Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Infinite PV's / hash extraction repetition problem

Author: Dann Corbit

Date: 20:07:19 02/24/03

Go up one level in this thread


On February 24, 2003 at 22:06:51, Robert Hyatt wrote:

>On February 24, 2003 at 20:38:31, Sune Fischer wrote:
>
>>On February 24, 2003 at 20:23:07, Robert Hyatt wrote:
>>
>>>On February 24, 2003 at 20:17:05, Sune Fischer wrote:
>>>
>>>>Hey there.
>>>>
>>>>I've dumped the triangle PV and are extracting from hash now.
>>>>I'm wondering if what's happening here is normal....
>>>>
>>>>Sometimes when it finds a repetition it seems to go in a loop creating
>>>>infinitely long PV's. What is the right way to handle this, should I cut the PV
>>>>after 1st or 2nd rep, or just let it go nuts?
>>>
>>>Think about it.  If you go thru one loop, you will go thru an infinite number.
>>>:)
>>
>>Yeah thought that was reasonable, but I couldn't figure out if maybe the depths
>>on the entries played a role here, but guess they don't.
>>
>>>Cut the PV at the _first_ repetition.  :)
>>
>>Okay, so do you print the first repeated move, as in
>>1.Qa4+ Bd7 2.Qb3 Be6 3.Qa4+ or just
>>1.Qa4+ Bd7 2.Qb3 Be6 ?
>>
>>>Or else find a way to do int pv[infinity].
>>>
>>>:)
>>
>>Hang on, I need a moment to reflect on that one.. :)
>>
>>-S.
>
>It's a matter of taste of course.  But I lop the thing off after making
>and printing the move, and then noticing it is repeated.  That "closes"
>the PV so that I can see what is happening.
>
>You have to do this if you print EGTB PVs as well, obviously.  Been there
>and debugged that problem too.  :)

How about showing the 3-fold repetition?  Obviously, there is no need to repeat
past that point.



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.