The site has moved to a new server, and there are now some issues to fix. Please report anything needing fixing with a comment to the homepage.



The Chess Variant Pages




[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

EarliestEarlier Reverse Order Later
[Subject Thread] [Add Response]
Sam Trenholme wrote on 2007-07-20 UTC
Thanks for looking in to this. I think the two of us will come up with different ideas about what makes an ideal starting position, since we have different criteria about what a good starting position entails.

For example, I only think color balance for pieces matter if the pieces are colorbound. It doesn't matter to me what colors non-colorbound pieces end up on, since those pieces can change color at will.

As another point, I no longer think it's essential that each and every pawn in the opening setup is defended. I think it's a good idea for white to be unable to threaten mate on his first move, since otherwise Black can be prevented from making natural developing moves in the opening; having all pawns defended stops these kinds of threats. One of my proposals posted in this thread, 'Narcotic chess' (RQNBKMBNAR), for example, has an undefended flank pawn, but appears to be a perfectly playable variant. The original Carerra setups (RANBQKBNMR and RMNBQKBNAR) have the same undefended flank pawn 'problem', but again appear perfectly playable.

I am not sure every pawn around the king has to be defended two times or more. FIDE chess has had, for over 500 years, the King Bishop's pawn defended by only the king, and this has not stopped FIDE chess from becoming the most popular Chess variant that we will ever have. However, I can see why one may not want these weakly defended pawns in a Capa setup, since there is 18 pawns more power (2 more pawns, the archbishop, and the matshall/chancellor) on the board than in FIDE Chess.

One thing I like to see in an opening setup is a Chess-like arrangement of the minor pieces. One problem with, say, RNBQKAMBNR, is that moving the center pawns forward two squares blocks the diagonals of the bishops, and it is difficult to make the knights active players in the game. The nice thing about, say, RQNBKABNMR, is that the knights, bishops, and center pawns are naturally developed without getting in each other's way.

So, in conclusion, since I have my own ideals about the initial position of the pieces, I will come up with a different opening setup than other chess variant inventors may decide upon,

- Sam


Derek Nalls wrote on 2007-07-21 UTC

'I only think color balance for pieces matter if the pieces are colorbound. It doesn't matter to me what colors non-colorbound pieces end up on since those pieces can change color at will.'

The color-bound pieces imbalance (e.g., queen and archbishop both on dark or light spaces) is measurably a much more efficacious fault than the color-changing pieces imbalance (e.g., chancellor and archbishop both of dark or light spaces) although I classified both as 'minor faults' in the overall scheme consisting of only 4 distinct faults.

As measured by the movement capabilities for each power piece taken one-at-a-time on an otherwise empty board at its opening position [an ideal, potential maximum evaluation to reveal space-based imbalances rather than an actual, obstructed condition existing upon the very first move of the game by white], a color-bound pieces imbalance often results in a number for one color that is appr. twice as high as the other color in CRC. For example, 39 dark spaces can be occupied while only 20 light spaces can be occupied on the first move. So, it is NOT merely a trivial fault (instead of a minor fault).

A color-changing pieces imbalance typically only throws the numbers (from a total of appr. 55-59) out of balance by 2-4 more than they would otherwise be since a perfect balance never exists anyway with any CRC positions. Still, when you have a vast number of positions to choose from (12,000+ according to Reinhard Scharnagl), why tolerate this imbalance, either. Logically, I cannot tolerate the former fault. So, I will not tolerate the latter fault of a similar nature, either. Thereby, I maintain consistent standards for the model.

_______________________________________________________

'... I no longer think it's essential that each and every pawn in the opening setup is defended. I think it's a good idea for white to be unable to threaten mate on his first move, since otherwise Black can be prevented from making natural developing moves in the opening; having all pawns defended stops these kinds of threats.'

Yes but ...

An undefended pawn can, with perfect play by white (the player with the first-move-of-the-game advantage) over a number of moves irrefutably result in a stolen pawn despite perfect play by black. [{proven to us by someone}* using a powerful, multi-CPU computer with one proposed CRC opening setup.] Ultimately, this material disadvantage is probable to lead to the eventual defeat of black. This is too unfair to black. The potential amplification of the pre-existing and marginally, unacceptably-high advantage for white is the reason that an undefended pawn is unconditionally classified as a fatal fault for any CRC position under my system.

__________

'I am not sure every pawn around the king has to be defended two times or more. FIDE chess has had, for over 500 years, the King Bishop's pawn defended by only the king, and this has not stopped FIDE chess from becoming the most popular Chess variant that we will ever have.'

Yes but ...

Standard Chess is the most stable FRC position available.

[You can quickly verify this fact by creating a select FRC scheme in a likewise manner as you created a select CRC scheme.]

______________________________________________________

'However, I can see why one may not want these weakly defended pawns in a Capa setup, since there is 18 pawns more power (2 more pawns, the archbishop, and the marshall/chancellor) on the board than in FIDE Chess.'

Exactly.

My analysis of fault-free and faulty positions exists entirely within the relative context of what resources are available with a given piece set and class of games. Consequently, I demand higher standards of the CRC piece set on the 10x8 board defensively than of the FRC piece set on the 8x8 board.

