Article View: gnu.chess
Article #4808Re: KILLT and AGING
From: Kong Sian
Date: Tue, 27 Sep 1994 10:28
Date: Tue, 27 Sep 1994 10:28
24 lines
1618 bytes
1618 bytes
> > Is pl73 better with or without KILLT and AGING defined? > > R. Luebbert > luebberw@lp.musc.edu > Defining KILLT is always good. The KILLT heuristics is actually a move ordering heuristics. It favors moves towards the centre and will try these out first. The effect of defining KILLT is to reduce the size of the search tree without any loss of move quality. I am however, unfamiliar with the AGING algorithm. I recently talked to Roger Uzun who told me that the AGING implementation in Gnuchess isn't correctly implemented. He has been exchanging lots of e-mail with Dan Oetting (who implemented the AGING feature), and they have agreed on a better way of doing it. Roger has sent me their version of AGING which I am currently trying to understand. I'll try and send a mail to Dan to see if he has any input (perhaps he might read this message). For the time being, I guess you might like to leave AGING out. Kong Sian
Technical Details
Message-ID:
<9409271025.AA04888@aba.nsrc.nus.sg>Path:
arc1-ca.pugleaf.net!bga.com!news.sprintlink.net!redstone.interpath.net!ddsw1!godot.cc.duq.edu!toads.pgh.pa.us!newsfeed.pitt.edu!uunet!cis.ohio-state.edu!nsrc.nus.SG!kongsianReferences:
<009850DF.C668B660.32400@lp.musc.edu>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 10d71d,e2119e0d023084d0 X-Google-Attributes: gid10d71d,public X-Google-ArrivalTime: 1994-09-28 00:11:51 PST Path: arc1-ca.pugleaf.net!bga.com!news.sprintlink.net!redstone.interpath.net!ddsw1!godot.cc.duq.edu!toads.pgh.pa.us!newsfeed.pitt.edu!uunet!cis.ohio-state.edu!nsrc.nus.SG!kongsian From: kongsian@nsrc.nus.SG (Kong Sian) Newsgroups: gnu.chess Subject: Re: KILLT and AGING Date: 27 Sep 1994 06:28:29 -0400 Organization: GNUs Not Usenet Lines: 25 Sender: daemon@cis.ohio-state.edu Distribution: gnu Message-ID: <9409271025.AA04888@aba.nsrc.nus.sg> References: <009850DF.C668B660.32400@lp.musc.edu>