Author: Eelco de Groot
Date: 17:10:50 08/02/05
Go up one level in this thread
On August 02, 2005 at 08:33:54, Eelco de Groot wrote:
>On August 02, 2005 at 00:52:11, Eelco de Groot wrote:
>
>>On August 01, 2005 at 15:16:55, Peter Berger wrote:
>>
>>>ROTFL - sorry for the subject line .. - hope I can still post sth of at least
>>>limitted interest.
>>>
>>>1.) Toying around with Fruit 2.1 currently - that would be change no 1 - now
>>>there is a free open source amateur engine that is just as strong as the
>>>strongest commercials on ordinary PCs . No, that's no news anymore. But just as
>>>others I had to see it for myself, despite various previous reports.
>>>
>>>2.) Second one is about openings. Dann Corbit built an interesting book for
>>>Fruit IMHO. I guess it's completely automated, but he chose a good point to
>>>leave book, usually about 2 or 3 moves before known theory ends, with something
>>>like 10-15 moves of opening theory on average. And in several games Fruit finds
>>>interesting and valid alternatives to grandmaster analysis.
>>>
>>>Just a little example ( my sources might be slightly dated as I don't play the
>>>Ruy from either side).
>>>
>>>[Event "50 3"]
>>>[Site "Centrino 1.6- PIV 3.2"]
>>>[Date "2005.08.01"]
>>>[Round "11"]
>>>[White "Junior 9.0.0.3"]
>>>[Black "Fruit 2.1"]
>>>[ECO "C80"]
>>>[Result "0-1"]
>>>
>>>1. e4 {0s} e5 {+0.01/1 1s} 2. Nf3 {3s} Nc6 {+0.01/1 0s}
>>>3. Bb5 {2s} a6 {+0.01/1 0s} 4. Ba4 {2s} Nf6 {+0.01/1 0s}
>>>5. O-O {3s} Nxe4 {+0.01/1 0s} 6. d4 {2s} b5 {+0.01/1 0s}
>>>7. Bb3 {2s} d5 {+0.01/1 0s} 8. dxe5 {3s} Be6 {+0.01/1 0s}
>>>9. Nbd2 {3s} Nc5 {+0.01/1 0s} 10. c3 {3s} d4 {+0.01/1 0s}
>>>11. Ng5 {3s} dxc3 {+0.01/1 0s} 12. Nxe6 {4s} fxe6 {+0.01/1
>>>0s} 13. bxc3 {4s} Qd3 {+0.01/1 0s} 14. Bc2 {4s} Qxc3
>>>{+0.42/13 1:16m} 15. Nb3 {3s}
>>>
>>>A famous line of the Open Ruy known to be busted (11. .. dxc3?!) - that's
>>>Kasparov-Anand, New York PCA WCh (10) 1995, 14. Bc2 is ! and 15. Nb3!! . NCO and
>>>ECO have a look at 15. ...Nxb3 and 15. ...Rd8 and prove a clear white advantage.
>>>Fruit, ever sceptical, just takes on e5 (and how come this move doesn't even get
>>>mentioned?).
>>>
>>>[D]r3kb1r/2p3pp/p1n1p3/1pn1P3/8/1Nq5/P1B2PPP/R1BQ1RK1 b kq -
>>>
>>
>>
>>Pro Deo 1.1 Vulcan {9k10i} would have gobbled up the pawn too, but prefers to do
>>that with the Knight. No deep strategic considerations I suppose, and I think
>>Black is taking some chances taking the pawn right now. The White King is
>>sitting comfortably in his castle warming his feet at the fire, sipping some
>>good Spanish wine perhaps while his colleague from the other side is rather in
>>the cold, no matter whether he stays in the center or moves to his left castle
>>or his right castle.. After 15 minutes the evaluation is already not equal
>>anymore. Qc3xe5 is number thirteen on the list by then. But I guess your point
>>is that Kasparov's play against Anand is not exactly refuted by Fruit's opening,
>>but Black deviates just in time here from the 1992 game... I have not checked
>>what Vulcan would play confronted with Qxe5!?
