Houdini 1.5 - increase in strenght at LTC

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

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
beram
Posts: 1187
Joined: Wed Jan 06, 2010 2:11 pm

Houdini 1.5 - increase in strenght at LTC

Post by beram » Sun Dec 19, 2010 10:14 pm

Increase in strenght at LTC for Houdini 1.5 against DR4

I ve done an expiriment with doubling the TC for the matches between Houdini 1.5 and Rybka 4 with same testpositions.
What I observed is that the gap in ELO doesn't diminish, as some Rybka fans keep on believing, but instead even becomes bigger; from 58% at 3min2sec to 60% at 6min4sec, to 62 % at TC 12min 8sec

So I challenge Jeroen Noomen to replay a 60 games (30 Noomenpositions) match under same TC conditions, 1hour a game each, as he earlier on did between Houdini 1.03a and Rybka 4, but this time with Houdini 1.5

Hat off for Robert Houdart !

Bram-PC, Blitz AMDx6 each 4cores, Blitz 3m+2s,
60 games Noomen testsuite+30 games privat testsuite

1 Houdini 1.5 x64 +36/=34/-20 58.89% 53.0/90
2 Deep Rybka 4 x64 +20/=34/-36 41.11% 37.0/90


Bram-PC, Blitz 6m+4s AMDx6 each 4cores
50 games privat testsuite

1 Houdini 1.5 x64 +16/=28/-6 60.00% 30.0/50
2 Deep Rybka 4 x64 +6/=28/-16 40.00% 20.0/50


Bram-PC, Blitz 12m+8s AMDx6 each 4cores
50 games privat testsuite

1 Houdini 1.5 x64 +20/=22/-8 62.00% 31.0/50
2 Deep Rybka 4 x64 +8/=22/-20 38.00% 19.0/50

beram
Posts: 1187
Joined: Wed Jan 06, 2010 2:11 pm

Re: Houdini 1.5 - increase in strenght at LTC

Post by beram » Tue Dec 21, 2010 6:36 pm

The test continues and again Houdini 1.5 shows his strenght in a long TC games match. After the match with 12m 8 sec I continued with 1 hour games
after 14 games H15 leads with 9,5 - 4,5 !

I give 4 games where H15 wins with white and black:
This position occurs in the first

[D] 3r1rk1/1b1n2p1/p2q3p/1p1P4/4P3/1PN2p1B/3Q1P1P/3R1R1K w - - 0

Code: Select all

27. Rg1 $1 {[%eval 28,20] [%emt 0:00:14] (Bxd7) the expected Bxd7 would give black better play,
but H15 plays the good move Rg1} Ne5 {[%eval -4,15] [%emt 0:00:11]} 28. Rg3 $1
{[%eval 33,21] [%emt 0:00:13] (Na2) again DR4 expects a worse and bad move for
white Na2 would displace the knight} Bc8 {[%eval 15,15] [%emt 0:00:21]} 29.
Rdg1 {[%eval 48,20] [%emt 0:00:14] (Bxc8)} g5 {[%eval 30,17] [%emt 0:00:25]}
30. Bxc8 {[%eval 48,19] [%emt 0:00:14]} Rxc8 {[%eval 30,17] [%emt 0:00:02]} 31.
Nd1 {[%eval 35,21] [%emt 0:00:51]} b4 $6 {[%eval 37,16] [%emt 0:00:48] (a5) 
this b4 is a bad move. DR4 trusts on a rook for knight sac on f5, but only few
moves later H15 shows he evals better the expected move a5 could probably hold
after} (31... a5 $1 32. Ne3 (32. Qxa5 Rf4 33. Ne3 Nd3 34. Qd2 Nc5) 32... Qb4
33. Qxb4 axb4 34. Nf5 Kh7 35. d6 Rc2 36. Rh3 Nf7) 32. Ne3 {[%eval 49,20] [%emt
0:00:29]} Rc3 {[%eval 37,15] [%emt 0:00:13]} 33. Nf5 {[%eval 60,20] [%emt 0:00:
12]} Rxf5 {[%eval 37,15] [%emt 0:00:04]} 34. exf5 {[%eval 60,19] [%emt 0:00:00]
} Rd3 {[%eval 73,14] [%emt 0:00:13]} 35. Qc2 {[%eval 66,20] [%emt 0:00:31]} Kf7
{[%eval 79,16] [%emt 0:00:49]} 36. Qc8 {[%eval 68,21] [%emt 0:00:13]} Rxd5 {
[%eval 98,17] [%emt 0:00:30]} 37. Rc1 {[%eval 77,20] [%emt 0:00:18] (Qb7+)} Rd4
{[%eval 93,15] [%emt 0:00:22]} 38. h3 {[%eval 135,19] [%emt 0:00:12] (Qh8)} a5
{[%eval 138,16] [%emt 0:00:33] (Kf6)} 39. Rgg1 {[%eval 332,19] [%emt 0:00:20]}

[Event "Bram-PC, Blitz 12m+8s"]
[Site "AMD X6 1090T 3200 Mhz, (Fritz"]
[Date "2010.12.19"]
[Round "47"]
[White "Houdini 1.5 x64"]
[Black "Deep Rybka 4 x64"]
[Result "1-0"]
[ECO "D15"]
[Annotator "0.24;0.05"]
[PlyCount "93"]
[EventDate "2010.??.??"]
[TimeControl "720+8"]

