Author: Dann Corbit
Date: 12:18:44 05/04/01
Go up one level in this thread
On May 04, 2001 at 15:04:23, Miguel A. Ballicora wrote: >On May 04, 2001 at 14:39:22, Dann Corbit wrote: > >>On May 04, 2001 at 14:22:23, Gian-Carlo Pascutto wrote: >> >>>On May 04, 2001 at 13:20:51, Dann Corbit wrote: >>> >>>>Prophecy: >>>>You will win the next WCCC[*] >>> >>>No Billion NPS will save the game if you come out of book -1 or -2 >>>Not having a 'cooked' book like the commercials have should be >>>quite a disadvantage. >>> >>>Alternatively you could use a narrow book with high thresholds and >>>let the hardware figure out the rest. Now _that_ could provide for >>>some interesting stuff. Especially if you get singular extensions >>>working again. >> >>Crafty's computer book is tiny. And well weeded. I think with that machine, >>you might not need one. I know it is a stupid thing to say, but I would suspect >>you might be ahead to simply do away with it. > >With that speed I would try to take out of book the computer opponent >very early creating a book with openings like 1.b3, 1.b4 etc, that are not bad >but not heavily analyzed. Then, I will see who does better in the opening with >no book with such a difference in speed. 1. b4 is my favorite opening. You might be surprised how well analyzed it is. Have you seen Marek's 1. b4 encyclopedia? http://www.algonet.se/~marek/b4.htm The first CAP project was a detailed analysis of 1. b4 My favorite game in the last Paderborn contest was the Quark Orangutan game against Zchess! What a beauty!!! >Other chance is to go to semiclosed positions where a small disadvantage is >never decisive, particularly against a computer that you are outsearching. >Deep Blue Junior losing against Fritz should be an example. Going to, or >allowing the Lasker-Pelikan (a.k.a. Sveshnikov) variation in the Sicilian is >silly. When you are out of book, you can be winning of losing. That even happen >to GMs! >That's even true for human play if you are a better player, you don't wanna go >for a theory fight unless you are also better in opening preparation >(Polugaievsky, Kasparov etc. could be examples) With horsepower like the Alpha machine, I think careful testing would be needed to find any sort of workable strategy. Which is why I would self-test only, if I were behind the wheel.
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.