>>
>>00:00:00.3 1,96 1 1968 Nxb3 Qh5+ Kd7 Qf7+ Be7 Bxb3 Qxa1
>>00:00:00.3 3,29 2 4617 Nxb3 axb3 Qxa1 Qh5+ g6 Bxg6+ Kd8
>>00:00:00.3 0,87 3 9027 Nxb3 Bxb3 Rd8 Qg4
>>00:00:00.3 0,98 3 12767 Nxe5 Bd2 Qb2
>>00:00:00.3 1,32 3 16650 Rd8 Bd2 Qxe5
>>00:00:00.3 0,95 4 19379 Rd8 Bd2 Qxe5 Qf3
>>00:00:00.4 0,91 5 46743 Rd8 Bd2 Qxe5 Qf3 Rd5
>>00:00:00.4 0,87 6 80799 Rd8 Bd2 Qxe5 Qf3 Nd4 Nxd4 Rxd4
>>00:00:00.5 0,66 7 193273 Rd8 Bd2 Qxe5 Re1 Qd5 Bf4 Nxb3 Qxd5
>>00:00:00.9 0,45 8 661030 Rd8 Bd2 Qxe5 Re1 Qd5 Bf4 Be7 Qxd5 Rxd5
>>00:00:02.0 0,51 8 1794728 Nxe5 Bd2 Qc4 Qh5+ Nf7 Nxc5 Qxc5 Qxc5 Bxc5 Rad1
>>00:00:05.1 0,37 9 4995726 Nxe5 Be3 Nxb3 axb3 Rd8 Qe2 Bd6 Rxa6
>>00:00:10.8 0,19 10 11126891 Nxe5 Be3 Rd8 Bd4 Qc4
>>00:00:15.9 0,22 10 16981113 Nxb3 axb3 Qxa1 Qf3 Nd8 Qxa8 Qxe5
>>00:00:30.7 0,13 11 33760108 Nxb3 Bxb3 Rd8 Qg4 Qxe5 Bf4 Rd4 Qf3 Qe4 Qxe4 Rxe4
>>00:01:16.4 0,00 12 87001311 Nxb3 Bxb3 Qxa1 Qh5+ g6 Qf3 Nd8 Rd1
>>00:01:45.7 0,08 12 122773787 Nxe5 Bf4 Bd6 Rc1 O-O-O Be4 Qb2 Nxc5 Bxc5
>>00:05:55.0 0,03 13 422795898 Nxe5 Bd2 Qc4 Be3 Ncd3 f4 Nb2 Qh5+ Nf7 Bxh7 Qd5 Qg6
>>00:15:51.9 -0,08 14 1143840139 Nxe5 Bd2 Qc4 Nxc5
>
>I let Pro Deo Vulcan {9k10i} testversion calculate longer and just now it comes
>with the following (7 hours 41 minutes), Athlon 3200+, evaluations from Black's
>point of view
>
>00:54:36.7 -0,48 15 3940673939 Nxe5 Bd2
>01:52:47.1 -0,30 15 3828937285 Nxb3 Bxb3 Rd8 Qh5+ g6
>02:25:48.1 -0,13 15 1963521336 Rd8 Qh5+
>03:52:56.0 -0,69 16 4048985325 Rd8 Bd2 Qxe5 Re1 Qf6 Nxc5 Rxd2
>04:20:22.8 -0,41 16 1755002296 Nxb3 Bxb3 Qxa1 Qh5+ g6 Qf3 Nd8
>07:41:27.6 -0,36 16 3646376805 Qxe5 Re1
>
>It seems I have to give Fruit 2.1 some more credit for playing its novelty
>15..Qc3xe5! What exactly the new PV from Vulcan is I'm not sure though, if it
>would be the line as in the game it would be a transposition of the first PV at
>16 ply, Rd8 Bd2 Qxe5 Re1 Qf6 Nxc5 Rxd2 and that got an eval of -0.69 after 3
>hours 52 min. If that line is more or less forced, what does Vulcan see now that
>it did not see in the transposition?
>
>Maybe a nice position to do more analysis?
>
> Eelco
>
Part three of analysis, I think we have to postpone for a while writing a letter
to John Nunn about oversights in the analysis of this line; at 17 ply the the
eval for 15..Qxe5 drops again to -0.69, what one would suspect to be the case if
the PV was a transposition of 15.Rd8. (Tuning remark: Pro Deo 1.1 Vulcan {9k10i}
has searchparameter [Selective Search = 80], which when I use this parameter
sometimes seems a bit ubstable in the PVs and then I change it back to [Adjusted
Selectivity =] again, hoping that will work wonders :))
10:18:06.5 -0,69 17 2281362615 Qxe5 Re1 Rd8
12:16:07.3 -0,62 17 2340410590 Nxb3 Bxb3 Qxa1 Qh5+ g6
Looking at the position after 15..Qxe5 16.Re1 Rd8 17.Bd2 Qf6 18.Nxc5
[D] 3rkb1r/2p3pp/p1n1pq2/1pN5/8/8/P1BB1PPP/R2QR1K1 b k -
the score drops further if White then replies 19.Rxe6 to 18..Rxd2
00:00:01 9.00 -0.60 18..Rxd2 19.Rxe6 Qxe6 20.Nxe6 Rxd1 21.Rxd1
Bd6 22.Nxg7 Kf7 23.Nf5 Be5 24.Bb3 Kf6 25.Ne3
Nd4 26.Nd5
00:00:03 10.00 -0.92 18..Rxd2 19.Ne4 Rxd1 20.Nxf6 gxf6 21.Raxd1
Nd8 22.Bb3 e5 23.Kh1
00:00:08 11.00 -0.82 18..Rxd2 19.Ne4 Rxd1 20.Nxf6 gxf6 21.Raxd1
e5 22.Bb3 Na5 23.Bc2 Nc6 24.Bb3 Na5 25.Bc2
00:00:23 12.00 -0.74 18..Rxd2 19.Ne4 Rxd1 20.Nxf6 gxf6 21.Raxd1
e5 22.Re3 Nb4 23.Be4 c5 24.Rh3 h6 25.Bg6
00:01:32 13.00 -0.81 18..Rxd2 19.Rxe6 Be7 20.Qe1 Qb2 21.Bb3 Rd6
22.Nxa6 Kd8 23.Rb1 Qd4
00:05:03 14.00 -1.12 18..Rxd2 19.Rxe6 Qxe6 20.Nxe6 Rxd1 21.Rxd1
Bd6 22.Nxg7 Kf7 23.Nf5 Ke6 24.Nxd6 cxd6 25.f4
d5 26.f5
00:13:51 15.00 -1.15 18..Rxd2 19.Rxe6 Qxe6 20.Nxe6 Rxd1 21.Rxd1
Nb4 22.Nxc7 Ke7 23.Bb1
00:49:56 16.00 -1.14 18..Rxd2 19.Rxe6 Qxe6 20.Nxe6 Rxd1
Eelco
>>
>>
>>
>>
>>
>>
>>>Whatever the absolute thruth, Fruit got very decent compensation for the
>>>exchange in the game and in fact had little trouble to reach a major advantage
>>>against the world champion, who lacked counterplay and pawn levers. Most moves
>>>even looked very obvious. Similar happened in several other Fruit games I have
>>>watched. I am getting more and more sceptical of classical opening books.
>>>
>>>--
>>>
>>>Unrelated, still .. - Fruit probably blew the huge advantage it reached later
>>>as it allowed transformation to a drawish endgame.
>>>
>>>[D]5b2/8/5k2/p4r2/Bp3K2/2p5/P4P2/4R3 w - -
>>>
>>>Here Junior played 49. Kg4 which should be ?? , while 49. Kg3 probably holds. I
>>>can only explain with wrong knowledge about bishop with opposite colour endings,
>>>and it's certainly not the first example from Junior I have seen covering this
>>>theme. I don't think a human would consider giving up the pawn on f2 .
>>>
>>>Everything with a huge grain of salt as usual :) .
