In No castling Chess what Engine would be the King or the best?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
lkaufman
Posts: 3873
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: In No castling Chess what Engine would be the King or the best?

Post by lkaufman » Wed Dec 18, 2019 6:04 am

Metaphysician wrote:
Tue Dec 17, 2019 7:31 pm
@GM Kaufman: I am persuaded by your arguments. Still, I think the basic premise — that the large number of draws at the top level is a problem — is open to question. Many draws are satisfying, well-played games.
While I wouldn’t want a WC title to be decided by a rapid or blitz game, there is something to be said for shortening up time controls generally. The result would be more blunders and fewer draws, which would appeal to many chess fans.
Certainly many draws are well-played games, and I wouldn't have any objection to 50% draws in World Championships or high level correspondence play or engine tournaments. I can even make an argument that up to 50% draws is a good thing, because it makes each game the equivalent of a two game match of a version with no draws, since in such a version half the two game matches would be drawn 1 to 1 (assuming equal level players). But there is nothing good to say about a draw frequency much above 50%, it just indicates that the draw margin is too wide. There are many top level chess games where it is clear that one player outplayed the other, but just couldn't push it past the drawing margin. As long as the rules remain as they are today, I favor reducing time limits at the top level to whatever is needed to bring the draw percentage down to 50% (along with other measures like requiring TD approval for a draw before say move 60 and using number of wins as the first tiebreak). But of course reducing time limits won't work in correspondence play.
Komodo rules!

MikeB
Posts: 3807
Joined: Thu Mar 09, 2006 5:34 am
Location: Pen Argyl, Pennsylvania

Re: In No castling Chess what Engine would be the King or the best?

Post by MikeB » Fri Dec 20, 2019 12:46 am

Nordlandia wrote:
Sun Dec 15, 2019 5:53 pm
Perhaps St. Louis can arrange a crazyhouse event and invite the best online players.

Crazyhouse is on par 960 on lichess. Crazyhouse is not "silly"

Crazyhouse or bughouse community has always been big on the internet and OTB. I personally think some of the online variants can be introduced OTB.

Crazyhouse is my favorite variant and there are almost no draws , even between engines. I think the name Crazyhouse is misfortunate, probably Shogi-Chess would have been better. ( Shogi also allows a captued pierce to be introduced back to the board) You need to have a double chess set to play OTB ( or just the right extra pieces).

User avatar
Nordlandia
Posts: 2545
Joined: Fri Sep 25, 2015 7:38 pm
Location: Sortland, Norway

Re: In No castling Chess what Engine would be the King or the best?

Post by Nordlandia » Fri Dec 20, 2019 11:47 am

MikeB wrote:
Fri Dec 20, 2019 12:46 am
Nordlandia wrote:
Sun Dec 15, 2019 5:53 pm
Perhaps St. Louis can arrange a crazyhouse event and invite the best online players.

Crazyhouse is on par 960 on lichess. Crazyhouse is not "silly"

Crazyhouse or bughouse community has always been big on the internet and OTB. I personally think some of the online variants can be introduced OTB.

Crazyhouse is my favorite variant and there are almost no draws , even between engines. I think the name Crazyhouse is misfortunate, probably Shogi-Chess would have been better. ( Shogi also allows a captued pierce to be introduced back to the board) You need to have a double chess set to play OTB ( or just the right extra pieces).
Crazyhouse is somewhat inconvenient OTB because of the
piece trading. But yeah, Crazyhouse draw are unheard of.

User avatar
Nordlandia
Posts: 2545
Joined: Fri Sep 25, 2015 7:38 pm
Location: Sortland, Norway

Re: In No castling Chess what Engine would be the King or the best?

Post by Nordlandia » Fri Dec 20, 2019 6:36 pm

What about testing this "double" fischer random position. Since white moves first, white king squares are swapped to even out the first move advantage.


lkaufman
Posts: 3873
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: In No castling Chess what Engine would be the King or the best?

Post by lkaufman » Fri Dec 20, 2019 7:30 pm

Nordlandia wrote:
Fri Dec 20, 2019 6:36 pm
What about testing this "double" fischer random position. Since white moves first, white king squares are swapped to even out the first move advantage.

Are you assuming normal chess castling (so White can't castle here) or FRC castling rules? With normal castling, I'm pretty sure Black is slightly better, the right to castle is more important than moving first. With FRC castling, the switching of K and Q doesn't hurt White at all, he remains better due to first move. This does suggest an FRC variant that allows for the entire White position to be reversed, so doubling (I think) the number of positions to 1920. But some might be too favorable for White, I don't know.
Komodo rules!

User avatar
Nordlandia
Posts: 2545
Joined: Fri Sep 25, 2015 7:38 pm
Location: Sortland, Norway

Re: In No castling Chess what Engine would be the King or the best?

Post by Nordlandia » Fri Dec 20, 2019 8:37 pm

What about w/o castling for white here and normal castling for black (FRC style)


lkaufman
Posts: 3873
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: In No castling Chess what Engine would be the King or the best?

Post by lkaufman » Fri Dec 20, 2019 8:48 pm

Nordlandia wrote:
Fri Dec 20, 2019 8:37 pm
What about w/o castling for white here and normal castling for black (FRC style)

Well, if Black can castle either side and White can't castle at all, that should favor Black clearly. It's not obvious what the reversal of K and Q favors one side or the other if castling both sides is still legal. I suppose analysis would show it makes some difference, but castling remains more important than first move.
Komodo rules!

User avatar
Nordlandia
Posts: 2545
Joined: Fri Sep 25, 2015 7:38 pm
Location: Sortland, Norway

Re: In No castling Chess what Engine would be the King or the best?

Post by Nordlandia » Sat Dec 21, 2019 6:25 am

This is quite balanced according SF and Komodo. Position is in equilibrium state of matter. It appears that disabling short castling, somehow equalize the first move advantage. White can castle on the queenside whereas black can castle either side FRC style.


lkaufman
Posts: 3873
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: In No castling Chess what Engine would be the King or the best?

Post by lkaufman » Sat Dec 21, 2019 2:36 pm

Nordlandia wrote:
Sat Dec 21, 2019 6:25 am
This is quite balanced according SF and Komodo. Position is in equilibrium state of matter. It appears that disabling short castling, somehow equalize the first move advantage. White can castle on the queenside whereas black can castle either side FRC style.

Yes, it makes sense that just losing short castling should roughly balance the first move, it's not too difficult for White to castle long in general. But it's not clear that making the chances even is a good thing; most likely that will just mean more draws.
Komodo rules!

Post Reply