Not logged inRybka Chess Community Forum
1 2 3 Previous Next  
Parent - - By SpiderG (***) [us] Date 2009-03-22 16:01
right. I wonder if that is the right score then...?
Parent - By Christian Packi (****) [de] Date 2009-03-22 16:06
3rr3/p4p2/2p5/P1P2p2/1Pkp3p/3N1P1P/2PR2P1/5K2 w - - 4 38


Look at blacks pawn structure but its completely winning. Rybka is a wizzard.
Parent - - By Richbike [us] Date 2009-03-22 16:06
At this time, Rybka score is 3.46...victory secured!
Parent - - By SpiderG (***) [us] Date 2009-03-22 16:08
Yes!! I'm very sure Rybka will win!
Thanks,   SpiderG
Parent - By SpiderG (***) [us] Date 2009-03-22 16:09
I was getting scared there... :)
Parent - - By Uri Blass (*****) [il] Date 2009-03-22 16:14
3.46 score secures nothing.
As long as there is no mate Rybka can resign and lose the game.
Another way for rybka to lose is by disconnecting and losing on time.

Uri
Parent - By Dragon Mist (****) [hr] Date 2009-03-22 16:15
We're already seeing mate in the horizon Uri :-)
Parent - - By Nick (*****) [gb] Date 2009-03-22 16:28
[Event "ICC 50 3 u"]
[Site "Internet Chess Club"]
[Date "2009.03.22"]
[Round "-"]
[White "Fruit-CCT11"]
[Black "Rybka"]
[Result "0-1"]
[ICCResult "White checkmated"]
[WhiteElo "2000"]
[BlackElo "2813"]
[Opening "Caro-Kann: advance variation"]
[ECO "B12"]
[NIC "CK.02"]
[Time "11:05:00"]
[TimeControl "3000+3"]

1. e4 c6 2. d4 d5 3. e5 c5 4. dxc5 Nc6 5. Nf3 Bg4 6. Bb5 Qa5+ 7. Nc3 e6 8.
Bd2 Qc7 9. Bxc6+ bxc6 10. Na4 Bxf3 11. Qxf3 Qxe5+ 12. Be3 Be7 13. O-O Bf6
14. Rad1 Ne7 15. Bd4 Qf5 16. Qxf5 Nxf5 17. Bxf6 gxf6 18. h3 e5 19. Rd3 Nd4
20. Rxd4 exd4 21. Rd1 O-O-O 22. f3 Rdg8 23. Rxd4 Rg5 24. Kf1 Rhg8 25. Rd2 h5
26. Nc3 h4 27. a4 Kc7 28. Na2 Rb8 29. b3 d4 30. b4 Rd5 31. a5 Kb7 32. Nc1
Ka6 33. Nd3 Re8 34. Nf4 Rdd8 35. Nh5 f5 36. Nf4 Kb5 37. Nd3 Kc4 38. Rf2 Kc3
39. Kg1 Re3 40. Kh2 Rxd3 41. cxd3 Kxd3 42. Ra2 Ke3 43. g4 d3 44. Ra3 Kf2 45.
Ra1 d2 46. Rd1 Ke2 47. Rxd2+ Rxd2 48. Kg2 Ke3+ 49. Kf1 Kxf3 50. Ke1 Re2+ 51.
Kf1 fxg4 52. hxg4 h3 53. Kg1 h2+ 54. Kh1 Kg3 55. b5 Re1# {White checkmated}
0-1
Attachment: fruit-rybka.txt (30k)
Parent - - By M ANSARI (*****) [kw] Date 2009-03-22 16:32
Excellent display of chess domination.  Can't wait to get my hands on Rybka 4 !!!  Good job Nick, Lukas, Jeroen and Vas ... Rybka was never in trouble in any of the games.  Rybka's play was fantastic and as if anyone here expected any less.  Some of these games need some serious analysis to see what has changed in the engine room, but it looks very promising.  Kudos to Crafty team for being the only one to hold Rybka to a draw (chesswise).

One thing to note though is that I found following the games extremely tedious ... there must be a better way to do this with more thought to how spectators can enjoy the games.
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 16:34
My thoughts exactly. Next Rybka will be a stunner!
As for this "broadcast", I follow the games on ICC throu Winboard. "#$%& thing does not supply moves, only FEN, hence messy outlook.

Edit: Rybka now takes the lead alone with 6.5/8
Parent - By Dragon Mist (****) [hr] Date 2009-03-22 16:46
Situation on top going into last round:

Place Name                        Feder Rtg Loc Score M-Buch. Buch.

