Author: blass uri
Date: 08:50:46 05/20/00
Go up one level in this thread
On May 20, 2000 at 11:31:23, Dann Corbit wrote: >On May 20, 2000 at 04:29:16, Bruce Moreland wrote: > >>On May 19, 2000 at 23:43:50, Peter Kappler wrote: >> >>>On May 19, 2000 at 23:06:26, Jerry Adams wrote: >>> >>>> >>>> Hi >>>> >>>> I'm just looking for some opinions, I think the Question is interesting. What >>>>I am asking is can a program beat a 2200 rated master if the master is allowed 2 >>>>hours for 40 moves and the program 5 sec a move. Lets Assume the hardware is >>>>what ssdf uses AMD K62 450. >>> >>> >>>I'd expect the 2200 player to win pretty handily. >>> >>>My guess is it would get interesting at around 20-25 seconds a move. >>> >>>--Peter >> >>I never tried this with standard time controls, but I tried it with radically >>shortened thinking times in blitz games. >> >>For several months I ran Ferret on a P5/133 at maybe 1/4 second per move, with >>no pondering. It played only rated 5 0 blitz and had a pretty good record >>against IM's on ICC, with an average rating of perhaps 2300 and a peak of over >>2500. >> >>One thing is that when you move so fast, you put tremendous pressure on your >>opponent. >> >>I think these programs would crush a master at 20 seconds per move. > >THe master gets 40/2 time control. The computer gets 20 seconds. I'm not so >sure. I think you did not read the question completely. >;-) I think that he read the question but the master will play weaker when the opponent plays very fast becasue the master will have less time to think on the opponent's time. I think the 2200 player is going to lose even if the computer has only 5 seconds per move except cases when the 2200 player has experience in playing serious games against chess programs. I think that most of the 2200 players did not train by playing serious games against chess programs. 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.