Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / 10 beautiful testpositions! Rybka-Cluster challenge....
1 2 3 Previous Next  
Parent - By Uly (Gold) Date 2010-12-30 19:43

> The tough part on that problem was the amazing King journey g2>f2>e1>d2 and again back d1>e1>f2>g2>h3>g4>h5>h6>g7>f8>e8>d7>c6!! That journey had to be done or else the position can't be reached.


I see, the King needs to be at d1 here so that Bd2 doesn't checkmate.

1. d3 Nf6 2. Nf3 Nc6 3. Ng5 Na5 4. Ne6 dxe6 5. Bg5 Nd5 6. Qd2 h6 7. Nc3 hxg5 8. Qf4 gxf4 9. Kd1 Nb3 10. Ke1 Na5 11. Kd2 Ne3 12. Nd5 Nd1 13. Ne3 fxe3+ 14. Kc1 Bd7 15. f4 Rh3 16. Kb1 Rf3 17. Kc1 Rf2 18. Kb1 Bc6 19. Kc1 Qd4 20. Kb1 Rd8 21. Kc1 Rd5 22. Kb1 Rh5 23. Kc1 Rh3 24. Kb1 Rhf3 25. g3 Be4 26. Bh3 Bh7 27. Bf5 Bg8 28. Bh7 Rf1 29. Kc1 Re1 30. Kb1 Nb3 31. Rg1 Nd2+ 32. Kc1 Nf1 33. Kb1 Qe4 34. Kc1 Qf5 35. Kb1 g6 36. Rh1 Bg7 37. Rg1 Bc3 38. Rh1 Bd2 39. Rg1 Bc1 40. Rh1 Nf2 41. Rg1 Ne4 42. Rh1 Nc5 43. Rg1 Nb3 44. Rh1

4k1b1/ppp1pp1B/4p1p1/5q2/5P2/1n1PprP1/PPP1P2P/RKb1rn1R b - -
Parent - - By SpiderG (***) Date 2011-01-05 23:09
Well then of course the cluster wouldn't know this was a draw... Do you think Einstein would implement a way for the cluster to know it was 50-move rule draw from the start of the position? :roll:
Parent - - By George Tsavdaris (****) Date 2011-01-07 20:00

>Do you think Einstein would implement a way for the cluster to know it was 50-move rule draw from the start of the position? :roll:


Yes.:lol:
Parent - By SpiderG (***) Date 2011-01-08 02:52
Well, I disagree :yell:

:smile:
Parent - - By Felix Kling (Gold) Date 2011-01-10 10:19
Einstein was a physicist, I don't think he would have been a good programmer or chess player.
Parent - - By George Tsavdaris (****) Date 2011-01-10 12:18
Yes because an unbelievably clever person can't become a good Chess programmer if he wants to.:roll:
And as such an exceptionally smart person he doesn't have any chances to become a good Chess player too.:lol:
Parent - - By Felix Kling (Gold) Date 2011-01-10 12:52
being smart as a physicist doesn't necessarily mean that you are smart as chess player or programmer. Actually the most important part of abilities is that you are really interested in something. Nobody knows if Einstein would have been interested in such stuff :)
Parent - - By George Tsavdaris (****) Date 2011-01-10 13:21

>being smart as a physicist doesn't necessarily mean that you are smart as chess player or programmer.


Yes, but it doesn't mean the opposite too, that you rushed to speculate: :wink:
"Einstein was a physicist, I don't think he would have been a good programmer or chess player."

Also:
You don't think? Where are your arguments?
And what i just said is that being an exceptionally smart person adds to the possibility of yours to become a good Chess programmer or even a good Chess player.
It doesn't make it a sure fact that you will be a good programmer or Chess player, but the probability for that is increased.

Yet you believe the opposite will be true. Why?

>Actually the most important part of abilities is that you are really interested in something. Nobody knows if Einstein would have been interested in such stuff :)


He was somewhat interested in Chess.
Parent - - By Felix Kling (Gold) Date 2011-01-10 13:33
Because we know that Einstein was quite interested in physics, so he would be today. And that means he wouldn't have much time for such things.
Parent - - By tano-urayoan (****) Date 2011-01-11 10:10
Well Einstein and Lasker knew each other, their converstions Einstein wanting to talk about chess and Lasker about physics :wink:
Parent - By Banned for Life (Gold) Date 2011-01-11 22:46
Well Einstein and Lasker knew each other, their converstions Einstein wanting to talk about chess and Lasker about physics

And as a result, each thought the other an overrated moron!
Parent - - By Master Om (Bronze) Date 2011-01-14 04:50 Edited 2011-01-14 05:06
Your logic is wrong as your thought. Being smart has relations in a chess board game. Here is glimpse of Einsteins Art of Playing against a Known physicist and friend  Robert Oppenheimer. And for your kind information Einstein was a good mathematician (without the good knowledge of mathematics you can't be a good physicist ) and people good in maths are likely to be good in chess as chess is mostly a calculating game.

