Gromacs developer team and work with them to push for a threads-based Gromacs implementation. Both of these are coming along well and we are testing cores in house. You should hopefully see these cores "in the wild" (i.e. running on FAH) in a month or two, assuming that tests go well.


looks like its just an updated core.. but might solve a few of the issues with windows..


this is what im waiting for...
3) GPU3: Next generation GPU core, based on OpenMM. We have been making major advances in GPU simulation, with the key advances going into OpenMM, our open library for molecular simulation. OpenMM started with our GPU2 code as a base, but has really flourished since then. Thus, we have rewritten our GPU core to use OpenMM and we have been testing that recently as well. It is designed to be completely backward compatible, but should make simulations much more stable on the GPU as well as add new science features. A key next step for OpenMM is OpenCL support, which should allow much more efficient use of new ATI GPUs and beyond.
I'm very excited about these new advances. It really should fundamentally improve the key science software behind FAH as well as making the donor experience much more smooth on our more experimental clients (i.e. on GPU and SMP).


i really hope it fixes the ATI issues so i can finally use my 4870 512mb card for something..
 
And hopefully a Native Linux GPU client.....Free us from the wrath of WDDM!!!
 
sirmonkey1985 said:
this is what im waiting for...

[...]

i really hope it fixes the ATI issues so i can finally use my 4870 512mb card for something..
Click to expand...
I agree, the GPU improvements especially to the Ati client is more compelling to me than the SMP client updates, even though I have more of the latter than the former currently running. :cool:
 
So the WinSMP client is going to be more on par with LinSMP and as a Linux user I won't have any more reason to be smug? Blasphemy!
Click to expand...
If they ditch MPI, then maybe. If not, then most likely not. Here's hoping they ditch MPI...

And hopefully they will allow Windows users to run 8-core WUs as well. Much less hassle...and potentially more ppd.

I agree, the GPU improvements especially to the Ati client is more compelling to me than the SMP client updates, even though I have more of the latter than the former currently running.
Click to expand...
What ATI improvements? My 4870 512MB is still folding at 3100ppd, exactly the same as when I got it 6 months ago.
 
Mr. Pedantic said:
What ATI improvements? My 4870 512MB is still folding at 3100ppd, exactly the same as when I got it 6 months ago.
Click to expand...
That's because nothing has been updated. If you read the link all the improvements have not been released yet.
 
nitrobass24 said:
Is the client written differently or is it still the same SMP client?
To me this just looks like SMP2 is really just an updated Core
If so wont we have the same shitty issues on Windows?
Click to expand...
Getting rid of MPI means a whole new client. Of course it means a new core as well, since without MPI everything must be coded differently. And MPI is the main reason why WinSMP is so shitty.
 
jeremyshaw said:
So I'm not the only one that hates Windows drivers?
Click to expand...
There's nothing wrong with Windows drivers. WDDM especially does a great job of handling things like crashes and driver updates. I've never gotten a graphics-related BSOD in Vista. However, WINE is a bit of a pain in the ass for using the GPU client through Windows, and it also causes a bit of a performance hit. A native Linux GPU client would be much more convenient.
 
Zero82z said:
There's nothing wrong with Windows drivers. WDDM especially does a great job of handling things like crashes and driver updates. I've never gotten a graphics-related BSOD in Vista. However, WINE is a bit of a pain in the ass for using the GPU client through Windows, and it also causes a bit of a performance hit. A native Linux GPU client would be much more convenient.
Click to expand...

WINE....
is not emulated?

I never got WINE to do what I wanted, even for just basic apps. It was typically workaround after workaround, and I got really sick of it.
 
WINE....
is not emulated?

I never got WINE to do what I wanted, even for just basic apps. It was typically workaround after workaround, and I got really sick of it.
Click to expand...
Same here. Either I just got open-source equivalents, rebooted into Windows to use them, or just did without. WINE was not at all a satisfactory solution for running Windows programs on Linux.

That's because nothing has been updated. If you read the link all the improvements have not been released yet.
Click to expand...
I know. That's what I was trying to point out.
 
Mr. Pedantic said:
Same here. Either I just got open-source equivalents, rebooted into Windows to use them, or just did without. WINE was not at all a satisfactory solution for running Windows programs on Linux.
Click to expand...

It is a pain in the ass to set up initially, but once you get it set up it usually works for most things except for a lot of games.
 
jeremyshaw said:
WINE....
is not emulated?

I never got WINE to do what I wanted, even for just basic apps. It was typically workaround after workaround, and I got really sick of it.
Click to expand...
Mr. Pedantic said:
Same here. Either I just got open-source equivalents, rebooted into Windows to use them, or just did without. WINE was not at all a satisfactory solution for running Windows programs on Linux.
Click to expand...
The only thing I ever used WINE for was to run my GPU client from within Linux, and for that purpose it worked fine once I set it up according to the guide. I have no interest in using Linux outside of F@H.
Mr. Pedantic said:
I know. That's what I was trying to point out.
Click to expand...
Well you did a poor job of that. There have been no major changes in the ATI GPU client since around February of this year, so obviously you wouldn't have seen any change in your PPD.
 
I am really looking forward to seeing if they can improve the performance for ATi cards if I start folding on my new 5870's.
 
I just hope he gets to rolling this out before my next 6 month deployment.

Murphy's law says he'll roll it out the day after I deploy and my boxen wouldn't fold a paper hat for 6 months.

 
jfb, figure out how to remote control your stuff

Ive got mine all set to be controlled completl'y on my phone, so I can control my rigs from anywhere
Important for us military based folders :)
 
I use LogMeIn and have all my folding rigs set to auto-login on reset so I can reboot and still have access, but VNC would work too, I use LMI because it intergrates perfectly with my phone, so I can do it anywhere, although I usually use my laptop for major changes
 
all my machines are behind the same router, so I could only port forward to one. I suppose I could use Remote Desktop to acess the others. But I've heard remote desktoping into a GPU folding rig crashes the GPU folding, I suppose exiting by reboot would at least get things rolling again.
 
jfb9301 said:
all my machines are behind the same router, so I could only port forward to one.
Click to expand...
I don't see why that would be the case. You can forward a different port to each machine on your network and use that to differentiate between the different machines.
 
Oh and there is no internet connection to a Submarine.

So to sum it up I suppose I could VNC into my machines, every couple months, when I hit a port, if I can find an internet connection.

This would be the best deal I have, so I had betting get figuring out that VNC install and setup. a bad scenario is still better than the alternative.
 
I suggest you use WHS, it provides a TS gateway and all you need is the interwebz :)
 
You must log in or register to reply here.
Back
Top