Author: Stuart Cracraft
Date: 15:34:44 09/30/04
Go up one level in this thread
On September 30, 2004 at 18:14:23, Robert Hyatt wrote: >On September 30, 2004 at 15:58:34, Stuart Cracraft wrote: > >>On September 30, 2004 at 12:06:53, Robert Hyatt wrote: >> >>>On September 30, 2004 at 05:00:31, Uri Blass wrote: >>> >>>>On September 30, 2004 at 04:52:12, martin fierz wrote: >>>> >>>>>On September 30, 2004 at 00:46:26, Uri Blass wrote: >>>>> >>>>>>On September 29, 2004 at 23:52:15, Stuart Cracraft wrote: >>>>>> >>>>>>>[D] 4r1k1/p1qr1p2/2pb1Bp1/1p5p/3P1n1R/1B3P2/PP3PK1/2Q4R w - - bm Qxf4; >>>>>>> >>>>>>>In this position I had everything turned on and got the solution >>>>>>>in a little more than 1 1/2 minutes: >>>>>>> >>>>>>> 1/11 g2f1 0.01 -953 945 g2f1 f4d5 >>>>>>> g2f1 f4d5 >>>>>>> 2/12 g2f1 0.01 -953 1644 >>>>>>> g2f1 f4d5 c1g5 >>>>>>> 3/12 g2f1 0.02 -953 5064 >>>>>>> g2f1 f4d5 c1g5 d5f6 >>>>>>> 4/20 g2f1 0.09 -953 20655 >>>>>>> g2f1 f4d5 b3d5 c6d5 c1c7 d6c7 f1g1 >>>>>>> 5/22 g2f1 0.65 -953 168943 >>>>>>> g2f1 b5b4 b3a4 f4d5 f6g5 d5e7 >>>>>>> 6/26 g2f1 2.59 -953 620310 >>>>>>> g2f1 b5b4 mtmt >>>>>>> 7/32> g2f1 60.58 -552 14192153 g2f1 e8c8 c1b1 f4d5 b1e4 d6b4 f6e5 c7b6 >>>>>>> g2f1 e8c8 c1b1 f4d5 b1e4 d6b4 >>>>>>> 7/34 c1f4 99.42 5113 24537823 c1f4 e8e6 f4g5 d7e7 b3e6 e7e6 h1d1 d6e7 f6e7 >>>>>>> c1f4 e8e6 f4g5 d7e7 b3e6 e7e6 h1d1 >>>>>>> >>>>>>> >>>>>>>I turned off null move (R=2) and got the solution in about 11 seconds: >>>>>>> >>>>>>>Alpha=-1332 Beta=-531 Maxdepth=9999999 MaxTime=99999999 >>>>>>> 1/11 g2f1 0.01 -953 908 g2f1 f4d5 >>>>>>> g2f1 f4d5 >>>>>>> 2/12 g2f1 0.01 -953 1565 >>>>>>> g2f1 f4d5 c1g5 >>>>>>> 3/14 g2f1 0.07 -953 20084 >>>>>>> g2f1 f4d5 c1g5 d5f6 >>>>>>> 4/22 g2f1 0.60 -953 131543 >>>>>>> g2f1 f4d5 b3d5 c6d5 c1c7 d6c7 f1g1 >>>>>>> 5/26>g2f1 6.80 -552 1607444 >>>>>>> g2f1 b5b4 b3a4 f4d5 f6g5 d5e7 >>>>>>> 5/36 c1f4 10.70 2260 2466497 c1f4 d6f4 h4h5 g6h5 h1h5 f4h6 h5h6 c7g3 g2g3 d7d6 >>>>>>> >>>>>>> c1f4 d6f4 h4h5 g6h5 h1h5 f4h6 h5h6 c7g3 g2g3 d7d >>>>>>> >>>>>>>So now my question is, would it make sense to consider an idea of >>>>>>>disabling null move under additional circumstances if those >>>>>>>circumstances can be identified. >>>>>>> >>>>>>> endgame >>>>>>> side to move in check >>>>>>> inside principal variation >>>>>>> last move a null move >>>>>>> >>>>>>>These are the ones I disable for -- I don't disable null move for >>>>>>>any material-related or alpha/beta related measures but perhaps >>>>>>>I should. Are any in common use? >>>>>>> >>>>>>>Stuart >>>>>> >>>>>>It is better to disable your recapture extensions. >>>>>>The problem here is not null move pruning and null move pruning willnot change >>>>>>the depth that you solve the problem when you will implement correctly checks in >>>>>>the qsearch. >>>>>> >>>>>>The target of chess programs is not to solve 141 faster but to play better. >>>>>>I am almost sure that if only the recapture extension save you many plies in 141 >>>>>>then you implement it in a way that is counter productive for games. >>>>>> >>>>>> >>>>>>I wonder how do you get depth 5 without null move pruning >>>>>> >>>>>>Can you post the remaining depth after every move in the line >>>>>>1.Qxf4 Bxf4 2.Rxh5 gxh5 3.Rxh5 Bh6 4.Rxh6 Rd6 Rh8# >>>>>> >>>>>>There are no checks in this line except the mate and you do checks in the >>>>>>qsearch so without recapture extensions you need at least 9 plies for it(after 8 >>>>>>plies you enter qsearch after Rd6 and do