Author: Dieter Buerssner
Date: 15:58:13 11/10/03
Go up one level in this thread
On November 10, 2003 at 17:37:22, martin fierz wrote: >i use a similar approach in my checkers program (hashtable etc.). the problem >for my chess program is that this means i have to write a PGN-parser to read the >lines i enter in chessbase (or the games i select from chessbase). and i just >hoped i could do without that ;-) Hey Martin, don't be that lazy! With your experience, it will take you less than a day, to code a decent PGN-parser. Of course, you could implement (if you have not done it already - perhaps also a day or less of effort) the UCI protocol, and be happy with the CB-interface. The GUI would have no problem, to use the lines you "enter(ed) in Chessbase". And your engine would use those lines. But it is certainly less fun. Much of the enjoyment I had while writing my engine, was because of the "native" book code inside. I got feedback about it (especially by Carlos Pesce, Mogens Larsen and Jan Kan), implemented some of those ideas, got more feedback, etc. I even implemented some rather obscure functions, that convert some PGN-like input to another form (say with coordinate notation), perhaps appending to white moves !!, and similar things. I guess, that Carlos only used this to create books for the possible opponents. Still, I liked to do it, and enjoyed it. Also, it was a nice task for me. Just clearly defined - much different to when you want to try some new heuristics in search or evaluation. The later may be very difficult to test and typically needs much more time (even when coding is easy). Cheers, Dieter
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.