![[de]](/mwf/v12/flags/de.png "Germany")
Who will hold out this? It´s really pain. Vas, please have a closer look to your calendar. Okay, some versions of Rybka for Linux and mobile apparatus in the near future. But then? October or November will be a nice time to present Rybka 3.0 UCI with some nice improvements (like wrong bishop). I think, you reserved this half year because of parallelism of Rybka. I´m not sure, if the customers will pay for this. It´s to early. Nobody of the customers will have more than 4 cores this (and the next) year and 2.3.2a works fine! It´s only because of this damn match in Mexico against Junior (which isn´t fixed), you reserved the time? You can also need time for adapting Rybka to the new Convekta GUI. But Rybka UCI has to go on.
Vas, please give us good reasons, why the next Rybka vesion will come in six month (three weeks holiday for Iweta and you are included)?
Vas, please give us good reasons, why the next Rybka vesion will come in six month (three weeks holiday for Iweta and you are included)?
PS: Arno wrote , that the prize fund arrieved at CSS. So you can plan four weeks holiday :-).
![[br]](/mwf/v12/flags/br.png "Brazil")
Roland, Vas wrote at some time ago that Rybka 3.0 will be launched
in the end of the year. Now at the and of the year or earlier january.
Man, for me only in february, because Vas wrote that the gui will be better
than Chessbase gui, they will do a dificult job.
For while you can apreciate the nice Rybka endgame and to think that anothers
good engines will be launched, I think that it will be stimulate for Vas.
Paulo Soares
in the end of the year. Now at the and of the year or earlier january.
Man, for me only in february, because Vas wrote that the gui will be better
than Chessbase gui, they will do a dificult job.
For while you can apreciate the nice Rybka endgame and to think that anothers
good engines will be launched, I think that it will be stimulate for Vas.
Paulo Soares
![[hu]](/mwf/v12/flags/hu.png "Hungary")
Yes, we're going into a cave now with Rybka. The plan is to come out with something amazing.
On the engine side, there will be a lot more than just MP improvements. I certainly won't stop working on the general search efficiency. Larry is going to have the entire eval to work with - yes, this includes king safety, and endgames.
There will also be Convekta's GUI.
Vas
On the engine side, there will be a lot more than just MP improvements. I certainly won't stop working on the general search efficiency. Larry is going to have the entire eval to work with - yes, this includes king safety, and endgames.
There will also be Convekta's GUI.
Vas
![[us]](/mwf/v12/flags/us.png "United States")
Yes sir.
I can't wait.
Rybka is going to rule the planet.
I can't wait.
Rybka is going to rule the planet.
![[ro]](/mwf/v12/flags/ro.png "Romania")
"Rybka is going to rule the planet".
Serious ?????? Maybe the whole galaxy/universe ?????? Maybe Rybka was touched by the Holy Spirit ??????? Maybe Rybka is a Hell's creation ?????? Maybe............... Who knows ?????????
I think Rybka is just a good chess engine !
Regards,
SilvianR
Serious ?????? Maybe the whole galaxy/universe ?????? Maybe Rybka was touched by the Holy Spirit ??????? Maybe Rybka is a Hell's creation ?????? Maybe............... Who knows ?????????
I think Rybka is just a good chess engine !
Regards,
SilvianR
> I think Rybka is just a good chess engine !
I think:
exactly that's what h1a8 said, perhaps with a 'very' between 'a' and 'good'.
Quap
![[ca]](/mwf/v12/flags/ca.png "Canada")
Hello Vasik,
I read the article in "Chess" magazine by Harald Fietz and was
very interested about the king safety issue.
In the game: Rajlich-Kingscrusher, where Black lost,
I could not find counterplay for Black after 16 ... hxg6 .
However, my experiments show that 16 ... h6
is maybe a better move for Black.
I followed a line starting with this move up to
23. f4 (and beyond), at which point Rybka 2.3.2a gives
a 32 centipawns advantage to Black at a depth of 19 plies
after 23. f4 .
