• Advertisement
Sign in to follow this  

ideas to implement on a powerful system

This topic is 3357 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I'm an information technology student, and at my university I have access to a QS21 Cell Blade system. Soon it will be upgraded to a dual QS22. I have already implemented a simple path tracer on it, but I`m looking for other interesting ideas to implement on this system besides (or instead of) the path tracer. I have collected some ideas (e.g. fluid simulation), but I would like to get new ideas and opinions on what would be worth implementing on this powerful system. Any ideas and opinions are welcome. Thanks

Share this post


Link to post
Share on other sites
Advertisement
How about an indistinguishably perfect virtual world? ;)

Okay, seriously. How about a raytracer with support for parametric and volumetric 3D textures, and for voxel NURBS surfaces? These are all pretty math-intensive operations whose next bottleneck, after raw CPU power, is random-access memory latency.

Share this post


Link to post
Share on other sites
Another interesting area that needs huge resources is Monte Carlo methods applied to computer go (look for ("UCT" or "MCTS") and "computer-go").

Share this post


Link to post
Share on other sites
Thank you for your replies!

I like raytracing, so parametric and volumetric 3D textures and voxel NURBS surfaces seem to be quite interesting.

I was thinking about applying MC methods to chess, but go seems to be a better idea.

Now I'll look into these topics.

P.S. Concerning the indistinguishably perfect virtual world: I'll wait with that until I get a quantum computer :)

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement