Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fruit 2.2.1 tablebase mystery - may be a bug?

Author: William Penn

Date: 09:19:01 11/05/05

Go up one level in this thread


On November 05, 2005 at 08:35:08, Kurt Utzinger wrote:

>On November 05, 2005 at 07:55:12, Vincent Lejeune wrote:
>
>>That confirm what I saw one time : I configured F2.2.1 to play on playchess.com
>>and when coming in endgame ( TB wich I had on my disk) I saw no Tb hits !
>>
>>I stopped and restart with same configuration and TB works !
>>
>>Very likely a bug ...
>>
>
>      Hi Vincent
>      After testing this and other positions
>      I can confirm this behaviour of Fruit 2.2.1
>      under all GUI's. First analysis will always
>      be OK, but after stopping and again starting,
>      no TB hits for a very long time and very
>      low search depth.
>      Regards
>      Kurt

I haven't seen that problem. For example:

[d]5b2/8/6p1/3k4/6K1/8/8/8 w - - 0 1

Segment of the analysis by Fruit 2.2.1 UCI in CB GUI:
49.Kg5
  =  (0.00)   Depth: 7/3   00:00:00, tb=5
49.Kg5
  =  (0.00)   Depth: 8/3   00:00:00, tb=6
49.Kg5
  =  (0.00)   Depth: 9/3   00:00:00, tb=7
49.Kg5
  =  (0.00)   Depth: 10/3   00:00:00, tb=8
... etc

stopping then restarting the analysis still gives tablebase access OK

Segment of the analysis by Fruit 2.2.1 in Shredder Classic GUI:
  1/01	 0:00 	-4.41 	49.Kf3 (2)
  2/03	 0:00 	  -M14	49.Kf3 (29)
  2/03	 0:00 	 0.00 	49.Kg5 (35)
  3/03	 0:00 	 0.00 	49.Kg5 (45)  TB:2
  4/03	 0:00 	 0.00 	49.Kg5 (56)  TB:3
  5/03	 0:00 	 0.00 	49.Kg5 (67)  TB:4
  6/03	 0:00 	 0.00 	49.Kg5 (78)  TB:5
  7/03	 0:00 	 0.00 	49.Kg5 (89)  TB:6
... etc

again, stopping then restarting the analysis still gives tablebase access OK

So maybe it only happens when there are more pieces on the board(?).
WP



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.