[Event "Princeton USA"]
[Site "Princeton USA"]
[Date "1933.??.??"]
[EventDate "1933.??.??"]
[Round "?"]
[Result "1-0"]
[White "Albert Einstein"]
[Black "Robert Oppenheimer"]
[ECO "C70"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "47"]

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 b5 5.Bb3 Nf6 6.O-O Nxe4 7.Re1
d5 8.a4 b4 9.d3 Nc5 10.Nxe5 Ne7 11.Qf3 f6 12.Qh5+ g6 13.Nxg6
hxg6 14.Qxh8 Nxb3 15.cxb3 Qd6 16.Bh6 Kd7 17.Bxf8 Bb7 18.Qg7
Re8 19.Nd2 c5 20.Rad1 a5 21.Nc4 dxc4 22.dxc4 Qxd1 23.Rxd1+ Kc8
24.Bxe7 1-0
Attachment: EvsO.aqs (89k)
Parent - By Labyrinth (*****) Date 2011-01-14 06:29
I've seen this game before, and while I have the utmost respect for its supposed players, I just found the game too "cute" to be real. Anyone else get that impression? I'd certainly like to believe it is legitimate..
Parent - - By Felix Kling (Gold) Date 2011-01-14 06:54
Of course you have to know mathematics well as physicist. But "people good in maths are likely to be good in chess" is not founded I think. Or do you have any stats? I hear that so often that chess improves your math skills and vice versa, but I think only people not playing chess themselves on a decent level say that. The truth is that you are only good in what you train a lot, if you train math it doesn't help you much in chess. math and chess are two different things. Maybe if you have talent in math, it's more likely that you have some talent for chess, but I don't think there's a strong correlation. Kasparov is a lousy physicist I assume :)
Parent - - By Labyrinth (*****) Date 2011-01-14 08:51
It would be easier to argue something along the lines of the same type of intelligence (spatial is a good candidate) used in mathematics can be molded for chess with great effect.

Still a huge mess though.
Parent - By Felix Kling (Gold) Date 2011-01-14 11:23
That "types of intelligence" theory is nothing I would support. If you want to learn something and are interested in it, you can learn everything. In general intelligence is not just something "inside", which can't be changed and describes your abilities to learn something. It's a mixture of experience, training and interest that makes you look clever.
I never felt my chess training improved my mathematical abilities, although my general approach to life (ie. thinking about things and trying to figure them out, being sceptical about things being said) may have influenced my interest and therefore my learning speed.
Of course there's something like the "health" of your brain. If your mother drinks alcohol when she's pregnant, you won't be as clever as someone who had a careful mother :) . Also someone doing boxing will damage his brain and so on. I also think watching television hurts your brain :)
There may also be some genetic difference for the capacity of your brain, which is basically what the IQ test tries to measure. But this is just one of the many factors and therefore the whole idea is somehow wrong I think.
- - By Uly (Gold) Date 2010-12-24 04:02
Oh, let me add this one, it would be trully incredible if the Cluster solves it as it's also a long string of difficult moves to find the win.

4q1kr/p6p/1prQPppB/4n3/4P3/2P5/PP2B2P/R5K1 w - - 0 1


1. Qxe5!! is the winning move.
Parent - By urodets (**) Date 2010-12-24 06:19
This is fantastic example!! engines are clueless here!
Parent - - By Master Om (Bronze) Date 2010-12-24 08:21
whats the solution ?
Parent - - By George Tsavdaris (****) Date 2010-12-24 10:20
1.Qxe5!! fxe5 2.Rf1 and then 3.Bd1! and then a plan of 4.Bb3 with huuuuge complications that white wins in every variation.
Try it. It's fun.:grin:
Parent - - By Catalyst_Kh Date 2010-12-25 03:07
Please tell me, how white is going to win after, for example:
1. Qxe5 fxe5 2. Rf1 Rc7 3. Bd1 Re7 4. Bb3 b5
Black will advance all queenside pawns and trade one or two, black will never move g6 to g5 and after Rf7(by white) will just return his queen to e8 to attack f7 rook (and force it back), so black can move queen over 8 line freely, avoiding zugzwang. Maybe it is finally a draw?
Parent - By tano-urayoan (****) Date 2010-12-25 04:48
Well after Rc7 white could even play Bb5
Parent - - By Labyrinth (*****) Date 2010-12-25 07:56 Edited 2010-12-25 08:04
I think what you are missing is that the queen moving to e8 does not force the rook back.

The rook just captures on e7, and after Qxe7 black still has a zugzwang problem because if the queen moves away, e7+ is fatal.

As long as white does not play Rf7 until the black queen has moved away this win is a sure thing.

