Author: Jonathan Lee
Date: 11:07:08 10/18/99
Go up one level in this thread
On October 17, 1999 at 20:19:13, leonid wrote: >Hi! > >I was very surprised when I found that in move ordering putting the checking >moves at the head of the moves chain is something unusual. It can be. I wrote me >logic alone and maybe in a wrong way. At least, somebody explained me why this >way is so wrong. I went to my game and tryed to find what is true. Disconnected >in 2 plys from 8 plys deposition of the checking moves at the head of the moves >chain. Speed of the logic went down between 10 to 40%. So putting the moves that >make check at the head of the chain have more that simple good sense. > >Leonid. I have 3 reasons why checking moves first are unneccessary or no difference. 1. (queen-king vs. king) or (rook-king vs. king) don't need checks just checkmate. 2. I have Genius at home and the high selective depth (12) will find (all short run) "forced moves", checks or no checks. 3. Computers must be absolute skeptics (in the short run) because there are "exceptions to uniformity". The efficiency in the short run will decide it's overall tactics. When I mean "the short run", it is the finite number of moves found per finite time. I hope this clears up the matter. It seems to me that the latest 1999 machines are beating much more than 99 percent of humans. Jonathan
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.