Update On GPU3, ATi Version Will Be Delayed

[LCB001]

Prof. Pande just updated his blog with news about GPU3 and looks like the ATi version won't make the initial release. Looks like GPU3 and SMP2 might end up sharing some code...

Some more details on the GPU3 core, regarding OpenCL

Here is an update and some more details on the third generation GPU core. As I stated before, this core is built on OpenMM, which brings the science further along and allows us to make easier updates. OpenMM supports OpenCL in beta form in its 1.0 release (scheduled for late January 2010), but it is important to stress that the OpenCL support is very early and so we will not be relying on OpenCL in the first releases of the GPU3 core.
This means that the core will only roll out for NVIDIA first. ATI has depreciated Brook, but does not have a fully-functioning OpenCL implementation, so we are stuck in between support on the ATI side. Once the OpenCL implementation matures (on both ATI and NVIDIA), we will be able to finish and optimize our OpenCL code (we can't reliably optimize code until the implementations are more finalized), and then the OpenCL portions will go into QA.
So, if you are interested in OpenCL (especially in terms of ATI support), this will not be in the initial release, but once the environment has matured, we will push forward. By the way, once OpenCL support for multi-core boxes has matured, we will also see about porting the GPU3 code as a new SMP2 style core as well (i.e. threads-based SMP support). If the performance is strong, it's appealing to think that we can go back to having a single dominant code base for much of FAH's calculations.

:canadianwave:

 

[Zero82z]

Well that's just lovely.

 

[_dangtx_]

they kinda abandoned ati big time....

 

[Zero82z]

they kinda abandoned ati big time....

Looks like someone didn't bother to read the post...

 

[geokilla]

Delayed or not, I like how GPU3 and SMP2 is soon around the corner. The only thing that I'm mad (not really) at is that all the folding the world does, I haven't seen much articles indicating how it's helped solve diseases and find a cure and everything. The points don't matter much to me.

 

[Arinoth]

We're the unsung heroes

 

[SugarJ]

What we do pinpoints how the proteins fold, it's interpreting what effect that has that takes guys with PhD's years and years, and can't be done by computers. We're providing them with the research for their theories. At least that's how I interpret it.

 

[sswilson]

What we do pinpoints how the proteins fold, it's interpreting what effect that has that takes guys with PhD's years and years, and can't be done by computers. We're providing them with the research for their theories. At least that's how I interpret it.

Exactly.... it's not like SETI@Home where your particular computer might be "the one".... FAH produces the Data that other folks can use. I think I've only seen one full fledged paper based solely on FAH data, but then again, I'm not exactly familiar with the circles that this information would be used for so there might be much more out there.

 

[Zero82z]

I think I've only seen one full fledged paper based solely on FAH data

Folding@home - Papers

 

[Soultribunal]

Like I explained a while back. It takes a lot of time to interpret the Data that we create.
In the 70's a chemist that my mother worked for then, and again in 2000 did work and his thesis at Stanford for Protien Folding and they were using special printouts and goggles to 'Fold' the protien.

It takes a lot of time and effort. The good news is each and every WU we complete helps shave the research time and development time.
A lot of research comes to dead ends but regardless this is truely something unique to do and I had it explained to me by someone in that profession why it is one of the greatest causes we can embark on.

ST