  1   Rybka,                                    6.5      28.0  36.0
2-3  Bright/Swaminathan,                       6        28.0  37.5
      Ikarus/IkarusX,                           6        25.0  33.0
4-6  Fruit/Fruit-CCT11,                        5.5      31.0  40.0
      Thinker/ChessThinkerX,                    5.5      29.5  39.0
      Arasan/ArasanX,                           5.5      28.0  34.0
7-8  Crafty,                                   5        27.0  37.0
      Twisted Logic/TwistedLogi,                5        25.5  33.5
9-14  Glaurung/GlaurungCCT11,                   4.5      32.5  43.0
      Deep Sjeng/Sjeng,                         4.5      30.0  39.5
      The Baron/thebaron,                       4.5      28.0  38.0
      Scorpio/dshawul,                          4.5      25.0  32.5
      Deuterium,                                4.5      23.5  32.0
      Diep,                                     4.5      22.5  31.0
Parent - - By Bouddha (****) [ch] Date 2009-03-22 17:06
Rybka was in trouble against Crafty. Against an other Rybka, Rybka may have lost this game.

regards
Parent - - By Kappatoo (*****) [de] Date 2009-03-22 17:14
On which move?
Parent - - By Bouddha (****) [ch] Date 2009-03-22 18:56
[Event "CCT 11"]
[Site "Internet Chess Club"]
[Date "2009.03.22"]
[Round "6"]
[White "crafty"]
[Black "Rybka"]
[Result "1/2-1/2"]
[ECO "B18"]
[WhiteElo "2749"]
[BlackElo "2813"]
[PlyCount "161"]
[EventDate "2009.??.??"]
[TimeControl "3000+3"]

1. e4 c6 2. d4 d5 3. Nd2 dxe4 4. Nxe4 Bf5 5. Nc5 Nd7 6. Bd3 e6 7. Nxb7 Qc7 8.
Bxf5 exf5 9. Qe2+ Be7 10. Nc5 Nxc5 11. dxc5 Nf6 12. Be3 O-O 13. O-O-O Qa5 14.
Qc4 Rfb8 15. Nf3 Rb4 16. Bd2 Rxc4 17. Bxa5 Ra4 18. Nd4 Rxa5 19. Nxc6 Rxc5 20.
Nxe7+ Kf8 21. Rhe1 a5 22. a3 f4 23. g3 Re8 24. b4 axb4 25. axb4 Rc4 26. Nf5
Rxe1 27. Rxe1 g6 28. Nd6 Rxb4 29. f3 Rd4 30. Nb5 Rd5 31. Nc3 Rh5 32. Re2 Rh3
33. Rg2 fxg3 34. hxg3 h5 35. Nd1 h4 36. gxh4 Nd5 37. Rf2 Rxh4 38. Nb2 Ke7 39.
c4 Nf4 40. Kd2 g5 41. Ke3 Ne6 42. Rc2 Ng7 43. Re2 Kd6 44. Rd2+ Ke6 45. Kf2 f6
46. Kg2 Nf5 47. Rd3 Rf4 48. c5 Nd4 49. Rc3 Nc6 50. Nc4 f5 51. Nd6 Rd4 52. Rc2
Rd5 53. Kg3 Kf6 54. Rc3 Rd1 55. f4 g4 56. Nc4 Nd4 57. c6 Nxc6 58. Ne3 Rd6 59.
Nxg4+ fxg4 60. Kxg4 Nd8 61. Re3 Rd1 62. Re2 Ne6 63. Re4 Nd4 64. Re3 Rh1 65. Ra3
Rg1+ 66. Rg3 Rxg3+ 67. Kxg3 Nf3 68. Kxf3 Kg7 69. Ke4 Kf8 70. Kd5 Kf7 71. Ke5
Ke7 72. f5 Kf7 73. Kd6 Kf6 74. Kd7 Ke5 75. Ke8 Kf6 76. Kf8 Ke5 77. Kg8 Kf6 78.
Kh8 Ke5 79. Kg8 Kf6 80. Kh8 Ke5 81. Kg8 {Game drawn by repetition} 1/2-1/2

After 21 move, Rybka position was not favorable. This type of positions are positions where computer may evaluate as = but white has the edge in the long run due to its favorable pawn structure & ending.
Parent - - By Kappatoo (*****) [de] Date 2009-03-22 19:05
So where do you think did white go wrong then? The position is very concrete. Evaluation is 90% due to tactics here.
It isn't just the pawn structure - white is a pawn up. But at the moment, white's knight is stuck on e7. Rybka utilized this fact to win back the pawn, after which a totally drawn ending arose. So you think white could have prevented this?
Parent - - By Bouddha (****) [ch] Date 2009-03-22 20:05
22.f4 was certainly better.

