Author: leonid
Date: 09:37:08 02/10/01
Go up one level in this thread
On February 10, 2001 at 12:20:23, Enrique Irazoqui wrote: >On February 10, 2001 at 11:53:07, leonid wrote: > >>On February 10, 2001 at 11:17:19, Heiner Marxen wrote: >> >>>On February 10, 2001 at 10:36:39, leonid wrote: >>> >>>>On February 10, 2001 at 08:23:58, Uri Blass wrote: >>>> >>>>>On February 10, 2001 at 07:42:35, leonid wrote: >>>>> >>>>>>Hi! >>>>>> >>>>>>If you like to solve a mate, try this one: >>>>>> >>>>>>[D]3R2K1/q3q1B1/2bQ3Q/QQ3NQ1/2NnQQB1/Qq1qq1nQ/2nqr3/2nkbRq1 w - - >>>>>> >>>>>>Please, indicate your result. >>>>>> >>>>>>Thanks, >>>>>>Leonid. >>>>> >>>>>easy? >>>> >>>>When I found this position, yesterday, it was solved by selective search 9 moves >>>>deep. I expect that shortest mate is even closer. Move that lead to mate is >>>>checking move. This is why I wrote "easy". >>>> >>>>Leonid. >>> >>>There is no mate in 8 according to Chest (55 mins K7/600, 350MB hash). >> >>Thanks! Now it is for sure 9 moves position. >> >>Leonid. > >On a mobile P750 with 128MB hash, Junior 7 beta finds a mate in 9 in 2 minutes. >Junior is good at solving this sort of freak mates, while Deep Fritz refuses to >compute them and crashes. > >Enrique If you will refresh my memory about Junior program, will be thankful. Who wrote? What language? For what system? How he solve mate positions? Special engin for this? Choice between selective and brute force (complete) search? Thanks, Leonid. > >New position >3R2K1/q3q1B1/2bQ3Q/QQ3NQ1/2NnQQB1/Qq1qq1nQ/2nqr3/2nkbRq1 w - - 0 1 > >Analysis by Junior 7 beta: > >1.Qfxg3 Qdxc4+ 2.Qxc4 Qxc4+ 3.Qad5 Q3xg5 4.Qexe7 > +- (9.87) Depth: 3 00:00:00 341kN >1.Qexc6 Qf7+ 2.Kh7 Qxf4 3.Nxd4 > +- (11.59) Depth: 3 00:00:00 469kN >1.Qxd3 Qf7+ 2.Kh7 Qexd3 3.Qaxd2+ > +- (16.58) Depth: 3 00:00:00 718kN >1.Qxd2+ Qdxd2 2.Nfxe3+ Qdxe3 3.Bxe2+ Kxe2 4.Qexe3+ > +- (22.12) Depth: 3 00:00:00 975kN >1.Nfxe3+ Q2xe3 2.Bxe2+ Kxe2 3.Qxe1+ Nxe1 4.Qfxe3+ > +- (22.50) Depth: 3 00:00:00 1235kN >1.Nfxe3+! > +- (22.80) Depth: 6 00:00:01 2392kN >1.Nfxe3+! Qgxe3 2.Qbxb3 Qxc4+ 3.Qxc4 Qxg5 4.Qxd2+ Qxd2 5.Qhxg3 Bxe4 > +- (27.04) Depth: 6 00:00:04 6839kN >1.Nfxe3+! > +- (27.34) Depth: 9 00:00:21 33305kN >1.Nfxe3+! Qgxe3 2.Rxe1+ Qxe1 3.Qxe1+ Nxe1 4.Qbxb3+ Ndxb3 5.Bxe2+ Ncxe2 6.Qxb3+ >Kc1 7.Qexe3+ Qexe3 8.Qxe3+ Qdxe3 9.Qbd1# > +- (#9) Depth: 9 00:02:00 108950kN > >(Irazoqui, Cadaqués 10.02.2001) > > >> >>>>>I am interested to know how much time do humans need to solve it. >>>> >>>>>Please indicate your result,the tree that you generated to prove the mate and >>>>>your time. >>>>> >>>>>I am not going to do it because I am too lazy but I can give you a hint that it >>>>>is a mate in at most 9 moves so you do not need to search longer lines >>>>> >>>>>Uri >>> >>>I stared at it for some minutes, myself. I did like Nb2+ until I found >>>that the knight at c4 is pinned! Uuuh. That's computer stuff ;-) >>> >>>Heiner
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.