Author: Uri Blass
Date: 00:52:33 11/11/02
Go up one level in this thread
On November 10, 2002 at 18:37:02, martin fierz wrote: >On November 10, 2002 at 17:23:44, Uri Blass wrote: > >>On November 10, 2002 at 16:32:45, martin fierz wrote: >> >>>On November 10, 2002 at 11:40:28, Bob Durrett wrote: >>> >>>>What Will Happen to Fritz When an Amateur Program Becomes #1? >>>> >>>>Especially, if the Amateur program can be downloaded like Crafty? >>>> >>>>[The question has to do with Fritz's marketability.] >>>> >>>>I like Fritz. Please don't kill it. : ( >>>> >>>>Maybe the Fritz people will defend themselves by making their programs more >>>>useful to ordinary users. Sort of like CM. That would be good. : ) >>>> >>>>Bob D. >>> >>>first, fritz is already much more useful to "normal" users. for example, there >>>is no single file i know of with a strong freeware program like >>>crafty/ruffian/yace which i can just install on my PC, and which then just >>>works. you need to fiddle with command-line options of winboard, or winboard >>>adapters, which is no problem to computer-savvy users, but i think the majority >>>of computer users is not able to install these programs easily or at all. >>>second, fritz is, at the moment, clearly stronger than all those amateurs. and i >>>see no way for that to change. being able to work on your program all day >>>instead of stealing a couple of hours from wife & kids is by far too big an >>>advantage... >>> >>>aloha >>> martin >> >>You assume that the opponents are unable to work >>all day on their chess program. >> >>It is not clear. > >if an amateur is working all day on his program then i agree with you, he can >produce an engine to rival the commercials. if not, it is hard to imagine. > >>I do not know how much time amateurs give for their programs >>but I suspect that the new amateurs gives more >>time for their chess program relative to >>the old amateurs. > >right, i guess, for multiple reasons. new amateurs may be young and have no >family, and therefore, more time for programming. also, it is new and exciting. >i think new amateurs can quickly get up to a high level, because many "tricks of >the trade" are well published (e.g. crafty's source is freely available. you >don't have to clone it, just look at it for ideas). but after that, when you >have to produce original ideas to improve your engine, the going gets tough... > >aloha > martin 1)I do not think that thinking of ideas is the hard thing and the hard thing is to implement ideas without bugs. There are ideas that are so complicated that almost no programmer plan to try to implement them. For example I read that aristarch does not ponder only on the expected move but ponders about all moves when it gives different time for different moves. I think that it is a good idea but I do not plan to implement it in the near future. 2)I still have not a good implementation of hash tables. I know that it may be better to use hash tables to save generating legal moves but for me generating a fast function to check if a move is legal is not a trivial task(I do it today only by generating the list of the legal moves and checking if the move is in the list)and in that case I also need to store the number of legal moves in the hash tables because I use that information in my search rules and in my mobility evaluation so I decided to test other ideas that are simpler to implement and are only change of few lines in the source code. Uri
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.