Author: Eric Jensen
Date: 20:51:25 08/24/04
Go up one level in this thread
On August 24, 2004 at 03:39:28, Uri Blass wrote: >On August 24, 2004 at 00:23:38, Eric Jensen wrote: > >>I have recently come across the following position, >> >>/r5k/pp2qppp/2b/n/2pPr/P1P1PQ1P/1B4P/1B2RRK/ >> >>with white to move. It will be white's 21st move. I have let crafty run and >>analyze this position for over 2 hours, and it returns a value of over +2 for >>white. However, I don't think that crafty understands the position, because >>I've had it sort of play against itself (I was using SCID for analysis, and >>pulled up the engine analysis with crafty) by making it's best move for each >>side after about a min or so each time, and it routinely wins with black. >> >>I would be interested to see a Deep Position evaluation of hte position from >>Shredder or Junior or something Stronger than Crafty, but also I would be >>interested in seeing games from a short engine match with this start position, >>just to see if the evaluations can be trusted. I don't really have the software >>to pull this off on my own, but if it piques anyone's interest and they'd like >>to post a follow-up, please do! > > >When you have FEN you need to have 8 places in every rank on the board and I see >that your ranks have less places. > >I guess that you mean > >[D]r5k1/pp2qppp/2b5/n7/2pPr3/P1P1PQ1P/1B4P1/1B2RRK1 w - - 0 1 > >I got this position simply by adding the number of empty places at the end of >every rank. > >I wonder why you omit the last number in every rank. > >Uri Oh, sorry, just because crafty doesn't recquire the last number.
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.