Are there any members here interested in our own Man vs. Machine Match. It can be a members only thing. We pick a system / program and see if we can do as well or better than the others. When I say Man I mean "WE" vote on a move to play vs. the Machine.
Your a Great Chess Player, but I play Great Chess.
I like the idea. As for the specific settings for the match, here's a suggestion:
<UL TYPE=SQUARE><LI>2-3 days/move for us humans to decide on a move. We could set it up as a poll (although a non-poll topic might be better) and individuals could present analysis for each candidate move.
<LI>30 minutes/move for the computer. We should probably use Rybka since its strongest, or maybe we could even use the "triple brain" feature to let Fritz and Shredder help out.
<LI>You should probably host the engine, AdminX, (if that's okay with you, of course) since you have the best selection of engines.
<LI>We humans shouldn't use engines for our analysis--after all, it's man vs. machine.
<LI>A six game match would be good, but that could take over a year, so maybe we could keep multiple games going simultaneously. Also, we could move through the opening quickly (i.e. decide in advance what to play. So, if the computer is white and plays 1. d4, we could vote for King's Indian, which would take us through at least the first 4 moves.[/list]
Great Idea!
Sounds cool to me also.
__________________________________________________ ____________________________
Challenge me at Gameknot, RedHotPawn or FICS
Good question, that's why I would like for us to hammer out the rules 1st. How do you think we should handle this?Originally posted by ebutaljib:
Who makes the final decision in case of a tie between two or more candidate moves?
Man, are we going to lose
Your a Great Chess Player, but I play Great Chess.
I don't think that I should vote since I am to be the operator. However I would like to give my opinion on moves to be considered.But I should have no vote in anyway what so ever.
PS: How bout running two at same time one White the Black?
Your a Great Chess Player, but I play Great Chess.
I don't think a simple vote is actually the best way to do it. Here's one scenario: for nost of the voting period, we all think move X is best and it gets 10 votes. 6 hours before the move is made, though, someone finds the refutation, but there isn't enough time for everyone to outvote it. Another scenario: GeorgievaMaria comes in and votes 10 times for a bad move with different accounts, and 10 patzers who just signed up vote for bad moves without looking at the analysis.
My suggestion on this point is that we discuss several candidates, and anyone who wants a say in the final move needs to give evidence that their move is best (i.e. analysis or a positional evaluation). All analysis has to be considered correct unless it can be refuted, while evaluations of the position can be disputed. Obviously, if we can't come to a concensus between two or move moves, we would have to have a vote.
One more point I'd like to bring up: Should the humans be allowed to consult opening books and/or databases? My inclination is that we should get to use opening theory because the computer has its opening book (which was created by humans), and that if we aren't allowed to look up the moves, the computer shouldn't be either. However, I don't think access to databases would be fair, since the computer can't look at one.
On the note of Computer Opening Book, If we still wish to use Rybka as the engine then I will not be using any opening book with it. It plays a pretty good game without it.Originally posted by c1arinetboy:
One more point I'd like to bring up: Should the humans be allowed to consult opening books and/or databases? My inclination is that we should get to use opening theory because the computer has its opening book (which was created by humans), and that if we aren't allowed to look up the moves, the computer shouldn't be either. However, I don't think access to databases would be fair, since the computer can't look at one.
Your a Great Chess Player, but I play Great Chess.
How does two days discussion sound and on the third day we vote. Fourth day move is made and the program (Rybka without book) is given 30 minutes think time to reply. Is this okay with everyone? If Tie we do a heads or tails for Tiebreak. I will take the candidate moves give it a heads or tails tag, post the choices in a password protected Zip or Rar file for download and release the password once the vote is in.
PS: When I say vote I am talking about c1arinetboy's idea. "That we discuss several candidates, and anyone who wants a say in the final move needs to give evidence that their move is best (i.e. analysis or a positional evaluation). All analysis has to be considered correct unless it can be refuted, while evaluations of the position can be disputed."
Your a Great Chess Player, but I play Great Chess.
If all is agreed then I will try to post moves between 6-8pm New York EST. Is this Monday (12/12/2005) a good start date?
Your a Great Chess Player, but I play Great Chess.
Posting Permissions
AdministratorX
Reply With Quote