>>>
>>>Peter
>>>
>>>[Event "50 3"]
>>>[Site "Centrino 1.6- PIV 3.2"]
>>>[Date "2005.08.01"]
>>>[Round "11"]
>>>[White "Junior 9.0.0.3"]
>>>[Black "Fruit 2.1"]
>>>[ECO "C80"]
>>>[Result "0-1"]
>>>
>>>1. e4 {0s} e5 {+0.01/1 1s} 2. Nf3 {3s} Nc6 {+0.01/1 0s}
>>>3. Bb5 {2s} a6 {+0.01/1 0s} 4. Ba4 {2s} Nf6 {+0.01/1 0s}
>>>5. O-O {3s} Nxe4 {+0.01/1 0s} 6. d4 {2s} b5 {+0.01/1 0s}
>>>7. Bb3 {2s} d5 {+0.01/1 0s} 8. dxe5 {3s} Be6 {+0.01/1 0s}
>>>9. Nbd2 {3s} Nc5 {+0.01/1 0s} 10. c3 {3s} d4 {+0.01/1 0s}
>>>11. Ng5 {3s} dxc3 {+0.01/1 0s} 12. Nxe6 {4s} fxe6 {+0.01/1
>>>0s} 13. bxc3 {4s} Qd3 {+0.01/1 0s} 14. Bc2 {4s} Qxc3
>>>{+0.42/13 1:16m} 15. Nb3 {3s} Qxe5 {+0.26/14 4:13m} 16. Re1
>>>{1:42m} Rd8 {-0.18/15 8s} 17. Bd2 {17s} Qf6 {-0.24/14
>>>1:17m} 18. Nxc5 {1:45m} Rxd2 {-0.21/15 0s} 19. Ne4 {2:32m
>>>(Txe6+)} Rxd1 {+0.04/16 1:21m} 20. Nxf6+ {47s} gxf6
>>>{+0.07/17 57s} 21. Raxd1 {42s} e5 {+0.11/17 1:00m} 22. Re3
>>>{19s (a4)} Bd6 {+0.15/16 1:40m} 23. Kf1 {1:24m} h5
>>>{+0.25/15 5s} 24. Bg6+ {1:44m (a4)} Ke7 {+0.30/15 1:12m}
>>>25. h4 {11s (f4)} Ke6 {+0.44/16 1:00m} 26. Bc2 {3s} f5
>>>{+0.41/16 4:00m} 27. g3 {4s (Lb3+)} Kf6 {+0.82/17 1:22m}
>>>28. Bd3 {3s (Lb3)} Nd4 {+1.00/17 1:19m} 29. Be2 {1:13m} a5
>>>{+1.12/17 2s} 30. Rb1 {1:05m (Kg2)} b4 {+1.00/17 1:16m}
>>>31. Bd1 {2:46m (Td1)} c5 {+1.18/17 1:01m} 32. Rc1 {1:12m}
>>>e4 {+1.17/18 0s} 33. Kg2 {1:39m} Rh7 {+1.17/17 15s} 34. Kf1
>>>{2:00m (Lb3)} Rc7 {+1.21/16 1:50m} 35. Bxh5 {1:48m} c4
>>>{+1.30/18 0s} 36. Bd1 {46s (Td1)} c3 {+1.78/16 1:02m}
>>>37. Ba4 {1:09m} Bf8 {+1.89/17 0s} 38. g4 {4:07m} Rh7
>>>{+2.08/18 0s} 39. g5+ {55s (Td1)} Ke5 {+2.19/16 45s}
>>>40. Rh3 {3s} Nf3 {+2.27/16 1:41m} 41. Kg2 {3:38m (Ke2)}
>>>Rxh4 {+2.33/17 1:00m} 42. Rxf3 {3s} exf3+ {+2.50/17 36s}
>>>43. Kxf3 {34s} Rf4+ {+2.46/18 32s} 44. Ke2 {1:03m} Rg4
>>>{+2.37/18 26s} 45. Rf1 {1:30m (Te1)} f4 {+2.52/17 45s}
>>>46. Kf3 {4s (Lb3)} Rxg5 {+2.41/17 42s} 47. Re1+ {9s} Kf6
>>>{+2.39/17 1:15m} 48. Kxf4 {3s} Rf5+ {+2.37/17 1:20m}
>>>49. Kg4 {2s (Kg3)} Rxf2 {+3.88/17 43s} 50. Bb3 {1:01m} a4
>>>{+4.37/18 0s} 51. Re6+ {42s} Kg7 {+4.32/20 0s} 52. Bxa4
>>>{22s} Rxa2 {+4.36/18 21s} 53. Bb3 {3s} Ra3 {+4.50/18 23s}
>>>54. Bd1 {54s} b3 {+4.78/19 0s} 55. Rc6 {31s} Bb4 {+4.78/17
>>>0s} 56. Rc4 {44s} Ba5 {+5.01/18 0s} 57. Rc5 {38s} b2
>>>{+5.03/18 0s} 58. Bc2 {1:19m} Ra1 {+5.04/18 0s} 59. Rg5+
>>>{41s (Kf4)} Kf6 {+5.36/14 25s} 60. Rf5+ {3s (Tc5)} Ke7
>>>{+5.58/15 40s} 61. Re5+ {24s (Tf2)} Kd6 {+8.85/17 50s}
>>>62. Re2 {30s} b1=Q {+9.11/18 8s} 63. Bxb1 {24s} Rxb1
>>>{+9.17/19 0s} 0-1
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.