Setup this position an let play Rybka vs Rybka in 3min blitz games.

Let it play 30 games and share the results with us.

Thank you

regards
Parent - - By Kappatoo (*****) [de] Date 2009-03-22 20:15
:) :)

That's a very kind offer, thank you.
Parent - By Bouddha (****) [ch] Date 2009-03-22 20:39
Your welcome :)
Parent - - By Nelson Hernandez (Gold) [us] Date 2009-03-24 22:08
This is opening book domination, not engine domination.  Rybka hardly could show her stuff in this!  By the 9th move white was already up against it and the only move with prospects for anything was 9.b4, which leads to a very likely draw.  Instead it played Bxc6+?! which I have only seen once before, played by a 2300-level player in the engine room. 
Parent - - By Vasik Rajlich (Silver) [hr] Date 2009-03-31 09:59
We've had an opening book revolution in computer chess over the past 12 months or so. At some point we probably won't be able to get at advantage in every single game any more - we'll see.

Vas
Parent - - By Nelson Hernandez (Gold) [us] Date 2009-03-31 11:36
Yes, your people came around to our principles and way of thinking!  :)
Parent - - By Vasik Rajlich (Silver) [pl] Date 2009-04-04 20:59
Is that supposed to be a compliment? :)

Vas
Parent - - By Nelson Hernandez (Gold) [us] Date 2009-04-04 22:28
Actually it was only half-serious.  Your original comment is a bit hard to quantify because you could be making any kind of mental comparison.  We've certainly been testing heavily on our end and there is a world of difference in some of the latest books that are coming out from those that were deemed cutting-edge a year ago.  But that is hardly surprising with so many of them heavily laden with Playchess R3 games (and clearly a lot of internal work) and with so many new entrants to the bookmaker glory stakes.  Some of the books in circulation now are pretty good pieces of work for .ctg format; at least they do pretty well against our tournament lines.  The key as I see it is that they are perfectly willing to exit book in dead-drawn positions and don't overreach.  That tactic can be lethal to the sharpies.  Looks like the balance of power is tilting back to those with deep pockets and hardware savvy!
Parent - - By Vasik Rajlich (Silver) [pl] Date 2009-04-05 06:22
Yeah, I agree. As with most things, it's easier for an expert to crush a novice than for a super-expert to crush an expert. Somehow our team hasn't felt this yet in the tournaments, but most likely at some point we will.

Vas
Parent - - By Nelson Hernandez (Gold) [us] Date 2009-04-05 10:59
I've been considering hooking up with some very, very weak engine--take your pick--and having them play our tournament book.  The results would be hilarious.  We'd exit them into very good positions, even winning positions in some cases, and they'd then proceed to blow game after game.  That would nonetheless make us pretty happy as a true hard-core bookmaker doesn't care that much about game outcomes as much as book-exit positions.
Parent - - By Vasik Rajlich (Silver) [pl] Date 2009-04-06 08:11

> We'd exit them into very good positions, even winning positions in some cases, and they'd then proceed to blow game after game.


Until you played us - then you would exit them into a lost position. :)

Seriously, though, you should do this. Find the best engine you can and go for it. Why not?

Vas
Parent - - By Nelson Hernandez (Gold) [us] Date 2009-04-06 15:03
Against you we would exit into dead-drawn positions time after time.  I'd see to it.  (Unless you wanted to bring the monster into a slightly unfavorable position and fight your way to an advantage through brute hardware and a superior engine, which makes perfect strategic sense.)

The delight for me is more in finding really awful engines with formerly awful books and having them punch way above their weight class for a dozen or two dozen moves.  Then our merits would be obvious and there would be nothing more to say.  I'm not as interested in helping a top-tier engine because I don't want to hear any lame excuses as to why they blew the middle-game from a perfectly good book-exit.
Parent - - By Vasik Rajlich (Silver) [pl] Date 2009-04-07 08:13
There is no glory Nelson in being the best player on a weak team. Don't you watch sports?

