WCCC Schedule + Book Rules + programmers representative

[Rémi Coulom]

Hi,

I have just put the list of Hotels and schedule on the web site of the tournament:
http://www.grappa.univ-lille3.fr/icga/event.php?id=43

The tournament is still as long as before. IIRC, at the time of the 8-core rule discussion, it was a rather unanymous complaint that the tournament is too long. If programmers still agree with that, maybe we could get them to shorten it. I would remove the WCSC, and shorten the WCCC to 9 rounds. Please tell me if you have suggestions.

Another point is the book rule. In the past, that rule was applied:
http://www.grappa.univ-lille3.fr/icga/n ... .php?id=35
It is a bit in contradiction with the originality rule. If you have an opinion about whether shared books should be allowed, then please also tell me.

This year, there will be a new election for ICGA programmers representative. I won't be a candidate. I hope another programmer will have the motivation to get involved with the association.

Thanks for your feedback.

Rémi

[marcelk]

The tournament is still as long as before. IIRC, at the time of the 8-core rule discussion, it was a rather unanymous complaint that the tournament is too long. If programmers still agree with that, maybe we could get them to shorten it. I would remove the WCSC, and shorten the WCCC to 9 rounds. Please tell me if you have suggestions.

Another point is the book rule. In the past, that rule was applied:
http://www.grappa.univ-lille3.fr/icga/n ... .php?id=35
It is a bit in contradiction with the originality rule. If you have an opinion about whether shared books should be allowed, then please also tell me.

I don't see a reason why an 11-round computer chess tournament can't be completed in 4 days, for example by using the 90 minutes time control of the DOCCC to schedule 3 games/day with ease. Eight playing days is over the top for a yearly event.

For example, scheduling it on 2 weekend days and 2 working days means a very low threshold for the amateur engine authors who are living just a few hours travel from Tilburg. Taking a full week off means the tournament competes with holidays or means a lot of unpaid time off for them.

Even with the planned time control some days have two rounds: why not then hold two rounds every day so it fits in 6 days, still a considerable difference.

This might not be an issue for commercial entries or academia, but I know for me it is a huge obstacle, even for this year where the travel time is absent. (It would also not be an obstacle if the tournament were held on a nice tropical island of course, but also then the competition with holidays is gone)

Regarding book: It is an integral part of the chess program, therefore the author is part of the team. The same book should not be used by multiple entries. Opening preparation is too important for the game outcome and it is an exciting part of the competition.

[Harvey Williamson]

Regarding book: It is an integral part of the chess program, therefore the author is part of the team. The same book should not be used by multiple entries. Opening preparation is too important for the game outcome and it is an exciting part of the competition.

I completely agree with this.

[hgm]

Main source of trouble is of course that they have scheduled the WCCC and the WCSC in parallel, so that the time available in each day has to be divided over the two tournaments, stretching out both. In the first 3 days there are already 8 rounds played: 2 on Saturday and 3on Sunday and Monday. But they are 4 for the WCCC and 4 for the WCSC.

This, of course, is pretty stupid. The ploy is no doubt to force more people to stay around for the entire duration of the event, but in practice it will of course only make more people stay at home. If the WCCC and WCSC would have been after each other, with the speed Chess in between, people for which an 8-days event is a problem can then just register for one of the two (+ speed chess, if they want).

I noticed that the usual day off for excursion directly after the conference(i.e. on Wednesday) has also been dropped.

[Harvey Williamson]

Regarding book: It is an integral part of the chess program, therefore the author is part of the team. The same book should not be used by multiple entries. Opening preparation is too important for the game outcome and it is an exciting part of the competition.

I completely agree with this.

I will add to this as well as the same book not being used by more than 1 team. The same book author can not provide a book for 2 teams.

[michiguel]

Hi,

I have just put the list of Hotels and schedule on the web site of the tournament:
http://www.grappa.univ-lille3.fr/icga/event.php?id=43

The tournament is still as long as before. IIRC, at the time of the 8-core rule discussion, it was a rather unanymous complaint that the tournament is too long. If programmers still agree with that, maybe we could get them to shorten it. I would remove the WCSC, and shorten the WCCC to 9 rounds. Please tell me if you have suggestions.

Another point is the book rule. In the past, that rule was applied:
http://www.grappa.univ-lille3.fr/icga/n ... .php?id=35
It is a bit in contradiction with the originality rule. If you have an opinion about whether shared books should be allowed, then please also tell me.

This year, there will be a new election for ICGA programmers representative. I won't be a candidate. I hope another programmer will have the motivation to get involved with the association.

Thanks for your feedback.

Rémi

This is misleading:

"An endgame database contains the (presumed) absolute truth about all the positions in that database, and these truths have no creative input from the person(s) who compiled the database."

If tablebases are allowed for everybody even if they are from the same author, it is fine, but do not use a wrong justification. The key is not how the tables are compiled with "no creative input", but the probing code, which include elaborated algorithms with caches and compression. Speed is everything.

The contradiction is that using external TBs involves using code (a lot of code). Just say that TBs are allowed because the competitors agreed.

Miguel

[hgm]

It is actually extremely dubious that people are allowed to use tablebase formats they did not design themselves, and probing code they did not write themselves. That you would allow this in a 'basement tournament' werea lot of engines run on a single computer, rather than having to install umptydifferent tablebase sets is obvious, but in an OTB tourney where each engine plays on its own computer, there is no justification.

It is OK if there is a commonly agreed API for tablebase services, makingit easy to make any engine use any tablebase format, but they would really have to use their own DLL to implement this API with their own tablebase format, if we want the WCCC to be taken seriously.

[bob]

Hi,

I have just put the list of Hotels and schedule on the web site of the tournament:
http://www.grappa.univ-lille3.fr/icga/event.php?id=43

The tournament is still as long as before. IIRC, at the time of the 8-core rule discussion, it was a rather unanymous complaint that the tournament is too long. If programmers still agree with that, maybe we could get them to shorten it. I would remove the WCSC, and shorten the WCCC to 9 rounds. Please tell me if you have suggestions.

Another point is the book rule. In the past, that rule was applied:
http://www.grappa.univ-lille3.fr/icga/n ... .php?id=35
It is a bit in contradiction with the originality rule. If you have an opinion about whether shared books should be allowed, then please also tell me.

This year, there will be a new election for ICGA programmers representative. I won't be a candidate. I hope another programmer will have the motivation to get involved with the association.

Thanks for your feedback.

Rémi

I like the book rule because I have seen games won before leaving the book. That takes a lot of preparation. Should multiple programmers have to face the book by a single commercial author? Most (except for those wanting to share a commercial book) would say no, and have in the past said no.

I'd certainly like to see 9 rounds. 2 a day, no break, and end it in 5 days. Or even 3 rounds on one day and end it in 4.

[Rémi Coulom]

I like the book rule because I have seen games won before leaving the book. That takes a lot of preparation. Should multiple programmers have to face the book by a single commercial author? Most (except for those wanting to share a commercial book) would say no, and have in the past said no.

Hi Bob,

Thanks for replying. I am afraid I don't understand what you mean.

What is the rule you like? The current ICGA rule or the rule that each program must use an original book?

Also, "multiple programmers have to face the book by a single commercial author": isn't it the case whatever the rule? Did you mean "a programmer has to face the same book multiple times"?

Rémi

[CRoberson]

Hi Remi,

On the book, each program should have a unique book from all the others and each of the books should have a different author. No book author should be on more than one team ( a program could have a team of book authors).

No individual program should have to face the same book more than once in a normal swiss or round robin tournament where you play an opponent only once.

Charles