Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / Rybka4 database of games where Rybka doesn't play good endga
1 2 3 4 Previous Next  
Parent - - By M ANSARI (*****) [kw] Date 2010-07-03 08:18
You know, if you go through an endgame book you will find that there is some information which can easily be added as knowledge.  Certain criteria such as which file and what rank the extra pawn is and where the opposing King is and whether the opposing King can reach the pawn or not.  It seems pretty clear cut and I would have thought it would be easy to add this information where it can quickly change the evaluation.  But I guess such information might slow down the search so much as to make it play worse.  But if a lookup table could be made then things might not be so slow.  Once you consider how many games end up as rook endgames, it seems the effort is well worth it.
Parent - - By Vasik Rajlich (Silver) [pl] Date 2010-07-03 09:46
Yes, of course, this is not difficult. I didn't do it because it's handled by tablebases, which the remote version will always access.

Vas
Parent - - By Wayne Lowrance (***) Date 2010-07-03 17:35
I think that is fair enough and a good decision. If TB's can figure it out
why slow the search down with duplicity. I don't consider this making
the lil girl dumb.
Wayne
Parent - - By Vasik Rajlich (Silver) [pl] Date 2010-07-04 05:55
Yes.

Vas
Parent - - By Banned for Life (Gold) Date 2010-07-04 10:10
Philosophically, don't you think a really good search and eval should be able to solve a KPk endgame? The only good thing about this is that it is the best example I have that the best engines don't play anywhere near perfectly in other parts of the game either.
Parent - By Vasik Rajlich (Silver) [pl] Date 2010-07-05 08:17
That's kind of a special case, it's basically a fortress (meaning: static advantage but no way to make progress). This theme is pretty rare in middlegames.

But I agree with the conclusion. :smile:

Vas
Parent - By Chess_uci [it] Date 2010-07-24 16:09
pls uci wow? back www
Parent - - By M ANSARI (*****) [kw] Date 2010-07-04 10:04
Well a lot of the rook endgames would need information even if 6 piece EGTB's are there.  I guess you must be talking about 7 piece EGTB's.  I guess that is an idea, but really am not sure how large these EGTB's are for rook endgames or that they have even been generated.  I would think they could be generated and it might be worth the effort to make rook endgames play perfectly.
Parent - - By Zarkon (***) [th] Date 2010-07-04 11:05
The EGTB I'd most like to see is KRPPKRP! Any idea how big that would be if it existed?
Parent - By M ANSARI (*****) [kw] Date 2010-07-04 13:15
I remember reading that some 7 piece tables were generated but not sure which.  I can think if you can cover all rook endgames that have 7 pieces, that would be a dramatic improvement and an easy ELO gain to whatever engine that uses.  It would also be a great analysis tool.  Of course those EGTB's would have to be on very fast media to make sure they don't affect search speed too much.  What would be ideal is to have a system where bit bases are used for search and then once a good move is found back off to hard disk.  Bitbases are maybe 10 or 15 X smaller so their footprint would be small enough to fit in fast RAM.
Parent - - By Vasik Rajlich (Silver) [pl] Date 2010-07-05 08:18
I'm pretty sure that will be the most important 7-man EGTB.

Vas
Parent - By dr_zied_haddad (**) [fr] Date 2010-07-05 13:36
Hi All
I'm happy to see that the discussions here are important, with many items and ideas shared by all of us. Just a remindre, the main objective of these lines are to collect positions where Rybka doesn't play optimally in endings. Don't forget to feed these lines with positions.

I'd like to Thank all of you, because your ideas and questions gave me again the passion for computer chess that i tended to keep private for years.
Parent - - By patrick delaurentis (**) Date 2010-07-29 17:42
i read vytron and banned for life stating that according to their analysis (MOVE QUALITY) games that they don't find rybka 4 reliable and vytron stated he has other main engines he relies upon- what engines are the the most reliable then in expert opinion if i cant depend on my rybka 4 that much- also vytron stated some of his main engines were rybka human and dynamic, are they more reliable?

i appreciate opinions alot and have another question i was hoping someone could answer-

what chess engine from their experience produces the best analysis further in to the notation then any other-meaning that on rybka 4 i notice an awful amount of better moves being found when i let her reach a depth of 23, copy the notation and then it seems that i can check the 3rd move from the starting point of the analysis which according to experts for every single move the ply drops by 1-
but i can check that move as i was saying and 60% of the time the depth shoots right up to 21-22-and a different move is found.
i would love an engine with a more precise (spread in depth) in its evaluations over the broad spectrum of 15 moves and rybka 4 does not seem so sure in that aspect.
Parent - By Uly (Gold) [mx] Date 2010-07-29 18:21
Please note that while I said relying on Rybka 4 for move choosing was dangerous, I also said relying on a single engine is dangerous, there's no engine I could say I would rely on any given game, I would use it as main engine for that game, or position, but assisted by other engines to make sure the engine knows what's going on (very often, they don't).

As for my core repertoire of main engines:

Zappa Mexico II
Naum   
Critter   
Rybka (3, Dynamic, Human, 4)
Stockfish
Komodo
Hiarcs Paderborn 2007

Recently left the list: Shredder 11, Fruit/Toga and Spark (though I have a long list of engines that I check when I run out of ideas on the position, being biased towards the engines with high rating).
Parent - By dr_zied_haddad (**) [fr] Date 2010-08-09 18:58
Hope u all show Vas that this thread will not be very short, to give him sufficient motivation to improve Rybka.
- - By Hurnavich (****) Date 2010-07-07 07:56
[Event "R4 v N4, Blitz 3m"]
[Site "Hurnavich-PC"]
[Date "2010.07.07"]
[Round "1"]
[White "Deep Rybka 4 SSE42 x64"]
[Black "Naum 4.2 x64"]
[Result "1/2-1/2"]
[ECO "C65"]
[WhiteElo "3264"]
[BlackElo "3186"]
[Annotator "0.15;0.03"]
[PlyCount "268"]
[TimeControl "180"]

