Author: Zheng Zhixian
Date: 06:41:46 10/27/05
Go up one level in this thread
On October 27, 2005 at 05:30:59, Uri Blass wrote: >On October 27, 2005 at 04:41:49, Alessandro Damiani wrote: > >>On October 27, 2005 at 04:39:22, Uri Blass wrote: >> >>>suppose that I give you 100 random positions from games. >>> >>>How much do you need to calculate fruit2.1's static evaluation of all positions >>>with no computer help. >>> >>>Uri >> >>Is this the first step to distributed human search engines? :-) >> >>Alessandro > >I see that I forgot the word fime in the first post. What does 'fime' mean? >I meant how much time do you need for calculating fruit's evaluation with no >computer help. > >Note that I doubt if humans can complete even one ply search with fruit's >evaluation at tournament time control without losing on time unless they play >correspondence games(but maybe I am wrong and I know that some human can >multiply big numbers very fast so maybe it is possible that some humans can do >it. Sounds silly, why would humans want to do exactly what computers do for evalution? I suppose when people say fruit's evalution is simple, they mean simple compared to other programs. In fact I suspect while a human grandmaster cannot tell you why he decides to evalute certain features more highly than others, as a result it looks like he has very "simple" evalution compared to a computer. in fact behind it is a very complicated subconcious calculation, pattern matching etc that is not available to him directly. And of course simple doesn't necessarily mean inaccurate. >I think that talented humans can complete one ply search and play in tournaments >like a computer but only with a simpler evaluation than fruit's evaluation. It would probably have to be a computer programmer. And it would have to be pretty simple. Material only sure. Let's see probably have to keep track of mobility , piece square tables? >It will be interesting to know what rating can the best humans achieve against >humans when they are forced to play like a computer with definitive algorithm >(of course their oppoents should not know the algorithm because playing like a >computer is enough disadvantage) Sounds like you want a human to play a full game by acting as a computer? If the former such a test did happen (though the human knew it was playing a simulation of a chess program). In fact , I seen to recall reading that Alan Turing did it by hand simulating a simple chess program back before there werent real computers that could run the program. I think he did more than 1 ply of course. Way simpler than modern chess programs of course, but it was complicated enough that he made several mistakes following his algo and he had to often redo it. If you restrict it to one ply, it would simplify things a lot. Then the question would be if a computer is doing only one ply searches (no qsearch, no extensions, just one ply), can you beat it? I should think so, I doubt evalution no matter how superior can avoid tactical mistakes if you are limited to 1 ply (without q search). > >When I said the best humans I do not mean to the best chess players but to the >best players in tournament when every human is forced to play like a computer >and if after the game it turned out that the player did a mistake in >implementing the algorithm that he decided to use he gets a loss(of course good >algorithm can say to play checkmate if it is possible and this part is easy for >chess players so there is not going to be mistakes of missing mate in 1). I don't see what the point of the test really. All it shows is that some chessplayers are better at programming and/or following logical steps and keeping track of the resulting data. It probably be so mechanical, it wouldn't feel much, if at all like playing chess. >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.