Talk:Grand chess

From Wikipedia, the free encyclopedia

	This article is within the scope of WikiProject Strategy games, an effort by several users to improve Wikipedia articles on strategy games. For more information, visit the project page, where you can join the project and/or contribute to the discussion.	SG Portal
???	This article has not yet received a rating on the assessment scale.
Strategy games To-do: edit  Create: Close Combat: The Russian Front • Close Combat: Battle of the Bulge • Close Combat: Invasion Normandy • List of units and buildings in StarCraft Expand: HeroQuest (board game) • Boku (game) • Qubic • Death Stacks • Entropy (board game) • Fitchneal • Five Field Kono Copyedit: Crossings (game) • Gonnect • Quoridor • Company of Heroes • Dark Reign • Myth (computer game) Cleanup: Abalone (board game) • Age of Mythology • Battle Realms • Outlive • Stronghold (2001 game) • Fleet Command • Combat Mission (computer game series) Cite Sources: Colonization (computer game) • MacHack (chess) Update: List of strategy video games More...

Is it OK if I put a link to something on my own web page, which was written by the world champion of Grand Chess. It is his extensive (and I mean extensive) notes on a world championship game he played of Grand Chess against the reining world champion, and is available here. I think the link is relevent to this page, but will not put the link in the encyclopedia entry myself because of the "no self-promotion" rule. Samboy 21:17, 6 Aug 2004 (UTC)

Since no one has said anything either way for over a month, I have just added the link in question. If anyone removes it because they feel it is self-promotion, I will respect their wishes. Samboy 23:42, 20 Sep 2004 (UTC)

The link doesn't work correctly in Internet Explorer. Can you please check? Andreas Kaufmann 11:33, 23 Sep 2004 (UTC)

Fixed (IE really ought to honor the content type instead of trying to guess it itself; Apache really ought to not send out content types for documents with unknown extensions--a consequence of the old UNIX tradition of "if it doesn't have a special extension, it's probably a plain text file". Samboy 21:19, 23 Sep 2004 (UTC)

[edit] Capablanca Chess and Grand Chess

I undone move of this article into Capablanca Chess. These variants are very different, there is no sense to place it there. Andreas Kaufmann 15:37, 6 Nov 2004 (UTC)

How does this compare to Gothic Chess(improved version of Capablanca chess), also a very popular variant. It has the same new pieces (with a different name). The placement of the pieces are different and Grand Chess also has the extra rank behind the group of pieces.

[edit] Spirit

Define "spirit". CRC as invented by Scharnagl is related ONLY to the 10x8 form of the games invented by Capablanca and known collectively as Capablanca Chess. Since there was a 10x10 form of Capablanca Chess, the link to that game should remain. Other games that do not have 10x10 forms do not belong here. --AceVentura

Perhaps spirit was not the best term. The fact is that they are both very similar ideas- expansion of the board with the addition of new species which are formed by combining the abilites of standard pieces. JoshuaZ 18:20, 15 April 2006 (UTC)

The fact that the other 2 games use the same pieces as Grand Chess is insufficient to justify links since all of these pieces move differently between 10x10 and 10x8 boards. I realize the other 2 games are related in origin but not closely enough in my judgment. Still, as long as I remain outnumbered by knowledgeable editors who disagree, I will leave it alone. --AceVentura

Well, right now, its only one other editor who disagrees. Does anyone else have an opinion on this? JoshuaZ 19:01, 15 April 2006 (UTC)

The link to Capablanca chess should stay. I don't have any strong feeling about two other games, they are accesible from Capablanca chess article anyway. Andreas Kaufmann 19:48, 15 April 2006 (UTC)

[edit] Piece values

I removed link to Grand Chess | Material Values of Pieces. The values given there seems to be wrong: 20 pawns / queen, 5 pawns /knight, 7 pawns /bishop - obviously wrong. Or I misunderstand something? Andreas Kaufmann 19:27, 15 April 2006 (UTC)

The piece valuations method by Nalls is reputable and fully explained in conceptual and mathematical detail within a 29-page paper. Read it. Scharnagl has endorsed and published Nalls' values even as they differ slightly from those generated using his own acclaimed method.

