Author: Albert Bertilsson
Date: 13:11:23 12/15/03
Go up one level in this thread
On December 15, 2003 at 15:54:44, Dann Corbit wrote: >On December 15, 2003 at 15:48:50, Albert Bertilsson wrote: > >>On December 15, 2003 at 15:41:27, Dann Corbit wrote: >> >>>On December 15, 2003 at 15:37:01, Albert Bertilsson wrote: >>> >>>>Hi great counters! >>>> >>>>Today the project progressed some 20%, dispite being in the toughest phase, from >>>>now on I guess it is downhill! Hopefully the project will be completed in two >>>>days. >>>> >>>>Of yesterdays new participants sfarrel (Scott Farell) proved to be a rocket, >>>>already has he completed close to 200 problems. >>>> >>>>A warm welcome to those who are new and as a note to Benny, you do know how to >>>>pick your problems (just completed monster problem that took a freaky 15 hours!) >>>> >>>>As the project have more participants it is getting a very nice pace, not as >>>>dependent on a few particular members any more. Of course even more participants >>>>would be nice, to really have a go at perft(12) over the Christmas holiday. >>>> >>>>So tell your friends to follow this link: >>>>http://www.albert.nu/programs/dperft >>>> >>> >>>Suggestion for another project: >>>Perform the same perft search, but spit out every checkmate and draw. >>>It would be useful theoretically to trim branches of the tree. >> >>Please explain in further detail what to use it for. > >I don't know what I might use it for. What I am thinking is that it may close >off whole branches of the tree. So I can know (for instance) there is no use >analyzing some particular position. It is, for all intents and purposes, dead. > Draws are not possible, as I have no draw detection in the perft code. Checkmates are detected (obviously for it to work) but I'm not sure about how to solve the communication problem (telling the server and storing it). I'll think about it. /Regards Albert > >>And what should be >>outputed. > >An EPD record. > >>Are there any draws that early? > >I am guessing that there are 3-move repeats with the king side open. Some are >probably blunders (a draw by ignorance), but some are probably forced (every >move is forced by one side).
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.