{AMD Phenom(tm) II X6 1090T Processor 3207 MHz W=19.4 plies; 7.308kN/s B=16.2
plies; 329kN/s} 1. d4 d5 2. Nf3 Nf6 3. c4 c6 4. Nc3 a6 5. Ne5 Nbd7 {Both last
book move} 6. Bf4 {[%eval 24,20] [%emt 0:00:29]} dxc4 {[%eval 5,18] [%emt 0:01:
19] (Nh5)} 7. Nxc4 {[%eval 24,21] [%emt 0:00:31]} b5 {[%eval 0,16] [%emt 0:00:
33]} 8. Ne5 {[%eval 16,22] [%emt 0:00:22]} Bb7 {[%eval 0,17] [%emt 0:00:39]
(Nxe5)} 9. Qc2 {[%eval 8,19] [%emt 0:00:41] (e3)} e6 {[%eval 0,17] [%emt 0:00:
50] (Nd5)} 10. Rd1 {[%eval 14,20] [%emt 0:00:17]} c5 {[%eval 10,16] [%emt 0:00:
23]} 11. Nxd7 {[%eval 16,20] [%emt 0:00:17]} Nxd7 {[%eval 10,17] [%emt 0:00:10]
} 12. d5 {[%eval 11,22] [%emt 0:00:17]} c4 {[%eval 7,17] [%emt 0:01:38] (Qf6)}
13. g3 {[%eval 18,21] [%emt 0:00:21] (h3)} e5 {[%eval 5,17] [%emt 0:00:29]
(Qa5)} 14. Be3 {[%eval 19,23] [%emt 0:00:21]} Bc5 {[%eval 11,18] [%emt 0:00:19]
} 15. Ne4 {[%eval 13,22] [%emt 0:00:29]} Bb4+ {[%eval 11,19] [%emt 0:00:29]}
16. Bd2 {[%eval 11,24] [%emt 0:00:17]} Bxd2+ {[%eval 4,19] [%emt 0:00:37]} 17.
Qxd2 {[%eval 5,22] [%emt 0:00:26]} O-O {[%eval 1,19] [%emt 0:00:22]} 18. Bg2 {
[%eval 12,23] [%emt 0:00:21]} Qb6 {[%eval 1,19] [%emt 0:00:28]} 19. O-O {
[%eval 6,23] [%emt 0:00:47]} Rad8 {[%eval 0,18] [%emt 0:00:33] (f5)} 20. b3 {
[%eval 15,22] [%emt 0:00:15]} cxb3 {[%eval -2,18] [%emt 0:00:16]} 21. axb3 {
[%eval 16,20] [%emt 0:00:00]} h6 {[%eval -2,19] [%emt 0:00:13]} 22. Nc3 {
[%eval 16,21] [%emt 0:00:16]} f5 {[%eval -2,18] [%emt 0:00:26] (Rc8)} 23. e4 {
[%eval 23,20] [%emt 0:00:16]} f4 {[%eval -2,16] [%emt 0:00:16]} 24. gxf4 {
[%eval 21,20] [%emt 0:00:26] (Rb1)} exf4 {[%eval 0,18] [%emt 0:00:19]} 25. Kh1
{[%eval 25,22] [%emt 0:01:22] (Qd4)} f3 {[%eval -2,17] [%emt 0:00:39]} 26. Bh3
{[%eval 18,19] [%emt 0:00:00]} Qd6 {[%eval 0,18] [%emt 0:00:28] [D] 3r1rk1/
1b1n2p1/p2q3p/1p1P4/4P3/1PN2p1B/3Q1P1P/3R1R1K w - - 0 27} 27. Rg1 $1 {[%eval
28,20] [%emt 0:00:14] (Bxd7) the expected Bxd7 would give black better play,
but H15 plays the good move Rg1} Ne5 {[%eval -4,15] [%emt 0:00:11]} 28. Rg3 $1
{[%eval 33,21] [%emt 0:00:13] (Na2) again DR4 expects a worse and bad move for
white Na2 would displace the knight} Bc8 {[%eval 15,15] [%emt 0:00:21]} 29.
Rdg1 {[%eval 48,20] [%emt 0:00:14] (Bxc8)} g5 {[%eval 30,17] [%emt 0:00:25]}
30. Bxc8 {[%eval 48,19] [%emt 0:00:14]} Rxc8 {[%eval 30,17] [%emt 0:00:02]} 31.
Nd1 {[%eval 35,21] [%emt 0:00:51]} b4 $6 {[%eval 37,16] [%emt 0:00:48] (a5)
this b4 is a bad move. DR4 trusts on a rook for knight sac on f5, but only few
moves later H15 shows he evals better the expected move a5 could probably hold
after} (31... a5 $1 32. Ne3 (32. Qxa5 Rf4 33. Ne3 Nd3 34. Qd2 Nc5) 32... Qb4
33. Qxb4 axb4 34. Nf5 Kh7 35. d6 Rc2 36. Rh3 Nf7) 32. Ne3 {[%eval 49,20] [%emt
0:00:29]} Rc3 {[%eval 37,15] [%emt 0:00:13]} 33. Nf5 {[%eval 60,20] [%emt 0:00:
12]} Rxf5 {[%eval 37,15] [%emt 0:00:04]} 34. exf5 {[%eval 60,19] [%emt 0:00:00]
} Rd3 {[%eval 73,14] [%emt 0:00:13]} 35. Qc2 {[%eval 66,20] [%emt 0:00:31]} Kf7
{[%eval 79,16] [%emt 0:00:49]} 36. Qc8 {[%eval 68,21] [%emt 0:00:13]} Rxd5 {
[%eval 98,17] [%emt 0:00:30]} 37. Rc1 {[%eval 77,20] [%emt 0:00:18] (Qb7+)} Rd4
{[%eval 93,15] [%emt 0:00:22]} 38. h3 {[%eval 135,19] [%emt 0:00:12] (Qh8)} a5
{[%eval 138,16] [%emt 0:00:33] (Kf6)} 39. Rgg1 {[%eval 332,19] [%emt 0:00:20]}
Nd7 {[%eval 186,14] [%emt 0:00:10] (Rd2)} 40. Rge1 {[%eval 358,17] [%emt 0:00:
00]} Qb8 {[%eval 253,14] [%emt 0:00:03] (Qf8)} 41. Qc6 {[%eval 412,16] [%emt 0:
00:01] (Qa6)} Qd6 {[%eval 336,14] [%emt 0:00:12]} 42. Qa8 {[%eval 410,15]
[%emt 0:00:01]} Qb8 {[%eval 468,14] [%emt 0:00:06]} 43. Qxf3 {[%eval 468,14]
[%emt 0:00:00]} Kg8 {[%eval 526,14] [%emt 0:00:18] (Nf6)} 44. Re7 {[%eval 706,
13] [%emt 0:00:00] (Qh5)} g4 {[%eval 660,15] [%emt 0:00:10]} 45. hxg4 {[%eval
685,12] [%emt 0:00:00] (Qc6)} Ne5 {[%eval 715,11] [%emt 0:00:08] (Qd8)} 46. Qe2
{[%eval 760,11] [%emt 0:00:00]} Rxg4 {[%eval 715,10] [%emt 0:00:04]} 47. Rxe5 {
[%eval 760,10] [%emt 0:00:00]} 1-0

[Event "Bram-PC, Blitz 12m+8s"]
[Site "AMD X6 1090T 3200 Mhz, (Fritz"]
[Date "2010.12.19"]
[Round "48"]
[White "Deep Rybka 4 x64"]
[Black "Houdini 1.5 x64"]
[Result "0-1"]
[ECO "D15"]
[Annotator "0.17;-0.06"]
[PlyCount "81"]
[TimeControl "720+8"]