{Intel(R) Core(TM) i7 CPU         950  @ 3.07GHz 4206 MHz  W=25.6 plies; 493kN/
s  B=32.6 plies; 9,446kN/s} 1. e4 {0.15/16 6 Both last book move} e5 {0.03/19 4
} 2. Nf3 {0.18/17 10 (Nc3)} Nc6 {0.15/18 4} 3. Bb5 {0.13/16 12 (Nc3)} Nf6 {0.
27/18 9} 4. O-O {0.23/15 2} Be7 {0.26/18 3 (Bc5)} 5. Re1 {0.36/15 5 (Nc3)} d6 {
0.35/17 5} 6. d4 {0.34/15 4} O-O {0.29/17 2 (exd4)} 7. Bxc6 {0.66/15 5 (d5)}
bxc6 {0.29/6 0} 8. dxe5 {0.66/16 2} Ng4 {0.43/17 3} 9. Bf4 {0.64/15 2 (exd6)}
dxe5 {0.58/18 7 (g5)} 10. Qxd8 {0.77/17 3} Rxd8 {0.65/19 4} 11. Nxe5 {0.77/18 3
} Nxe5 {0.65/19 3} 12. Bxe5 {0.89/14 0} Rd7 {0.64/19 4 (Be6)} 13. Bc3 {0.88/18
4 (Nc3)} Ba6 {0.69/18 5} 14. Nd2 {0.88/18 3} Rad8 {0.84/18 10} 15. Nb3 {0.95/
18 2} Bb7 {0.97/19 11} 16. Na5 {0.99/18 5} Ba8 {0.97/20 2} 17. b3 {1.06/16 3
(b4)} c5 {0.81/18 3} 18. f3 {1.14/15 2 (Nc4)} Bg5 {0.83/17 4 (f6)} 19. Kf2 {1.
20/16 7 (Nc4)} h5 {0.94/16 2 (h6)} 20. Nc4 {1.35/16 4} Bb7 {0.97/17 2} 21. Nb2
{1.31/15 5 (Ke2)} Be7 {1.01/16 6} 22. Rad1 {1.34/16 4} h4 {1.09/17 5 (Rxd1)}
23. Rxd7 {1.47/16 3} Rxd7 {1.09/6 0} 24. Nc4 {1.47/16 3 (a4)} Kf8 {1.16/17 2
(h3)} 25. Re2 {1.50/14 3 (a4)} Ba6 {1.17/17 3 (Bg5)} 26. Rd2 {1.54/18 3 (g3)}
Rxd2+ {1.14/19 1} 27. Bxd2 {1.54/19 3} Ke8 {1.12/20 1} 28. Ke3 {1.54/17 3 (Ba5)
} f6 {1.11/18 2 (g5)} 29. Be1 {1.53/18 3 (g3)} g5 {1.16/19 1} 30. g3 {1.52/18 1
} h3 {1.17/21 4} 31. f4 {1.52/18 3 (Bc3)} g4 {1.17/19 2 (Kf7)} 32. Ba5 {1.87/
18 3 (Bc3)} Bxc4 {1.15/21 1} 33. bxc4 {1.98/18 0} c6 {1.18/22 1} 34. Bd2 {1.98/
22 1 (Kf2)} Kf7 {1.62/22 2} 35. f5 {2.08/24 1} Bd6 {1.77/24 1} 36. Kf2 {2.08/
24 0} Be5 {1.78/25 1} 37. Be3 {2.08/25 1} Ke8 {1.78/25 1 (a6)} 38. Bxc5 {2.08/
26 2} Bb8 {1.78/26 2 (a6)} 39. Bd4 {2.08/23 2 (Kf1)} Ke7 {1.78/26 1} 40. Kg1 {
2.08/23 1 (Kf1)} a6 {1.78/26 1} 41. Be3 {2.08/24 1 (c5)} Ke8 {1.78/25 1 (Kf7)}
42. Kf2 {2.08/26 2 (Kf1)} Ke7 {1.78/26 1 (Be5)} 43. c5 {2.08/25 2 (Bd4)} Be5 {
1.78/26 1} 44. Kf1 {2.09/25 1} Kd8 {1.78/26 1 (Ke8)} 45. Bf2 {2.09/28 2} Kd7 {
1.78/28 1 (Ke8)} 46. Ke1 {2.09/29 1 (Ke2)} Ba1 {1.78/28 1 (Kc7)} 47. Ke2 {2.09/
28 1} Be5 {1.78/28 1} 48. Kd3 {2.09/27 1} Ke7 {1.78/28 1 (Kc8)} 49. Kc4 {2.09/
26 1 (a4)} Bb8 {1.78/27 1 (Kd8)} 50. Bg1 {2.09/27 2 (Kb4)} Kf7 {1.78/27 1 (Be5)
} 51. Kd4 {2.09/27 1 (Kd3)} Be5+ {1.78/28 1 (Ke7)} 52. Kd3 {2.09/28 1 (Ke3)}
Bc7 {1.78/27 0 (Bb8)} 53. Bf2 {2.09/28 1} Kg7 {1.78/28 1 (Be5)} 54. Kc4 {2.09/
27 1 (a3)} Be5 {1.78/27 1 (Kf7)} 55. Kb4 {2.09/26 1 (Be1)} Bc7 {1.78/27 1} 56.
Kc3 {2.09/26 1} Kf7 {1.78/28 1} 57. Kc4 {2.09/27 1 (a4)} Be5 {1.78/28 1} 58.
Kb3 {2.09/27 1 (Kb4)} Kg7 {1.78/28 1 (Bc7)} 59. Bg1 {2.09/27 1 (a4)} Bc7 {1.78/
27 1 (Kf7)} 60. Kc4 {2.09/27 1} Kf8 {1.78/27 0 (Bb8)} 61. Kd4 {2.09/26 1 (a4)}
Bb8 {1.78/27 0 (Be5+)} 62. Kd3 {2.09/26 1 (a4)} Be5 {1.78/28 1 (Kg7)} 63. Ke3 {
2.09/25 1 (a4)} Ke7 {1.78/28 1 (Kg7)} 64. Bf2 {2.09/27 1 (Kd2)} Kd7 {1.78/27 1}
65. Ke2 {2.09/27 1 (a4)} Ke8 {1.78/27 0 (Ke7)} 66. Ke1 {2.09/27 1 (Bg1)} Bc3+ {
1.78/27 1 (Ke7)} 67. Kd1 {2.09/27 1} Ba1 {1.78/27 1 (Kd7)} 68. Kd2 {2.09/26 1
(a4)} Be5 {1.78/28 0} 69. Bg1 {2.09/27 1 (a4)} Kf8 {1.78/27 0 (Kd7)} 70. Ke2 {
2.09/27 1 (a4)} Bc7 {1.78/26 0 (Ke7)} 71. Ke3 {2.09/25 1 (a4)} Be5 {1.78/29 1}
72. Kf2 {2.09/26 0 (a4)} Bc7 {1.78/27 1 (Ke7)} 73. Kf1 {2.09/26 1 (Ke2)} Be5 {
1.78/27 1} 74. Ke1 {2.09/28 1 (a4)} Kf7 {1.78/29 1 (Ke7)} 75. Kf2 {2.09/27 0
(Ke2)} Ba1 {1.78/28 1 (Bd4+)} 76. Kf1 {2.09/27 0 (a4)} Ke8 {1.78/29 1 (Be5)}
77. Ke1 {2.09/26 0 (a4)} Bb2 {1.78/26 0 (Kf7)} 78. Kd2 {2.09/25 0} Be5 {1.78/
27 0} 79. Ke2 {2.09/26 0} Ba1 {1.78/26 0 (Kf7)} 80. Kd3 {2.09/24 1} Be5 {1.78/
27 0} 81. Kc4 {2.09/25 0 (a4)} a5 {1.78/26 1 (Ke7)} 82. Bf2 {2.38/24 0 (a4)}
Bc7 {1.78/24 1 (Ke7)} 83. Kd4 {2.43/22 0 (a3)} Kd7 {1.94/25 1 (Be5+)} 84. Be1 {
2.58/22 1} Be5+ {2.07/25 2 (Kc8)} 85. Ke3 {2.61/22 0} Bb8 {2.09/24 1 (Kc8)} 86.
a4 {3.06/16 0} Bc7 {2.09/24 0 (Be5)} 87. Kf2 {3.33/21 0} Ke7 {2.09/25 0 (Ke8)}
88. Kf1 {3.33/23 0 (Bc3)} Bd8 {2.09/25 1 (Kd7)} 89. Bd2 {3.33/24 0 (Bc3)} Bc7 {
2.09/26 0} 90. Kg1 {3.33/25 0} Kd7 {2.09/26 0} 91. Bf4 {3.33/25 0 (Bc3)} Be5 {
2.09/25 0} 92. Bxe5 {3.33/25 0} fxe5 {2.09/6 0} 93. Kf1 {3.33/37 0 (c3)} Ke7 {
2.09/40 0} 94. Kf2 {3.33/39 0 (Ke2)} Kf7 {2.09/43 0 (Kf6)} 95. Ke3 {3.33/39 0
(Ke2)} Ke7 {2.09/46 0 (Kg8)} 96. c3 {3.33/37 0} Kf6 {2.09/48 0} 97. Kf2 {3.33/
37 0 (Kd3)} Ke7 {2.09/48 0 (Kf7)} 98. Ke1 {3.33/37 0 (Ke2)} Kf6 {2.09/52 0} 99.
Kd1 {3.33/30 0 (Ke2)} Kf7 {2.09/54 0 (Ke7)} 100. Kc2 {3.33/37 0} Kf6 {2.09/56
0 (Ke7)} 101. Kd2 {3.33/38 0 (Kb3)} Kf7 {2.09/55 0 (Ke7)} 102. Kd1 {3.33/37 0
(Ke2)} Kf6 {2.09/56 0} 103. Kc2 {3.33/38 0} Kg5 {2.09/59 0 (Ke7)} 104. Kc1 {3.
33/38 0 (Kb3)} Kf6 {2.09/60 0} 105. Kb2 {3.33/38 0 (Kb1)} Kg5 {2.09/60 1 (Ke7)}
106. Kb3 {3.33/37 0 (Ka2)} Kf6 {2.09/59 0 (Kh6)} 107. Ka2 {3.33/38 0 (Kc4)} Kg5
{2.09/60 0 (Ke7)} 108. Kb1 {3.33/38 0 (Kb2)} Kf6 {2.09/60 0} 109. Ka1 {3.09/37
0 (Kc1)} Kg5 {2.09/60 0 (Ke7)} 110. Kb2 {3.09/38 0} Kf6 {2.09/60 0} 111. Kb3 {
3.09/33 0 (Kb1)} Kg5 {2.09/59 1 (Ke7)} 112. Kc4 {3.09/31 0 (Ka2)} Kf6 {2.09/59
0} 113. Kd3 {3.09/25 0} Kg7 {2.09/59 1 (Ke7)} 114. c4 {3.09/36 0 (Ke2)} Kf6 {
2.02/57 0 (Kg8)} 115. Kd2 {3.09/37 0 (Ke2)} Kg5 {2.02/57 0} 116. Kc2 {3.09/24
0 (Ke1)} Kf6 {2.02/60 0} 117. Kb3 {3.09/23 0 (Kb1)} Ke7 {2.02/58 0 (Kg5)} 118.
Kb2 {3.09/37 0 (Ka3)} Kf6 {2.02/56 0} 119. Kc1 {3.09/19 0 (Kc2)} Kg5 {2.02/59
0 (Kf7)} 120. Kd2 {3.09/38 0 (Kd1)} Kf6 {2.02/59 0} 121. Kd1 {3.09/37 0 (Kc2)}
Kf7 {2.02/58 0 (Kg5)} 122. Ke2 {3.09/25 0 (Ke1)} Ke8 {2.02/44 0 (Kf6)} 123. Ke1
{3.09/31 0 (Ke3)} Kd7 {2.02/45 0 (Ke7)} 124. Kd1 {3.09/36 0} Ke8 {2.02/56 1
(Ke7)} 125. Kc2 {3.09/24 0 (Ke2)} Ke7 {2.02/57 0 (Kf7)} 126. Kb3 {3.09/37 0}
Kf7 {2.02/57 1} 127. Ka2 {0.00/34 0} Kf6 {2.02/60 0 (Ke7)} 128. Kb2 {0.00/34 0
(Ka1)} Kg5 {2.02/34 0 (Ke7)} 129. Kc1 {3.09/7 0 (Ka1)} Kh5 {2.02/34 0 (Kf6)}
130. Kb2 {3.09/16 0 (Kd1)} Kg5 {0.00/60 0} 131. Kb3 {0.00/17 0 (Ka1)} Kf6 {2.
02/60 0} 132. Ka3 {0.00/40 0 (Ka2)} Kg5 {0.89/53 0 (Kf7)} 133. Ka2 {0.00/26 0}
Kf6 {0.00/37 0} 134. Kb2 {0.00/42 0 (Ka1)} Kg5 {0.00/60 0 Draw accepted}
1/2-1/2

Rybka shows 3+ advantage ??? untill late in the game when she accepts it's a draw.
Parent - - By Arrière Pensée (Gold) Date 2010-07-07 16:45
Okay! What settings did you use for - Normal Move Time & Max Move Time? Just curious!
Parent - - By Hurnavich (****) Date 2010-07-07 20:27
3 min total aside all default settings,
Parent - By Arrière Pensée (Gold) Date 2010-07-07 20:55
Thanks!
Parent - By dr_zied_haddad (**) [fr] Date 2010-07-18 09:47
Please continue to feed this post with yours positions. It's an important issue for all the customers for future Rybka development.
- - By Yoav Dothan (**) [il] Date 2010-07-19 07:36
This endgame is a draw, Rybka 4 gives +3.19

4K3/5P1k/5Bn1/8/8/8/8/8 w - - 0 10
Parent - - By dr_zied_haddad (**) [fr] Date 2010-07-20 06:47
Posted by JohnL

Couldn't resist posting two of todays end positions from the Sparkassen tournament. Rybka thinks there is about half-pawn resp. whole pawn advantages which is pretty big if you compare with same evaluation in the opening.
But in both cases players agreed a draw. The better side didn't even bother playing on, despite no risk.

