gothic chess fifty move rule comments

[Uri Blass]

http://www.gothicchess.com/rules.html

"For each reversible move made in cascade, increment the counter. If any irreversible move is made, reset the counter to 0."

later it is explained

"if a piece capture occurs, the move is irreversible, so the counter is reset to 0. If a players castles, the move is irreversible, so again, the counter would be reset to 0. A pawn move of any kind is irreversible. If none of these instances occur, and one player has made 50 reversible moves, then the game is declared a draw."

My comments:

1)A situation when a player lose his castling rights(not by castling) is not considered as irreversible move inspite of the fact that a side cannot get back his castling right.

2)In chess both players need to make 50 moves
I see that in gothic chess even if one player made 50 irreversible moves and another player made 49 moves the game is declared as a draw.

Am I coorect?

[Dariusz Orzechowski]

If any analogy with the FIDE rules is meant, then certain rules have been mixed up. According to FIDE Handbook, 50 consecutive moves without the movement of any pawn and without any capture are required to declare a draw. There is nothing about castling there.

On the other hand, any move resulting in losing castling rights or ep capture should be considered irreversible but should not reset the 50 move counter. They should be, however, taken into account in the 3-fold repetition.

[George Tsavdaris]

http://www.gothicchess.com/rules.html

"For each reversible move made in cascade, increment the counter. If any irreversible move is made, reset the counter to 0."

later it is explained

"if a piece capture occurs, the move is irreversible, so the counter is reset to 0. If a players castles, the move is irreversible, so again, the counter would be reset to 0. A pawn move of any kind is irreversible. If none of these instances occur, and one player has made 50 reversible moves, then the game is declared a draw."

My comments:

1)A situation when a player lose his castling rights(not by castling) is not considered as irreversible move inspite of the fact that a side cannot get back his castling right.

2)In chess both players need to make 50 moves
I see that in gothic chess even if one player made 50 irreversible moves and another player made 49 moves the game is declared as a draw.

Am I coorect?

You are correct.
Ed has mixed rules up in a wrong way and i have told him that exact thing once, but it seems he forgot it or didn't pay attention to it.

50 move rule does not have nothing to do as Dariusz said, with castling rights or en-pasant capture. Only about Pawn-moves and captures.

And also as you say in your 2) EACH player should make 50 moves without piece capture AND Pawn movement. Not just ONE player.

And also another mistake i have told him there is, in the page you gave:
If one of the (correct) criteria of 50 move rule appears, then the game is NOT draw automatically. The player having the move should ask about it.
If he doesn't want to ask it then the game can continue.

[Uri Blass]

If any analogy with the FIDE rules is meant, then certain rules have been mixed up. According to FIDE Handbook, 50 consecutive moves without the movement of any pawn and without any capture are required to declare a draw. There is nothing about castling there.

On the other hand, any move resulting in losing castling rights or ep capture should be considered irreversible but should not reset the 50 move counter. They should be, however, taken into account in the 3-fold repetition.

I know the fide chess rules(inspite of not mentioning them in the case of
irreversible moves).
I understood that the gothic chess rules are different and I am not sure if it was done on purpose.

My comments were only on things that seem to me illogical and not on things that seem to me different than normal chess.


1)It is not illogical to have a rule to reset the 50 move rule after irreversible moves so I did not complain about the decision to have a rule that is different than normal chess and decided to assume that it was done intentionally.

The target of the 50 move rule is to prevent cases when there is no progress and irreversible move mean some type of progress so it is logical to reset the counter not only after castling but also after losing castling rights and after other moves that prevent repetition like sacrifice that is not a capture that force the opponent to capture in the next move
or check that force the opponent to lose castling rights in the next move.

resetting the 50 move rule only after castling and not resetting them after different irreversible moves does not seem to me logical and this is the reason for comment 1

2)I see no logical reason to have a rule that say that the game is drawn if one of the player made 50 irreversible moves unlike normal chess when you need both players to make 50 irreversible moves and this is the reason for comment 2.

Uri

[Matthias Gemuh]

I simply use FIDE rules for Gothic chess in ChessGUI and ArcBishop.

Matthias.

[George Tsavdaris]

I simply use FIDE rules for Gothic chess in ChessGUI and ArcBishop.

Matthias.

Hi, do you know how to make Arcbishop and Chancellor to play Gothic Chess in WinboardF GUI by H.G.Muller(and Tim Mann and Alessandro Scotti)?

I mean one solution is to load ArcBishop or Chancellor to play Capablanca Chess and then paste the Gothic Chess FEN. But aside from this, is anything better?
When in the Winboard.INI line:
ArcBishop /boardWidth=10 /variant="capablanca" /firstInitString="variant capablanca\nnew\nrandom\n"
.....i set gothic instead of capablanca, it says that the variant is not supported....

[Matthias Gemuh]

My latest engine versions inform the GUI about supported variants with
feature variants="normal,fischerandom,capablanca,8x8_capablanca,gothic,bird,caparandom"
and expect "variant gothic" from GUI to play Gothic chess.

I have not tested in Winboard_F yet, but thought everything should work.
Maybe Winboard_F does not parse the "feature variants" info the way I expect.

Matthias.

[GothicChessInventor]

You are correct.
Ed has mixed rules up in a wrong way and i have told him that exact thing once, but it seems he forgot it or didn't pay attention to it.

50 move rule does not have nothing to do as Dariusz said, with castling rights or en-pasant capture. Only about Pawn-moves and captures.

And also as you say in your 2) EACH player should make 50 moves without piece capture AND Pawn movement. Not just ONE player.

And also another mistake i have told him there is, in the page you gave:
If one of the (correct) criteria of 50 move rule appears, then the game is NOT draw automatically. The player having the move should ask about it.
If he doesn't want to ask it then the game can continue.

Yes, I did make a mistake. The 50 move rule is supposed to be exactly as in chess. There is one difference:

In computer vs. computer tournaments, the 50 move rule MAY be waived if one side has properly announced mate in X where X + the previous non-pawn non-capture movecount would exceed 50 moves.

If it has happens that there is no mate in X, and the number of moves played is > X, the game cannot be won by the program making the initial claim.

Examples:

A) You make 40 moves towards the 50-move rule count, and then a program announces mate in 11. The game will be allowed to continue, even though 40 + 11 > 50 and would otherwise draw.

Suppose it turns out to be a mate in 12? The program that made the announcement is "penalized" and automatically cannot win. It can offer or accept a draw from the opponent. And, if it miscues, it can lose, of course.

B) Even if the 50-move ruler counter is at some low number, say 4, if a program announces mate in 117, it should be allowed to play this out! In one game with SMIRF from 2006, Vortex did announce mate in 141 in an ending of 6 pawns vs. knight and 3 pawns, Vortex the side with only pawns! SMIRF misplayed at the game ended much sooner, but the evaluation without tablebases showed Vortex at -2 pawns! That means the "smarter" version allowed material to be won in order to announce this long range mate. This type of play should be encouraged, and not constrained to some arbitrary movecount ruled created primarily for human players.

This addendum is so that programs don't say "mate in 300" on move 1 of the game and try and abuse the rule.

I see no reason to enforce the 50-move rule in computer contests for a few reasons:

1. In the case of one program pounding another program, it's always rewarding to see an extravagant mate being executed.

2. Gothic Chess endings can be rather long ones. Rather than arbitrarily add to the 50-move rule some constant which might not be sufficient either, don't penalize human players from having to play longer, yet don't detract from computer programs that don't get tired either. In cases where there is no mate, notice the 50-move rule is played by default.

3. Encourage programs to have fast leaf node evaluations to encourage mate solutions as the endgame approaches. That way, more interesting games can result.

4. Encourage tablebase generation on the 80-square board. This has always fascianted me, as George can tell you!

And to answer George: I didn't forget what you said, but now people will find my quote on discussion boards and might keep replicating my same initial error!

Sorry for the confusion. I should really get this changed.

[George Tsavdaris]

My latest engine versions inform the GUI about supported variants with
feature variants="normal,fischerandom,capablanca,8x8_capablanca,gothic,bird,caparandom"
and expect "variant gothic" from GUI to play Gothic chess.

I have not tested in Winboard_F yet, but thought everything should work.
Maybe Winboard_F does not parse the "feature variants" info the way I expect.

No it was just my stupidity.

I was using:
ArcBishop /boardWidth=10 /variant="gothic" /firstInitString="variant capablanca\nnew\nrandom\n"

While the correct is that it is working:
ArcBishop /boardWidth=10 /variant="gothic" /firstInitString="variant gothic\nnew\nrandom\n"