{AMD Phenom(tm) II X6 1090T Processor 3207 MHz W=17.3 plies; 355kN/s B=21.2
plies; 7.699kN/s} 1. d4 d5 2. Nf3 Nf6 3. c4 c6 4. Nc3 a6 5. Ne5 Nbd7 {Both
last book move} 6. g3 {[%eval 17,17] [%emt 0:01:47]} e6 {[%eval -6,21] [%emt 0:
00:20]} 7. f4 {[%eval 9,17] [%emt 0:00:28]} dxc4 {[%eval -18,20] [%emt 0:00:33]
(h5)} 8. Nxc4 {[%eval -11,17] [%emt 0:00:25] (a4)} b5 {[%eval -34,22] [%emt 0:
00:18]} 9. Ne3 {[%eval -14,19] [%emt 0:00:44] (Ne5)} Bb7 {[%eval -36,22] [%emt
0:00:35]} 10. Bg2 {[%eval -28,18] [%emt 0:00:15]} Qc7 {[%eval -35,21] [%emt 0:
00:26] (Qb6)} 11. O-O {[%eval -28,16] [%emt 0:01:22] (Qc2)} Rd8 {[%eval -48,23]
[%emt 0:00:30] (h5)} 12. Qc2 {[%eval -38,15] [%emt 0:00:37] (f5)} c5 {[%eval
-60,22] [%emt 0:00:44] (h5)} 13. d5 {[%eval -40,15] [%emt 0:00:17]} c4 {[%eval
-58,22] [%emt 0:01:13]} 14. a4 {[%eval -52,16] [%emt 0:00:36]} Bc5 {[%eval -58,
22] [%emt 0:00:31]} 15. axb5 {[%eval -52,16] [%emt 0:00:19] (dxe6)} Qb6 {
[%eval -66,21] [%emt 0:00:29]} 16. Na4 {[%eval -62,16] [%emt 0:00:24]} Bxe3+ {
[%eval -66,19] [%emt 0:00:00]} 17. Kh1 {[%eval -62,17] [%emt 0:00:04]} Qa7 {
[%eval -93,21] [%emt 0:00:43] (Qxb5)} 18. dxe6 {[%eval -90,16] [%emt 0:00:15]}
Bxc1 {[%eval -92,21] [%emt 0:00:31] (Bxg2+)} 19. exd7+ {[%eval -84,17] [%emt 0:
00:23]} Rxd7 {[%eval -63,19] [%emt 0:00:00]} 20. Qxc1 {[%eval -84,18] [%emt 0:
00:29]} Bxg2+ {[%eval -122,21] [%emt 0:00:38]} 21. Kxg2 {[%eval -84,18] [%emt
0:00:00]} axb5 {[%eval -119,20] [%emt 0:00:00]} 22. Nc3 {[%eval -84,18] [%emt
0:00:05]} Qb7+ {[%eval -123,22] [%emt 0:00:16]} 23. Rf3 {[%eval -84,19] [%emt
0:00:12]} O-O {[%eval -121,23] [%emt 0:00:49]} 24. Ra5 {[%eval -84,18] [%emt 0:
00:08]} Re8 {[%eval -121,21] [%emt 0:00:00]} 25. Rxb5 {[%eval -97,18] [%emt 0:
00:28] (b4)} Rxe2+ {[%eval -147,21] [%emt 0:00:21]} 26. Nxe2 {[%eval -97,18]
[%emt 0:00:06]} Qxb5 {[%eval -117,19] [%emt 0:00:00]} 27. b3 {[%eval -127,17]
[%emt 0:00:23]} Qb7 {[%eval -132,22] [%emt 0:00:14]} 28. Qxc4 {[%eval -127,17]
[%emt 0:00:08]} Rd2 {[%eval -132,20] [%emt 0:00:00]} 29. Kf2 {[%eval -145,16]
[%emt 0:00:27]} Ng4+ {[%eval -129,20] [%emt 0:00:17]} 30. Ke1 {[%eval -145,17]
[%emt 0:00:03]} Rd8 {[%eval -128,21] [%emt 0:00:14]} 31. Rd3 {[%eval -145,17]
[%emt 0:00:07]} Re8 {[%eval -129,21] [%emt 0:00:18] (Qh1+)} 32. Rd5 {[%eval
-140,16] [%emt 0:00:35] (Kd2)} Qa7 {[%eval -153,21] [%emt 0:00:18] (Nxh2)} 33.
h3 {[%eval -196,16] [%emt 0:00:28]} Qg1+ {[%eval -180,21] [%emt 0:00:13] (Qf2+)
} 34. Kd2 {[%eval -196,15] [%emt 0:00:00]} Qe3+ {[%eval -180,20] [%emt 0:00:00]
} 35. Kc2 {[%eval -204,18] [%emt 0:00:41]} Qxe2+ {[%eval -177,20] [%emt 0:00:
01]} 36. Qxe2 {[%eval -204,18] [%emt 0:00:03]} Rxe2+ {[%eval -166,19] [%emt 0:
00:00]} 37. Kd3 {[%eval -207,19] [%emt 0:00:10]} Nf6 {[%eval -173,25] [%emt 0:
00:26]} 38. Ra5 {[%eval -207,19] [%emt 0:00:36]} Re1 {[%eval -187,23] [%emt 0:
00:24]} 39. b4 {[%eval -250,18] [%emt 0:00:52] (Ra7)} Kf8 {[%eval -215,23]
[%emt 0:00:20]} 40. Ra3 {[%eval -261,19] [%emt 0:00:22] (Ra7)} Ke7 {[%eval
-224,22] [%emt 0:00:13]} 41. Kc4 {[%eval -261,20] [%emt 0:00:17] (Ra7+) adjud.}
0-1

[Event "Bram-PC, Blitz 60m"]
[Site "AMD X6 1090T 3200 Mhz, (Fritz"]
[Date "2010.12.21"]
[Round "13"]
[White "Houdini 1.5 x64"]
[Black "Deep Rybka 4 x64"]
[Result "1-0"]
[ECO "A80"]
[Annotator "0.17;0.00"]
[PlyCount "97"]
[TimeControl "3600"]

{AMD Phenom(tm) II X6 1090T Processor 3207 MHz W=22.9 plies; 8.340kN/s B=19.
2 plies; 379kN/s} 1. d4 f5 2. h3 Nf6 3. g4 fxg4 4. hxg4 Nxg4 5. Qd3 g6 6. Nh3 {
Both last book move} d5 {[%eval 0,18] [%emt 0:01:22]} 7. Ng5 {[%eval 17,22]
[%emt 0:00:55]} Bf5 {[%eval 0,18] [%emt 0:01:17]} 8. Qb3 {[%eval 13,22] [%emt
0:01:07]} Nc6 {[%eval 0,18] [%emt 0:00:27]} 9. f3 {[%eval 13,22] [%emt 0:01:12]
} Nf6 {[%eval 0,18] [%emt 0:00:48]} 10. e4 {[%eval 32,22] [%emt 0:01:37]} dxe4
{[%eval 0,17] [%emt 0:01:37]} 11. Qf7+ {[%eval 26,21] [%emt 0:01:06] (fxe4)}
Kd7 {[%eval 0,9] [%emt 0:00:00]} 12. fxe4 {[%eval 27,21] [%emt 0:01:37]} Bg4 {
[%eval 0,19] [%emt 0:02:11]} 13. Be3 {[%eval 15,20] [%emt 0:01:46] (Be2)} Kc8 {
[%eval -24,18] [%emt 0:01:42]} 14. Qc4 {[%eval 24,21] [%emt 0:00:49]} Qd7 {
[%eval -24,17] [%emt 0:03:23] (h5)} 15. Nc3 {[%eval 22,21] [%emt 0:00:45] (Nd2)
} e6 {[%eval 0,17] [%emt 0:01:29]} 16. Be2 {[%eval 23,21] [%emt 0:01:34]} Bxe2
{[%eval 0,17] [%emt 0:01:05]} 17. Qxe2 {[%eval 19,22] [%emt 0:00:47] (Kxe2)} e5
{[%eval -10,18] [%emt 0:01:12]} 18. O-O-O {[%eval 13,22] [%emt 0:01:04]} exd4 {
[%eval 0,18] [%emt 0:01:02]} 19. Qc4 {[%eval 8,21] [%emt 0:01:13] (Nb5)} h5 {
[%eval -12,17] [%emt 0:00:39] (Bd6)} 20. Rhf1 {[%eval 57,21] [%emt 0:00:45]
(Bxd4)} Qe7 {[%eval 0,17] [%emt 0:01:02]} 21. Bxd4 {[%eval 57,19] [%emt 0:00:
00]} Nxd4 {[%eval 0,18] [%emt 0:00:18]} 22. Nf7 {[%eval 59,23] [%emt 0:00:43]
(e5)} Rh7 {[%eval 44,15] [%emt 0:00:23]} 23. Rxd4 {[%eval 59,21] [%emt 0:00:00]
} Rxf7 {[%eval 81,16] [%emt 0:00:44]} 24. e5 {[%eval 87,23] [%emt 0:01:21]} b6
{[%eval 106,17] [%emt 0:02:39]} 25. Qc6 {[%eval 184,22] [%emt 0:00:56]} Kb8 {
[%eval 106,18] [%emt 0:00:08]} 26. Rfd1 {[%eval 184,21] [%emt 0:00:00]} Bh6+ {
[%eval 105,17] [%emt 0:00:06]} 27. Kb1 {[%eval 183,20] [%emt 0:00:00]} Rf8 {
[%eval 105,18] [%emt 0:00:07]} 28. a4 {[%eval 204,24] [%emt 0:00:49] (Nb5)} Be3
{[%eval 166,17] [%emt 0:00:59]} 29. R4d3 {[%eval 204,22] [%emt 0:00:00] (a5)}
Qe8 {[%eval 177,20] [%emt 0:01:06] (a6)} 30. Rd8+ {[%eval 221,21] [%emt 0:00:
02]} Qxd8 {[%eval 177,7] [%emt 0:00:00]} 31. Rxd8+ {[%eval 210,20] [%emt 0:00:
00]} Rxd8 {[%eval 177,20] [%emt 0:00:00]} 32. Qxf6 {[%eval 341,22] [%emt 0:00:
50] (exf6)} Rc8 {[%eval 262,21] [%emt 0:01:05]} 33. Nd5 {[%eval 335,21] [%emt
0:00:00]} Bc5 {[%eval 262,21] [%emt 0:00:19]} 34. Qc6 {[%eval 335,20] [%emt 0:
00:00]} a5 {[%eval 262,21] [%emt 0:00:19]} 35. e6 {[%eval 336,19] [%emt 0:00:
02]} g5 {[%eval 279,21] [%emt 0:02:30] (Bd6)} 36. e7 {[%eval 323,18] [%emt 0:
00:00]} Bxe7 {[%eval 369,23] [%emt 0:00:58]} 37. Nxe7 {[%eval 324,17] [%emt 0:
00:00]} Rd8 {[%eval 369,23] [%emt 0:00:31]} 38. Nd5 {[%eval 477,24] [%emt 0:00:
45] (Qf3)} Ra7 {[%eval 384,19] [%emt 0:00:15]} 39. Qg6 {[%eval 477,22] [%emt 0:
00:00]} Ra8 {[%eval 384,21] [%emt 0:00:24] (Rd6)} 40. Qg7 {[%eval 508,27]
[%emt 0:00:43] (Qf7)} Rxd5 {[%eval 549,24] [%emt 0:01:14] (Rc8)} 41. Qg8+ {
[%eval 514,26] [%emt 0:00:21]} Ka7 {[%eval 549,25] [%emt 0:01:04]} 42. Qxd5 {
[%eval 528,25] [%emt 0:00:11]} Rh8 {[%eval 584,24] [%emt 0:00:47]} 43. Qxg5 {
[%eval 530,28] [%emt 0:00:50]} Rh7 {[%eval 584,24] [%emt 0:00:51] (Rc8)} 44.
Qg6 {[%eval 542,32] [%emt 0:00:42]} Re7 {[%eval 621,23] [%emt 0:02:59]} 45.
Qxh5 {[%eval 565,32] [%emt 0:05:54]} Re6 {[%eval 640,24] [%emt 0:00:29] (c5)}
46. Qh7 {[%eval 612,27] [%emt 0:00:33] (Qb5)} Rc6 {[%eval 678,26] [%emt 0:01:
10] (Kb7)} 47. Qf7 {[%eval 650,30] [%emt 0:00:45] (c3)} Rc5 {[%eval 678,24]
[%emt 0:00:08]} 48. c3 {[%eval 650,28] [%emt 0:00:00]} Kb7 {[%eval 678,23]
[%emt 0:00:32]} 49. Ka2 {[%eval 650,31] [%emt 0:00:28] (Qe6)} 1-0