Mamedyarov-Kramnik

8/5pkp/6p1/r7/8/3R1P2/5P1P/6K1 w - -
         
8         
7         
6         
5         
4         
3         
2         
1         
a b c d e f g h 

Start<>EndLookup

[-0.46]  d=20  32.Rd4 Kf6 33.Kg2 Rg5 34.Kh1 Ke6 35.h4 Rh5 36.Kg2 g5 37.hxg5 Rxg5 38.Kh3 h6 39.Re4 Kf5 40.Rg4 Rxg4 41.fxg4 Kf4 42.Kh4 f5 43.gxf5 (0:00:31) 360kN
[-0.48]  d=19  32.Rd4 Kf6 33.Kg2 Rg5 34.Kh1 Ke6 35.h4 Rh5 36.Kg2 g5 37.hxg5 Rxg5 38.Kh3 h6 39.Re4 Kf5 40.Rg4 Rxg4 41.fxg4 Kf4 42.Kh4 f5 43.gxf5 (0:00:26) 321kN
[-0.48]  d=18  32.Rd4 Kf6 33.Kg2 Rg5 34.Kh1 Ke6 35.h4 Rh5 36.Kg2 g5 37.hxg5 Rxg5 38.Kh3 h6 39.Re4 Kf5 40.Rg4 Rxg4 41.fxg4 Kf4 42.Kh4 f5 43.gxf5 (0:00:26) 321kN
[-0.48]  d=17  32.Rd4 Kf6 33.Kg2 Rg5 34.Kh1 Ke6 35.h4 Rh5 36.Kg2 g5 37.hxg5 Rxg5 38.Kh3 h6 39.Re4 Kf5 40.Re7 f6 (0:00:22) 321kN
[-0.49]  d=16  32.Rd4 Kf6 33.h4 Kg7 34.Kg2 Ra7 35.f4 f5 36.Kg3 Kh6 37.Rd3 Kh5 38.Rd1 Ra3 39.f3 Ra7 40.Rb1 Rd7 41.Kg2 (0:00:16) 321kN
[-0.49]  d=15  32.Rd4 Kf6 33.h4 Kg7 34.Kg2 Ra7 35.f4 f5 36.Kg3 Kh6 37.Rd3 Kh5 38.Rd1 Ra3 39.f3 Ra7 40.Rb1 Rd7 41.Kg2 Rc7 (0:00:15) 321kN

Naiditsch-Leko

8/8/5pk1/7p/4P2P/5KP1/4R3/4r3 b - -
         
8         
7         
6         
5         
4         
3         
2         
1         
a b c d e f g h 

Start<>EndLookup

[+1.07]  d=20  53...Rf1 54.Kg2 Rd1 55.e5 Rd7 56.Kf3 Kf5 57.exf6 Kxf6 58.Rb2 Rd3 59.Kf4 Rd4 60.Ke3 Rd5 61.Rb6 Kf5 62.Kf3 Rd3 63.Kg2 Rd5 64.Kh3 Ke4 65.Rh6 Rb5 66.Rg6 Ra5 67.Re6 Kf5 68.Rc6 Rb5 (0:01:03) 284kN
[+1.07]  d=19  53...Rf1 54.Kg2 Rd1 55.e5 Rd7 56.Kf3 Kf5 57.exf6 Kxf6 58.Rb2 Rd3 59.Kf4 Rd4 60.Ke3 Rd5 61.Rb6 Kf5 62.Kf3 Rd3 63.Kg2 Rd5 64.Kh3 Ke4 65.Rh6 Rb5 66.Rg6 Ra5 67.Re6 Kf5 68.Rh6 Ke4 (0:00:56) 212kN
[+1.07]  d=18  53...Rf1 54.Kg2 Rd1 55.e5 Rd7 56.Kf3 Kf5 57.exf6 Kxf6 58.Rb2 Rd3 59.Kf4 Rd4 60.Ke3 Rd5 61.Rb6 Kf5 62.Kf3 Rd3 63.Kg2 Rd5 64.Kh3 Ke4 65.Rh6 Rb5 66.Rg6 Ra5 67.Re6 Kf5 68.Rh6 Ke4 (0:00:49) 157kN
[+1.07]  d=17  53...Rf1 54.Kg2 Rb1 55.e5 Rb7 56.exf6 Rb6 57.Rf2 Rxf6 58.Rb2 Rf5 59.Rb6 Rf6 60.Rb5 Rf5 61.Rb6 Rf6 62.Rb5 Rf5 63.Rb6 Rf6 64.Rb5 Rf5 65.Rb6 Rf6 66.Rb5 Rf5 67.Rb6 Rf6 68.Rb5 Rf5 (0:00:30) 107kN
[+1.01]  d=16  53...Rf1 54.Kg2 Rc1 55.Re3 Kf7 56.Rb3 Kg6 57.Rb6 Rc2 58.Kf3 Rc3 59.Kf2 Kf7 60.Rb5 Kg6 61.e5 Rc2 62.Kf3 Rc3 63.Kg2 Rc2 64.Kh3 fxe5 65.Rxe5 Rc6 66.Rg5 Kh6 67.Rd5 Kg6 68.Kg2 Rf6 (0:00:13) 102kN
[+1.01]  d=15  53...Rf1 54.Kg2 Rb1 55.e5 Rb7 56.exf6 Rb6 57.Rf2 Rxf6 58.Rb2 Rf5 59.Rb6 Rf6 60.Rb5 Rf8 61.Rg5 Kh6 62.Rd5 Kg6 63.Kh3 Rf6 64.Rg5 Kh6 65.Rc5 Rb6 66.Kg2 Kg6 67.Rg5 Kh6 68.Rd5 Kg6 (0:00:11) 93kN

A little disappointing, one would figure that these quite common misevaluations would be not too difficult to add knowledge about.
Some kind of built-in pattern tablebase, perhaps?
Parent - - By Xaggard (**) Date 2010-07-20 08:10
4kb2/3n4/pn2N3/4p3/8/5b2/PPP5/1K1R1B2 w - -


This position is deadly drawn. Any engine except of Rykba shows here near a draw. Rybka gives about +1 for white :twisted:
Totally wrong misevaluation!
Parent - - By dr_zied_haddad (**) [fr] Date 2010-07-20 11:08
Hi
The evaluation of this position is very very complicated, and i don't thik that any engine's evaluation is worth to trust.
Let's talk a little bit:
White got a rook + pawn vs Bishop pair or Knight pair. Advantages for white: a chain of three pawns with possibility of advancing these pawns and promote, weaknesses: It throws Bishop pair, black got an isolated passed pawn.

The overall evaluation tends to tell that the position could be abouit equal if both sides plays accurately, but white advante with his extra pawn and chain of almost two passed pawns could become decisive.

In a matter of pure chess skills, white's chances to win are better, because it's position could improve easily in case of even slightly inferior black play. Black to improve his position and take advantage from his passed pawn is not easy. Without a black rook The rook gets the advantage, the bishop pair could be strong, but for that black must optimise his pieces positions, which is not an easy task.

The evaluation of Rybka in my opinion is the most realistic among other engines. But i don't think it diserves +1, may be + 0.55

Would like to see how cluster rybka evaluates this position
Parent - - By Xaggard (**) Date 2010-07-20 17:30
If you'll run eng-eng matches (like I did) where rykba plays for white you'll get nothing more than draws. Rykba will display you +1 or even +1.60 (like vs hIARCS) but this is all due to her misunderstanding of what is going on on the board.
Parent - By dr_zied_haddad (**) [fr] Date 2010-07-20 19:12
Hi
I haven't said that Rybka understood all subtilities of t his position, i said i think it's the closest to a correct evaluation, as it evaluates white is the better side.
In fact; when asking some grandmasters (i'm also a 2200 rated player, never got a FIDE rating but my national rating with many FIDE rated players), they think that white is slightly better, not close to have a clear advantage if black plays flowlessly.

It seems that Rybka's evaluation of Bishop pair is lower than the evaluation by the other engines. Could be also an evaluation in a different manner of pawn structure and passed pawns.
Vas could help in better understanding of what happens.
Parent - - By JohnL (***) Date 2010-07-20 19:05
OK thanks, forgot about this thread. Good idea.
Parent - By dr_zied_haddad (**) [fr] Date 2010-07-20 19:13
You are wellcome my friend.
Continue your work and feed this post by your excellent comments and positions
- - By Regularuser (***) [gb] Date 2010-07-20 10:23
There have been lots of examples on here where R4 misevaluates endgames but very few where it plays badly.  Indeed there have been examples in other threads where its evaluation is suspect but where it nonetheless plays very accurately.

Presumably misevaluation will lead to misplaying some of the time.   Could someone please post examples where they think this has happened, giving the moves as well as the position.   I would be particulalrly interested in examples of where R4 transitions to a rook ending in error because it over-evaluates the resulting positions. This has been mentioned numerous times in other threads so presumbly there are plenty of examples. 

