Author: Fernando Villegas
Date: 12:38:18 11/23/00
Go up one level in this thread
On November 23, 2000 at 12:42:19, Enrique Irazoqui wrote: >On November 23, 2000 at 12:17:50, Fernando Villegas wrote: > >>Bug, "feature", whatever the name, you does not expect that void in a 2000 year >>program. >>Fernando > >Dear Fernando, > >Expecting it or not is one thing, blaming beta testers is another. In any case, >you have too much experience with chess programs to play surprised by Tiger's >inability to mate with knight and bishop. There are many other shortcomings in >chess programs that don't belong in your "2000 year program", and much more >important than this mate, from recognizing bad bishops to being able to avoid >blockades and weaknessess in the castle. I think it's understandable if a >programmer has other priorities, and in my opinion Christophe has the right >ones, as proven by Gambit. Look, he could have solved this mate at the expense >of producing Gambit. What do you prefer? > >Enrique Dear Enrique: Chess is about to give mate and to give mate is about to handle typical mates: so, even if you ever get there, you must know about how to handle that stuff. I am sure that when you was taught about chess, they first explained you how to mate and not positional things like bad bishops, even if this last situation is more frequent and practically important. It is a matter of concepts, of basics. Any case I did not blame Christophe. I did make some question about how testers makes his test. Maybe it could be advisable for tester to be critiziced from time to time in order not not forget the elementals. Respect your question, it is fully rethoric and speculative. Oye compadre, apenas te creo lo que dices, sigamos esto en email.... Fernando
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.