[Event "Bram-PC, Blitz 60m"]
[Site "AMD X6 1090T 3200 Mhz, (Fritz"]
[Date "2010.12.21"]
[Round "14"]
[White "Deep Rybka 4 x64"]
[Black "Houdini 1.5 x64"]
[Result "0-1"]
[ECO "A80"]
[Annotator "0.00;0.12"]
[PlyCount "156"]
[TimeControl "3600"]

{AMD Phenom(tm) II X6 1090T Processor 3207 MHz W=18.2 plies; 365kN/s B=22.2
plies; 7.866kN/s} 1. d4 f5 2. h3 Nf6 3. g4 fxg4 4. hxg4 Nxg4 5. Qd3 g6 6. Nh3 {
Both last book move} d5 {[%eval 12,22] [%emt 0:01:01]} 7. Ng5 {[%eval 0,19]
[%emt 0:03:29]} Bf5 {[%eval 29,22] [%emt 0:02:59]} 8. Qb3 {[%eval 0,18] [%emt
0:01:55]} Nc6 {[%eval 29,22] [%emt 0:01:40]} 9. f3 {[%eval 0,18] [%emt 0:02:11]
} Nf6 {[%eval 12,22] [%emt 0:02:05]} 10. e4 {[%eval 0,18] [%emt 0:00:30]} Bd7 {
[%eval 6,22] [%emt 0:05:11] (dxe4)} 11. c3 {[%eval 0,18] [%emt 0:06:29] (exd5)}
e6 {[%eval 13,22] [%emt 0:02:09]} 12. Bd3 {[%eval -12,17] [%emt 0:01:12]} Nh5 {
[%eval 9,21] [%emt 0:00:57]} 13. exd5 {[%eval -12,18] [%emt 0:02:13]} exd5 {
[%eval 9,19] [%emt 0:00:00]} 14. Qxd5 {[%eval -12,18] [%emt 0:00:34]} Qe7+ {
[%eval -11,21] [%emt 0:01:29]} 15. Ne4 {[%eval -19,19] [%emt 0:03:19]} Be6 {
[%eval -16,23] [%emt 0:00:49]} 16. Qg5 {[%eval -19,19] [%emt 0:00:51]} O-O-O {
[%eval -11,22] [%emt 0:00:58]} 17. Qxe7 {[%eval -16,19] [%emt 0:01:03]} Nxe7 {
[%eval -15,23] [%emt 0:01:49]} 18. Kf2 {[%eval -19,19] [%emt 0:01:32] (Na3)}
Nd5 {[%eval -11,23] [%emt 0:00:36]} 19. Bc4 {[%eval -17,18] [%emt 0:01:42]
(Nbd2)} h6 {[%eval -14,22] [%emt 0:00:40] (Re8)} 20. Na3 {[%eval -21,18] [%emt
0:03:55]} Rh7 {[%eval -8,21] [%emt 0:01:03] (Re8)} 21. Bb3 {[%eval -4,17]
[%emt 0:01:01]} Bxa3 {[%eval -12,22] [%emt 0:00:40]} 22. bxa3 {[%eval -4,17]
[%emt 0:00:16]} Bf5 {[%eval -11,23] [%emt 0:00:33]} 23. Bd2 {[%eval -5,17]
[%emt 0:00:22]} Ndf4 {[%eval -12,23] [%emt 0:01:11]} 24. Bc4 {[%eval -10,17]
[%emt 0:01:47] (a4)} Nh3+ {[%eval -9,23] [%emt 0:00:44]} 25. Kg2 {[%eval -9,18]
[%emt 0:00:23]} N5f4+ {[%eval -8,24] [%emt 0:00:54]} 26. Kh2 {[%eval -9,19]
[%emt 0:00:57] (Kg3)} g5 {[%eval -15,22] [%emt 0:00:36] (Rf8)} 27. Rae1 {
[%eval -13,19] [%emt 0:01:08]} Rf8 {[%eval -17,22] [%emt 0:00:34]} 28. Rhf1 {
[%eval -11,18] [%emt 0:00:24] (Kg3)} Rg7 {[%eval -18,23] [%emt 0:01:12]} 29.
Nc5 {[%eval -26,18] [%emt 0:01:01] (Re3)} Rg6 {[%eval -16,24] [%emt 0:01:08]
(h5)} 30. Bf7 {[%eval -23,15] [%emt 0:00:46] (Re5)} Rf6 {[%eval -29,23] [%emt
0:00:38] (Rc6)} 31. Bc4 {[%eval -22,20] [%emt 0:01:03]} b6 {[%eval -20,23]
[%emt 0:00:45]} 32. Ne4 {[%eval -18,19] [%emt 0:00:33]} Rc6 {[%eval -21,22]
[%emt 0:00:35] (Rg6)} 33. Ba6+ {[%eval -28,18] [%emt 0:02:58]} Kd8 {[%eval -19,
23] [%emt 0:00:54] (Kb8)} 34. c4 {[%eval -39,18] [%emt 0:02:23] (a4)} Rg6 {
[%eval -22,22] [%emt 0:00:32]} 35. Bb4 {[%eval -29,17] [%emt 0:01:15]} Rf7 {
[%eval -18,24] [%emt 0:00:31]} 36. Bb5 {[%eval -29,17] [%emt 0:01:09]} Bd7 {
[%eval -13,23] [%emt 0:00:23]} 37. Bxd7 {[%eval -51,16] [%emt 0:00:28]} Kxd7 {
[%eval -22,23] [%emt 0:00:44]} 38. Re3 {[%eval -56,16] [%emt 0:00:09]} h5 {
[%eval -22,23] [%emt 0:00:29]} 39. d5 {[%eval -48,16] [%emt 0:00:11] (Bd2)} g4
{[%eval -30,23] [%emt 0:00:23]} 40. Be1 {[%eval -49,15] [%emt 0:00:08]} c6 {
[%eval -31,23] [%emt 0:00:38] (Ng5)} 41. fxg4 {[%eval -69,15] [%emt 0:00:17]}
hxg4 {[%eval -31,21] [%emt 0:00:00]} 42. Bh4 {[%eval -69,16] [%emt 0:00:14]}
Rf8 {[%eval -31,23] [%emt 0:00:28]} 43. dxc6+ {[%eval -88,15] [%emt 0:00:18]
(a4)} Kxc6 {[%eval -57,23] [%emt 0:00:34]} 44. Rfe1 {[%eval -97,16] [%emt 0:00:
13] (Nc3)} Rh8 {[%eval -62,24] [%emt 0:00:31]} 45. Kg3 {[%eval -97,16] [%emt 0:
00:04]} Nh5+ {[%eval -60,25] [%emt 0:00:35]} 46. Kg2 {[%eval -139,16] [%emt 0:
00:08]} N3f4+ {[%eval -82,25] [%emt 0:00:47]} 47. Kf2 {[%eval -139,18] [%emt 0:
00:11] (Kg1)} Rg7 {[%eval -91,22] [%emt 0:00:23] (Rf8)} 48. Rd1 {[%eval -131,
15] [%emt 0:00:16] (Kf1)} Nh3+ {[%eval -240,20] [%emt 0:00:17] (Rf8)} 49. Kf1 {
[%eval -271,14] [%emt 0:00:21]} Rf8+ {[%eval -251,21] [%emt 0:00:22]} 50. Ke1 {
[%eval -278,15] [%emt 0:00:09]} Re8 {[%eval -251,19] [%emt 0:00:00]} 51. Rd6+ {
[%eval -280,15] [%emt 0:00:16]} Kc7 {[%eval -251,19] [%emt 0:00:00]} 52. Rh6 {
[%eval -279,15] [%emt 0:00:04]} g3 {[%eval -251,18] [%emt 0:00:00]} 53. Rxh5 {
[%eval -282,16] [%emt 0:00:03]} g2 {[%eval -250,17] [%emt 0:00:00]} 54. Bg3+ {
[%eval -286,17] [%emt 0:00:02]} Rxg3 {[%eval -250,16] [%emt 0:00:00]} 55. Rxg3
{[%eval -306,19] [%emt 0:00:14] (Rh7+)} Rxe4+ {[%eval -239,15] [%emt 0:00:00]}
56. Kd2 {[%eval -306,20] [%emt 0:00:12]} g1=Q {[%eval -237,14] [%emt 0:00:00]}
57. Rxg1 {[%eval -325,20] [%emt 0:00:18] (Rh7+)} Nxg1 {[%eval -250,13] [%emt 0:
00:00]} 58. Kd3 {[%eval -325,20] [%emt 0:00:26]} Rf4 {[%eval -269,24] [%emt 0:
00:20] (Rg4)} 59. Rh7+ {[%eval -335,19] [%emt 0:00:22]} Kb8 {[%eval -269,22]
[%emt 0:00:00]} 60. a4 {[%eval -338,19] [%emt 0:00:19] (Rh1)} Rf3+ {[%eval
-332,25] [%emt 0:01:06]} 61. Kc2 {[%eval -348,21] [%emt 0:00:11]} Ra3 {[%eval
-337,24] [%emt 0:00:27]} 62. Rh5 {[%eval -348,21] [%emt 0:00:08] (Kb2)} Ne2 {
[%eval -339,22] [%emt 0:00:00] (Kb7)} 63. Kb2 {[%eval -363,20] [%emt 0:00:10]}
Rxa4 {[%eval -356,25] [%emt 0:00:34]} 64. Kb3 {[%eval -363,21] [%emt 0:00:14]
(Rh8+)} Ra6 {[%eval -361,24] [%emt 0:00:24]} 65. Rh8+ {[%eval -381,19] [%emt 0:
00:08] (c5)} Kc7 {[%eval -386,25] [%emt 0:00:22]} 66. Rh7+ {[%eval -411,19]
[%emt 0:00:12]} Kc6 {[%eval -390,24] [%emt 0:00:18] (Kd6)} 67. Rh4 {[%eval
-427,20] [%emt 0:00:21] (Rf7)} Kc5 {[%eval -415,25] [%emt 0:00:24]} 68. Rh5+ {
[%eval -427,21] [%emt 0:00:06]} Kd4 {[%eval -415,23] [%emt 0:00:00]} 69. Rh1 {
[%eval -437,20] [%emt 0:00:08]} Ra5 {[%eval -434,25] [%emt 0:00:24] (Ng3)} 70.
Rh4+ {[%eval -431,22] [%emt 0:00:11] (Rd1+)} Kd3 {[%eval -447,25] [%emt 0:00:
13] (Ke3)} 71. Rh3+ {[%eval -446,22] [%emt 0:00:10]} Kd2 {[%eval -447,23]
[%emt 0:00:00]} 72. Rh2 {[%eval -469,21] [%emt 0:00:07]} Rg5 {[%eval -473,25]
[%emt 0:00:30] (Ra6)} 73. Kb4 {[%eval -446,18] [%emt 0:00:06] (c5)} a5+ {
[%eval -505,24] [%emt 0:00:16]} 74. Ka3 {[%eval -503,19] [%emt 0:00:06] (Ka4)}
Rg1 {[%eval -579,23] [%emt 0:00:14] (Rg6)} 75. Rh5 {[%eval -503,18] [%emt 0:00:
02] (Kb2)} Rb1 {[%eval -628,22] [%emt 0:00:02] (Rg3+)} 76. Rh4 {[%eval -834,23]
[%emt 0:00:11]} Kd3 {[%eval -718,27] [%emt 0:00:14]} 77. Rh3+ {[%eval -828,22]
[%emt 0:00:15]} Kxc4 {[%eval -638,24] [%emt 0:00:00] (Ke4)} 78. Rh4+ {[%eval
-828,23] [%emt 0:00:02]} Nd4 {[%eval -926,23] [%emt 0:00:10]} 0-1