Vas
Parent - By Nelson Hernandez (Gold) [us] Date 2009-04-07 11:26
No.
Parent - - By Vempele (Silver) [fi] Date 2009-03-22 16:38
Sjeng lost to Arasan.
Parent - - By M ANSARI (*****) [kw] Date 2009-03-22 16:47
Ahh, I thought this was already over.
Parent - By Dragon Mist (****) [hr] Date 2009-03-22 16:57
Round 9 we have white against IkarusX.
Parent - - By Nick (*****) [gb] Date 2009-03-22 17:09
Yes one more round Majd, followed by a 9 round blitz tourney (time control 5+3), which we'll give a go too.
Parent - - By M ANSARI (*****) [kw] Date 2009-03-22 17:13
Now that would be a challenge ... getting the cluster to work at 5_3 blitz will be interesting to say the least :).  Can you really play at such fast time controls with the inherent latencies involved with cluster?
Parent - - By violina [rs] Date 2009-03-22 17:19
I want to make cluster of my 4 dual core. But HOW?
Parent - By M ANSARI (*****) [kw] Date 2009-03-22 19:17
Cluster is not available for public yet ... keep your fingers crossed though, it might show up soon.
Parent - By Nick (*****) [gb] Date 2009-03-22 17:21
We'll find out, if it's not working I'll play stock R3 from my local Linux quad, QX9650 @ 4ghz.
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:06
r2q1r1k/ppp1b1pp/2n2p2/3b4/3P4/P1P2N2/4BPPP/R1BQ1RK1 w - - 0 16


We are out of book, IkarusX still in.
Parent - - By Nick (*****) [gb] Date 2009-03-22 17:08
I am eating lunch, keep going Dragon!
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:09
No problem.

r2q1r1k/ppp1b1pp/2n5/3b1p2/3P4/P1P5/3NBPPP/R1BQ1RK1 w - - 0 17


IkarusX(C) whispers: +0.06/18  16. ... f5 17. c4

Rybka(C) kibitzes: +0.36 17.Nc4; depth 17; 77s;
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:13
It's not so much that Rybkas 17.Nc4 was not expected by IkarusX, but it was not expected by anyone on Earth! :-)
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:15
17. Nc4 f4 18. Bd3 g6

r2q1r1k/ppp1b2p/2n3p1/3b4/2NP1p2/P1PB4/5PPP/R1BQ1RK1 w - - 0 19


Rybka(C) kibitzes: +0.62 19.Nd2; depth 17; 68s;

now we know where that Knight was heading for: back to d2 :-)
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:18
IkarusX(C) kibitzes: +0.37/17  19. ... Rb8 20. Re1

Rybka(C) kibitzes: +0.67 20.Re1; depth 19; 0s;

1r1q1r1k/ppp1b2p/2n3p1/3b4/3P1p2/P1PB4/3N1PPP/R1BQR1K1 b - - 0 20
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:22
IkarusX position is a mess anywhere you look, both queen and king side does not look good. In centre ..., well in centre maybe acceptable. Nc4 did the trick, weakened black over limits.

IkarusX(C) kibitzes: +0.78/18  21. ... a6 22. a4

1r1q1r1k/2p1b2p/p1n3p1/1p1b4/3P1p2/P1PB4/3N1PPP/1RBQR1K1 w - - 0 22


Rybka(C) kibitzes: +1.18 22.a4; depth 16; 24s;
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:25
22. a4 bxa4 23. Rxb8 Nxb8 (+1.07 IkarusX) 24. Qxa4 (+1.49 Rybka)

1n1q1r1k/2p1b2p/p5p1/3b4/Q2P1p2/2PB4/3N1PPP/2B1R1K1 b - - 0 24
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:27
Out of 64 squares, black is better on maybe ... 3? :-)
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:34
Don't know why IkarusX is taking over 10 minutes for next move?!
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:38
1n1q1r1k/2p4p/p5p1/3bR3/Q1PP1p2/3B4/5PPP/2B3K1 b - - 0 28


IkarusX(C) kibitzes: +2.23/19  27. ... Bxe5 28. Rxe5

Rybka(C) kibitzes: +2.63 28.Rxe5; depth 19; 0s;

MoroGamgee kibitzes: So instead of Nf3-e5 we went Nf3-d2-c4-d2-c4-e5 ;-)

IkarusX(C) kibitzes: +2.63/19  28. ... Bf7 29. d5

Rybka(C) kibitzes: +3.19 29.d5; depth 18; 0s;
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:42
Renzer kibitzes: why is chess so easy when Rybka plays it..all makes sense
Parent - - By Dragon Mist (****) [hr] Date 2009-03-22 17:43
1n1q1rk1/2p2b1p/p7/3P2p1/2P1Rp2/3B4/2Q2PPP/2B3K1 b - - 0 31


Rybka(C) kibitzes: +4.42 31.Qc2; depth 16; 24s;

BigMomma(C) kibitzes: still not too late to play f3 :)
1 2 3 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill