Hi Simplex0,

Well, the current test is on Help Cure Cancer to GPU and is getting closer to being released to production. It will run concurrent with the CPU version of this project. The organization that runs this project from Toronto under the Help Cure Cancer auspices has much broader application of the findings of this research, so this project, running since 2007 and nearing it's 5th anniversary, has been drawing in many -to be processed- crystal data sets from around the world.

This will be the first GPU supported project at WCG, and likely the only one for a longer while. If successful, the word will surely be spread as it cuts years out of these researches and or allows for much greater resolution processing in the same amount of time a CPU only based project would need. The application must to be suitable for porting to GPU and fit and few are at this time.

For the moment beta testing uses known task results already processed on CPU to verify that they produce the exact identical output, which is critical of course.

Thak you SekeRob & Mikey.
I truly hope that at least ON project in BOINC eventually will accomplish what Folding@home has done for years, get applications that can make good use of BOTH AMD and Nvidia cards but it seams to take forever.

MOST gpu projects CAN use both but most are ALSO tweaked to prefer one brand over the other. It is kind of like a Windows/Mac/Linux thing, which one you prefer means the things you do will be quicker on that platform. There are lots of other Boinc projects out there, such as: PrimeGrid, Collatz, DistRTgen, Moo, Seti, GPUGrid, Einstein, Albert, MilkyWay, Poem and I am sure there are others I am just not thinking of right now! ALL of them use Boinc as the basis for the crunching and just like WCG are pretty easy to use. The MAIN difference is the tech support here, when those other projects have problems the tech support/feedback from the Project folks can be NON EXISTENT AT BEST!! WCG has, as long as I have been here which is since May 2009, been EXCELLENT even on a bad day!!! If you look SekeRob is EVERYWHERE and willing to try and help EVERYONE!! Other Admins help ALL THE TIME too!! LawrenceHardin is an admin and someone posted that after a new Boinc install that Boinc doesn't work, LH was the FIRST post asking the person for more info so he could help them! As an example several years ago Seti went down for over a week, NO COMMENTS, no units, NO webpage, NO NOTHING ANYWHERE! After things were fixed they came back up and said 'whoops our bad'! They then said there excuse for NOT communicating about the problem was 'would you rather have us working on fixing the problem or telling you we are working on fixing the problem?' When WCG has a problem they tell other websites and essentially 'put the word out' that there is a problem and they ARE working on it! WCG also seems to be able to bring up their own website up saying 'we are having problems and are working on it', other Projects DO NOT do that!

The dividing line that cleanly separates the 'means' (models that can be used to find cures) camp from the 'ends' (find cures) camp is not only blurred, the members of those camps, as it were, also keep changing their allegiance. One can say the same thing for technology as a whole. Finding a solution toward an end is inseparable from finding the means to be able to find a solution. Put another way, an effort can not just throw resources toward ends and call it a day; how much chances that effort would have in successfully achieving its ends is dependent on how well the means allow the effort to be successful. In short, finding cures for diseases can only be as successful and limited up to what the means allow. This is one thing.

Another thing is that grid-computing is essentially a tool to aid those types/areas of research that lend (or can be made to lend) themselves to the computational approach. Grid-computing is not about searching, developing, or designing a specific product, process, or technology to meet a need/demand; those are for commercial companies (and their research arm/links) to concern themselves with.

Taken together, grid-computing output lays the foundation for other entities/groups/companies to have the best possible chance of succeeding in making effective cures, solutions, products, or even new technologies.
;
; edit1_2012.09.17Mo.1925.utc
; -- readability/presentation improvements
;

Simplex0,

Scrub and let me try to explain: Beta tests [the forum we are in and exchanging thought] usually take 1-2 runs [1-2-3 weeks], then they are moved to production. The Help Conquer Cancer [HCC1] GPU application is an exception. Many new technical issues to overcome, but we are now quite close to speed up this research 18-20 times per analysis run processed, compared to the CPU version that has been running for years now.

The theoretical models and simulations and analysis lead directly or indirectly to either fundamental knowledge, which is key to finding cures [HPF2 Human Proteome Folding e.g.], or in the case of for instance Help Fight Childhood Cancer [HFCC], FightAIDS@Home [FA@H], Discover Dengue Drugs Together [DDDT2] etc, directly identify compounds that are wet-lab tested. HFCC on our grid for instance has picked out several dozen chemical combinations out of a library of 3 million that the scientists in Chiba have been testing in the past 2 years, which has give them knew knowledge and indicators which other targets to also process on our grid [T10 for instance]... computational chemistry and biology. Many promising leads have resulted from our contributions and many papers of new knowledge have been written that have made it to the most respected medical journals. The path from lab-test to medicine or vaccine is long, no illusions. We're just helping shortening this path by years and years.