Denzwell
Posts: 40
Joined: Fri Mar 05, 2010 1:22 pm

Re: Houdini 1.5 - increase in strenght at LTC

Post by Denzwell » Tue Dec 21, 2010 8:24 pm

Can you share please the games Pgn ?

Uri Blass
Posts: 8796
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Houdini 1.5 - increase in strenght at LTC

Post by Uri Blass » Tue Dec 21, 2010 8:50 pm

I think that it is better to have the same type of time control for comparison.

3+2 6+4 12+8 are the same type

60+0 is not the same type and same type is only 3x+2x for x>4 so
24+16 or 36+24 are good for comparison but not 60+0.

Robert Flesher
Posts: 1240
Joined: Tue Aug 18, 2009 1:06 am

Re: Houdini 1.5 - increase in strenght at LTC

Post by Robert Flesher » Tue Dec 21, 2010 9:22 pm

beram wrote:Increase in strenght at LTC for Houdini 1.5 against DR4

I ve done an expiriment with doubling the TC for the matches between Houdini 1.5 and Rybka 4 with same testpositions.
What I observed is that the gap in ELO doesn't diminish, as some Rybka fans keep on believing, but instead even becomes bigger; from 58% at 3min2sec to 60% at 6min4sec, to 62 % at TC 12min 8sec

So I challenge Jeroen Noomen to replay a 60 games (30 Noomenpositions) match under same TC conditions, 1hour a game each, as he earlier on did between Houdini 1.03a and Rybka 4, but this time with Houdini 1.5

Hat off for Robert Houdart !

Bram-PC, Blitz AMDx6 each 4cores, Blitz 3m+2s,
60 games Noomen testsuite+30 games privat testsuite

1 Houdini 1.5 x64 +36/=34/-20 58.89% 53.0/90
2 Deep Rybka 4 x64 +20/=34/-36 41.11% 37.0/90


Bram-PC, Blitz 6m+4s AMDx6 each 4cores
50 games privat testsuite

1 Houdini 1.5 x64 +16/=28/-6 60.00% 30.0/50
2 Deep Rybka 4 x64 +6/=28/-16 40.00% 20.0/50


Bram-PC, Blitz 12m+8s AMDx6 each 4cores
50 games privat testsuite

1 Houdini 1.5 x64 +20/=22/-8 62.00% 31.0/50
2 Deep Rybka 4 x64 +8/=22/-20 38.00% 19.0/50

Indeed!

I have been playing 30min and 60min games and Houdini is destroying Deep Rybka SSE42x64 4.0. In any event Houdini is living up to it's name and making Rybka 4.0 disappear from the top of the rating lists.

Perhaps, and only now Vas will unleash the fabled "cloud" Rybka ? The fans have surmised that it does exist. Let's see if they were right, or if "cloud" Rybka was just another bedtime story to passify the kids.
Time will tell :arrow:

Uri Blass
Posts: 8796
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Houdini 1.5 - increase in strenght at LTC

Post by Uri Blass » Wed Dec 22, 2010 4:14 am

Based on the rybka forum I read a claim that Rybka is better at very long time control(and results at 40/40 with 8 cores based on the rybka forum is +3 =6 -1 for rybka4)

I do not know if the results are correct and of course we will need to check the games for errors(I remember cases not only about houdini when one program got very small depth at long time control in part of the moves so it was clear that something is wrong)

12+8 may be better than 6+4 and 6+4 may be better than 3+2 for houdini
but the interesting question is if this continue to slower time control of the same type(24+16 or 36+24 or 60+40).

I also wait for the results in 24 hours/game from SCCT.

beram
Posts: 1187
Joined: Wed Jan 06, 2010 2:11 pm

Re: Houdini 1.5 - increase in strenght at LTC

Post by beram » Wed Dec 22, 2010 6:20 pm

Latest standings in my Houdini1.5 – Rybka4 match at LTC, 1 hour each

Houdini 1.5 leads after 27 games of 50games with: +9 =16 -2 = 63%
So after 3m2s 58%; 6m4s 60%; 12m8s 62 % I now have 63 % at 1 hour games with the same privat testsuite.

No longer there will be excuses for the Rybka fans as, “it is the contempt factor” or “Rybka is better at LTC” or “program is specially tuned for Rybka”, this Houdini 1.5 is simply better and probably is going to stay number one for a long time.

p.s to Max Anthony I will post the games after the match

User avatar
AdminX
Posts: 5515
Joined: Mon Mar 13, 2006 1:34 pm
Location: Acworth, GA
Contact:

Re: Houdini 1.5 - increase in strenght at LTC

Post by AdminX » Wed Dec 22, 2010 9:40 pm

I am currently running a short 20 game match at Tournament (Long) Time Controls. Only 6 games completed at this time. Current standing is 4 draws, 2 wins, and 0 losses in favor of Houdini.

In the sixth game Deep Rybka was holding her own untill she took a bite of a poisoned h-pawn at her 32nd move. Then things started to go downhill as four moves later Houdini started driving home the point.

Ponder was on, and each engine used 2 cores.

[d]5r1k/4b1pp/3p4/3Bp1P1/qP2Pr1P/p1PRQ3/K2R4/8 b - - 0 32


[Event "Intel Core 2 Quad Q9000, 120m/40+60m/20"]
[Site "Intel Core 2 Quad (Q9000)"]
[Date "2010.12.22"]
[Round "6"]
[White "Houdini 1.5 x64"]
[Black "Deep Rybka 4 x64"]
[Result "1-0"]
[ECO "B99"]
[Annotator "0.05;0.09"]
[PlyCount "91"]
[TimeControl "40/7200:20/3600:1800"]

{Intel(R) Core(TM)2 Quad CPU Q9000 @ 2.00GHz 1994 MHz W=22.3 plies; 1,
367kN/s; 31 TBAs; Opening Tree.ctg B=19.2 plies; 103kN/s; 218 TBAs; Opening
Tree.ctg} 1. e4 {[%eval 0,0] [%emt 0:00:00]} c5 {[%eval 0,0] [%emt 0:00:00]} 2.
Nf3 {[%eval 0,0] [%emt 0:00:00]} d6 {[%eval 0,0] [%emt 0:00:00]} 3. d4 {[%eval
0,0] [%emt 0:00:00]} cxd4 {[%eval 0,0] [%emt 0:00:00]} 4. Nxd4 {[%eval 0,0]
[%emt 0:00:00]} Nf6 {[%eval 0,0] [%emt 0:00:00]} 5. Nc3 {[%eval 0,0] [%emt 0:
00:00]} a6 {[%eval 0,0] [%emt 0:00:00]} 6. Bg5 {[%eval 0,0] [%emt 0:00:00]} e6
{[%eval 0,0] [%emt 0:00:00]} 7. f4 {[%eval 0,0] [%emt 0:00:00]} Be7 {[%eval 0,
0] [%emt 0:00:00]} 8. Qf3 {[%eval 0,0] [%emt 0:00:00]} Qc7 {[%eval 0,0] [%emt
0:00:00]} 9. O-O-O {[%eval 0,0] [%emt 0:00:00]} Nbd7 {[%eval 0,0] [%emt 0:00:
00]} 10. g4 {[%eval 0,0] [%emt 0:00:00]} b5 {[%eval 0,0] [%emt 0:00:00]} 11.
Bxf6 {[%eval 0,0] [%emt 0:00:00]} Nxf6 {[%eval 0,0] [%emt 0:00:00]} 12. g5 {
[%eval 0,0] [%emt 0:00:00]} Nd7 {[%eval 0,0] [%emt 0:00:00]} 13. f5 {[%eval 0,
0] [%emt 0:00:00] Both last book move} O-O {[%eval 9,18] [%emt 0:05:31]} 14.
fxe6 {[%eval 5,20] [%emt 0:06:00] (Rg1)} fxe6 {[%eval 10,20] [%emt 0:10:45]}
15. Qg3 {[%eval 3,24] [%emt 0:00:00] (Nxe6)} Nc5 {[%eval 0,19] [%emt 0:05:48]}
16. a3 {[%eval -1,23] [%emt 0:05:23]} Rb8 {[%eval -5,19] [%emt 0:00:02]} 17. b4
{[%eval -10,23] [%emt 0:08:05]} Na4 {[%eval -8,20] [%emt 0:00:00]} 18. Nxa4 {
[%eval -10,21] [%emt 0:00:00]} bxa4 {[%eval -28,19] [%emt 0:03:00]} 19. Bh3 {
[%eval -9,23] [%emt 0:01:09]} e5 {[%eval -28,19] [%emt 0:01:21]} 20. Ne6 {
[%eval -13,23] [%emt 0:04:23]} Bxe6 {[%eval -8,20] [%emt 0:00:00]} 21. Bxe6+ {
[%eval -13,21] [%emt 0:00:00]} Kh8 {[%eval 0,21] [%emt 0:05:37]} 22. Kb1 {
[%eval -11,23] [%emt 0:02:21] (Rhf1)} a5 {[%eval -30,19] [%emt 0:11:31]} 23. c3
{[%eval -10,25] [%emt 0:00:00]} Rf4 {[%eval -21,20] [%emt 0:04:41]} 24. Bd5 {
[%eval -8,24] [%emt 0:07:45] (Rhe1)} Rbf8 {[%eval -8,18] [%emt 0:04:04] (Bxg5)}
25. Rd3 {[%eval -8,25] [%emt 0:06:59]} axb4 {[%eval -7,20] [%emt 0:00:00]} 26.
axb4 {[%eval -8,23] [%emt 0:00:01]} Rf2 {[%eval 0,19] [%emt 0:05:18]} 27. Rhd1
{[%eval 0,25] [%emt 0:09:55]} Qd7 {[%eval 0,21] [%emt 0:00:00] (Qa7)} 28. h4 {
[%eval 24,24] [%emt 0:05:36] (g6)} a3 {[%eval 14,20] [%emt 0:06:28]} 29. R1d2 {
[%eval 30,26] [%emt 0:21:56]} Rf1+ {[%eval 18,22] [%emt 0:00:00] (R2f4)} 30.
Ka2 {[%eval 25,19] [%emt 0:00:00]} Qa4 {[%eval 24,22] [%emt 0:15:56]} 31. Qh3 {
[%eval 48,25] [%emt 0:00:00]} R1f4 {[%eval 26,19] [%emt 0:04:09]} 32. Qe3 {
[%eval 50,24] [%emt 0:06:20] (Qe6)} Rxh4 {[%eval 23,18] [%emt 0:05:54] (Qe8)}
33. Qg3 {[%eval 57,22] [%emt 0:04:07] (Qb6)} Rhf4 {[%eval 52,19] [%emt 0:04:38]
} 34. Rh2 {[%eval 58,24] [%emt 0:00:00]} h6 {[%eval 52,20] [%emt 0:02:03]} 35.
Rdd2 {[%eval 54,24] [%emt 0:09:24]} Qa7 {[%eval 52,19] [%emt 0:00:00] (Qd7)}
36. g6 {[%eval 148,22] [%emt 0:03:08]} Bd8 {[%eval 289,17] [%emt 0:00:00]} 37.
Rdg2 {[%eval 431,19] [%emt 0:01:35]} Bf6 {[%eval 363,17] [%emt 0:01:50]} 38.
Rh5 {[%eval 502,19] [%emt 0:00:00]} Rf1 {[%eval 423,19] [%emt 0:06:45]} 39. Bc4
{[%eval 588,22] [%emt 0:05:10]} Rf4 {[%eval 429,20] [%emt 0:00:29] (Bg5)} 40.
Rgh2 {[%eval 611,16] [%emt 0:00:02] (Qh3)} Bh4 {[%eval 720,17] [%emt 0:03:55]
(Rh4)} 41. R5xh4 {[%eval 813,20] [%emt 0:00:54]} Rxh4 {[%eval 720,18] [%emt 0:
01:07]} 42. Rxh4 {[%eval 783,21] [%emt 0:00:00] (Qxh4)} Qe7 {[%eval 585,19]
[%emt 0:09:03] (Qc7)} 43. Bf7 {[%eval 856,21] [%emt 0:04:31] (Kxa3)} Rxf7 {
[%eval 651,17] [%emt 0:03:05]} 44. gxf7 {[%eval 856,21] [%emt 0:00:00]} Qxf7+ {
[%eval 652,19] [%emt 0:04:38]} 45. Kxa3 {[%eval 890,22] [%emt 0:12:24]} Qf1 {
[%eval 661,20] [%emt 0:00:00]} 46. Rh2 {[%eval 897,21] [%emt 0:02:49]} 1-0
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers

User avatar
AdminX
Posts: 5515
Joined: Mon Mar 13, 2006 1:34 pm
Location: Acworth, GA
Contact:

Re: Houdini 1.5 - increase in strenght at LTC

Post by AdminX » Thu Dec 23, 2010 10:43 am

Just woke up, score is now +3 =4 -1 in favor of Houdini 1.5
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers

beram
Posts: 1187
Joined: Wed Jan 06, 2010 2:11 pm

Re: Houdini 1.5 - increase in strenght at LTC

Post by beram » Thu Dec 23, 2010 8:08 pm

AdminX wrote:I am currently running a short 20 game match at Tournament (Long) Time Controls. Only 6 games completed at this time. Current standing is 4 draws, 2 wins, and 0 losses in favor of Houdini.

In the sixth game Deep Rybka was holding her own untill she took a bite of a poisoned h-pawn at her 32nd move. Then things started to go downhill as four moves later Houdini started driving home the point.

Ponder was on, and each engine used 2 cores.

[d]5r1k/4b1pp/3p4/3Bp1P1/qP2Pr1P/p1PRQ3/K2R4/8 b - - 0 32

I had a closer look at the position as well and come to the same conclusion. With Qe8 (as expected by H15) DR4 could probably hold the position, now it suddenly cracked especially after the nice move g6. On my modest Centrino Dualcore, T8100, 2100Mhz, 32 bit 2CPU, Win XP, Fritzmark 6,29, it took Rybka4 about 13 minutes to find the strong 36 g6! H15 took 90 seconds and Stfish 1.9 about 8 minutes.

perhaps worth another diagram:
[D] 5r1k/q3b1p1/3p3p/3Bp1P1/1P2Pr2/p1P3Q1/K2R3R/8 w - - 0 36

have a nice day
grts Bram

Post Reply