Thnaks in advance.
Parent - By dr_zied_haddad (**) [fr] Date 2010-07-20 11:34 Edited 2010-07-20 11:39
Hi
Your statement gives in fact a part of the truth.

Misunderstanding a position (wrong evaluation) could be followed by an accurate play, especially if the position contains only one or two moves that doesn't loose the game or worsens it till loss.

But misunderstanding a position is in it self complicated to interpret. In fact, the engine will loose the right lines before, because it chooses other lines that are possibly inferior. Especially for those who use engines for analyses, this is an important issue.

So apart from misunderstanding subtle positions, it's the overall quality of play that is important to reach certain positions.
- By Yoav Dothan (**) [il] Date 2010-07-21 11:35
In this endgame
8/BB6/3P4/n1K2k2/1r1P4/8/2n5/8 w - - 0 1


Rybka gives -4.31 after depth 20 while it is a draw. ({Deep Rybka 4 w32y:} 1. Bc8+ Kf6 2. Kd5 Rb5+ 3. Ke4 Nb7 4. d7 Ke7 5.
Bb8 Nd8 6. Ba7 Nc6 7. Bc5+ Kd8 8. Kd3 Ne1+ 9. Ke4 Rb1 10. Ba6 Nb8 11. Be2 Nxd7
12. Bd6 Nc2 13. Kd3 Nb4+ 14. Kc4 Nb6+ 15. Kc5 Nc2 16. Bd3 {-4.31/20})

After playing the moves 1. Bd5  Nb7+ 2. Kc6 Nxd4+ 3. Kc7 Ke5 4. Bxb7 Nb5+ 5. Kb6 Nxd6+ 6. Kc6 Nxb7 7. Bb6 Nd6 8. Bc7
Rd4 rybka sees the draw
- By JohnL (***) Date 2010-07-21 17:19 Edited 2010-07-21 17:38
Some examples from today's elite games. In both positions below the players agreed a draw, the one material ahead not even bother trying. But Rybka 4 thinks that was a big mistake.

Vachier-Lagrave 2723 - Nguyen 2617 1/2-1/2

8/8/6pk/7p/4KP2/2R5/8/6r1 b - -


[-1.33]  d=20  1...Re1 2.Kf3 Kg7 3.Rc6 Kf7 4.Rc5 Kf6 5.Ra5 Re8 6.Rc5 Re7 7.Ra5 Kf7 8.Rg5 Ra7 9.f5 Kf6 10.Rxg6 Kxf5 11.Rc6 Ra5 (0:02.08)
[-1.33]  d=19  1...Re1 2.Kf3 Kg7 3.Rc6 Kf7 4.Rc5 Kf6 5.Ra5 Re8 6.Rc5 Re7 7.Ra5 Kf7 8.Ra8 Kg7 9.Ra6 Rb7 10.f5 Rb3 11.Kf4 Rb4 12.Kf3 gxf5 13.Ke3 h4 (0:01.45)
[-1.33]  d=18  1...Re1 2.Kf3 Kg7 3.Rc6 Kf7 4.Rc5 Kf6 5.Rc8 Re6 6.Rf8 Kg7 7.Rb8 Rc6 8.Rb3 Ra6 9.Ke4 Kf6 10.Kf3 Ra5 11.Ke4 h4 12.Rb6 Kg7 13.Kf3 Ra3 14.Kg4 h3 15.Rb2 Rd3 16.Rh2 (0:00.50)
[-1.33]  d=17  1...Re1 2.Kf3 Kg7 3.Rc6 Kf7 4.Rc5 Kf6 5.Rc8 Re6 6.Rf8 Kg7 7.Rc8 Ra6 8.Rc3 Kf6 9.Rb3 Ra5 10.Ke4 h4 11.Rc3 Ra2 12.Rc6 Kg7 (0:00.37)
[-1.33]  d=16  1...Re1 2.Kf3 Kg7 3.Rc6 Kf7 4.Ra6 Re6 5.Ra5 Rc6 6.Ke4 Kf6 7.Ra3 Rc4 8.Kf3 Rc5 9.Ke4 h4 10.Ra6 Kg7 11.Ra3 Kh6 12.Rh3 Kh5 (0:00.27)
[-1.33]  d=15  1...Re1 2.Kf3 Kg7 3.Rc6 Kf7 4.Ra6 Re6 5.Ra5 Rc6 6.Ra7 Ke6 7.Ra1 Kf5 8.Ra5 Kf6 9.Ra3 Rc5 10.Ke4 Rc4 11.Kf3 Rc5 12.Ke4 Rc4 13.Kf3 Rc5 14.Ke4 Rc4 15.Kf3 Rc5 16.Ke4 Rc4 (0:00.11)

Le Quang 2681 - Kramnik 2790 1/2-1/2

6k1/5p2/4pbp1/3r4/8/5QP1/4PPK1/8 b - -


[+1.06]  d=20  1...Kg7 2.Qe4 Rd8 3.f4 Rc8 4.e3 Rc3 5.Kf3 Rc1 6.Kf2 Rc3 7.Qb4 Rc1 8.Qd2 Rc8 9.Qd7 Rd8 10.Qb7 e5 (0:00.21)
[+1.06]  d=19  1...Kg7 2.Qe4 Rd8 3.f4 Rc8 4.e3 Rc3 5.Kf3 Rc1 6.Kf2 Rc3 7.Qb4 Rc1 8.Qd2 Rc8 9.Qd7 Rd8 10.Qb7 e5 (0:00.11)
[+1.06]  d=18  1...Kg7 2.Qe4 Rd8 3.f4 Rc8 4.e3 Rc3 5.Kf3 Rc1 6.Kf2 Ra1 7.Qc2 Ra3 8.Qc7 Rd3 9.Qb8 (0:00.09)
[+1.06]  d=17  1...Kg7 2.Qe4 Rd2 3.f4 Rb2 4.Kf3 Rb3 5.e3 Rc3 6.Kf2 Rb3 7.Kg2 Rc3 8.Kf2 Rb3 9.Kg2 Rc3 10.Kf2 Rb3 11.Kg2 Rc3 12.Kf2 Rb3 13.Kg2 Rc3 14.Kf2 Rb3 15.Kg2 Rc3 16.Kf2 Rb3 (0:00.04)
[+1.05]  d=16  1...Kg7 2.Qe4 Rd2 3.f4 Rd4 4.Qc2 Rb4 5.Qc5 Rb3 6.Qc4 Ra3 7.Kf2 Rc3 8.Qe4 Rb3 9.Qa4 Rc3 10.Qa7 e5 (0:00.01)
[+1.05]  d=15  1...Kg7 2.Qe4 Rd2 3.f4 Rd4 4.Qc2 Rb4 5.Qc5 Rb3 6.e3 Rc3 7.Qa7 e5 (0:00.01)
- - By JohnL (***) Date 2010-07-21 17:36
Some old examples from this forum:

Sideline from Topalov-Anand match, dead draw.

7K/1p4BP/p5k1/P2b4/1P6/8/8/8 b - -


[+1.31]  d=35  1...Be6 2.Bd4 Bd5 3.Bc5 Kf7 4.Be3 Kg6 5.Bg5 Be6 6.Bc1 Ba2 7.Be3 Bb3 8.Bf4 Bd5 9.Bg5 Be6 10.Bc1 Ba2 11.Be3 Bb3 12.Bf4 Bd5 13.Bg5 Be6 14.Bc1 Ba2 15.Be3 Bb3 16.Bf4 Bd5 (0:01.10)
[+1.31]  d=34  1...Be6 2.Bd4 Bd5 3.Bc5 Kf7 4.Be3 Kg6 5.Bg5 Be6 6.Bc1 Ba2 7.Be3 Bb3 8.Bf4 Bd5 9.Bg5 Be6 10.Bc1 Ba2 11.Be3 Bb3 12.Bf4 Bd5 13.Bg5 Be6 14.Bc1 Ba2 15.Be3 Bb3 16.Bf4 Bd5 (0:00.46)
[+1.31]  d=33  1...Be6 2.Bd4 Bd5 3.Bc5 Kf7 4.Be3 Kg6 5.Bg5 Be6 6.Bc1 Ba2 7.Be3 Bb3 8.Bf4 Bd5 9.Bg5 Be6 10.Bc1 Ba2 11.Be3 Bb3 12.Bf4 Bd5 13.Bg5 Be6 14.Bc1 Ba2 15.Be3 Bb3 16.Bf4 Bd5 (0:00.26)
[+1.31]  d=32  1...Be6 2.Bd4 Bc4 3.Bc3 Bb3 4.Bd2 Be6 5.Bc1 Ba2 6.Be3 Bb3 7.Bf4 Bd5 8.Bg5 Be6 9.Bc1 Ba2 10.Be3 Bb3 11.Bf4 Bd5 12.Bg5 Be6 13.Bc1 Ba2 14.Be3 Bb3 15.Bf4 Bd5 16.Bg5 Be6 (0:00.19)
[+1.31]  d=31  1...Be6 2.Bd4 Bc4 3.Bc3 Bb3 4.Bd2 Be6 5.Bc1 Ba2 6.Be3 Bb3 7.Bf4 Bd5 8.Bg5 Be6 9.Bc1 Ba2 10.Be3 Bb3 11.Bf4 Bd5 12.Bg5 Be6 13.Bc1 Ba2 14.Be3 Bb3 15.Bf4 Bd5 16.Bg5 Be6 (0:00.13)
[+1.31]  d=30  1...Be6 2.Bd4 Bc4 3.Bc3 Bb3 4.Bd2 Be6 5.Bc1 Ba2 6.Be3 Bb3 7.Bf4 Bd5 8.Bg5 Be6 9.Bc1 Ba2 10.Be3 Bb3 11.Bf4 Bd5 12.Bg5 Be6 13.Bc1 Ba2 14.Be3 Bb3 15.Bf4 Bd5 16.Bg5 Be6 (0:00.06)

Granda - Nakamura, supposedly a drawn position (and the game was drawn later).