not find the move Rh8 mate. >>>>>> >>>>>>If recapture extensions lead to 4 plies extension then it seems clear that your >>>>>>new program will be clearly weaker in games. >>>>>> >>>>>>I suspect that you extend 4 plies in this line because you extend 1 ply every >>>>>>time that the last 2 moves are captures. >>>>>> >>>>>>It is clearly a bad idea to do it and it is clear that even with more reasonable >>>>>>restriction of the recapture extension it is not productive for a lot of >>>>>>programs. >>>>>> >>>>>>I suggest that you test the version with the recapture extension in games >>>>>>against the version without the recapture extension if you do not believe me >>>>>>that your recapture extension is a bad idea. >>>>>> >>>>>> >>>>>>Uri >>>>> >>>>>hi uri, >>>>> >>>>>where do you see recapture extensions influencing WAC.141? the main line has no >>>>>single "recapture" if you define it as "capture of a piece of the same value, on >>>>>the same square, on two successive plies". i think that is the normal definition >>>>>of recapture. >>>>> >>>>>what am i missing? >>>>> >>>>>cheers >>>>> martin >>>> >>>>It seems that >>>>Stuart Cracraft defines it as 2 consecutive captures. >>>> >>>>He probably extend the line that I showed by 4 plies for that reason: >>>> >>>>1.Qxf4 Bxf4 =>extension >>>>1...Bxf4 2.Rxh5=>extension >>>>2.Rxh5 gxh5=>extension >>>>2...gxh5 3.Rxh5=>extension >>>> >>>>I tried to understand how can he get Wac141 at depth 5 without null move pruning >>>>when he is using no checks in the qsearch and no mate threat extensions and this >>>>was my conclusion and based on his response I understand that my conclusion was >>>>correct. >>>> >>>>Uri >>> >>> >>>If he really does that (IE N consecutive captures triggers N-1 extensions) it is >>>broken and will hurt performance no matter what due to over-extending on >>>sequences of captures, which results in drastically reduced search depth on >>>other lines. >> >>I thought I was supposed to extend on a recapture because it is obvious! >> >>How would you (Uri and Bob) extend for this sequence? >> >>>>1.Qxf4 Bxf4 =>extension >>>>1...Bxf4 2.Rxh5=>extension >>>>2.Rxh5 gxh5=>extension >>>>2...gxh5 3.Rxh5=>extension > > >I am not sure I would extend _any_ of those. The idea behind the recapture >extension is to prevent a simple capture from being used to hide a threat that >can be pushed over the horizon and out of sight. I only extend when the >capture/recapture are equal pieces. IE Qxf4 Bxf4 is not a simple recapture, as >the material score changes. My requirements for recapture are > >(a) two consecutive captures on the same square; >(b) the first capture and the second capture are "even trades" such as BxN NxB, >or RxR QxR, or PxP NxP, etc. the Qf4 move can't start a recapture sequence >since BxQ is not an equal trade, black wins material. >(c) I didn't trigger the recapture extension at the previous ply which avoids >over-extending on a series such as BxN BxB NxB NxN. I only want _two_ >extensions there, not three... Thanks -- I can do a+b fairly easily. To try layering on c I'll need to keep track of previous extensions at previous plies which I don't do. All decisions about extending are made without reference to previous plies. That's a mistake. I need to keep arrays like recapext[MAXPLY], checkext[MAXPLY], etc. and set the levels accordingly and to prevent extension ply after ply after ply.
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.