A Forensic Challenge for those with the KQPKQP EGT

[guyhaw]

ROOKIE -v- THE BARON: Round 8, Game 29 of the ICGA World Computer Chess Championship 2011, Tilburg, the Netherlands http://www.digitalgametechnology.com/Ch ... 11/tfd.htm

The game reached this KQPKQP position:

6Q1/3k4/5q2/8/8/1p4P1/8/6K1 b - - 0 65

and THE BARON correctly announced ‘mate in 30’. The arbiter, Jan Krabbenbos asked that the game be played out – a correct call, as THE BARON shortly and intermittently lost touch with its inner KQPKQP EGT. It is thought that the most likely reason is that one or more of the 12 constituent files of the KQPKQP EGT had got corrupted.

So, the question for those with the KQPKQP EGT, files ‘0-9’ and ‘a-b’, is:

Which constituent file should THE BARON have been consulting before each Black move ... from moves 65 to 89? The game continued:

65. ... Qd4+ 66. Kg2 b2 67. Qg6 Qd2+ 68. Kh3 Kc7 69. Qe4 Kd6 70. Qf5 Qh6+
71. Kg2 Qc1 72. Qf6+ Kc5 73. Qe7+ Kb5 74. Qb7+ Ka4 75. Qa6+ Kb3
76. Qb7+ Ka2 77. Qd5+ Ka1 78. Qa8+ Kb1 79. Qd5 Qc2+ 80. Kh3 Qe2
81. Qf7 Kc1 82. Qf4+ Kd1 83. Qf5 Ke1 84. Kh4 Qe3 85. Kh5 Qb3
86. Qe4+ Kf2 87. Qd4+ Kg2 88. Qd2+ Kh3 89. Qd7+ Kxg3

at which point THE BARON dialled into its KQPKQ EGT and won in DTM-optimal fashion.

g

[jshriver]

Interesting will take a look. I finally have have the full 3-4-5-6 set thanks to this project.
If I get any results will post a response.

-Josh

Edit: if this is due to corrupt egtb's this should be a note for people using them to systematically do md5sum checks against your dataset when competing.
I'm paranoid enough I have 2, 2TB drives with the dataset mirrored and use "tripwire" as a means to check for bit rot. Maybe less paranoia and more a method to make sure nothing goes awry.

[guyhaw]

Ok, thanks Josh.

What I would have done, if I''d had the EGT is to show my engine just one of the twelve component files at a time ... and see whether how it set about 'evaluating' or 'looking up' each position.

I have to close down and boot up FRITZ each time to show it a different set of EGTs - bit of a bore that - but maybe that's the way to do it.

I look forward to your findings.

g

[Les]

Hi JS,

Listen when you say you now have all the 3-4-5-6 piece egtb's does that also contain the 5-1??

Thx,

Les