R7/P6k/6p1/r7/4p1KP/4P3/8/8 w - -


[+3.20]  d=26  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Kg7 11.Kb6 Rb1 12.Kc5 Ra1 13.Kd5 Ra4 14.Kc6 Ra3 15.Kb7 Rb3 16.Ka6 (0:00.43)
[+3.20]  d=25  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.28)
[+3.20]  d=24  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.17)
[+3.20]  d=23  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.11)
[+3.20]  d=22  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.08)
[+3.20]  d=21  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.04)
[+3.20]  d=20  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.02)
[+3.20]  d=19  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.01)
[+3.20]  d=18  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kb4 Ra1 10.Kb5 Ra3 11.Kb4 Ra1 12.Kb5 Ra3 13.Kb4 Ra1 14.Kb5 Ra3 15.Kb4 Ra1 16.Kb5 (0:00.01)
[+3.20]  d=17  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kc6 Kg7 10.Kd5 Ra4 11.Ke5 Kh7 12.Kd5 Kg7 13.Ke5 Kh7 14.Kd5 Kg7 15.Ke5 Kh7 16.Kd5 (0:00.00)
[+3.20]  d=16  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kc6 Kg7 10.Kd5 Ra4 11.Ke5 Kh7 12.Kd5 Kg7 13.Ke5 Kh7 14.Kd5 Kg7 15.Ke5 Kh7 16.Kd5 (0:00.00)
[+3.20]  d=15  1.Kf4 Ra4 2.Ke5 Kg7 3.Kd5 Kh7 4.Kc6 Ra3 5.Kb7 Rb3 6.Ka6 Ra3 7.Kb6 Rb3 8.Kc5 Ra3 9.Kc6 Kg7 10.Kd5 Ra4 11.Ke5 Kh7 12.Kd5 Kg7 13.Ke5 Kh7 14.Kd5 Kg7 15.Ke5 Kh7 16.Kd5 (0:00.00)

Carlsen - Bacrot. Draw agreed in this position. Rybka 4 thinks white has pretty good chances to win.

6k1/7p/2b5/1R4p1/4P3/2B5/5KPP/8 b - -


[+0.58]  d=30  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ba6 4.Kd5 Bb7 5.Ke5 Kg6 6.Bd4 h6 7.Bc3 Bc6 8.Bb4 Kf7 9.g3 Kg7 10.Bd2 Kg6 11.Bc3 Bb7 12.Bb4 Kg7 13.Bd6 Bc6 14.Kd4 Bb7 15.h4 Kg6 16.hxg5 Kxg5 (0:02.43)
[+0.58]  d=29  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ba6 4.Kd5 Bb7 5.Ke5 Kg6 6.Bd4 h6 7.Bc3 Bc6 8.Bb4 Kf7 9.g3 Kg7 10.Bd2 Kg6 11.Bc3 Bb7 12.Bb4 Kg7 13.Bd6 Bc6 14.Kd4 Kf7 15.h4 gxh4 16.gxh4 Kg6 (0:01.44)
[+0.58]  d=28  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ba6 4.Kd5 Bb7 5.Ke5 Kg6 6.Bd4 h6 7.Bc3 Bc6 8.Bb4 Kf7 9.g3 Kg7 10.Bd2 Kg6 11.Bc3 Bb7 12.Bb4 Kg7 13.Bd6 Bc6 14.Kd4 Kf7 15.h4 gxh4 16.gxh4 Kg6 (0:01.16)
[+0.58]  d=27  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.Bb4 Kf7 7.Bc5 Kg7 8.Ke5 Kg6 9.Bd4 Bb7 10.Be3 Ba8 11.Bb6 Bc6 12.g4 Bd7 13.h3 Bb5 14.Kd6 Bf1 15.e5 (0:00.41)
[+0.58]  d=26  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.Bb4 Kf7 7.Bc5 Kg7 8.Ke5 Kg6 9.Bd4 Bb7 10.Be3 Ba8 11.Bb6 Bc6 12.Ba7 Kg7 13.Bf2 Kg6 14.Ba7 Kg7 15.Bf2 Kg6 16.Ba7 Kg7 (0:00.28)
[+0.58]  d=25  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.Bb4 Kf7 7.Bc5 Kg7 8.Ke5 Kg6 9.Bd4 Bb7 10.Be3 Ba8 11.Bb6 Bc6 12.Ba7 Kg7 (0:00.19)
[+0.58]  d=24  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.Bb4 (0:00.11)
[+0.58]  d=23  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.Bb4 Kf7 7.Bc5 Kg7 8.Ke5 Kg6 9.Bd4 h5 10.Be3 (0:00.07)
[+0.57]  d=22  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.h4 Kf6 7.hxg5 hxg5 8.Bb4 Bb7 9.Ba5 Kg6 10.Bb6 Kf6 11.Bd8 Kg6 12.Ba5 Bc6 13.Bc3 Kf7 14.Ke5 Kg6 15.Bd4 Kh5 16.Kf5 Bd7 (0:00.05)
[+0.57]  d=21  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.h4 Kf6 7.hxg5 hxg5 8.Bb4 Bb7 9.Ba5 Kg6 10.Bb6 Kf6 11.Bc7 (0:00.03)
[+0.57]  d=20  1...Bxb5 2.Ke3 Kf7 3.Kd4 Ke6 4.Bd2 h6 5.g3 Bc6 6.h4 Kf6 7.hxg5 hxg5 8.Bb4 Bb7 9.Ba5 Kg6 10.Bb6 Kf6 11.Bc7 Bc6 12.Bd8 Kg6 13.e5 Kf5 (0:00.02)
Parent - - By dr_zied_haddad (**) [fr] Date 2010-07-22 07:34
Yeah
in fact those positions are a draw. May be rook ending 2 vs 1 pawn with the h5 passed pawn and rook on g1 square is more difficult to draw. I think it needs much technique +++, and play without any weakness to draw it.
Parent - By dr_zied_haddad (**) [tn] Date 2010-07-24 09:04
From my summer holidays
please continue feeding this post with your positions
- - By JohnL (***) Date 2010-07-25 13:43
Some more R4 examples from current top tournaments:

Kramnik - Mamedyarov

4r1k1/5ppp/5b2/8/4N3/4P3/4KPPP/1R6 w - -


R4 wants to play Nxf6?!, but Kramnik didn't agree and I think that black has good drawing chances after that move.
So this could be an example where bad endgame evaluation affects the actual outcome.

Multi variation(3):
[+1.91]  d=19  1.Nxf6 gxf6 2.Rb5 Kg7 3.g4 Ra8 4.h4 Ra2 5.Kf3 Rc2 6.Rb7 Rc3 (0:03.13)
[+1.37]  d=19  1.f4 Be7 2.Rb7 g6 3.Kf3 Kg7 4.Nc3 Rc8 (0:03.17)
[+1.31]  d=19  1.Rb7 Be5 2.g3 h6 (0:03.54)

Even in the end position where Mamedyarov resigned, R4 thinks the score is lower than after Nxf6?! until the search finds the forced win sequence.

6k1/4r1pp/8/1R3p2/3P1P2/3K2P1/7P/8 b - -


[+2.50]  d=20  5...g6 6.Re5 Rc7 7.d5 Kf7 8.Re2 Rc1 (0:02.52)
[+2.62]  d=19  5...g6 6.Re5 Rc7 7.d5 Kg7 8.Kd4 (0:01.59)
[+2.24]  d=19  5...g6 6.Re5 Rd7 7.d5 Kf7 (0:01.00)
[+1.97]  d=19  5...g6 6.Re5 Rd7 7.d5 Kf7 (0:00.51)
[+2.09]  d=18  5...g6 6.Re5 Ra7 7.d5 Ra4 8.d6 Kf7 9.Re7 (0:00.45)
[+2.25]  d=18  5...g6 6.Re5 Ra7 7.d5 Kf7 8.Kc4 Ra4 9.Kc5 Ra5 10.Kb6 (0:00.32)
[+2.10]  d=18  5...g6 6.Re5 Ra7 7.d5 Kf7 8.Kc4 Ra4 9.Kc5 Ra5 10.Kb6 (0:00.23)
[+1.95]  d=17  5...g6 6.Re5 Ra7 7.d5 Kf7 8.Kc4 Ra4 9.Kc5 Ra5 10.Kb6 (0:00.18)
[+1.98]  d=17  5...g6 6.d5 Kf7 7.Rb2 Ra7 (0:00.15)
[+1.83]  d=16  5...g6 6.d5 Kf7 7.Rb2 Ra7 (0:00.10)
[+1.83]  d=16  5...g6 6.d5 (0:00.09)
[+1.68]  d=15  5...g6 6.d5 (0:00.07)
[+1.93]  d=15  5...g6 6.d5 Kf7 7.Kd4 Kf6 8.Rb6 Kf7 9.h4 Re4 10.Kc5 h5 11.d6 Re3 12.Rb8 Rd3 13.Kc6 (0:00.04)
[+1.78]  d=15  5...g6 6.d5 Kf7 7.Kd4 Kf6 8.Rb6 Kf7 9.h4 Re4 10.Kc5 h5 11.d6 Re3 12.Rb8 Rd3 13.Kc6 (0:00.01)
[+1.63]  d=14  5...g6 6.d5 Kf7 7.Kd4 Kf6 8.Rb6 Kf7 9.h4 Re4 10.Kc5 h5 11.d6 Re3 12.Rb8 Rd3 13.Kc6 (0:00.00)
[+1.58]  d=13  5...g6 6.d5 (0:00.00)
[+1.58]  d=12  5...g6 6.d5 (0:00.00)
[+1.58]  d=11  5...g6 6.d5 (0:00.00)
[+1.58]  d=10  5...g6 6.d5 (0:00.00)
[+1.58]  d=9  5...g6 6.d5 (0:00.00)

Le Qu - Le Ko

Game was drawn which probably was correct. R4:s analysis (with 5-men TB) shows no increased score up to at least d35. It the score doesn't increase with depth I think Rybka should simply lower it.

