Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: transposition tables: recognising repetitive positions

Author: José Carlos

Date: 00:13:18 08/10/01

Go up one level in this thread


On August 09, 2001 at 12:36:35, TEERAPONG TOVIRAT wrote:

>>
>>  I don't use the TT to detect repetitions. Instead, I check in the move list,
>>looking for the hash key of the actual position on the board.
>>  And I don't store draw-by-repetition scores in the TT, because they are messy
>>to handle.
>
>
>Hi,
>
>Accidentally, this post lead to the same question as mines below.
>This is in my mind. I plan to avoid storing draw by repetition scores
>in hashtable. Did you run some tests about this ? How about the result?
>Please,let me know.
>
>Thanks,
>Teerapong

  From reading Miguel's post, I think I dindn't express correctly (sorry). What
I meant is that I do not store the position where the repetition is detected,
but I hash the moves that lead to it. I do this because it's (i think)
impossible to know if the value returned back along the path is caused _only_ by
the repetition, is _affected_ by it, or has nothing to do with it.
  For example, suppose you have an endgame position where many searchs lead to
draw by insuf material. Other lines can be draws by stalemate. You can't know
where the 0 value comes from.
  But it's even worse. Suppose you have a line where there are some mate
threats. You reach a position where you cut the search due to a draw by
repetition, and return the 0 value. Then you say "I will ignore it, because in
case of no repetition, black is mated", but it is possible that the repetition
can not be avoided in that path due to previous errors (or simply bad move
ordering). Then you miss a line where you can draw.
  It's possible to imagine many of these examples. At the end, I think, it's
better to close your eyes and ignore all of that.

  José C.



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.