2009 ACCA World Computer Rapid Chess Championships

[Gian-Carlo Pascutto]

2 entrants by same author? Ahem...

[Harvey Williamson]

2 entrants by same author? Ahem...

Yes this does not seem right - all authors could make multiple entries if this is allowed. I have probably 1000+ engines written by Mark Uniacke can i enter all of those?

[swami]

2 entrants by same author? Ahem...

Yes this does not seem right - all authors could make multiple entries if this is allowed. I have probably 1000+ engines written by Mark Uniacke can i enter all of those?

http://www.taccl.org/ACCAWCRCC/2009ACCAWCRCC/WCRCC.html

Code: Select all

# An author may enter more than one program as long as they are substantially different programs as opposed to a different version of the same program. 

Participants List:

http://www.taccl.org/ACCAWCRCC/2009ACCA ... CPart.html

Engine "Telepath" by Charles Roberson
Engine "Noonian chess" by Charles Roberson

[Harvey Williamson]

Code: Select all

# An author may enter more than one program as long as they are substantially different programs as opposed to a different version of the same program. 

Who is the judge of this?

[BubbaTough]

Code: Select all

# An author may enter more than one program as long as they are substantially different programs as opposed to a different version of the same program. 

Who is the judge of this?

I think the system in previous events has been: The tournament director makes this (and all other) decisions, and delegates whatever portion of the decision he wants at whatever point he decides to. Which is as it should be in my opinion, since he is doing all the organization work. In previous events Charles himself has entered two of his own engines. This rule has been around since the very first ACCA event, and we have not had any real problems with it, so let's just let it go.

-Sam

[CRoberson]

2 entrants by same author? Ahem...

I wish to hear all thoughts on this by all, so type them up. I want to hear reasoning not just "I don't like it".

I've exchanged much email with Allard about this. The programs are planned to be run by two different operators (Swami and Kenny). I've
warned Allard that both programs could be disqualified if there is any collaborative efforts during the tournament. That is to say one of them
losing to the other via manual intervention.

The rule has allowed me to enter Telepath (my current effort) and NoonianChess (my long halted effort). The two programs are
considerably different in every way especially strength. So, they are unlikely to be paired together. I even run them on very
disparate (speed) hardware. I've noticed that we get more entrants when I enter NoonianChess. It usually finishes near the bottom if not the bottom.

The rule was put in place to allow authors that are in a similar situation with me to enter two programs. Example: TheBaron used to be much stronger than CTD.

Spark is running in Olivier's openwar event. It seems weaker than Bright, but it is early in the event.

[Gian-Carlo Pascutto]

My reasons against this:

1) The more entries you have, the better the odds of winning. So there's an incentive to enter as many times as possible.
2) If there is ever a suspicion of collusion, it will be hard to prove or disprove what happened. But the problem could have been avoided entirely (by not allowing it).
3) Allowing multiple entries per author might increase the number of participating programs (=irrelevant) but does nothing to increase the number of participating authors (=important).
4) "Substantially different" is a very vague description which can cause huge debate, is not well defined and tricky to judge.
5) It adds nothing to the tournament experience if you have to play the same guy twice.

[Harvey Williamson]

My reasons against this:

1) The more entries you have, the better the odds of winning. So there's an incentive to enter as many times as possible.
2) If there is ever a suspicion of collusion, it will be hard to prove or disprove what happened. But the problem could have been avoided entirely (by not allowing it).
3) Allowing multiple entries per author might increase the number of participating programs (=irrelevant) but does nothing to increase the number of participating authors (=important).
4) "Substantially different" is a very vague description which can cause huge debate, is not well defined and tricky to judge.
5) It adds nothing to the tournament experience if you have to play the same guy twice.

I agree with all the points from GCP.

[hgm]

I have always entered with two engines, Joker and micro-Max, with entirely different goals. I don't see why any of them should be discriminated against just because they happen to have been written by the same author. Both are currently active projects. I would also very much like to see both Telepath and Noonian enter.

[Zach Wegner]

I think the objection is valid, but it was really only brought up because Bright is a pretty strong engine. Nobody complained about Telepath/Noonian or Joker/umax in the past. The rules clearly allow it, as far as I understand Spark is a completely different engine.