Author: Uri Blass
Date: 17:40:21 09/30/05
Go up one level in this thread
On September 30, 2005 at 20:15:30, Albert Silver wrote: >On September 30, 2005 at 18:53:21, Uri Blass wrote: > >>On September 30, 2005 at 18:11:26, Steve B wrote: >> >>>The Midway point of the match >>>a rather unusual game in which the Res-Fruit is out of book as early as move 6 >>>and proceeds to play the next 5 moves on its own and each move is in the R40's >>>book! >>> >>>not sure if this is a testimony to the analytical ability of Fruit or a >>>compliment to the R40's book or both >>> >>>the game ended in a time scramble with a mate announcement in 5 >>> >>>GAME 4-Tasc R40 WINS >>> >>>Res Fruit WCCC 05-3.5 >>>Tasc R40 v2.5-1.5 >>> >>>[Date "2005.9.30"] >>>[Round "5"] >>>[White "Tasc R40 V2.5"] >>>[Black "Res.-FruitWCCC05"] >>>[Time Control "GAME 30 MIN"] >>>[Result [1-0] >>> >>>1.e4 c5 2.Nf3 d6 3.Bb5+ Bd7 4.Bxd7+ Qxd7 5.c4 Nc6 6.d4 >>> >>>[D] r3kbnr/pp1qpppp/2np4/2p5/2PPP3/5N2/PP3PPP/RNBQK2R b KQkq d3 0 6 >>> >>>from here until move 13, Fruit ,thinking on it own, comes up with the next 5 >>>moves(until move 11) and they are all in the R40's book >>>it does eat up some precious clock time though.. >>>at move 12 the R40 still has the full 30 minutes on its clock >>> >>>6.. Qg4 7.d5 Qxe4+ 8.Be3 Nd4 9.Qa4+ b5 10.Qa6 Nc2+ 11.Kd1 Rb8 12.Nbd2 Qd3 >>>13.Qxa7 Rd8 14.Rc1 Nxe3+ 15.fxe3 Nf6 16.Re1 Ng4 17.Rc3 Nf2+ 18.Kc1 Qg6 >>>19.e4 b4 20.Qa4+ Rd7 21.Qa8+ Rd8 22.Qc6+ Rd7 23.Rce3 Ng4 24.Qa8+ Rd8 25.Qa4+ Rd7 >>>26.R3e2 f6 27.Kb1 Qh5 28.Nf1 Qg6 29.Qc6 h5 30.h3 Nh6 31.Qa8+ Rd8 32.Qa4+ >>>Kf7 33.Qa5 Re8 34.N1d2 e5 35.dxe6ep+ Rxe6 36.Qa8 Kg8 >>>37.Qd5 Qe8 38.Nh4 Kh7 39.g3 g6 40.Ng2 Nf7 41.Nf4 Re7 >>>42.Qd3 Ra7 43.Nf3 Ne5 44.Nxe5 fxe5 45.Nd5 Qa8 46.Qb3 Rf7 47.g4 Bh6 >>>48.Rg2 hxg4 49.Qg3 Qc8 50.hxg4 Bf4 51.Qf3 Kg7 52.Rf1 Qa6 53.Qd3 Ra8 >>>54.b3 Raf8 55.g5 Ra7 56.Qh3 Rh8 >>> >>> >>>[D] 7r/r5k1/q2p2p1/2pNp1P1/1pP1Pb2/1P5Q/P5R1/1K3R2 w - - 0 57 >>> >>>now the Res is short of time with a little less then 90 Seconds left on the >>>clock >> >>I think that it is a bad time management by fruit. >>There is no jutification to use more than half of the timefor 30 moves that mean >>there is no justification to use more than 3/4 of the time for 60 moves. >> >>Note that >>55...Ra7 was the losing error and I wonder how fruit2.2 played that move >> >>My fruit2.2 can play that move only at depth 5 and not at higher depth and >>evaluates Rh8 as 0.00 if it gets deep enough >> >>New game - Fruit 2.2, 10'/40+10'/40+10'/40 >>[D]5r2/5rk1/q2p2p1/2pNp1P1/1pP1Pb2/1P1Q4/P5R1/1K3R2 b - - 0 1 >> >>Analysis by Fruit 2.2: >> >>55...Rh8 >> = (-0.12) Depth: 1/8 00:00:00 >>55...Rh8 56.Ka1 >> = (0.11) Depth: 2/13 00:00:00 >>55...Rh8 56.Ka1 Qc6 >> ² (0.44) Depth: 3/13 00:00:00 >>55...Kh8 56.Ka1 Rh7 >> ² (0.42) Depth: 3/13 00:00:00 >>55...Ra8 56.Qh3 Rh8 >> ² (0.35) Depth: 3/13 00:00:00 >>55...Ra8 56.Qf3 Rh8 57.Rd2 >> = (0.07) Depth: 4/13 00:00:00 >>55...Ra8 56.Qh3 Rh8 57.Qe6 Rff8 >> ² (0.41) Depth: 5/16 00:00:00 >>55...Rh8 56.Ka1 Rh5 57.Qf3 Rh8 >> ² (0.40) Depth: 5/16 00:00:00 >>55...Ra7 56.Qh3 Rh8 57.Qe6 Rf7 >> ² (0.28) Depth: 5/16 00:00:00 >>55...Ra7 56.Qh3 Rh8 57.Qe6 Rf7 58.Ka1 >> ² (0.50) Depth: 6/16 00:00:00 29kN >>55...Rh8 56.Ka1 Rh5 57.Rgf2 Rxg5 58.Qh3 >> = (0.22) Depth: 6/16 00:00:00 35kN > >>Uri > >It could easily have played it. Remember I published my comparative results on >my 624 MHz PDA as being roughly 15 times slower than my Athlon at 2GHz? This is >65+% slower than that same PDA. > >The problem is that you see the 0 seconds and somehow presume it must be similar >after all zero times 15 is still zero. My testing showed that those fractions of >a second can easily be 0.7 or 0.8 seconds for example, so multiply that by >40-50, add to that the time trouble, and it becomes VERY believable. > > Albert Note only that 35 knodes were enough for fruit2.2 to avoid Ra7 in my analysis. Note that the time trouble is probably because fruit was not tuned to no increasement with ponder off and I guess that changing the default parameters to tell fruit to play faster may be productive for fruit at time control of x minutes per game. Uri
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.