Move 1 poll: Everyone vs. Stockfish

[MooMoo2]

Here's your chance to select an opening move! Discussion can be found at: http://www.mersenneforum.org/showthread.php?t=20574

[gd_barnes]

I'm not sure where the discussion for the actual move should be. Note here: When was it decided (and by whom) that we would all play Stockfish and that we would be white? I looked in the other thread and there had not been a consensus. MooMoo it seems like you just decided to do this on your own. Perhaps that's why there are only 3 votes so far.

[Brian-E]

Originally Posted by gd_barnes

I'm not sure where the discussion for the actual move should be. Note here: When was it decided (and by whom) that we would all play Stockfish and that we would be white? I looked in the other thread and there had not been a consensus. MooMoo it seems like you just decided to do this on your own. Perhaps that's why there are only 3 votes so far.

There are times when it's best to just get on with things, especially if the potential participants are bickering about whether we've got any chance of beating a computer instead of discussing the practicalities. (I'm guilty there.) You have to get on with it before people start losing interest.

Thanks for organising this, MooMoo2.

[LaurV]

It says on the other thread in different places that:

1. We play a game as white, with about the same rules like the games we played with human teams, which include:
2. No engine should be used by humans.
3. A week of thinking per move for humans
4. Unclear (24 hours? one week?) of thinking for computer. We understand that MooMoo (or who?) has a life, and may not post some moves in 24 hours, and won't be upset of it, if some moves take days, but we would be upset if a move is not posted in few days, a week, etc. This is my addition/interpretation and is not agreed together, but it is common sense.
5. We play a second game (in the same time? after the first game?) as black, within the same conditions, except that we are allowed to use any chess engine except Stockfish. This because MooMoo believes that Stockfish is so strong it can't lose a game when it plays as white.

So, there are 2 games. This is the one we play white.

We took this challenge more for the last point, in spite of the fact that we are not owning a very good chess engine, we will use a weaker one in the second game to avoid blunders, and use internet opening databases, we want to beat MooMoo, but if we can not, then we will be satisfied to beat his chess program... . We are (wrongly?) convinced that a man who can think for a week in front of a chess board, for each move, can beat (or at least draw against) any computer/program, in spite of the fact that we hold computers and artificial intelligence at a very high esteem (we were a programmer for almost half a century and we believed for most of this period that computers can become more intelligent than humans, in spite of the axiom that "a program can only be as intelligent as the idiot who made it").

We use royal we.

Edit: crosspost, I was replying to Gary.

Edit2: actually here the strategy is completely wrong, we should let the computer start the other game first, so we can mimic the moves in this game and make at least one point per total. Let the freaking fish stock against himself...

[gd_barnes]

Originally Posted by Brian-E

There are times when it's best to just get on with things, especially if the potential participants are bickering about whether we've got any chance of beating a computer instead of discussing the practicalities. (I'm guilty there.) You have to get on with it before people start losing interest.

Thanks for organising this, MooMoo2.

I see. Point taken. There was such a buildup to the previous 3 games but this one started quickly so it surprised me. I guess since we're all playing against it, it doesn't matter. BTW, I will have time on-and-off to contribute as I am out of town a lot. I went ahead and voted already.

[gd_barnes]

Originally Posted by LaurV

Edit2: actually here the strategy is completely wrong, we should let the computer start the other game first, so we can mimic the moves in this game and make at least one point per total. Let the freaking fish stock against himself...

lol. I seem to remember a movie where someone did that by playing 2 games simultaneously against 2 grand masters and ended up winning one of them. Of course it would be kind of pointless and boring for us to do that.

Someone correct me if I'm wrong since much of this did not appear set in stone: We are not allowed to use computer analysis as white but we are allowed to use it as black? Now that would be interesting.

[paulunderwood]

When I played several games against GNU Chess, I noticed that it always played Nc6 as Black's first move. The only time I beat it was when I played f4 as the first move

[Batalov]

Originally Posted by Vasil

1.e4 e5 2.f4 exf4. We now know the exact outcome of this position, assuming perfect play, of course.
I know your next question, so I am going to preempt it: there is only one move that draws for White, and that is, somewhat surprisingly, 3.Be2.
Any other move loses by force.

Old but gold. ;-)
Take 3000 cores, press a few keys - and the position is solved (with 99.9999999% accuracy as we know)

[gd_barnes]

I've seen that. It is very interesting. No king's gambit please. :-) Serge, I see that you chose "something else" like Paul. Unlike him you did not say what your other move was. Can you say what it is? It appears that he is choosing 1. f4.

For everyone's reference: Here is a good opening database:

http://www.365chess.com/opening.php

[MooMoo2]

Originally Posted by LaurV

It says on the other thread in different places that:

1. We play a game as white, with about the same rules like the games we played with human teams, which include:
2. No engine should be used by humans.
3. A week of thinking per move for humans

That's all correct.

4. Unclear (24 hours? one week?) of thinking for computer. We understand that MooMoo (or who?) has a life, and may not post some moves in 24 hours, and won't be upset of it, if some moves take days, but we would be upset if a move is not posted in few days, a week, etc. This is my addition/interpretation and is not agreed together, but it is common sense.

Computer thinking time will vary from less than a minute for obvious moves to a few hours for positions where many different moves are almost equally good (or bad). Computer analysis time will never exceed 24 hours for a specific position, and in almost all cases, I'll post Stockfish's response within 24 hours after the polls close.

5. We play a second game (in the same time? after the first game?) as black, within the same conditions, except that we are allowed to use any chess engine except Stockfish. ...

So, there are 2 games. This is the one we play white.

That's right. I was planning to start the next game once this game ends, but I don't have a problem starting it now if many people want to.

MooMoo believes that Stockfish is so strong it can't lose a game when it plays as white. ...
We took this challenge more for the last point, in spite of the fact that we are not owning a very good chess engine, we will use a weaker one in the second game to avoid blunders, and use internet opening databases, we want to beat MooMoo, but if we can not, then we will be satisfied to beat his chess program... . We are (wrongly?) convinced that a man who can think for a week in front of a chess board, for each move, can beat (or at least draw against) any computer/program, in spite of the fact that we hold computers and artificial intelligence at a very high esteem (we were a programmer for almost half a century and we believed for most of this period that computers can become more intelligent than humans, in spite of the axiom that "a program can only be as intelligent as the idiot who made it").

I have absolute confidence that Stockfish will not lose a game when it plays as White, regardless of what engine you use or what computer you run it on. An increasing number of correspondence chess games are draws, and the odds of black winning is pretty much zero: http://serverchess.blogspot.com/2015...m-is-long.html

[gd_barnes]

Originally Posted by MooMoo2

That's all correct.


Computer thinking time will vary from less than a minute for obvious moves to a few hours for positions where many different moves are almost equally good (or bad). Computer analysis time will never exceed 24 hours for a specific position, and in almost all cases, I'll post Stockfish's response within 24 hours after the polls close.


That's right. I was planning to start the next game once this game ends, but I don't have a problem starting it now if many people want to.


I have absolute confidence that Stockfish will not lose a game when it plays as White, regardless of what engine you use or what computer you run it on. An increasing number of correspondence chess games are draws, and the odds of black winning is pretty much zero: http://serverchess.blogspot.com/2015...m-is-long.html

I take that as a huge challenge! Even playing as white, I think we all need to ban together with our best chess engines, run them on our fastest computers, and use some human analysis to differenciate between very equal moves to see if we can draw or beat this thing. We would at least have a small chance as white.