3R4/8/7r/3bp3/3k1p2/5P2/4BK2/8 w - -


[-1.33]  d=35  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Kg4 Rc5 8.Kf5 Ra5 9.Rd1 Rb5 10.Rd7 Rc5 11.Rd8 Ra5 12.Rd1 Rb5 13.Rd7 Rc5 14.Rd8 Ra5 15.Rd1 Rb5 16.Rd7 (1:47.15)
[-1.33]  d=34  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rb6 6.Rd8 Kc4 7.Rc8 Kb4 8.Kg4 Rb5 9.Rc1 Rc5 10.Rd1 Kc4 11.Rd7 Ra5 12.Kf5 Kc3 13.Kg4 Ra4 14.Kf3 Ra3 15.Rd5 Kb4 16.Kg4 (1:09.26)
[-1.33]  d=33  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Kg4 Rc5 13.Rd7 Kc4 14.Kf5 Ra5 15.Rd1 Rb5 16.Rd7 (0:46.35)
[-1.33]  d=32  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Kg4 Rc5 13.Rd7 Kc4 14.Kf5 Ra5 15.Rd1 Rb5 16.Rd7 (0:29.20)
[-1.33]  d=31  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Kg4 Rc5 13.Rd7 Kc4 14.Kf5 Ra5 15.Rd1 Kc3 16.Kg4 (0:17.00)
[-1.33]  d=30  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Kg4 Ra5 13.Kf5 Rc5 14.Rd7 Kc3 15.Kg4 Rb5 16.Rd1 (0:10.58)
[-1.33]  d=29  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Rc1 Ra5 13.Rc8 Ra3 14.Kg4 Rc3 15.Re8 Rc5 16.Rd8 (0:07.13)
[-1.33]  d=28  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Rc1 Ra5 13.Kg4 Rc5 14.Rd1 Rb5 15.Kf5 Ra5 16.Rc1 (0:04.58)
[-1.33]  d=27  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Rc1 Ra5 13.Kg4 Rc5 14.Rd1 Rb5 15.Kf5 Ra5 16.Rc1 (0:03.39)
[-1.33]  d=26  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Rc1 Ra5 13.Kg4 Rc5 14.Rd1 Rb5 15.Kf5 Ra5 16.Rc1 (0:03.22)
[-1.33]  d=25  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Rd6 4.Re8 Kd4 5.Kf3 Rc6 6.Rd8 Kc4 7.Rd1 Rc5 8.Rc1 Kb5 9.Rb1 Ka4 10.Rb7 Rb5 11.Rc7 Kb4 12.Rc1 Ra5 13.Kg4 Rc5 14.Rf1 Kc3 15.Rd1 Rc4 16.Kf3 (0:02.31)
[-1.33]  d=24  1.Bd3 Kc5 2.Be4 Bxe4 3.fxe4 Kc4 4.Rd5 Rh5 5.Kg1 Rg5 6.Kf2 Kc3 7.Kf3 Kb3 8.Rc5 Kb4 9.Rd5 Kc3 10.Kf2 Kc4 11.Ke2 Kb4 12.Kf2 Kc4 13.Ke2 Kb4 14.Kf2 Kc4 15.Ke2 Kb4 16.Kf2 (0:01.46)
[-1.33]  d=23  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=22  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=21  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=20  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=19  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=18  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=17  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
[-1.33]  d=16  1.Bd3 Kc5 2.Rc8 Kd6 3.Rd8 Ke6 4.Re8 Kf6 5.Kg2 Bc6 6.Rg8 Ke6 7.Rg7 Kd6 8.Bc2 Bd5 9.Ra7 Kc5 10.Ra5 Kd4 11.Ra4 Kc3 12.Be4 (0:00.00)
Parent - By dr_zied_haddad (**) [tn] Date 2010-07-27 15:21
Hi from my summer Holidays
I saw these two positions, very instructive.
Rybka is still far from the best humans. May be Rybka6 will approach such level of understunding of positions.
- - By JohnL (***) Date 2010-07-30 21:39 Edited 2010-07-30 21:54
Some more examples below:

I would very much have preferred the (distant) passed pawn as white and also the doubled pawns as black when no passed pawn, otherwise not. My judgement(somewhat calibrated with endgame books):
First position: Not too difficult draw (say +0.15).
The second: Theoretical win, very good chance to win(like +1.25).
The third: Some practical chances but probable draw(perhaps +0.50).
The fourth: Good winning chances(kind of +0.90).
The fifth: At least as good as the fourth (put +1.00)

Order: 2,5,4,3,1

Rybka 4 (with >1000 higher ELO...) including some search:
+1.58->1.92
+1.53->1.55
+1.18->1.18
+0.85->0.85
+0.69->0.97

Order: 1,2,3,4,5 (at low depth)

Who do you agree with? :wink:

r7/5pk1/5p1p/8/8/4PK2/1R3PPP/8 w - -

[+1.92]  d=17  1.Rb5 Ra2 2.g4 Kg6 3.h4 Ra7 4.h5 Kg7 5.Rb2 Kg8 6.Kg3 Kg7 7.Kf4 Rc7 8.f3 Kf8 9.Rb5 Kg7 10.e4 Rc1 11.Rb7 Rc3 12.Rd7 (0:00.02)
[+1.92]  d=16  1.Rb5 Ra2 2.g4 Kg6 3.h4 Ra7 4.h5 Kg7 5.Rb2 Kg8 6.Kg3 Kg7 7.Kf4 Rc7 8.f3 Kf8 9.e4 Kg7 10.Rb6 Rc1 11.Rb7 (0:00.02)
[+1.78]  d=16  1.Rb5 Ra2 2.g4 Kg6 3.h4 Rd2 4.Rb7 Ra2 5.Kg3 Re2 6.h5 Kg7 (0:00.01)
[+1.66]  d=16  1.Kf4 Ra5 2.e4 Kg6 3.g4 h5 4.f3 h4 5.h3 Ra1 6.Rb6 Rg1 7.e5 Rg3 8.Rxf6 Kg7 9.g5 Rxh3 10.Kg4 Rh1 (0:00.01)
[+1.66]  d=15  1.Kf4 Ra5 2.e4 Kg6 3.g4 h5 4.f3 h4 5.h3 Ra1 6.Rb6 Rg1 7.e5 Rg3 8.Rxf6 Kg7 9.g5 Rxh3 10.Kg4 Rh1 (0:00.01)
[+1.65]  d=14  1.Kf4 Ra5 2.h4 h5 3.f3 f5 4.Rd2 Kg6 5.g3 Ra4 6.Rd4 Ra6 7.Rd8 Ra4 8.Ke5 Ra5 9.Rd5 f6 10.Ke6 (0:00.01)
[+1.67]  d=13  1.Kf4 h5 2.e4 Kg6 3.f3 Ra5 4.Rd2 Ra4 5.g4 hxg4 6.fxg4 Ra1 7.h4 Rf1 8.Kg3 Re1 (0:00.00)
[+1.60]  d=13  1.g4 h5 2.gxh5 Ra5 3.h4 Rxh5 4.Kg4 Kg6 5.Rb6 Re5 6.Kf4 Rh5 7.Kg3 Ra5 8.e4 Ra2 9.f3 (0:00.00)
[+1.60]  d=12  1.g4 h5 2.gxh5 Ra5 3.h4 Rxh5 4.Kg3 Rc5 5.Rb7 Kg6 6.Rd7 Rc3 7.Kf4 Rc5 8.f3 (0:00.00)
[+1.61]  d=11  1.g4 h5 2.gxh5 Ra5 3.h4 Rxh5 4.Kg3 Rc5 5.Rb7 Kg6 6.Rd7 Rc3 7.Kf4 Rc5 8.Kg4 (0:00.00)
[+1.62]  d=10  1.g4 h5 2.gxh5 Ra5 3.Kg4 Rg5 4.Kf4 Rxh5 5.f3 Rh4 6.Kf5 Rh5 7.Ke4 Rh3 8.Rg2 Kf8 (0:00.00)
[+1.62]  d=9  1.g4 h5 2.gxh5 Ra5 3.Kg4 Rg5 4.Kh4 Kh6 5.Rb7 Rxh5 6.Kg3 Kg6 7.h4 f5 8.Rb6 (0:00.00)
[+1.62]  d=8  1.g4 h5 2.gxh5 Ra5 3.Kg4 Rg5 4.Kh4 Kh6 5.Rb7 Rxh5 6.Kg3 Rg5 7.Kf4 Rg2 8.f3 (0:00.00)
[+1.63]  d=7  1.g4 Kg6 2.Kf4 Ra4 3.e4 Ra3 4.Rc2 Rb3 (0:00.00)
[+1.66]  d=6  1.g4 Kg6 2.Kf4 Ra3 3.f3 Rc3 4.e4 h5 (0:00.00)
[+1.55]  d=6  1.Kf4 h5 2.h3 Kg6 3.e4 Ra1 4.Rd2 (0:00.00)
[+1.56]  d=5  1.Kf4 Kg6 2.g4 Ra1 3.f3 h5 4.gxh5 (0:00.00)
[+1.59]  d=4  1.Kf4 Kg6 2.g4 Ra3 3.h3 h5 4.gxh5 (0:00.00)
[+1.54]  d=3  1.Kf4 Kg6 2.g4 Ra3 3.e4 (0:00.00)
[+1.58]  d=3  1.Kf4 Ra7 2.e4 Kg6 (0:00.00)

