Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fun with statistical fluctuation

Author: Dann Corbit

Date: 15:48:38 01/01/02

Go up one level in this thread


On January 01, 2002 at 17:57:00, Gian-Carlo Pascutto wrote:

>After changing a few things in the way my engine communicates
>with WinBoard, I played a test game to see if I had not broken
>Fritz's WinBoard adapter support.
>
>To my surprise, the following happened:
>
>[Event "Test game"]
>[Site "GCP"]
>[Date "2002.01.01"]
>[Round "1"]
>[White "Sjeng 12.7"]
>[Black "Fritz 7"]
>[Result "1-0"]
>[ECO "C07"]
>[PlyCount "73"]
>
>1. e4 e6 2. d4 d5 3. Nd2 c5 4. Ngf3 Nc6 5. Bb5 dxe4 6. Nxe4 Bd7
>7. d5 exd5 8. Qxd5 Qa5+ 9. Nc3 Nf6 10. Qg5 Ne4 11. Qe3 O-O-O
>12. Qxe4 Re8 13. Qxe8+ Bxe8 14. O-O Bd6 15. Rd1 Nd4 16. Nxd4 cxd4
>17. Rxd4 Bxh2+ 18. Kxh2 Bxb5 19. Bf4 Bc6 20. Rad1 Qf5 21. Kg1 Re8
>22. f3 g5 23. Bg3 Qxc2 24. R4d2 Qf5 25. b4 b6 26. Rd6 Kb7
>27. a4 Re3 28. Nb5 Bxb5 29. axb5 Re7 30. R6d5 Qf6 31. Rd6 Qc3
>32. Rd8 Qe3+ 33. Kh2 a5 34. Rb8+ Ka7 35. Rdd8 Qf4 36. Bxf4 gxf4
>37. Ra8+ 1-0
>
>Great way to start the new year, don't you think :)

This does not look like any sort of accident.  A slow and steady buildup of
strength until the finishing blow.  By here, for sure Sjeng has already won:
[D]3R4/pk2rp1p/1p6/1P4p1/1P6/2q2PB1/6P1/3R2K1 b - -

But the amazing move is this one:
[D]1R6/k3rp1p/1p6/pP4p1/1P6/4qPB1/6PK/3R4 w - - bm Rdd8; am bxa5; c0 "WOW!";

Absolutely incredible that Sjeng found Rdd8, since the vastly inferior bxa5
looks so good to programs.  Now, bxa5 is a good enough move to probably win
eventually.  But Rdd8 is a positive death blow!

Which programs (after a good, long, think) choose Rdd8 over bxa5?

You have really done something wonderful here.



This page took 0.01 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.