The way these contests are usually designed, if you have the "right" (or one of the right) algorithms, it won't take more than a second or 2, and making it faster is a waste of time.
If you have the "wrong" algorithm, it will take you a few billion years, and optimizing it to take only a few million years probably won't help, either. CUDA may be interesting and useful in some cases in real life... but I'd say forget about it for programming contests. On Sep 9, 11:27 am, Vexorian <[email protected]> wrote: > Dear fellow codejammers, excuse the slightly OT question, but I have > been wondering about GPU and the possible improvements it could have > in these contests (or whether there would be an improvement or not). > These questions are mostly because I do not understand things like > CUDA or how the GPU architecture differs from CPU too well yet. It's > just a theoretical speculation, so lest ignore the discussion about > whether CUDA would be allowed or not in codejam, and instead about how > useful it could be. > > First, I think it is at least good for things like linear algebra and > linear programming maybe? Those problems were not uncommon last year. > So maybe solving the problems on your GPU could give you a strong > speed advantage on this sort of problems? > > What other common problems do you think could get improvements from > it? > > Also, does anyone know if something like running all N cases of an > usual problem simultaneously would take a good advantage from GPU's > better parallelism? --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "google-codejam" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/google-code?hl=en -~----------~----~----~----~------~----~------~--~---