r7/5pk1/5p1p/8/8/3P1K2/1R3PPP/8 w - -

[+1.55]  d=17  1.d4 Kg6 2.Rb3 Rd8 3.Ke3 Rd6 4.Rb7 Re6 5.Kf4 Rd6 (0:00.02)
[+1.55]  d=16  1.d4 Kg6 2.Rb3 Rd8 3.Ke3 Rd7 4.g3 Re7 5.Kf3 Ra7 6.h3 Rc7 7.Ke4 f5 8.Kf4 Rc4 9.Ke3 Rc2 10.d5 (0:00.01)
[+1.55]  d=15  1.d4 Kg6 2.Rb3 Rd8 3.Ke3 Rd7 4.g3 Re7 5.Kf3 Ra7 6.h3 Rc7 7.Ke4 f5 8.Kf4 Rc4 9.Ke3 Rc2 10.d5 (0:00.01)
[+1.55]  d=14  1.d4 Kg6 2.Rb3 Rd8 3.Ke3 Rd7 4.g3 Re7 5.Kf3 Ra7 6.h3 Rc7 7.Ke4 f5 8.Kf4 Rc2 9.Rb6 (0:00.01)
[+1.53]  d=13  1.d4 Kg6 2.Rb3 Rd8 3.Ke3 h5 (0:00.00)
[+1.54]  d=12  1.d4 Kg6 2.g3 Ra1 3.Ke3 Ra3 4.Kf4 Ra4 5.Ke4 Kg5 6.Rb5 Kg6 7.Rb7 Ra2 8.Ke3 Ra3 9.Kf4 Ra2 10.Ke3 Ra3 11.Kf4 Ra2 12.Ke3 Ra3 13.Kf4 Ra2 14.Ke3 Ra3 15.Kf4 Ra2 16.Ke3 (0:00.00)
[+1.57]  d=11  1.d4 Kg6 2.g3 Ra1 3.Ke4 Re1 4.Kf4 Rd1 5.Ke3 h5 6.f3 Re1 7.Re2 Rd1 (0:00.00)
[+1.57]  d=10  1.d4 Kg6 2.g3 Ra1 3.Ke4 Re1 4.Kf4 Rd1 5.Ke3 h5 6.f3 Re1 7.Re2 Rd1 (0:00.00)
[+1.52]  d=9  1.d4 Kg6 2.g3 Ra1 3.Rb7 Ra3 4.Kf4 Ra2 5.Ke3 Ra3 6.Ke2 Ra2 7.Kf3 Ra3 8.Kf4 Ra2 9.Ke3 Ra3 10.Ke2 Ra2 11.Kf3 Ra3 12.Kf4 Ra2 13.Ke3 Ra3 14.Ke2 Ra2 15.Kf3 Ra3 16.Kf4 (0:00.00)
[+1.54]  d=8  1.d4 Kg6 2.g3 Rd8 3.Rb4 Rd6 4.Kf4 Rd8 5.Rc4 h5 6.h3 (0:00.00)
[+1.53]  d=7  1.d4 Kg6 2.Ke4 Re8 3.Kd3 Rg8 4.f3 Rd8 5.Rb7 f5 (0:00.00)
[+1.58]  d=6  1.d4 Kg6 2.Rb7 Ra3 3.Ke2 Ra2 4.Ke3 Ra3 5.Ke4 Ra2 (0:00.00)
[+1.58]  d=5  1.d4 Kg6 2.Rb7 Ra3 3.Ke2 Ra2 4.Ke3 Ra3 5.Ke4 Ra2 (0:00.00)
[+1.59]  d=4  1.d4 Rd8 2.Ke3 Re8 3.Kf4 Rd8 4.Rb4 (0:00.00)
[+1.64]  d=3  1.d4 Rd8 2.Ke3 Re8 3.Kf4 Rd8 (0:00.00)
[+1.58]  d=2  1.d4 Rd8 2.Ke3 (0:00.00)
[+1.41]  d=2  1.Ke4 Re8 2.Kf5 Re5 3.Kf4 Rd5 (0:00.00)
[+1.47]  d=1  1.Ke4 Re8 2.Kf5 Re5 3.Kf4 (0:00.00)
[+1.35]  d=1  1.d4 (0:00.00)
[+1.53]  d=1  1.d4 (0:00.00)

8/5p2/r4kpp/8/8/4PK2/1R3PPP/8 w - -


[+1.18]  d=17  1.h4 Ra5 2.Rb6 Kg7 3.g3 Rf5 4.Ke2 Ra5 5.e4 f5 6.exf5 Rxf5 7.f3 Re5 8.Kf2 Ra5 9.g4 h5 10.Kg3 hxg4 11.fxg4 (0:00.01)
[+1.18]  d=16  1.h4 Ra5 2.Rb6 Ke7 3.g3 Rf5 4.Ke2 Re5 5.f3 Ra5 6.e4 Ra3 7.h5 Ra2 8.Ke3 Ra3 9.Kf4 (0:00.00)
[+1.19]  d=15  1.h4 Ke6 2.g4 g5 3.Rb5 f6 4.h5 Rc6 5.Rb7 f5 6.gxf5 Kxf5 (0:00.00)
[+1.18]  d=14  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Rb7 f5 5.Rg7 (0:00.00)
[+1.19]  d=13  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Rb7 f5 5.Rg7 (0:00.00)
[+1.19]  d=12  1.h4 Ke6 2.g4 g5 3.Rb5 f6 4.h5 Rc6 5.Rb7 f5 6.Rb5 fxg4 7.Kxg4 Rc4 8.f4 gxf4 9.exf4 Rc6 (0:00.00)
[+1.18]  d=11  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Rb7 f5 (0:00.00)
[+1.18]  d=10  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Rb8 Rc6 5.Re8 Kf6 6.Rd8 (0:00.00)
[+1.15]  d=9  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Ke4 Rc6 5.f4 Rc4 (0:00.00)
[+1.16]  d=8  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Ke4 Rc6 5.f4 Rc4 6.Kf3 gxf4 (0:00.00)
[+1.14]  d=7  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Ke4 Rc6 5.Rb7 Rc4 (0:00.00)
[+1.17]  d=6  1.h4 Ke6 2.g4 g5 3.h5 Rd6 4.Rb8 f6 5.Kg3 (0:00.00)
[+1.14]  d=6  1.g4 g5 2.Kg3 Ke5 3.Rb7 Ke6 4.e4 f6 5.f3 (0:00.00)
[+1.14]  d=5  1.g4 g5 2.Kg3 Ke5 3.Rb7 Ke6 4.Rb8 f5 (0:00.00)
[+1.08]  d=5  1.Rb5 g5 2.g4 Kg6 3.Kg3 Ra2 4.Rb7 Re2 5.Rb8 f6 (0:00.00)
[+1.03]  d=5  1.Rb7 Ke6 2.g4 Ra5 3.h4 f5 4.Rb6 Kf7 5.gxf5 Rxf5 6.Kg3 Re5 7.Kf4 Rf5 8.Kg3 Re5 9.Kf4 Rf5 10.Kg3 Re5 11.Kf4 Rf5 12.Kg3 Re5 13.Kf4 Rf5 14.Kg3 Re5 15.Kf4 Rf5 16.Kg3 (0:00.00)
[+1.18]  d=4  1.Rb7 Ke6 2.g4 Ra5 3.h4 f5 4.Rb6 Kf7 5.gxf5 Rxf5 6.Kg3 Re5 7.Kf4 Rf5 8.Kg3 Re5 9.Kf4 Rf5 10.Kg3 Re5 11.Kf4 Rf5 12.Kg3 Re5 13.Kf4 Rf5 14.Kg3 Re5 15.Kf4 Rf5 16.Kg3 (0:00.00)
[+1.18]  d=3  1.Rb7 Ra8 2.h4 Ra2 3.Rb6 Kg7 4.g4 Rd2 5.Kg3 h5 6.gxh5 (0:00.00)
[+1.19]  d=2  1.Rb7 Ra8 2.h4 Ra2 3.g4 h5 4.gxh5 gxh5 5.e4 Ke6 6.Ke3 (0:00.00)
[+1.18]  d=1  1.Rb7 Ra8 2.h4 Ra2 3.g4 h5 4.gxh5 gxh5 5.e4 Ke6 6.Ke3 (0:00.00)

