Author: leonid
Date: 19:26:22 02/14/00
Go up one level in this thread
On February 14, 2000 at 17:06:28, leonid wrote: >On February 14, 2000 at 10:48:55, Heiner Marxen wrote: > >>On February 13, 2000 at 20:52:57, leonid wrote: >> >>>On February 13, 2000 at 09:25:18, Heiner Marxen wrote: >>> >>>>On February 12, 2000 at 10:18:44, leonid wrote: >>>> >>>>>Hello! >>>>> >>>>>Please give your position that contain mate between 4 and 8 moves. Would like to >>>>>know the speed at which usually logic solve mates. I did this four years ago, >>>>>but then I was alone. Now data could be more diversified and useful for me as >>>>>for somebody else. >>>>> >>>>>Until now positions presented here were almost all the time beyond 9 moves (18 >>>>>plies). This make solution too long to find. >>>>> >>>>>Please indicate your hardware and your game. >>>>> >>>>>I will give, for start, one very simple position that contain mate in 6 moves. >>>>>White to go. >>>>> >>>>>[D]1q2kr2/rB2bbP1/3p1n2/1NpP2BN/P4P1p/1P2Q3/6PP/R3K2R w >>>> >>>>Are you sure this is the intended position? According to Chest there is >>>>no mate in 6. What is the intended main variation (PV) ? >>>> >>>>Heiner >>>> >>>>>Thanks for your next position! >>>>>Leonid. >>> >>> >>>Very Sorry!!!!! My mistake! I is good that I came to the Club. Was all the day >>>in my writing. I must for sure next time use, like you, graphics. It can reduce >>>the mistakes. >>> >>>ON G7 MUST STAY WHITE PAWN. Now it is black. >> >>Ok, I corrected the diagram above, and viola: there is a solution :-)) >>Chest on a P/133 with 10 MB hash table needs 522.44 seconds to find: >> >> Nxf6+ Kd8 Qxe7+ Kxe7 Nd7+ Kxd7 Bc6+ Kc8 gxf8Q+ Be8 Qxe8# >> >>For a complete solution tree it needs 571.51 seconds, and find e.g. that >>2.gxf8Q+ and 2.gxf8R+ are also solutions. >> >>>Today are so many good positions to try but it is too late for me. Will come >>>back tomorrow to try few. >>> >>>Once again my excuses, >>>Leonid. >> >>You are welcome! Thanks for the problem! It is not exactly easy for Chest. >> >>Heiner > > >Thanks for putting this in good graphics! If I will find the time will look how >long it will take for Rebel of Genius. Genius I used a lot when tried my mate >solving logic and I like this game very much. It make almost no mistake and take >any position. Rebel I bought recently and was amazed by openness of "his talking >to people" and also by his stability. I asked him to solve only few positions >for mate, some fantasy position that make other excellent game die on the spot, >but nothing happened to him. Rebel, like Genius, had no allergic reaction. > >Leonid. Just wanted to find the time of the last position on Rebel or Genius but did something else. Reached one that I tried in 1995 on few games and was curious to see at what time, this time, Rebel 10 will solve the problem. Could never find final solution since had not enough time. Find only time for Rebel 10 in 5 moves (37min 51 sec) when the position is mate in 6 moves. For 6 moves I waited up to 40 minutes (AMD 400). Had no more time to wait. Position is good to try the stability of the logic. I don't give the time for my logic, when position is created on my "quick mate solving logic". This could give some destorted image of my game speed. Could appear as too quick. When I tried my logic for speed I used the positions from different books and the positions created on other games. /8/8/2Q1n3/1Q1nqn2/KQnq1q1k/1Q2q3/8/1Q6/ Black to start and mate in 6 moves. Leonid.
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.