Author: Robert Hyatt
Date: 06:33:04 02/13/99
Go up one level in this thread
On February 12, 1999 at 22:43:50, Dann Corbit wrote: >Let's suppose that some position has 20 legal moves. > >For each of those new positions (current position + legal move) we analyze for >one second. > >For each of these analysis results, we choose winning position and analyze it >for one second. We do this 20 times. For the 20 positions, this would take 400 >seconds. After this process, we look at the 20 end results, and choose the best >one. > >How would this compare to 400 seconds worth of analysis from the initial >position? > >I ask because of Uri's super-clever problem and Jeremiah's ability to locate the >answer once a single correct step is made. > >Comments? Thoughts? Boots aimed at my cranium? That's one way to do selective search. But you start off with 20 moves, and use 1 second searches to eliminate 19 of them. What if one of them wins, but you only see this if you search for 2 seconds? You miss this. What if the one that looks best after 1 second looks bad when you follow it and do the next one second search?
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.