Author: Pete Galati
Date: 16:04:12 01/16/01
Go up one level in this thread
On January 16, 2001 at 17:26:23, Uri Blass wrote: >On January 16, 2001 at 15:46:49, Pete Galati wrote: > >>On January 16, 2001 at 13:53:05, Dann Corbit wrote: >> >>>On January 16, 2001 at 10:20:06, Pete Galati wrote: >>>[snip] >>>>Cool, that has some valuable info on it. How did you determine what the best >>>>moves were and what the scores for various moves should be? Last night before I >>>>dozed off, I ran Comet on position 4, and Comet said b4 at zero seconds, and >>>>stayed with that for 7 hours, never changed it's mind even once. You have that >>>>listed as a score of 7 fwiw. >>> >>>I didn't determine a thing. The positions were from a magazine article where a >>>group of GM's decided what the values should be for each choice. >> >>Ok. I like that idea better than a bunch of programs doing the deciding. I just >>installed another 64mb mem into this disposable lighter computer, so I'll have >>to run position 4 again and see if the results change. >> >>Decided to order the mem over the phone because I went back to the chain >>computer store where the puter came from, and they didn't have any in stock and >>I was too stupid to call first and check. It's cheaper to order by phone or >>online than it is to feed the V8. >> >>Pete > >I totally dislike the idea that some GM's decide about the scores and the moves. > >I have a bad opinion about GM's analysis and I believe that the move of >computers after long analysis is usually better. > >The main problem is the fact that GM's have no motivation to be correct in >analysis because they do not get more money or better rating for better >analysis. > >I found in the past tactical mistakes in GM's analysis because they did not use >computers to help them and it proved that they do not do serious analysis in >tactical positions. I suppose the ideal thing would be to have the GMs with their insight deside what are good moves and then double check with strong programs, but like we often see, there's often a difference in opinion between the strong programs. The problem with relying too heavilly on programs is that they have no insight whatsoever (a mild exageration) > >It leads me to suspect that they do not do serious analysis also in positional >problems. > >I am also not sure if GM's are responsible for all the analysis and I suspect >that part of the analysis was done by weaker players. > >Uri I'm far too weak a Chess player to judge that. Pete
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.