There are 3 obvious reasons to me that pawns in this game carry a lower relative value than you would expect in chess:

1. The pawns are much more limited in potential, probable material gain thru promotion under the rules of Grand Chess. Read them.

2. The 10 x 10 board makes pawns notably weaker compared to all pieces of unlimited range (versus the 8 x 8 board).

3. The 10 x 10 board makes pawn promotion more difficult. Although the distance covered and number of moves required to achieve promotion is about the same, the additional maneuvering options for pieces of unlimited range to use positionally to pick-off pawns getting close to promotion makes success (for pawns) less likely.

Unless you can discredit the calculation method by Nalls (and probably, Scharnagl as well), you should not destroy this useful, tactical reference. It will be restored. Sometimes, what is wrong is not obvious.--AceVentura

I don't have time to check all calculations, but there should be a mistake there. 20 pawns / queen is radicuoulus. Add only 10 (not 20!) pawns to white and remove a white's queen. White wins without any troubles. Are piece values provided for grand chess were confirmed by somebody else? Link would be nice. Thanks! Andreas Kaufmann 20:41, 15 April 2006 (UTC)

This seems to be related to the general problem of how to estimate piece values. A good example of the problems associated with this sort of thing- on a very large board, n rooks will generally beat n-k queens, even when the point value is fine (in fact, this starts to occur for pretty small board sizes). In fact, one can get a similar thing to occur even if it is rooks v. amazons or rooks v. krakens. Point values are incredibly hard to estimate in a useful fashion in chess variants. I'm inclined to agree with Andreas in this case. It seems to be an example of where somewhat took a highly formulaic approach to calculate relative value. JoshuaZ 20:51, 15 April 2006 (UTC)

I checked Nall's results and there are also flaws in values for other variants. For example, in chess the rook is 5.7 pawns and not 5 as generally excepted. The rook is also totally overvalued for Capablanca random chess, where it is more valuebale then Archbishop! Any more or less experienced 8x10 chess player will confirm that this is wrong, see for example piece values in Gothic Chess. Andreas Kaufmann 21:05, 15 April 2006 (UTC)

I actually agree with most of the recent remarks by Andreas Kaufmann andJoshuaZ. You both think deeply like very good players who understand that the levels of depth and irony within the tactics and strategy of chess variants go to such extremes that every rule has an exception and every exception has an exception.

The point being that relative piece values should not be taken literally and to the extreme. They should only be used as tactical guidelines with limitations as an important consideration toward deciding in most (but not all) cases which exchanges to force or void. Furthermore, they should be used with caution, mainly during the opening game and midgame. Wherever game-winning objectives are attainable thru material sacrifices (esp. during the endgame), positional play must be used primarily (for survival or victory) with exchange values used only secondarily.

For example, if in chess, black removes the queen (valued at appr. 9 points) and white removes all 8 pawns (valued at 8 points- 1 point x 8), you might predict on paper that white has a 1 point advantage toward winning the game. Notwithstanding, playing without half of its pieces, white is going to lose! Does that mean the well-established, relative piece values for chess are also wrong? No. This is not a case of contradiction. Instead, it is a case of nuance. An all-or-nothing expectation from any piece valuation method is unrealistic esp. when examples are used that push game conditions to the extremes. Besides, how often can one realistically expect an exchange 10-20 pieces deep to occur via forced lines of play? If you are a player involved in such a dangerous escalation, then you had better consider game-winning conditions as well as exchange values.

To the credit of both Nalls and Scharnagl, they adequately explain within their published works that game-specific, positional values are as important as material values. Moreover, Scharnagl actually uses his own piece valuations method within his SMIRF program with obvious success since it plays numerous chess variants strongly and quickly. So, I have no good reason to assume their works are without any substance or merit. --AceVentura

You should not take the universally known piece valuations- rounded-off to the nearest integer!- for chess too seriously. I have seen about a dozen rational systems for determining piece values in chess and their range of values is such that Nalls' results are not out-of-range. --AceVentura