1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c6 5. e3 Nbd7
6. Qc2 Bd6 7. g4 dxc4 8. Bxc4 b5 9. Be2 O-O
10. e4 Be7 11. g5 Ne8 12. Bf4 Nd6 13. O-O-O Nb6 14. Bd3 Ndc4
15. h4 f6 16. g6
{16 ... h6 17. Nd2 e5 18. Bxc4+ Nxc4 19. dxe5
Nxe5 20. Nf3 Qe8 21. Nxe5 fxe5 22. Bxe5 Bc5 23. f4 } =/+
16 ... hxg6
I used infinite analysis with the Fritz GUI to continue this
variation for many days of CPU time.
I copy below all the moves until checkmate by Black in case
anyone is interested.
Rajlich-Kingscrusher with 16 ... h6 variation:
--------------------------------------------------
1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c6 5. e3 Nbd7 6.
Qc2 Bd6 7. g4 dxc4 8. Bxc4 b5 9. Be2 O-O 10. e4 Be7 11. g5 Ne8 12. Bf4 Nd6 13.
O-O-O Nb6 14. Bd3 Ndc4 15. h4 f6 16. g6 h6 17. Nd2 e5 18. Bxc4+ Nxc4 19. dxe5
Nxe5 20. Nf3 Qe8 21. Nxe5 fxe5 22. Bxe5 Bc5 23. f4 Be3+ 24. Kb1 Bxf4 25. Bxf4
Rxf4 26. Rd6 Be6 27. Ka1 Qxg6 28. Qc1 Raf8 29. Rxc6 Qg4 30. Nxb5 Rxe4 31. Nc3
Re5 32. Ra6 Rf7 33. Ra4 Qg3 34. Qd1 Kh7 35. Ne4 Qf3 36. Qxf3 Rxf3 37. b3 Bd5
38. Nd6 Rf6 39. Nc4 Re2 40. Rd1 a6 41. Nd2 Rff2 42. Nb1 Bb7 43. Rc1 Re6 44. Rd4
Rh2 45. Nc3 h5 46. Rg1 Re3 47. Nd5 Bxd5 48. Rxd5 Rxh4 49. Rdg5 Re7 50. R1g2
Rh1+ 51. Kb2 Rh3 52. Ka3 Rh4 53. Kb2 Rf7 54. Kc3 Rc7+ 55. Kd2 Ra7 56. Rc5 Rh1
57. Ke3 Re7+ 58. Kf2 Rd7 59. Ke2 Rd6 60. Rc7 Rg6 61. Rxg6 Kxg6 62. Rc6+ Kf5 63.
Kf3 Rh3+ 64. Kg2 Rd3 65. Rxa6 h4 66. Rb6 g5 67. Rb8 h3+ 68. Kh2 g4 69. Rf8+ Ke6
70. Re8+ Kf7 71. Rh8 Rd2+ 72. Kh1 Rxa2 73. Rh4 Rg2 74. b4 Kg6 75. Rh8 Kg5 76.
Rh7 Kf4 77. Rd7 Rc2 78. Rd5 g3 79. Rd4+ Ke3 80. Rd1 h2 81. Rb1 Kf3 82. b5 Kg4
83. Rb4+ Kh3 84. Rh4+ Kxh4 85. b6 Rc1+ 86. Kg2 h1=Q# 0-1
Regards,
David Bernier
I read the article in "Chess" magazine by Harald Fietz and was
very interested about the king safety issue.
In the game: Rajlich-Kingscrusher, where Black lost,
I could not find counterplay for Black after 16 ... hxg6 .
However, my experiments show that 16 ... h6
is maybe a better move for Black.
I followed a line starting with this move up to
23. f4 (and beyond), at which point Rybka 2.3.2a gives
a 32 centipawns advantage to Black at a depth of 19 plies
after 23. f4 .