8/5p2/r4kpp/8/8/3P1K2/1R3PPP/8 w - -


[+0.85]  d=15  1.Rb7 Rd6 2.Ke3 Re6 3.Kd2 g5 4.h3 (0:00.02)
[+0.87]  d=14  1.Rb7 Rd6 2.Ke3 Re6 3.Kd2 g5 4.g3 Ra6 5.h3 Kg6 (0:00.02)
[+0.85]  d=13  1.Rb7 Ke6 (0:00.00)
[+0.84]  d=12  1.Rb7 Ke6 2.Ke3 Ra1 3.h3 Re1 4.Kf4 Re2 5.Kf3 (0:00.00)
[+0.84]  d=11  1.Rb7 Ke6 2.Ke3 Ra1 3.h3 Re1 4.Kf4 Re2 5.Kf3 Ra2 (0:00.00)
[+0.88]  d=10  1.Rb7 Ke6 2.g4 Ra3 3.Ke3 Ra1 4.Rb6 Kd5 5.Rb5 Kc6 6.Rb8 Re1 7.Kf3 (0:00.00)
[+0.92]  d=9  1.Rb7 Ke6 2.Ke3 f5 (0:00.00)
[+0.92]  d=8  1.Rb7 Ke6 2.Ke3 f5 3.Rh7 h5 4.Rb7 Kf6 5.h4 Ra2 6.Rb6 (0:00.00)
[+0.84]  d=7  1.Rb7 Ke6 2.Ke3 f5 3.Rh7 h5 4.Rb7 Kf6 5.f4 Ra2 (0:00.00)
[+0.90]  d=6  1.Rb7 Ke6 2.Ke3 f5 3.Rh7 h5 4.Rb7 h4 5.Rh7 (0:00.00)
[+0.81]  d=5  1.Rb7 Ke6 2.Ke3 Ra2 3.Rb6 Ke5 4.d4 Kd5 5.Rb5 Ke6 (0:00.00)
[+0.87]  d=4  1.Rb7 Ra2 2.d4 Ke6 3.Rc7 f6 4.Rc6 (0:00.00)
[+0.78]  d=4  1.d4 Ke6 2.Ke4 f5 3.Ke3 Ra3 4.Ke2 (0:00.00)
[+0.79]  d=4  1.d4 Rd6 2.Ke4 Re6 3.Kf4 g5 4.Kf3 (0:00.00)
[+0.94]  d=3  1.d4 Rd6 2.Ke4 Re6 3.Kf4 g5 4.Kf3 (0:00.00)
[+0.98]  d=2  1.d4 Rd6 2.Ke4 Re6 (0:00.00)
[+0.90]  d=1  1.d4 Ke6 2.Kf4 Ra4 (0:00.00)
[+0.85]  d=1  1.Rb7 Ke6 2.Ke3 (0:00.00)

8/5p2/r4kpp/8/8/2P2K2/1R3PPP/8 w - -


[+0.97]  d=14  1.Ke4 Rc6 2.Rc2 Re6 3.Kd5 Re5 (0:00.01)
[+0.97]  d=13  1.Ke4 Rc6 2.Rc2 Re6 3.Kd5 Re5 4.Kd4 Re1 5.c4 Rd1 (0:00.01)
[+1.05]  d=13  1.Ke4 Rc6 2.Rc2 Rc7 3.c4 Ke6 4.c5 Rc8 5.Kd4 (0:00.01)
[+0.93]  d=12  1.Ke4 Rc6 2.Rc2 Ke6 3.c4 (0:00.00)
[+0.93]  d=12  1.Ke4 (0:00.00)
[+0.81]  d=11  1.Ke4 Rc6 2.Rc2 Re6 3.Kd5 Re5 4.Kd4 (0:00.00)
[+0.76]  d=10  1.Ke4 Rc6 2.Rc2 Re6 3.Kd5 Re5 4.Kd4 Rh5 5.h3 Rg5 6.f3 Re5 (0:00.00)
[+0.73]  d=9  1.Ke4 Rc6 2.Rb3 Re6 3.Kd3 Rd6 4.Ke2 Rc6 5.Ke3 (0:00.00)
[+0.73]  d=8  1.Ke4 Rc6 2.Rb3 Re6 3.Kd3 Rd6 4.Ke2 Rc6 5.Ke3 Re6 6.Kf4 Rc6 7.Ke4 Re6 8.Kd3 Rd6 9.Ke2 Rc6 10.Ke3 Re6 11.Kf4 Rc6 12.Ke4 Re6 13.Kd3 Rd6 14.Ke2 Rc6 15.Ke3 Re6 16.Kf4 (0:00.00)
[+0.77]  d=7  1.Ke4 Rc6 2.Kd4 Rd6 3.Kc5 Rd1 4.c4 Ke5 5.Rb7 f6 6.Re7 (0:00.00)
[+0.73]  d=7  1.Rb3 Rc6 2.Ke3 Re6 3.Kf4 Rc6 4.Ke4 Re6 5.Kd3 Rd6 6.Ke4 Rd2 7.Ke3 (0:00.00)
[+0.78]  d=6  1.Rb3 Rc6 2.Ke3 Re6 3.Kf4 Rc6 4.Ke4 Re6 5.Kd3 Rd6 6.Ke4 Rd2 7.Ke3 (0:00.00)
[+0.73]  d=6  1.Ke4 Rc6 2.Rb3 Re6 3.Kd3 Rd6 4.Ke3 Rc6 5.f4 (0:00.00)
[+0.76]  d=5  1.Ke4 Rc6 2.Kd4 Rd6 3.Kc5 Rd3 4.c4 Ke5 5.Re2 (0:00.00)
[+0.82]  d=4  1.Ke4 Rc6 2.Kd4 (0:00.00)
[+0.75]  d=4  1.c4 Rc6 2.Rb4 Kg5 3.Rb5 (0:00.00)
[+0.90]  d=3  1.c4 Rc6 2.Rb4 Kg5 3.Rb5 (0:00.00)
[+0.91]  d=3  1.c4 (0:00.00)
[+0.79]  d=2  1.c4 Rc6 2.Rb4 Ke5 3.Ke3 (0:00.00)
[+0.69]  d=2  1.Ke3 Re6 2.Kf4 Rc6 3.Rb3 (0:00.00)
Parent - - By dr_zied_haddad (**) [fr] Date 2010-07-30 23:25
i AGREE with your evaluation. I'm going to be more radical than you. Positions 1 and 3 seems to lead to draws. Even if diagram 1 is still a little bit more favourable to white because of the doubled pawns, from a positional kind of view. But white must find a way to take advantage from it, which i.m.h.o is hard to achieve. But because of this, i prefer to play this position as white and to loose a half point not to play it with black and save the half point.

Positions 2 and 4: a slight but constant advantage for White. May be evaluation is between 0,35 and 0,5. Black must play very accurately to achieve the draw. His plan is simple, and he must avoid a few tactics (not to loose another pawn or weaken his pawns structure or exchange the rooks). The doubled pawns seems not to be a major issue, but in such an ending i tend to evaluate white's position as between 0,5 and 0,7 because doubled pawns are usually more difficult to protect, and there are no possibilities like sacrificing one of the doubled pawns to weaken or double pawns for white.

The 5th diagram seems to be a clear advantage for white, because of what john already said. He's analyses of all the positions are really very good.
Parent - - By patrick delaurentis (**) Date 2010-07-31 18:13
i hear a lot of disappointed people when it comes to rybka 4 and came up with a solution... if we all want stronger engines we need to stop buying unless huge improvements are made. if we keep buying the next rybkas with improvements as slim as rybka 3 to rybka 4 then we will keep getting junk!!!
Parent - - By Uly (Gold) [mx] Date 2010-07-31 18:41
Actually, the improvement of 40 elo over Rybka 4 is huge, apparently the error was to allow the 100 elo jump of Rybka 3. Had it been 30 elo jump for Rybka 3, then 30 for Rybka 4 (and we could be at Rybka 6 by now), users wouldn't be complaining. In reality they should be happy for the extraordinary product they got at the time, Rybka 3.
Parent - - By keoki010 (Silver) [us] Date 2010-07-31 19:25
I'm happy :yell: But I'm disappointed because I was promised bug fixes and still don't have either of the 2 that were promised. Here it is the last day of July and still no patch for R4's analysis bug. :cry:
Parent - - By Uly (Gold) [mx] Date 2010-07-31 19:31
I'm still happy that Vas has his priorities right, and that instead of taking care of his engine he's taking care of his child. I'd have done the same (but I wouldn't have promised anything...)

They say that after you have a kid, you become a totally different person, maybe Vas's schedule will change drastically.
Parent - - By Banned for Life (Gold) Date 2010-07-31 19:49
I'm not sure I understand this, or maybe I'm just too old to see eye-to-eye with you. Babies are in the mommy domain. Daddies need to be bringing home the bacon. If mommy needs additional help, she should enlist her mother or other female relatives. I hope my post won't be stricken because of blatant sexism...
Parent - - By Uly (Gold) [mx] Date 2010-07-31 22:38
I'm a feminist :razz:

Babies should be a 50-50 responsibility of the mother and father, and there's nothing wrong with mommies bringing home the bacon, if they wish so.
Parent - - By Banned for Life (Gold) Date 2010-07-31 23:42
I have no problem with mommies bringing home the bacon, but last time I checked, only mommies had babies. Since they have to go through that very demanding step, I figure they should just own the whole process! :lol:
Parent - By keoki010 (Silver) [us] Date 2010-07-31 23:52
+1, concur.  That's the way it was always done until .....Political correctness.  Nothing wrong with papa taking a turn in the middle of the night to change a dirty diaper either. :red:
Parent - - By irulats (****) [ie] Date 2010-08-01 07:01

> Babies are in the mommy domain. Daddies need to be bringing home the bacon. If mommy needs additional help, she should enlist her mother or other female relatives.


That all goes without saying but in today's Europe, most new families must go it alone in an atmosphere which is becoming daily more anti-family. I, like Vytron, encourage him on this one! :smile:
Parent - By Uly (Gold) [mx] Date 2010-08-01 15:39
Thanks, I think that if the girl just gave birth, she could use some help, I just hope Vas is providing it :smile:
Parent - By JohnL (***) Date 2010-07-31 19:28
Yes, just because there are some weaknesses doesn't mean the entire product is crap. And it seems like most other programs aren't better in the same areas.

If 40 ELO corresponds to almost twice the speed it is quite significant in my world. When you are analysing you wait a certain time to get a certain quality of the moves.
Now assume that you can get the same quality after half the time, this means a significant speed up in your analysis process.

BTW, I finally managed to get IDeA going and it is actually a quite nice tool. And I seem to belong to the minority who thinks randomizer/monte carlo is useful as well.
Up Topic Rybka Support & Discussion / Rybka Discussion / Rybka4 database of games where Rybka doesn't play good endga
1 2 3 4 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill