Author: Joshua Lee
Date: 10:52:42 08/15/01
Go up one level in this thread
On August 15, 2001 at 12:26:31, Bruce Moreland wrote: >On August 15, 2001 at 11:39:38, Joshua Lee wrote: > >>Nothing because Computers Can't! >> >>But there are some positions with the incorrect solution As i said before. > >There are 879 positions in the ECM suite, and mine finds over 500 in less than >one second. There is also "Brilliant Ways To Checkmate", almost all of which >will be found in under a second. That's an easy suite, but could any human do >the whole book (1001 positions) in under five or ten minutes? > >My point here is that arguments of the form, "Computers are not GM's because >they can't solve *this* position in tournament time", are unsound. They can >solve many other hard problems very quickly. An argument that states that >computers must be stronger than humans because they can solve these problems >more quickly than humans is just as sound as your argument -- meaning not sound. > >GM strength is based upon something. I suggest that it's based upon ability to >score points in chess games, rather than the ability to solve some small >fraction of tactical problems in tournament time. > >bruce Good Point there with the Tactical Problems (there still are many that are not solved in Tournament Time Control Other than the Nolot Positions. But Tactical Problems are Calculating which Computers do Best There are still Positions in the Endgame that will take a comp an extreemly long time just pick up a copy of Shereshevsky's Endgame Strategy and you will see that all the 5 and 6man endgame Tablebases in the world will not make the Comp as Good as Some Players in the Endgame Also Look at the Positional Test Set's Bratko-Kopec , Think Like a GM, Test Your Positional Play Etc. I am sure you will see exactly what you're program Doesn't know. As i said before you made a good point about "Tactical" Problems , The same Holds True for alot of the Endgame Problems ECE, MES. I have a problem with the Norms computers have made Deep Junior Had very fast 4.13Ghz and earned 2702 while Chess Tiger Get's it's norm on a 866Mhz Computer well the difference in hardware should have the two separated by around 120 Also Deep Junior Played GM's who were top 100 in the world while Chess Tiger Played against Some GM's the Average was under 2580 in fact 2467 so not even IM Strength. I've Read that a GM or IM Should win 3 out of 5 against a 2400 or 2500 (i wish i remembered where i saw this) which comes out to 2480 or 2580 being IM or GM So while CT played against people with both titles it only had average oposition of near IM Strength. I personally view DJ performance as more impressive , Both are Great Programs and should be expected to perform like that on a 4Ghz machine Sorry about getting off the topic.... Thankyou
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.