
August 3rd, 2002, 10:16 AM
|
|
|
|
Join Date: Apr 2007
Posts: 2
Time spent in forums: < 1 sec
Reputation Power: 0
|
|
|
Future Suggestion
Hi Matt,
I like the concept of the current contest. Unfortuneatly though, I don't have time to enter this one (though I did enjoy posting a solution to the previously held klondike contest).
I do have a suggestion for future contests of this type. I would suggest that you should have really turned this into two simultaneously run contests, one for the client (current contest) and one for the game server. (or at least, they'd be back to back contests).
Basically, you would (as you have already) write the rules/specs for the contest, but the implication would be that coders would submit the game engines that follow those rules. Obviously, you would still allow client submitions as well (either now or in the following contest).
I guess the reasoning is this... Everyone who wants to compete in this contest will basically have to build their own game server anyway to test their A.I. Might as well have this be a contest in and of itself. You would simply judge the game engines based on rules compliance, elegance and speed (or whatever).
Obviously, the game engine contest could encompass more sophisticated PHP programming (which might entice the more advanced crowd to join in). For instance, maybe part of the game engine contest would be to write a socket listening server and provide the PHP client library to connect to it. Then, the beginner PHP programmer would just implement the library in their robot client and all the messaging would have already been programmed by the game engine designer.
I guess one of my basic thoughts about this contest is that it's pretty cheesy. The game itself (as described), doesn't really push my buttons too much (woo, drop a sticky turd and teleport trap near my flag, run like hell for opponents flag). I would suggest at least using some techniques in your contests that push people to try new things. The game messaging (plain text files) is quite boring, elementary and just plain wrong from a good architecture point of view.
I think the only reason most coders get involved with coding contests is to try new things and learn new techniques along the way. There's no way I would spend volunteer time on a coding project just to do the things I've already mastered.
Sorry for the verbose posting. Keep up the good work and I look forward to any replies to this.
Adam
|