Sample line:

1.Qxe5 fxe5 2.Rf1 Rc7 3.Bd1 Re7 4.Bb3 b5 5.Bd5 a5 6.Kg2 b4 7.c4 b3 8.a3 a4 9.Rf2 Qd8 10.Rf7 Qe8 11.Rxe7 Qxe7 12.Kh3 g5 13.Kg4 Qf6 14.e7+ Qf7 15.e8Q#
Parent - By Catalyst_Kh Date 2010-12-25 16:12
Thanks Labyrinth, that idea is exactly what i missed.
Parent - - By MoKy (**) Date 2011-01-07 11:23
Hi George, are you 100% sure that white can win?
Parent - - By George Tsavdaris (****) Date 2011-01-07 19:59 Edited 2011-01-07 21:21
I'm sure. I always trust my analysis even if this had been with very old programs.
Here is my old analysis and if you can find a flaw please tell us. :lol:

[Event "?"]
[Site "?"]
[Date "2004.07.09"]
[Round "?"]
[White "Qxe5!!!!! Version-12"]
[Black "?"]
[Result "*"]
[Annotator "George"]
[SetUp "1"]
[FEN "4q1kr/p6p/1prQPppB/4n3/4P3/2P5/PP2B2P/R5K1 w - - 0 1"]
[PlyCount "19"]

1. Qxe5 fxe5 2. Rf1 Rc7 (2... Qe7 $3 3. Bd1 b5 (3... Rc4 4. Bb3 b5 5. a4 $3 {
With the same continuation with the main line of 2...Qe7. That is 3.Bd1 b5 4.
Bd3 Rc4 5.a4!!}) 4. Bb3 Rc4 5. a4 $3 (5. Kg2 {Perhaps this wins too but it
doesn't matter.} a6 6. Bxc4 bxc4 7. b3 cxb3 8. axb3 Qe8 9. c4 Qe7 10. b4 Qxb4
11. Rf7 Qb2+ (11... a5 12. Rd7 $3 $18) 12. Kg3 Qc3+ 13. Kg4 $18) (5. Bxc4 {
Perhaps this wins too but it doesn't matter.} bxc4 6. b4 cxb3 (6... Qd6 7. a4
Qe7 8. b5 Qd6 9. a5 $18) (6... Qe8 7. a4 a6 8. b5 $18) 7. axb3 a6 $2 (7... a5
$1 8. Kg2 (8. b4 axb4 9. cxb4 Qxb4 10. Rf7 Qd4+ 11. Kf1 Qc4+ 12. Kf2 Qa2+ 13.
Kg3 Qb3+ 14. Rf3) 8... Qd6) 8. b4 g5 (8... Qa7+ 9. Rf2 Qb6 (9... g5 10. Kf1 Qg7
(10... Qe7 11. Rf5 $18) 11. Bxg7 Kxg7 12. Rf5 $18) 10. Kf1 Qd6 11. Kg2 $18 Qe7
12. Rf3 $1 g5 13. Rf5 Qg7 14. Bxg7 Kxg7 $18 15. e7 Re8 16. Rxe5 Kf6) 9. h4 Qd6
10. Rf2 $18) 5... a6 6. Kg2 $1 {Ba2! first is about the same.} Qe8 (6... g5 7.
Ba2 Qd8 (7... Qd6 8. b4 $3 $18) (7... g4 8. b4 $18) 8. h3 $3 Qa8 (8... Qe7 9.
b4 $18 Qg7 10. Bxg7 Kxg7 11. Bxc4 bxc4 $18) 9. Kg3 Qe8 10. Rf3 Qe7 11. b4 Qd6
12. Bxc4 bxc4 13. b5 axb5 14. a5 $3 $18) 7. Ba2 $1 Qa8 (7... Qe7 8. b4 $18) 8.
Kg3 $1 Qe8 (8... Qd8 9. Rf3 Qd6 (9... g5 10. b4 $18) 10. b4 $18 g5 11. Bxc4
bxc4 12. b5 axb5 13. a5 $18) 9. Rf3 $3 Qe7 10. b4 $1 $18) 3. Bd1 b5 (3... Rc4
4. Bb3 b5 5. a4 a6 $18 {Transposition of 3...b5 4.Bb3 Rc4 5.a4 a6}) (3... a6 4.
Bb3 Re7 5. Bd5 b5 (5... a5 6. a4 $18) (5... Qb8 6. Rf7 $18) 6. b3 $18 a5 7. c4
bxc4 8. bxc4 $18) (3... a5 4. Bb3 Re7 5. a4 $18) (3... Re7 4. Bb3 a6 (4... a5
5. Bd5 Qb8 (5... b5 6. b3 $18) (5... a4 6. b3 b5 7. c4 $18 axb3 8. axb3 bxc4 9.
bxc4 $18) 6. Rf7 $18) (4... b5 5. Bd5 a6 (5... a5) 6. b3 Qb8 (6... a5 7. c4 $18
) (6... b4 7. c4 a5 8. c5 $18) 7. c4 Qe8 (7... bxc4 8. bxc4 $18) 8. c5 $18) (
4... Qb8 5. Rf7 $18) 5. Bd5 a5 (5... b5 6. b3 a5 (6... b4 7. c4 $18) 7. c4 $18)
6. a4 $18) (3... Rg7 4. Bb3 Re7 (4... Qe7 5. Kg2 $18) 5. Bd5 b5 (5... a6 6. a4
$18) (5... a5 6. a4 $18) (5... Qb8 6. Rf7 Qe8 7. Rxe7 Qxe7 8. Kg2 $18) 6. b3 a5
(6... a6 7. a4 bxa4 (7... b4 8. c4 $18) (7... Qb8 8. c4 $18) 8. bxa4 Qb8 9. c4
$18) 7. c4 bxc4 (7... b4 8. c5) 8. bxc4) 4. Bb3 Rc4 (4... Re7 5. Bd5 b4 (5...
a5 6. b3 Qb8 (6... b4 7. c4 $18) 7. c4 a4 (7... bxc4 8. bxc4 g5 9. c5 $18) 8.
cxb5 $18) 6. c4 Qc8 7. Kg2 Qe8 8. c5 $18) 5. a4 a6 6. Ba2 $3 Qd8 (6... Qe7 7.
b4 g5 (7... Qa7+ 8. Kg2 Qa8 9. Bxc4 Qxe4+ (9... bxc4 10. Kg3 $18) 10. Kh3 $18)
8. axb5 axb5 9. Bxc4 bxc4 10. b5 $18) (6... Qa8 7. b4 $3 $18 Qa7+ 8. Kg2 Qa8 9.
Bxc4 $18) (6... g5 7. b4 $18 Qe7 8. axb5 axb5 9. Bxc4 bxc4 10. b5 $18) 7. b4 $3
Qe7 (7... Qe8 8. axb5 axb5 9. Bxc4 bxc4 10. b5 $18) 8. axb5 axb5 9. Bxc4 bxc4
10. b5 $18 *
Parent - - By Lazy Frank (****) Date 2011-03-10 16:21 Edited 2011-03-11 20:03
Very nice position!

I'm try deep analyse.
George, you are absolute right!
Mainline Qxe5 fe Rf1 with following set bishop to a2-g8 diagonale win!

But im test some more lines!

Qxe5 fe Rf1 and now black answer a6!
Tought white choise between Bxa6/Bd3/Kg2/Rf2/h3.

In my opinion there best move is Rf2!! (protection from Qe7-c5+ maneuver)
Parent - - By Lazy Frank (****) Date 2011-03-16 21:40 Edited 2011-03-16 22:03
This quiet move (Rf2) before Bd1 advantage is:
1) protection from Qe7-c5+ maneuver
2) set 2-line pawns protection
3) makes possible moves Rd2/Rg2
4) await whats stay on square e7 - rook or queen
5) flexible position for queen back sacrifice from black side (amazing, 4 variations)
6) decrease problems later when both (white/black pieces)  freeze on right side (e1-e8-h8-h1)
:razz::red::roll::sad:
Disadvantage : lost square d1 control (possible Qd8-d1+ maneuver)

So, full line of play:
1. Qxe5!! fe 2. Rf1! a6! 3.Rf2!! Rc7 4.Bd1! Rg7 5.Bb3! Re7 6.a4!! (clear win position) ...
Parent - By Peter Grayson (****) Date 2015-09-18 17:15
I see the position has come up again on Talkchess

http://www.talkchess.com/forum/viewtopic.php?t=57674&sid=f69441f4ac3f251358bbb51eefdda399

I note that Komodo 9.2 disagrees with your line ...

> So, full line of play:
> 1. Qxe5!! fe 2. Rf1! a6! 3.Rf2!! Rc7 4.Bd1! Rg7 5.Bb3! Re7 6.a4!! (clear win position) ...


3...Qd8 4.Rf7 Rc7 5.Bc4 Rxc4 6.Rg7+ Kf8 7.Rxh7+ Kg8 8.Rg7+ Kf8 9.Rh7+(0.00)   Depth: 52   00:03:32  7750MN Komodo 9.2

Do you know if there was any further analysis elsewhere?

PeterG
Parent - - By George Tsavdaris (****) Date 2010-12-24 10:17 Edited 2010-12-24 10:24

>Oh, let me add this one, it would be trully incredible if the Cluster solves it as it's also a long string of difficult moves to find the win.


I have already asked for this position in the past:
Rybka Cluster challenge. 24 very hard testpositions!!

But the last year's cluster couldn't find the win. Here is what Lukas had posted:
Lukas Reply.... (Rybka Cluster challenge. 24 very hard testpositions!!)

Analysis by Rybka 3w91 clusterJB:

1.Qd5 Rxe6 2.Bd1 Nf7 3.Ba4 Qe7 4.Qa8+ Qd8 5.Qxd8+ Nxd8 6.Bb3
  =/+  (-0.60 !)   Depth: 8   00:00:00  318kN
1.Qd5 Rxe6 2.Bd1 Qd7 3.Qxd7 Nxd7 4.Bb3 Kf7 5.Rd1 Nc5 6.Bxe6+ Kxe6 7.Kf2 Nxe4+ 8.Kf3 Nd6 9.Re1+ Kf7
  -/+  (-0.76)   Depth: 8   00:00:00  359kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bd2 Qe7 4.Qxe7 Rxe7 5.Rf1 Rxe4
  =/+  (-0.45 !)   Depth: 8   00:00:00  459kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qb8+ Re8 6.Qxb6 Qe6 7.Qc7 Qe7 8.Qc4+ Qf7 9.Qd4 Qe6 10.Rf1 Kf7
  =/+  (-0.41)   Depth: 8   00:00:00  546kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qb8+ Re8 6.Qxb6 Qe6 7.Qc7 Qe7 8.Qc4+ Qf7 9.Qd4 Qe6 10.Rf1 Kf7 11.Bg5 f5 12.Qa7+ Kg8 13.Bh6 Qe7
  =/+  (-0.43)   Depth: 9   00:00:00  702kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qa8+ Re8 6.Qc6 g5 7.Rd1
  =/+  (-0.31 !)   Depth: 10   00:00:00  994kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qa8+ Re8 6.Qc6 g5 7.Rd1 g4 8.Qxb6 Qxa2 9.Qxf6 Qa7+ 10.Kh1 Qb7+ 11.Kg1
  =  (-0.16 !)   Depth: 10   00:00:00  1233kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qa8+ Qe8 6.Qb7 Re7 7.Qd5+ Re6 8.Re1 Kf7 9.Rxe6 Qxe6 10.Qb7+ Qe7 11.Qd5+ Ke8 12.Qc6+ Qd7 13.Qxf6 Qg4+ 14.Kh1 Qe4+ 15.Kg1 Rg8 16.Qxb6
  =  (-0.16)   Depth: 10   00:00:00  1424kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qa8+ Qe8 6.Qb7 Re7 7.Qd5+ Re6 8.Re1 Kf7 9.Rxe6 Qxe6 10.Qb7+ Qe7 11.Qd5+ Ke8 12.Qc6+ Qd7 13.Qxf6 Qg4+ 14.Kh1 Qe4+ 15.Kg1 Rg8 16.Qxb6
  =  (-0.16)   Depth: 11   00:00:00  1620kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bxf7+ Qxf7 5.Qa8+ Qe8 6.Qb7 Re7 7.Qd5+ Re6 8.Re1 Kf7 9.Rxe6 Qxe6 10.Qb7+ Qe7 11.Qd5+ Ke8 12.Bg7 Qxg7 13.Qa8+ Kf7 14.Qd5+ Ke7 15.Qb7+ Kf8 16.Qc8+
  =  (-0.04 !)   Depth: 12   00:00:01  2444kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 12   00:00:02  5402kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 13   00:00:02  5874kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4 Rxe4 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 14   00:00:03  7192kN
1.Qa3 Rxe6 2.Qxa7 Nf7 3.Bc4[] Rxe4 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 15   00:00:03  9666kN
1.Qa3 Rxe6[] 2.Qxa7 Nf7 3.Bc4[] Rxe4[] 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 16   00:00:05  14194kN
1.Qa3 Rxe6[] 2.Qxa7 Nf7 3.Bc4[] Rxe4[] 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 17   00:00:09  28482kN
1.Qa3 Rxe6[] 2.Qxa7 Nf7 3.Bc4[] Rxe4[] 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 18   00:00:17  53907kN
1.Qa3 Rxe6[] 2.Qxa7 Nf7 3.Bc4[] Rxe4[] 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 19   00:00:29  92587kN
1.Qa3 Rxe6[] 2.Qxa7 Nf7 3.Bc4[] Rxe4[] 4.Bb3 g5 5.Qb7 Re1+ 6.Rxe1[] Qxe1+ 7.Kg2 Qe2+ 8.Kg3 Qd3+ 9.Kg2 Qe2+ 10.Kg3 Qd3+ 11.Kg2 Qe2+ 12.Kg3 Qd3+ 13.Kg2 Qe2+ 14.Kg3 Qd3+ 15.Kg2 Qe2+ 16.Kg3
  =  (0.00)   Depth: 20   00:01:00  205mN
1.Qa3 Rxe6[] 2.Qxa7 Qe7 3.Qa8+ Qe8 4.Qa7 Qe7 5.Qa8+ Qe8 6.Qa7 Qe7 7.Qa8+ Qe8 8.Qa7 Qe7 9.Qa8+ Qe8 10.Qa7 Qe7 11.Qa8+ Qe8 12.Qa7 Qe7 13.Qa8+ Qe8 14.Qa7 Qe7 15.Qa8+ Qe8 16.Qa7
  =  (0.00)   Depth: 21   00:02:09  403mN
1.Qa3 Rxe6[] 2.Qxa7 Qe7 3.Qa8+ Qe8 4.Qa7 Qe7 5.Qa8+ Qe8 6.Qa7 Qe7 7.Qa8+ Qe8 8.Qa7 Qe7 9.Qa8+ Qe8 10.Qa7 Qe7 11.Qa8+ Qe8 12.Qa7 Qe7 13.Qa8+ Qe8 14.Qa7 Qe7 15.Qa8+ Qe8 16.Qa7
  =  (0.00)   Depth: 22   00:04:01  757mN
1.Qa3 Rxe6[] 2.Qxa7 Qe7 3.Qa8+ Qe8[] 4.Qa7 Qe7 5.Qa8+ Qe8[] 6.Qa7 Qe7 7.Qa8+ Qe8[] 8.Qa7 Qe7 9.Qa8+ Qe8[] 10.Qa7 Qe7 11.Qa8+ Qe8[] 12.Qa7 Qe7 13.Qa8+ Qe8[] 14.Qa7 Qe7 15.Qa8+ Qe8[] 16.Qa7
  =  (0.00)   Depth: 23   00:07:12  1327mN
1.Qa3 Rxe6[] 2.Qxa7 Qe7 3.Qa8+ Qe8[] 4.Qa7 Qe7 5.Qa8+ Qe8[] 6.Qa7 Qe7 7.Qa8+ Qe8[] 8.Qa7 Qe7 9.Qa8+ Qe8[] 10.Qa7 Qe7 11.Qa8+ Qe8[] 12.Qa7 Qe7 13.Qa8+ Qe8[] 14.Qa7 Qe7 15.Qa8+ Qe8[] 16.Qa7
  =  (0.00)   Depth: 24   00:14:13  2595mN
1.Qa3 Rxe6[] 2.Qxa7 Qe7 3.Qa8+ Qe8[] 4.Qa7 Qe7 5.Qa8+ Qe8[] 6.Qa7 Qe7 7.Qa8+ Qe8[] 8.Qa7 Qe7 9.Qa8+ Qe8[] 10.Qa7 Qe7 11.Qa8+ Qe8[] 12.Qa7 Qe7 13.Qa8+ Qe8[] 14.Qa7 Qe7 15.Qa8+ Qe8[] 16.Qa7
  =  (0.00)   Depth: 25   00:25:33  4648mN

can't find the winning move - but how is the line for that win?
Parent - - By Uly (Gold) Date 2010-12-24 22:22
Thanks, I see, so the Cluster was already defeated by a position (is that from an actual game? I think that hasn't happened yet.)
Parent - - By George Tsavdaris (****) Date 2010-12-24 22:41

>is that from an actual game?


[Event "Moscow-ch"]
[Site "Moscow"]
[Date "1951.??.??"]
[EventDate "?"]
[Round "5"]
[Result "1-0"]
[White "Yuri S Gusev"]
[Black "Yuri Averbakh"]
[ECO "B70"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "73"]

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 g6 6.Be2 Nc6 7.Nb3
Bg7 8.O-O Be6 9.f4 Rc8 10.f5 Bd7 11.g4 Ne5 12.g5 Ng8 13.Nd5 f6
14.Be3 b6 15.Nd4 Kf7 16.c3 Qe8 17.Ne6 Bxe6 18.fxe6+ Kf8
19.Nxf6 Nxf6 20.gxf6 Bxf6 21.Bh6+ Kg8 22.Rxf6 exf6 23.Qxd6 Rc6
24.Qxe5 fxe5 25.Rf1 Rc8 26.Bd1 Rc4 27.Bb3 b5 28.Bxc4 bxc4
29.b3 a5 30.bxc4 Qe7 31.Kg2 Qa3 32.Rf2 Qe7 33.Rf1 g5 34.Rf5 g4
35.c5 Qd8 36.c6 Qe7 37.c7 1-0

>Thanks, I see, so the Cluster was already defeated by a position (is that from an actual game? I think that hasn't happened yet.)


What hasn't happened yet?
There are countless of moves from actual games, that the cluster can't find.:lol:
Parent - By Felix Kling (Gold) Date 2010-12-26 23:16
Sure there are countless moves the cluster doesn't find, but most because they aren't any good :)
Parent - By Lukas Cimiotti (Bronze) Date 2010-12-25 09:59
2K3k1/1p6/R3p1p1/1rB1P1P1/8/8/1Pb5/8 w - - 0 1


Analysis by Rybka Cluster 248 Cores:

1.Rc6 bxc6 2.b4 Rxc5 3.bxc5 Be4 4.Kc7 Kg7 5.Kd6 Kf7 6.Kd7 Bd5 7.Kd8 Bf3 8.Kd7 Be4 9.Kd6 Bd5 10.Kd7
  -+  (-2.89)   Depth: 16   00:00:01
1.Rc6 bxc6 2.b4 Rxc5 3.bxc5 Kf7 4.Kd7 Be4 5.Kd8 Bd5 6.Kd7 Kf8 7.Kd8 Bg2 8.Kd7 Kf7
  -+  (-2.86)   Depth: 16   00:00:02  19147kN
1.Rc6 bxc6 2.b4 Rxc5 3.bxc5 Kf7 4.Kc7 Be4 5.Kd7 Bh1 6.Kd8 Bg2 7.Kd7 Bd5 8.Kd8 Bf3
  -+  (-2.86)   Depth: 17   00:00:03  31405kN
1.Rc6 bxc6 2.b4 Ba4 3.Bd6 Rd5 4.Kb7 Rd3 5.Kb6 Kf7 6.Bc7 Bb5 7.Kc5 Rg3 8.Bd8 Rg4 9.Kb6 Rxb4 10.Kc5
  -+  (-3.09)   Depth: 18   00:00:04  45437kN
1.Rc6 bxc6 2.b4 Ba4 3.Kd7 Kf7 4.Kc7 Rxc5 5.bxc5 Ke7 6.Kb6 Kd7 7.Kb7 Bb3 8.Kb6 Bc2 9.Ka5
  -+  (-3.21)   Depth: 19   00:00:05  55748kN

For this I used some contempt settings :wink:
Parent - - By Labyrinth (*****) Date 2010-12-24 10:32
8/p5p1/1pP3p1/p5p1/k3p3/4p3/K7/8 w - - 0 1


White wins by force.
Parent - - By George Tsavdaris (****) Date 2010-12-24 11:09
Yes and the solution:

[Event "??"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "Mate in 35"]
[Black "-"]
[Result "1-0"]
[SetUp "1"]
[FEN "8/p5p1/1pP3p1/p5p1/k3p3/4p3/K7/8 w - - 0 1"]
[PlyCount "69"]
[EventDate "????.??.??"]

1. c7 e2 2. c8=Q e1=Q 3. Qc4+ Qb4 4. Qc6+ Qb5 5. Qxe4+ Qb4 6. Qd3 g4 7. Qd7+
Qb5 8. Qxg4+ Qb4 9. Qd7+ Qb5 10. Qd4+ Qb4 11. Qd3 g5 12. Qd7+ Qb5 13. Qd4+ Qb4
14. Qd3 g6 15. Qd7+ Qb5 16. Qd4+ Qb4 17. Qd3 a6 18. Qd7+ Qb5 19. Qd4+ Qb4 20.
Qd3 g4 21. Qd7+ Qb5 22. Qxg4+ Qb4 23. Qd7+ Qb5 24. Qd4+ Qb4 25. Qd3 g5 26. Qd7+
Qb5 27. Qd4+ Qb4 28. Qd3 g4 29. Qd7+ Qb5 30. Qxg4+ Qb4 31. Qd7+ Qb5 32. Qd4+
Qb4 33. Qd3 b5 34. Qc2+ Qb3+ 35. Qxb3# 1-0

That is a very good position for the cluster!! Since i don't see something very difficult only to see some moves ahead. Let's see if Lukas will have some time to all these....
Parent - - By Akbarfan (***) Date 2010-12-24 13:12 Edited 2010-12-24 13:17
It's a mate in 42

33... Qb2+  34.Kxb2 Kb4  35.Qc3+ Kb5  36.Qc7 Kb4  37.Kc2 a4  38.Qc6 a5  39.Kd3 Ka3  40.Qb5 Ka2  42.Kc2 a3  42.Qb1#
33... Qa3+  34.Qxa3+ Kb5  35.Qc3 a4  36.Ka3 a5  37.Qc7 Ka6  38.Kxa4 b5+  39.Kb3 a4+  40.Kb4 a3  41.Kc5 a2  42.Qb6#
Parent - - By George Tsavdaris (****) Date 2010-12-24 13:19
Yes of course my mistake.
Parent - - By RFK (Gold) Date 2010-12-24 17:26
6N1/6p1/1B2P3/6r1/6Pk/5ppP/6P1/7K w - - 0 1


1... Rd5 2. Ne7 Rd1+ 3. Bg1 f2 4. Ng6+ Kg5 5. h4+ Kxg6 6. h5+ Kg5 7. e7 fxg1=Q#
{(0:00:02) 294kN}{[%eval -29993,20,Deep Rybka 4 SSE42 x64][%meval 2s|TB:93]} *
Parent - - By George Tsavdaris (****) Date 2010-12-24 17:52
Don't tell me you entered the position manually by hand?
Because it's white to play(notice the little FEN tag saying w). :lol:

BTW all GUIs have the very easy paste position feature. So you just copy the position from here by selecting it and paste it there.
That's way too easier and you can avoid putting the wrong color....:lol:
Parent - - By RFK (Gold) Date 2010-12-24 18:17 Edited 2010-12-24 18:20
No pasted the position! Oddly enough analysis continues to move black first!?
Parent - - By George Tsavdaris (****) Date 2010-12-24 18:40

>No pasted the position! Oddly enough analysis continues to move black first!?


:eek::eek:
In what GUI this happened? That's a bug for sure if the FEN has white to move and the GUI says to the engine that is the black side to move.
Parent - By RFK (Gold) Date 2010-12-24 18:45
I turned the puzzle over to Fritz and the analysis was as you predicted!
Parent - By Labyrinth (*****) Date 2010-12-25 01:39
Yep, classic case of a problem that can be made simple (relatively) with the use of abstract concepts, but is incredibly difficult using linear, concrete calculation.

I still think that there must be a way for computers to look ahead without calculating linearly.
Parent - - By h1a8 (***) Date 2010-12-27 19:49
can you post or pm me the rest of the solution.
Parent - By Uly (Gold) Date 2010-12-28 11:21
From Howard E at some forbidden zone:

1. Qxe5 fxe5 2. Rf1 a6 3. Bd1 b5 4. Bb3 Rc4 5. a4 Qe7 6. axb5 axb5 7. Rf2 Qe8 8. Ba2 Qe7 9. b4

Or

1. Qxe5 fxe5 2. Rf1 Rc7 3. Bd1 Re7 4. Bb3 b5 5. a4 bxa4 6. Bxa4 Qd8 7. b4 Qb8 8. Bb3 a6 9. c4

Black has plenty of alternatives, but after playing with them it was shocking to see engines jumping from scores of -8.00 (black is winning by 8 pawns) directly to #10 (white is mating in ten moves :eek:) in the next iteration, and stuff like that.
- - By Weirwindle (***) Date 2010-12-30 19:23
Drunick posted a problem a couple years ago that has just been bumped in the forum.
Can rybka after long depth find this?
n2Bqk2/5p1p/5KP1/p7/8/8/2Q5/8 w - - 0 1
Parent - - By George Tsavdaris (****) Date 2010-12-30 22:33
Huh! That's my favorite testposition!!:lol:
This has been posted in CCC million times and i think i'm the first one that i've posted it, long long time ago. I have found it in Chessmaster and it was called "By Matous, 1975".
The Queen was on g4 i think in my version and you can find her on a6 also and on c3 like you gave, but i think a6 is the better square as it deceives more the computers. :lol:

The solution of course is(i remember it without even looking at the board-imagine how much i loved it:lol:) 1.Qc8! Kg8(forced) 2.Bc7!! Qxc8 3.gxf7+ Kh8 4.Be5 Qc5 5.Bb2! Nc7(and now both Knight and Queen are overloaded) 6.Ba1 a4 7. Bb2 a3 8.Ba1 a2 9.Bb2 a1Q 10.Bxa1 and now black can choose the way he dies.:grin:

The big hardware power of cluster here, perhaps it will not help Rybka, since if Rybka prunes the seemingly super bad-heavy-material-losing-move 2.Bc7, then it will not choose it no matter the ply depth. But perhaps i'm wrong.
Parent - By Weirwindle (***) Date 2010-12-30 22:40
Yes, I just posted a solution on the other topic using ChestUCI here http://rybkaforum.net/cgi-bin/rybkaforum/topic_show.pl?tid=7397
Parent - By Lazy Frank (****) Date 2011-03-21 05:51 Edited 2011-03-21 05:55
Position:

n2Bqk2/5p1p/5KP1/p7/8/8/2Q5/8 w - - 0 1


Task: found mate in 13 moves
Test options: without any predefined settings,
without tablebases, without on/off pruning function,
only clear engine alpha-beta research nodes algorithm.
Note: 10^3 hope is 1024.

Solution: not found.
Current depth=23/87
Research nodes=16028 mN
Conclusion: unfortunate Qc2-c5+ with 0.00 stay first, Qc2-c8 with 0.00 - second in MPV query.
But it is "normal" decision.
This puzzle is one of the biggest exclusion from chess/mate solutions.
Up Topic Rybka Support & Discussion / Rybka Discussion / 10 beautiful testpositions! Rybka-Cluster challenge....
1 2 3 Previous Next  

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill