Common stuff competitions

I've been trying to think of some coding competitions that could benifit everyone here. My thought was to make a 2 week long competition to make common library stuff.

Examples:
Best FPS Camera lib
Best Astar Lib
Best Game Menu lib
Best 3rd Person Chase Cam Lib
Best Entity Dropper

They would be judged by ease of use. I don't have anything to pony up for prises so I guess it would be braging rights or something.

Heck, I don't know. Let me know what you think.

Lighting and also occlusion too.

My internal occlusion system owns and I will not be submitting it as it is quite simply the best - although, it would be handy to see just how much better it is than anything else.

My lighting system is somewhat rough. I'm talking about realtime light that does not pass through solid objects (walls, etc) - theoretically, it isn't too hard to implement - especially now that I can offload this onto someone else to complete.

If anyone can wrap-up the lighting one inside two weeks then all the better.

Hah! You fight like a dairy farmer!

Shouldn't that be Best Path Finding Lib? I know that on the C64 or whatever there was a fairly sound performance argument in favor of using A-Star, but do people really have so little RAM these days that you can't just load up a Shortest Path Tree for each node? I mean it's not like Dijkstra's SPT algorithm takes very long on moderately modern hardware, even for rather complex graphs, and depending on how your data is setup, you can precalculate it all upfront.

Yes but Dijkstra is quite bad with you have to calculate similar paths for more than one object, at that point HDA* (why remain on A* if there are higher level versions of it who are really there to be used instead of looking nice?) will outperform Dijkstra.

One could as well ask: why use any of both if you could use potential field to search the path without grid / node nets at all? ;-)

These are all things you need, aren't they?

LOL @ PCS .... hmm ... I wonder

More seriously - people share useful code rather a lot on this site. Do we really need to make it competetive?

Its a good idea, one think to add that i havent seen any examples of is a really good controler config screen for setting up joypads. theres bit of it in the code archive but nothing that does the whole job.

I think I'm going to attempt to organized this.

I like the pathfinding change. I would agree that limiting it to Astar wouldn't be the best.

As for my reason for suggesting this competition, I think the shared code in the archive is good, but they seem to lack the polished finish that would make them really stand out as useful sometimes. And it's not like I can complain, it's free stuff, and for free their freakin' great! But I think some competition to improve the quality of some of these tools would really boost the community and inspire some innovation. I would love to see a working GTA cam system.

I'm trying to think what I could offer up as a prize. I'm an artist and can make game models that will work in Blitz 3d. I could probably make and exclusive model pack that only the winner has the rights to use. Then of course you run into "is it the style I need" issues. I don't know, I'll keep thinking about it.