Author: Uri Blass
Date: 13:52:16 03/01/04
Go up one level in this thread
On March 01, 2004 at 16:25:12, Stephen Ham wrote: >On March 01, 2004 at 16:11:01, Tansel Turgut wrote: > >>Dear Stephen, nice to hear from you! >> >>I just ordered Shredder 8. >> >>I tried the triple brain with Shredder 5.32, but didn't like it..Overall I >>believe that Shredder is a very good program. >> >>I don't like Fritz though... >> >>Ken frey likes DPA..I tried a little and didn't like it though.. >> >>best regards, >>Tansel > >Dear Tansel, > >It's a pleasure to chat with you again! > >Yes, it was indeed largely the comments of Kenneth Frey that caused me to >experiment with Deep Position Analysis. Who knows what he does with the data >from the final result? My one experience was simply confusing. But maybe this is >indeed more a function Shredder tha DPA itself. Perhaps if I'd chosen another CB >engine (I have F7 and H9) I might have more meaningful data. Perhaps I'll find >time to repeat the experiment this weekend. > >Do you use DPA in your correspondence games, Tansel? Do you find more meaningful >results than merely using the Infinite Analysis mode for extended periods? Do >you check your DPA results against the IA mode for accuracy? Do you think a lot >of CC players use DPA? I wonder if Uri Blass uses it? > >All the best, >Stephen I use deep position analysis with Fritz and not with shredder. The problem is that part of the evaluation of shredder is dependent on the root position so comparing the numbers that it has at different positions after search may be misleading. Fritz usually does not behave in that way so comparing results in different positions after search is not misleading. problem of wrong evaluation is no big problem if the program gives better score for better moves. Uri
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.