Author: blass uri
Date: 09:30:08 03/05/00
Go up one level in this thread
On March 05, 2000 at 09:30:14, Thorsten Czub wrote:
>[Event "millennium-tournament 1999-2000"]
>[Site "k6-400 40/120"]
>[Date "2000.02.29"]
>[Round "8"]
>[White "CSTal2.03"]
>[Black "Rebel-Tiger12.0E"]
>[Result "1/2-1/2"]
>
>1. Nf3 Nf6 2. d4 e6 3. c4 b6 4. g3 Ba6 5. b3 Bb4+ 6. Bd2 Be7 7. Bg2 c6 8. Bc3
>d5 9. Ne5 Nfd7 10. Nxd7 Nxd7 11. Nd2 O-O 12. O-O Rc8 13. e4 dxc4 14. bxc4 b5
>15. Re1 bxc4 16. Qa4 Bb5 17. Qc2 Re8 18. a4 {this is the first move cstal has to
>compute about} Ba6 19. Rab1 {with this move cstal throws rebel-tiger out of
>book} c5 {+0.34} 20. d5 {+0.59} exd5 {+0.26} 21.
>exd5 {+1.14} Rb8 {with this move tiger goes into minus a little, so very late
>getting the idea of the position. it considers a long time about Rb8} 22. Rxb8
>Qxb8 {-0.22} 23. Qe4 {+0.99} Nb6 {-0.02 rebel-tiger again believes this is
>nearly sero, even evaluates it +0.02 later. a mistake...} 24. Nf3 Bb7 {cstal
>believes this was a big mistake and it's score increases. rebel-tiger says -0.08
>in ply 11, doing arround 133.000 NPS. -0.14 in ply 12 and 13. the main-line
>shows the moves that will come, but it does not evaluate these moves right. it
>is: Bb7 Ng5 g6 Nxf7 Bxd5 Nh6+ Kf8 Qg4 Bxg2 Kxg2 Qa8+ f3 Bd6 Qg5 Nd7 Rxe8+
>Qxe8. what a pity! CSTal often has the opposite problem. it evaluates nicely,
>but has wrong main-lines :-)))} 25. Ng5 {+2.82} g6 {now rebel-tigers score
>decreases to -1.14 in ply 12. how can it oversee a move that was in it's
>main-line before? -1.70 in ply 13.} 26. Nxf7{+2.72 cstal still with very high
>speculative scores} Bxd5 {-1.66} 27.
>Nh6+ {+1.34} Kf8 28. Qg4 {+2.36} Bxg2 {-2.22} 29. Kxg2 {+2.80} Qb7+ 30. Kg1
>{+2.99} Bd6 31. Rd1 {+3.05} Re4 32. Qf3+ {Cstal likes those kind of positions
>much: >+3.59 pawns} Bf4 33.
>Rd8+ {+3.08} Ke7 34. Qd1{+5.06} Bc7 35. Ng8+ {+4.32} Ke6 36. Rf8 {>+3.92} Qd5
>37. Re8+ {3.98} Kd6 38. Qxd5+ {+3.96} Nxd5 39.
>Rxe4 {+2.55} Nxc3 40. Rxc4 {+1.77} Nd5 41. f4 {+1.61 the very speculative
>evaluations have gone, cstal now tries to stabilize the position. please see
>that the evaluations do not really 1:1 count for the position. they are only
>speculative and show the subjective point of view of cstal. it is euphorical.
>like a human beeing. it has wishes and dreams.}
I agree that humans have wishes and dreams but strong humans do not evaluate
positions as +5.06 like tal did at move 34 if they do not see a forced line to
win.
I am sure that they will prefer to go for positions like the following position
that is only 2 pawns advanatge but a clear win if they can.
[D]6r1/5kpp/8/8/8/8/3KPPPP/5R2 w - - 0 1
If tal does not evaluate the last position as more than +5.06 then something is
wrong with tal's evaluation.
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.