Since I have only developed one tool to date, the select CRC analysis tool, I realize this fact was unexplained and undemonstrated.

_____________________________________________________________

Whatever piece set and gameboard with 100's-1000's of possible opening setups is presented, my goal is to customize a method to it by:

1. Focusing upon a highly-selective set of positions based upon your well-defined, easily-used criteria as potentially most favorable, interesting, least asymmetrical and worthwhile to investigate in detail- existing in a number manageable for one person in terms of the time and work required.

2. Applying an extremely-selective filter so that only one to a maximum of a few of the very strongest positions available survive with a rating of fault-free.

The very best possible game(s) out of the vast number within a class should exist there. This is due to the fact that the number of pawn backups (in chess variants related to standard Chess) impose mathematical limitations upon what is possible, despite the vast number of tactical, offensive multi-move options available, to effectively imbalance the game to the further advantage of white (the player with the first move of the game).

*Edited - JG


Sam Trenholme wrote on 2007-07-21 UTC
First of all, Derek, I hope the posting I made did not, in any way, show disrespect towards your opinion, research, nor criteria for choosing the opening setup you did. The length and tone of your post indicates that you may have been personally offended by my posting. I did not mean any offense towards you; I was merely pointing out that I have different ideas about what makes an ideal opening setup.

I think I need to clarify some tthings so you can understand me better. Anything in italics like this is something you wrote. So, let me make some minor clarifications:

The color-bound pieces imbalance (e.g., queen and archbishop both on dark or light spaces)

'Colorbound', for me, has a very specific meaning. I use Betza's meaning for colorbound: A piece that, for the entire game, always has to be on the same color. A bishop. for example, that starts on the white squares will always be on the white squares for the entire game, since it can not make a move going from the white squares to the black squares. Netither the queen nor archbishop are colorbound; both pieces can reach any square on a blank board in two or three moves.

As an aside, I like your using 'what squares can the power pieces go were they the only pieces on the board' as an evaluation criteria for evaluating an opening setup.

you have a vast number of positions to choose from (12,000+ according to Reinhard Scharnagl)

Actually, we only have 72 positions to choose from (see the beginning of this thread again).

Edit: It looks like the beginning of this thread got eaten, so, again: I observed that all of the various Capablanca opening setups proposed over the centuries have the following three features:

  • Symmetrical with the rooks, knights, and bishops.
  • The rooks are either in the corners or one file away from the corners
  • The king is always in a center file
There are only 72 possible opening setups that meet all three criteria.(End edit)

An undefended pawn can, with perfect play by white (the player with the first-move-of-the-game advantage) over a number of moves irrefutably result in a stolen pawn despite perfect play by black

As I recall, the evidence for that assertion was very questionable. An undefended flank pawn will not result in a proven win for white. It might make the opening a little more tactical; for example, in Narcotic Chess (RQNBKMBNAR), black might be forced to develop his marshall side knight in order to defend his archbishop pawn.

Again, please do not take my postings personally, and thank you for your insights.

- Sam


Derek Nalls wrote on 2007-07-21 UTC
'I hope the posting I made did not, in any way, show disrespect towards
your opinion, research, nor criteria for choosing the opening setup you
did.'

No.  Definitely not.  My stoic, argumentative, hardened, informational
writing style has unintentionally and unnecessarily created some
cyber-enemies for me over the years that I never wanted at all and never
personally disliked.

Actually, I am somewhat relieved that you were not personally offended and
publicly hostile over the implication within my work (select CRC analysis
tool) that Schoolbook Chess is not fault-free due to its exclusion from
the select 48 games I analyzed in detail (along with all other king &
archbishop centered positions).

It is very significant that I did not see my way clear to devising
anything like the select CRC analysis tool until you first devised a
couple of simple rules for paring-down candidate CRC positions to a
manageable number.
__________________

''Colorbound', for me, has a very specific meaning. I use Betza's
meaning for colorbound: A piece that, for the entire game, always has to
be on the same color. A bishop. for example, that starts on the white
squares will always be on the white squares for the entire game, since it
can not make a move going from the white squares to the black squares.
Neither the queen nor archbishop are colorbound; both pieces can reach any
square on a blank board in two or three moves.'

Yes, my usage of color-bound in association with the queen and archbishop
was too vague and deserves clarification.  To be sure, I agree that the
queen and archbishop are NOT color-bound pieces.  Still, they will
contribute to the 'color-bound pieces imbalance' if both of them start
the game upon the same dark or light spaces.  This is due to the fact that
both composite pieces contain, in part, color-bound bishops.  That nuance
was absent in my explanation.  It is difficult to be concise without
losing vital completeness and accuracy.
_______________________________________

It is debatable that the rating system used with the select CRC analysis
tool is too strict.  Admittedly, it was designed to detect faults of
various importance within the vast majority of CRC positions presented to
it thereby eliminating all except one to a few.

4 comments displayed

EarliestEarlier Reverse Order Later

Permalink to the exact comments currently displayed.