Author: Albert Silver
Date: 15:12:21 10/04/04
Go up one level in this thread
>[D]rr4k1/5p2/3p2p1/3P3p/pnP1P2P/Bn2NR2/5PP1/qB1Q2K1 w - - 1 1
>
>43. e5 Qxa3
>
>[D]rr4k1/5p2/3p2p1/3PP2p/pnP4P/qn2NR2/5PP1/1B1Q2K1 w - - 0 1
>
>
>44. Rxf7 Kxf7 45. Qf3+ Kg8 46. Qf6
>Qa1 47. Qe6+ Kf8 48. Qxd6+ Ke8 49. Qxg6+ Kd8 50. d6 Kc8 51. Qf7 Qxb1+ 52. Kh2
>Na6 53. e6
>Nbc5 54. e7 Qe4 55. d7+ Nxd7 56. e8Q+ Qxe8 57. Qxe8+ Kc7 58. Nd5+ Kc6 59. Qe6+
>Kc5 60. Qxd7
>a3 61. Qf5 Kxc4 62. Ne3+ Kc3 63. Qc2+ Kb4 64. Qc4+ Ka5 65. Qc3+ Kb5 66. Qb3+ Kc6
>67. Qa4+
>Kb6 68. Qxa3 Rh8 69. Qd6+ Kb5 70. Qe5+ Nc5 71. Qb2+ Kc6 72. Qf6+ Kb5 73. Nd5
>Rhc8 74. Qb2+
>Kc6 75. Ne7+ 1-0
>
>Hello, I am curious if there are programs that can find Ferret's 43.e5! in the
>first diagram, in reasonable time? You probably have to have some sort of
>singular extensions to play 43.e5! even if then you then don't see 44.Rxf7 yet.
My current settings take a long time even on a 2 GHz box:
Analysis by Pro Deo S&D 5.9:
1.Bxb4 Nd4 2.Qe1 Qb2 3.Nc2 Nxc2 4.Bxc2 Rxb4 5.Bd1 Rxc4 6.Rg3 Re8 7.Re3
-+ (-2.19) Depth: 13 00:00:14 13936kN
1.Bxb4 Nd4 2.Qe1 Qb2 3.Bxd6 Nxf3+ 4.gxf3 Qxb1 5.Qf1
-+ (-2.52) Depth: 14 00:00:48 49393kN
1.Bxb4 Nd4 2.Qe1 Qb2 3.Bxd6 Qxb1 4.Qxb1 Rxb1+
-+ (-3.04) Depth: 15 00:06:39 399052kN
1.e5 Qxa3 2.e6 fxe6 3.Rf6 Qc1
-+ (-2.69) Depth: 15 00:10:31 641215kN
1.Nf5 Ra7 2.Rxb3 axb3 3.Bxb4 Rxb4 4.Nxd6
-+ (-2.40) Depth: 15 00:14:41 893039kN
It will probably find it the next ply, but I don't feel like waiting to see how
long.
>Rebel 12 could find e5 with its singular extensions enabled, but the searchtree
>in Rebel 12 would often grow out of control at ply>12 . Programs with singular
>extensions, like Bruce Moreland's Ferret have the advantage here, but is it
>findable with other extensions?
>
>My own current settings for Pro Deo do play 44.Rxf7! but only after 16 minutes,
>lots of nondefault extensions
>
>[Pruning = MISC_30]* extend singular checks , [Extensions (remaining)= 15] -not
>sure this last one is right as I remember these remaining Rebel extensions were
>bitsettings, not normal settings?-
I think I read that Ed had de-activated this last switch (Extensions remaining).
>and big minusscore on my Celeron 500MHz, 43.e4-e5 I can't get yet. (43.Ba3xb4
>and 43.Ne3-f5 are other choices but I'm afraid these only lead to a lost game).
>
>
>Pro Deo 1.0 at move 44, custom settings:
>00:16:44.8 -3,52 12 162748792 Rxf7 Kxf7 Qf3+ Kg8 Qe4 Qc1+ Kh2 Qxb1 Qxb1
>00:16:44 12.15 -3.52 44.Rxf7 Kxf7 45.Qf3 Kg8 46.Qe4 Qc1 47.Kh2 Qxb1
>48.Qxb1
My current settings find this at a similar depth though a bit faster, albeit on
an Athlon 2400XP+ (2 GHz):
Analysis by Pro Deo S&D 5.9:
2.e6 fxe6 3.Bxg6 Qc1 4.dxe6 Qxd1+ 5.Nxd1 Nd4 6.Re3 a3
-+ (-3.28) Depth: 11 00:00:04 2786kN
2.e6 fxe6 3.Bxg6 Qc1 4.dxe6 Qxd1+ 5.Nxd1 a3 6.Bf7+ Kh8 7.Rf5 a2 8.Rxh5+
-+ (-3.58) Depth: 12 00:00:08 7502kN
2.Nf5 gxf5 3.Rg3+ Kf8 4.Qxh5 Qc1+ 5.Kh2 Ke7 6.Qxf5 Qh6 7.h5 Nd4 8.exd6+ Qxd6
9.Qe4+ Ne6
-+ (-3.16) Depth: 12 00:00:19 18264kN
2.Rxf7
-+ (-3.16) Depth: 12 00:00:56 53111kN
2.Rxf7 Kxf7 3.Qf3+ Kg8 4.Qf6 Qa1 5.Qxg6+ Kf8 6.exd6
-+ (-2.03) Depth: 12 00:02:11 124321kN
A word of warning: the MISC_58 switch causes evaluation problems. Among some of
the quirks is a regular tendency to chuck a piece for pawns).
>Is Ferret the only one to find the winning continuation quickly ?
I'm curious to know what its evaluation was when it played e5 and Rxf7. BTW,
what was its hardware?
Anyhow, thanks for dredging up this position to add to my collection of tough
ones.
Albert
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.