Suggestions for new CCT rules

[garybelton]

I believe the analogy between humans and computers as to the role of the opening book is also false

I certainly agree with you on this, the computer can "remember" the line into a far greater percentage of the total moves of a game than a GM can.

[garybelton]

That said, a rule which requires a unique book per entrant should go a long ways towards preventing these kinds of results in the future

Yes I made this suggestion during the CCT weekend and am glad it got accepted.

[Harvey Williamson]

That said, a rule which requires a unique book per entrant should go a long ways towards preventing these kinds of results in the future

Yes I made this suggestion during the CCT weekend and am glad it got accepted.

Yes that is a good rule.

[garybelton]

Also Nelson Hernandez makes a good point on the Rybka forum about book limiting, some openings need more moves in books than others. Do you know how the CB "Import to relative ECO length" works? I think something like that is what is needed.

[Harvey Williamson]

(In fact, would it be possible for Vas to slam down the copyright hammer and declare this tournament null-and-void, because folks were using lines out of the copyrighted Rybka book without authorization?)



Ian

Lines from the copyrighted Rybka book were not being used. The top book cookers on Playchess do not use Jeroen's book anymore. Although it is a great book it is now a bit out of date. If someone was using his book they should be kicked from the tournament as it is his copyright.

I was chatting with Jeroen all the way through the tournament he had no issue with the lines that were being played.

The lines are from games played in public just as GM games are.

[mathmoi]

In the case of standard UCI + GUI, the engine algorithms are not even running during the "in book" section.

[...]

This is my first post to this forum and my opinion is approximately worthless, just sharing the perspective of a long-time chess and computer hobbyist.

Hi Ben, welcome to CCC,

Note that not every engine let the GUI choose their book move for themselves. For example MatMoi, my engine, generate it's own book and during game pick the move he'll played himself. I worked hard developing the heuristics that create the book and select moves from the books during the game the same way I developed the main search.

[Harvey Williamson]

In the case of standard UCI + GUI, the engine algorithms are not even running during the "in book" section.

[...]

This is my first post to this forum and my opinion is approximately worthless, just sharing the perspective of a long-time chess and computer hobbyist.

I Ben, welcome to CCC,

Note that not every engine let the GUI choose their book move for themselves. For example MatMoi, my engine, generate it's own book and during game pick the move he'll played himself. I worked hard developing the heuristics that create the book and select moves from the books during the game the same way I developed the main search.

Hiarcs in CCT did not use a GUI book. It used a book controlled by Hiarcs and learning code written by Mark.

[Highendman]

Also Nelson Hernandez makes a good point on the Rybka forum about book limiting, some openings need more moves in books than others. Do you know how the CB "Import to relative ECO length" works? I think something like that is what is needed.

Nelson made another smart observation on the Rybka forum - that the Hiarcs team outfoxed the Sjeng team, not allowing them to use the h/w advantage in that game.

It would have been sweeter had that line been improved by the Hiarcs team and lured Sjeng into defeat - instead what has likely happened that both Sjeng and Hiarcs books had that game imported from the same source into their book without someone explicitly analysing that game and opting for that line.

One more reason why I don't believe in the quantity vs. quality or game-collecting in order to mass-import into a book. (there are other good uses for game collecting that are not related to mass-importing them into books).

Nonetheless - good play by the Hiarcs team mitigating the Sjeng Cluster advantage through the book.

[garybelton]

Nelson made another smart observation on the Rybka forum - that the Hiarcs team outfoxed the Sjeng team, not allowing them to use the h/w advantage in that game.

My comment in the chat this weekend was "If I had the Sjeng hardware I would only book 1.b4 for white and 1 ... g6 lines for black." I don't think their past world champion book maker Erdo was too impressed with my suggestion

[Harvey Williamson]

Also Nelson Hernandez makes a good point on the Rybka forum about book limiting, some openings need more moves in books than others. Do you know how the CB "Import to relative ECO length" works? I think something like that is what is needed.

Nelson made another smart observation on the Rybka forum - that the Hiarcs team outfoxed the Sjeng team, not allowing them to use the h/w advantage in that game.

It would have been sweeter had that line been improved by the Hiarcs team and lured Sjeng into defeat - instead what has likely happened that both Sjeng and Hiarcs books had that game imported from the same source into their book without someone explicitly analysing that game and opting for that line.

One more reason why I don't believe in the quantity vs. quality or game-collecting in order to mass-import into a book. (there are other good uses for game collecting that are not related to mass-importing them into books).

Nonetheless - good play by the Hiarcs team mitigating the Sjeng Cluster advantage through the book.

Almost certainly the case that we both imported the same game. I was chatting with Suj earlier. Like us they were unable to use their normal tournament book. Unfortunately we both keep our books for official tournaments in ctg format. Something we cant reliably use on FICS.