Author: Roy Eassa
Date: 16:18:43 09/03/01
Go up one level in this thread
On September 03, 2001 at 18:59:29, Christian Kongsted wrote: >For me it is interesting to hear about your correspondance play because I have >been playing correspondance chess myself. I dont see what satisfaction you can >get out of only picking one of the computer programs moves - dont you want to >use your own creativity? > >Sometimes when I have been playing correspondance chess, I can feel that I am >playing against a computer program. It is not that difficult to spot, it is not >very interesting either, and my opinion is that the players that trust their >computers to much wont get very far against good, positional (human) players. > >I am very impressed by the new Junior 7, which I consider the best program >commercially available, At the moment, in the SSDF test, Junior 7 is being soundly defeated by Gambit Tiger 2.0 by the score of 17.5 to 11.5. I know it isn't over until the fat lady sings, but there is decent evidence that Junior 7 is NOT the best program commercially available. >but this program also has problems sometimes, f.ex. to >hit the right plan in a closed Kings Indian/closed French (or positions like >that), and endings that require not only calculation, but also planning and >understanding. Tactically, it is simply amazing, I have to admit that... > >My impression is that the strongest players are those that use their own >intuition to find the right moves, and after that - check his analysis with a >program. > >If correspondance chess is only a matter of picking which program that is >playing, and who lets the computer think the most time, then I think the idea is >dead (sorry). But I dont think it is - yet. > >I have been playing a few over the board-tournaments lately and used different >computer programs to analyse the games afterwards (Junior 7, Shredder 5.32 and >sometimes Hiarcs 7.32, I also have Fritz 6 but dont use it much) > >It was not a surprise for me that they were able to spot many tactical >opportunities, but in some positions they really dont get what is going on >positionally, even if they get a lot of time. Of course I cant state this as an >objective fact - I just believe that if I play against an IM or GM and we use a >lot of time and have a common understanding about the positional factors in the >positions, then there has to some truth in it... > >In the Deep Blue-Kasparov match it is obvious to many chess players that the >computer made some complete nonsense and tempoloss moves in game 1 of the >97-match. The program was _extremely strong_ and played fantastic chess in some >of the other games, but it complete failed to understand the positional aspects >in the early phase of game 1. Therefore I wouldn't trust it, and I wont blindly >trust todays programs, even though I know considerable impromevements have been >made in positions understanding > >The strength of the program vary to much, depending on the position they are set >to analyse. They can play like 2100 in one position and 2800 in another, and >this is what Deep Blue did as well...
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.