Women World Fischer Random Championship 2019 Next Month in USA

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.
Post Reply
jp
Posts: 1248
Joined: Mon Apr 23, 2018 5:54 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by jp » Mon Jan 13, 2020 1:06 am

Ovyron wrote:
Mon Jan 13, 2020 12:33 am
How a superset behaves has nothing to do with the superset being one of its sets or not. As an example I bring about this variant that also includes traditional chess as a subset:
Obviously if you like only one starting position, you'll dislike everything that has a different starting position or larger set of possible starting positions.

Your arguments amount to nothing. You should just say you don't like Chess960 on a totally emotional level and be done with it.

User avatar
Ovyron
Posts: 3803
Joined: Tue Jul 03, 2007 2:30 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by Ovyron » Mon Jan 13, 2020 1:16 am

jp wrote:
Mon Jan 13, 2020 1:06 am
Obviously if you like only one starting position, you'll dislike everything that has a different starting position or larger set of possible starting positions.
No, I already said expanded shuffled chess, with more than 3500 starting positions, is chess, and I like it. The difference is it doesn't have retarded castling.

jp
Posts: 1248
Joined: Mon Apr 23, 2018 5:54 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by jp » Mon Jan 13, 2020 3:55 am

If it doesn't have any castling, then obviously it's not current traditional chess, which does have castling.

You're of course welcome to enjoy your shuffle chess variant.

User avatar
Ovyron
Posts: 3803
Joined: Tue Jul 03, 2007 2:30 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by Ovyron » Mon Jan 13, 2020 4:18 am

jp wrote:
Mon Jan 13, 2020 3:55 am
If it doesn't have any castling, then obviously it's not current traditional chess, which does have castling.
Just like only 48 positions of Chess960, the rest made up some castling rules that aren't in traditional chess, that aren't unlike players holding a pawn in their pocket and being able to drop it as the game goes on, or any other eccentric rule that isn't part of chess.

jp
Posts: 1248
Joined: Mon Apr 23, 2018 5:54 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by jp » Mon Jan 13, 2020 4:21 am

Ovyron wrote:
Mon Jan 13, 2020 4:18 am
jp wrote:
Mon Jan 13, 2020 3:55 am
If it doesn't have any castling, then obviously it's not current traditional chess, which does have castling.
Just like only 48 positions of Chess960, the rest made up some castling rules that aren't in traditional chess
Again, the Chess960 castling rule is a generalization (but the only one possible) of the current castling rule of traditional chess. In positions where the K and R are on traditional positions, this Chess960 castling rule is exactly the same as the current traditional one. In positions where the K and R are not on traditional positions, it is an obvious generalization.

You still seem to be struggling with the notion of generalization.

Dropping pieces on the board is not a generalization of any chess rule.

User avatar
Ovyron
Posts: 3803
Joined: Tue Jul 03, 2007 2:30 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by Ovyron » Mon Jan 13, 2020 5:26 pm

jp wrote:
Mon Jan 13, 2020 4:21 am
In positions where the K and R are not on traditional positions, it is an obvious generalization.
It's not, I already brought up position 524 where K and Q swap places:



On this one when castling to the left, the King moves only 1 square, but when castling to the right, the King moves 3 squares. Why is this? Because in position 518 those are their landing squares? But 518 is just some random number.

It seems a better generalization would have been to just have the king move 2 squares towards the rook, then we can drop half the 960 positions because they're identical to their mirrored twins and their games are indistinguishable from the other ones when players don't castle.

Now, I want you to forget about who's saying this so you drop your biases and prejudices against me, and tackle this argument directly, saying "your argument is bogus" (without explaining why) or "you just don't like Chess960 and this is emotional" are just fallacious.

User avatar
hgm
Posts: 24143
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Full name: H G Muller
Contact:

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by hgm » Mon Jan 13, 2020 6:04 pm

The King could be closer than 2 squares to the Rook, so the generalization you mention would not allow castling in too many positions to be acceptable. But an alternative generalization that is just as acceptable as the Chess960 one could have the King end op on the b-file for A-side castling.

Chess without castling is just the continuation of one particular opening line of Chess, where you destroy castling rights by moving out the Knights, moving the Rooks, and then move everything back. So it is a thematic tournament, not a variant.

User avatar
Ovyron
Posts: 3803
Joined: Tue Jul 03, 2007 2:30 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by Ovyron » Mon Jan 13, 2020 7:31 pm

hgm wrote:
Mon Jan 13, 2020 6:04 pm
But an alternative generalization that is just as acceptable as the Chess960 one could have the King end op on the b-file for A-side castling.
Yes, that has my support, this is a flaw of castling that chess has had since it was implemented and I'd like to see some chess tournaments where long castling to b-file was implemented. Ironically though, I usually long castle with a bishop on d2/d7 where an enemy queen would be attacked by the rook if I moved the Bishop, so I'd miss that.

Yet another generalization would be to allow players to castle in 4 different ways: K to b file or to c file with the A-rook or K to f file or to g file with the H-rook. The advantage of this is that this would make automatically chess and chess960 a subset of this (so all Chess and Chess960 games ever played would be legal games of this where players just decided to ignore their other 2 castling options.)
hgm wrote:
Mon Jan 13, 2020 6:04 pm
Chess without castling is just the continuation of one particular opening line of Chess, where you destroy castling rights by moving out the Knights, moving the Rooks, and then move everything back. So it is a thematic tournament, not a variant.
But in lots of Chess960 starting positions it's not possible to get rid of castling. I'd not like that in an improved FRC without castling some were considered thematic and some were considered a variant just by how the starting position can get rid of castling or not.

jp
Posts: 1248
Joined: Mon Apr 23, 2018 5:54 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by jp » Tue Jan 14, 2020 3:28 am

Ovyron wrote:
Mon Jan 13, 2020 5:26 pm
jp wrote:
Mon Jan 13, 2020 4:21 am
In positions where the K and R are not on traditional positions, it is an obvious generalization.
It's not, I already brought up position 524 where K and Q swap places:
You are not understanding terminology or basic concepts, which makes it very hard to have a sensible discussion.

I have already said above multiple times that this generalization is not unique. Your statement "it's not" indicates you are confused. The only thing "it's not" is unique. It is a generalization. I have also already stated before that there are only a few simple generalizations (of the castling rule). Of course, these have also been looked at.
e.g. the generalization where the K always moves two squares in castling is called Chess480.

User avatar
Ovyron
Posts: 3803
Joined: Tue Jul 03, 2007 2:30 am

Re: Women World Fischer Random Championship 2019 Next Month in USA

Post by Ovyron » Tue Jan 14, 2020 3:35 am

jp wrote:
Tue Jan 14, 2020 3:28 am
I have already said above multiple times that this generalization is not unique.
No, what you said was:
jp wrote:
Mon Jan 13, 2020 4:21 am
Again, the Chess960 castling rule is a generalization (but the only one possible) of the current castling rule of traditional chess.
I guess you meant to say "(but NOT the only one possible)". Now everything makes sense :)

Post Reply