1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c6 5. e3 Nbd7
6. Qc2 Bd6 7. g4 dxc4 8. Bxc4 b5 9. Be2 O-O
10. e4 Be7 11. g5 Ne8 12. Bf4 Nd6 13. O-O-O Nb6 14. Bd3 Ndc4
15. h4 f6 16. g6
{16 ... h6 17. Nd2 e5 18. Bxc4+ Nxc4 19. dxe5
Nxe5 20. Nf3 Qe8 21. Nxe5 fxe5 22. Bxe5 Bc5 23. f4 } =/+
16 ... hxg6
I used infinite analysis with the Fritz GUI to continue this
variation for many days of CPU time.
I copy below all the moves until checkmate by Black in case
anyone is interested.
Rajlich-Kingscrusher with 16 ... h6 variation:
--------------------------------------------------
1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c6 5. e3 Nbd7 6.
Qc2 Bd6 7. g4 dxc4 8. Bxc4 b5 9. Be2 O-O 10. e4 Be7 11. g5 Ne8 12. Bf4 Nd6 13.
O-O-O Nb6 14. Bd3 Ndc4 15. h4 f6 16. g6 h6 17. Nd2 e5 18. Bxc4+ Nxc4 19. dxe5
Nxe5 20. Nf3 Qe8 21. Nxe5 fxe5 22. Bxe5 Bc5 23. f4 Be3+ 24. Kb1 Bxf4 25. Bxf4
Rxf4 26. Rd6 Be6 27. Ka1 Qxg6 28. Qc1 Raf8 29. Rxc6 Qg4 30. Nxb5 Rxe4 31. Nc3
Re5 32. Ra6 Rf7 33. Ra4 Qg3 34. Qd1 Kh7 35. Ne4 Qf3 36. Qxf3 Rxf3 37. b3 Bd5
38. Nd6 Rf6 39. Nc4 Re2 40. Rd1 a6 41. Nd2 Rff2 42. Nb1 Bb7 43. Rc1 Re6 44. Rd4
Rh2 45. Nc3 h5 46. Rg1 Re3 47. Nd5 Bxd5 48. Rxd5 Rxh4 49. Rdg5 Re7 50. R1g2
Rh1+ 51. Kb2 Rh3 52. Ka3 Rh4 53. Kb2 Rf7 54. Kc3 Rc7+ 55. Kd2 Ra7 56. Rc5 Rh1
57. Ke3 Re7+ 58. Kf2 Rd7 59. Ke2 Rd6 60. Rc7 Rg6 61. Rxg6 Kxg6 62. Rc6+ Kf5 63.
Kf3 Rh3+ 64. Kg2 Rd3 65. Rxa6 h4 66. Rb6 g5 67. Rb8 h3+ 68. Kh2 g4 69. Rf8+ Ke6
70. Re8+ Kf7 71. Rh8 Rd2+ 72. Kh1 Rxa2 73. Rh4 Rg2 74. b4 Kg6 75. Rh8 Kg5 76.
Rh7 Kf4 77. Rd7 Rc2 78. Rd5 g3 79. Rd4+ Ke3 80. Rd1 h2 81. Rb1 Kf3 82. b5 Kg4
83. Rb4+ Kh3 84. Rh4+ Kxh4 85. b6 Rc1+ 86. Kg2 h1=Q# 0-1
Regards,
David Bernier
Hi,
yes, probably 16. .. h6 is best just because 16. .. hxg6 runs into trouble. We were sure that Rybka wouldn't play this move, which made our analysis easier.
Anyway, I'm quite sure that white is somehow a bit better after 16. .. h6.
Vas
yes, probably 16. .. h6 is best just because 16. .. hxg6 runs into trouble. We were sure that Rybka wouldn't play this move, which made our analysis easier.
Anyway, I'm quite sure that white is somehow a bit better after 16. .. h6.
Vas
Hi Vas,
Rybka 3. Isn't in sync with the GUI release is it? I thought the GUI would be out there at least by the Holiday Season-cashregister Cashinger ringading Season.
Robert
Rybka 3. Isn't in sync with the GUI release is it? I thought the GUI would be out there at least by the Holiday Season-cashregister Cashinger ringading Season.
Robert
We'll release when it's ready. Our target audience isn't exactly Christmas shoppers anyway :)
Vas
Vas
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill