Junior - Hiracs

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

User avatar
AdminX
Posts: 6340
Joined: Mon Mar 13, 2006 2:34 pm
Location: Acworth, GA

Junior - Hiracs

Post by AdminX »

Junior played a sweet game versus Hiarcs today. Almost makes me feel like playing Eminem's "Guess Who's Back". The opening was a Sicilian Najdorf Variation.

B81 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. h3 e6 7. g4 b5 8. Bg2 Bb7 9. O-O

[d]rn1qkb1r/1b3ppp/p2ppn2/1p6/3NP1P1/2N4P/PPP2PB1/R1BQ1RK1 b kq - 0 9

Seems like I had just finished reading a artical by Alexey Kuzmin called "Renaissance thanks to a piece sacrifice" just the other day. :wink: However in the Junior game black followed a different path with 9. ... Nfd7 instead playing 9. ... b4 and allowing white to play Nd5 (Karjakin's idea).

To the best of my knowledge both programs were playing on similar hardware which is a Intel Xeon W5580 @ 3.2GHz x 8 PC. Someone correct this if I am wrong.

[Event "17th World Computer Chess Championship 2"]
[Site "Pamplona/ESP"]
[Date "2009.05.12"]
[Round "4"]
[White "Junior"]
[Black "Hiarcs"]
[Result "1-0"]
[ECO "B81"]
[PlyCount "63"]
[EventDate "2009.??.??"]

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. h3 e6 7. g4 b5 8. Bg2
Bb7 9. O-O Nfd7 10. f4 b4 11. Nce2 Nc5 12. Be3 Bxe4 13. a3 bxa3 14. b4 Bxg2 15.
Kxg2 Na4 16. Ng3 d5 17. f5 e5 18. Rxa3 Nb2 19. Qf3 Qd7 20. f6 g6 21. Re1 Nc4
22. Rd3 Qb7 23. Ndf5 gxf5 24. Rxd5 Qc6 25. Nxf5 Nd7 26. Bc5 Rc8 27. Red1 Rc7
28. Qe4 Bxc5 29. bxc5 Rg8 30. Kh2 Qxf6 31. Qxc4 Rg5 32. Qe4 1-0
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
jdart
Posts: 4366
Joined: Fri Mar 10, 2006 5:23 am
Location: http://www.arasanchess.org

Re: Junior - Hiracs

Post by jdart »

That's quite a game, although I'm not sure how much was book preparation vs. real-time searching.

--Jon
Uri
Posts: 473
Joined: Thu Dec 27, 2007 9:34 pm

Re: Junior - Hiracs

Post by Uri »

jdart wrote:That's quite a game, although I'm not sure how much was book preparation vs. real-time searching.

--Jon
Junior played this game like Kasparov played in his best days.
It's weird to see an old program like Junior 10 play so strong. Rybka 3 on my Intel core 2 Q8200 "saw" these moves only after a long time. I let Rybka 3 32 bit think for 20 minutes and it didn't see 20. f6. Only after 23 minutes if found this aggressive and winning move.
CThinker
Posts: 388
Joined: Wed Mar 08, 2006 10:08 pm

Re: Junior - Hiracs

Post by CThinker »

Uri wrote:
jdart wrote:That's quite a game, although I'm not sure how much was book preparation vs. real-time searching.

--Jon
Junior played this game like Kasparov played in his best days.
It's weird to see an old program like Junior 10 play so strong. Rybka 3 on my Intel core 2 Q8200 "saw" these moves only after a long time. I let Rybka 3 32 bit think for 20 minutes and it didn't see 20. f6. Only after 23 minutes if found this aggressive and winning move.
I think you just need to give Rybka a huge hash table (e.g., 2GB) and it will find the move much earlier.

Pepito finds the move in under a minute, and so does Inert-T with a big hash table (512 MB).
ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Re: Junior - Hiracs

Post by ernest »

Uri wrote:I let Rybka 3 32 bit think for 20 minutes and it didn't see 20. f6. Only after 23 minutes if found this aggressive and winning move.
Well, my Rybka 3 (32-bit, 512 MB hash on Core 2 Duo@3GHz) doesn't choose 20.f6, and thinks 22... Qb7 is a mistake by Hiarcs.
Junior - Hiarcs, 17th World Computer Chess Championship 2 2009
[d]rn2kb1r/3q1ppp/p7/3ppP2/1P1N2P1/R3BQNP/1nP3K1/5R2 w kq - 0 1

Code: Select all

Analysis by Rybka 3 32-bit:
20.f6 g6 21.Rb3 Nc4 
  =/+  (-0.33)   Depth: 6   00:00:00  9kN
20.f6 g6 21.Rb3 Nc4 22.Re1 h5 
  =/+  (-0.37)   Depth: 7   00:00:00  19kN
20.f6 g6 21.Rb3 Nc4 22.Re1 h5 
  =/+  (-0.37)   Depth: 8   00:00:00  28kN
20.Rb3 Nc4 21.f6 
  =/+  (-0.48)   Depth: 9   00:00:01  103kN
20.f6 
  =/+  (-0.28 !)   Depth: 10   00:00:08  694kN
20.f6 g6 21.Re1 
  =  (-0.18)   Depth: 10   00:00:09  774kN
20.f6 g6 21.Bf2 Nc4 22.Rd3 Bxb4 23.c3 Bc5 24.Nc2 Bxf2 25.Rxd5 Qe6 26.Rxf2 0-0 
  =  (-0.21)   Depth: 11   00:00:11  903kN
20.f6 g6 21.Bf2 Nc4 22.Rd3 h5 23.gxh5 Bxb4 24.Nb3 e4 25.Nxe4 Rxh5 26.Nec5 Qf5 27.Qe2+ Qe5 28.Qxe5+ Rxe5 29.Bg3 Re2+ 30.Rf2 
  =/+  (-0.31)   Depth: 12   00:00:16  1331kN
20.Nh5 Rg8 
  =  (-0.01 !)   Depth: 12   00:00:30  2503kN
20.f6 g6 21.Bf2 Nc4 22.Rd3 h5 23.gxh5 Bxb4 24.Nb3 e4 25.Nxe4 Rxh5 26.Nec5 Qf5 27.Qe2+ Qe5 28.Qxe5+ Rxe5 29.Bg3 Re2+ 30.Rf2 
  =/+  (-0.31)   Depth: 12   00:00:56  4984kN
20.Bf2 Nc4 21.Rd3 exd4 22.Re1+ Kd8 23.Rxd4 Qc6 24.Rxd5+ 
  =  (0.00)   Depth: 12   00:01:03  5555kN
20.Bf2 Nc4 21.Rd3 exd4 22.Re1+ Kd8 23.Rxd4 Qc6 24.Rxd5+ 
  =  (0.00)   Depth: 13   00:01:03  5555kN
20.Bf2 
  =  (0.20 !)   Depth: 14   00:01:43  8560kN
20.Bf2 Nc4 21.Rd3 Bxb4 22.f6 g6 23.Nb3 Qc6 24.Rxd5 0-0 25.c3 e4 26.Qxe4 Re8 27.Qf3 Bxc3 28.Rc1 Ne5 29.Qe4 
  =  (-0.06)   Depth: 14   00:01:51  9290kN
20.Bf2 Nc4 21.Rd3 Bxb4 22.f6 g6 23.Nb3 Qc6 24.Rxd5 0-0 25.c3 e4 26.Qxe4 Re8 27.Qf3 Bf8 28.Rfd1 Ne5 29.Qf4 Qxc3 30.Na5 Nbd7 31.Ne4 
  =  (-0.02)   Depth: 15   00:02:23  12574kN
20.Bf2 Nc4 21.Rd3 Bxb4 22.f6 g6 23.Nb3 Qc6 24.Rxd5 0-0 25.c3 e4 26.Qxe4 Re8 27.Qf3 Bf8 28.Rfd1 Ne5 29.Qf4 Qxc3 30.Na5 Nbd7 31.Ne4 
  =  (-0.02)   Depth: 16   00:03:30  19130kN
20.Bf2 
  =  (0.18 !)   Depth: 17   00:06:41  36814kN
20.Bf2 Nc4 21.Rd3 Bxb4 22.f6 g6 23.Nb3 Qc6 24.Rxd5 0-0 25.c3 Ba3 26.Rfd1 a5 27.Nd2 Nxd2 28.R1xd2 a4 29.Ne4 Re8 30.Rd1 
  =  (0.22)   Depth: 17   00:08:34  44192kN
20.Bf2 Nc4 21.Rd3 Bxb4 22.f6 g6 23.Nb3 Qc6 24.Rxd5 0-0 25.c3 Ba3 26.Rfd1 a5 27.Nd2 Nxd2 28.R1xd2 a4 29.Ne4 Re8 30.Rd1 
  =  (0.22)   Depth: 18   00:12:15  65282kN
20.Bf2 Nc4 21.Rd3 Qb7 22.Nb3 Bxb4 23.Rxd5 0-0 24.f6 g6 25.Rfd1 Nb6 26.Bxb6 Qxb6 27.c3 Ba3 28.Nd2 Nc6 29.Qe4 Rad8 30.Nc4 Qb3 31.R5d2 Rxd2+ 32.Rxd2 Bc1 33.Rd6 
  +/=  (0.38)   Depth: 19   00:40:55  218mN
20.Bf2 Nc4 21.Rd3 Qb7 22.Nb3 Bxb4 23.Rxd5 0-0 24.f6 g6 25.Rfd1 Nb6 26.Bxb6 Qxb6 27.c3 Ba3 28.Nd2 Nc6 29.Qe4 Rad8 30.Nc4 Qb3 31.R5d2 Rxd2+ 32.Rxd2 Bc1 33.Rd6 
  +/=  (0.38)   Depth: 20   01:04:06  345mN
20.Bf2 Nc4 21.Rd3 Qb7 22.Nb3 Bxb4 23.Rxd5 0-0 24.f6 g6 25.Rfd1 Nb6 26.Bxb6 Qxb6 27.Ne4 Nc6 28.c4 h6 29.c5 Qb8 30.Nd6 a5 31.Qe4 a4 32.Nxf7 Rxf7 33.Qxg6+ Kf8 34.Qxh6+ Kg8 35.Qg6+ 
  +/=  (0.41)   Depth: 21   02:07:44  691mN
ernest
Posts: 2041
Joined: Wed Mar 08, 2006 8:30 pm

Re: Junior - Hiracs

Post by ernest »

Multi-variation analysis gives the following picture (at depth 17, after more than 1 hour analysis):

Code: Select all

Analysis by Rybka 3 32-bit:
1. =  (0.14): 20.Bf2 Nc4 21.Rd3 Bxb4 22.f6 g6 23.Nb3 Qc6 24.Rxd5 0-0 25.c3 Ba3 26.Rfd1 a5 27.Nd2 Re8 28.Nxc4 Qxc4 
2. =  (0.00): 20.f6 g6 21.Rc3 Na4 22.Rd3 Nb2 23.Rc3 Na4 24.Rd3 Nb2 25.Rc3 Na4 26.Rd3 Nb2 27.Rc3 Na4 28.Rd3 Nb2 29.Rc3 Na4 30.Rd3 Nb2 31.Rc3 Na4 32.Rd3 Nb2 33.Rc3 Na4 34.Rd3 Nb2 35.Rc3 
3. =  (0.00): 20.Bg1 Nc4 21.Rc3 exd4 22.Re1+ Kd8 23.Rxc4[] dxc4 24.Qxa8[] Qc6+ 25.Qxc6[] Nxc6 26.Bxd4 Nxd4 27.Rd1[] Bxb4 28.Rxd4+ Ke7 29.Re4+ Kd7 30.Rd4+ Ke7 31.Re4+ Kd7 32.Rd4+ Ke7 33.Re4+ Kd7 34.Rd4+ Ke7 35.Re4+ 
4. =  (0.00): 20.Rc3 Na4 21.Rd3[] Nb2 22.f6 g6 23.Rc3 Na4 24.Rd3 Nb2 25.Rc3 Na4 26.Rd3 Nb2 27.Rc3 Na4 28.Rd3 Nb2 29.Rc3 Na4 30.Rd3 Nb2 31.Rc3 Na4 32.Rd3 Nb2 33.Rc3 Na4 34.Rd3 Nb2 35.Rc3