The Chess Variant Pages

A Critique of David Howe’s

CV Construction Set Proposal

By John William Brown


David Howe’s paper describes a game wherein physical tokens—representing certain powers of mobility and usurpation—are carried by tangible game pieces. These power tokens totally define a piece. Gaining or losing a token changes the character of a piece. Pre-priced tokens are purchased at the beginning of a game by contestants allotted an equal amount of cash. Players must also purchase parcels of board squares on which to initially place their pieces. With these resources in hand, each player constructs an army and, in turn, deploys it on its proper turf. Beyond that, a Chess Variant is played.

I interpret David’s description to be a metaphor for a computerized game wherein token transactions are performed concurrently with mnemonic-graphics updates. (Physical counters and tokens would lack the mnemonic qualities necessary for one to quickly discern a piece’s powers.) The computer would symbolically describe each piece’s attributes, either cartographically or iconicographically, and update these descriptions as powers are won and lost. The scope and versatility of current computer-graphics techniques would allow such a game engine to find expression in myrid forms.

Sim Army

It has long been said that chess is a symbolic form of warfare. This idea is supported by the fact that chess was especially popular during the Islamic revolution. History tells us that many a decisive battle was won by ragtag Bedouins who were fed by their religious zeal and entertained by their chess sets. In fact the first great chess-variants movement occurred at this time.

This is a very compelling image. But yesterday’s zealous hoard would have little chance against a modern general. Why? Because the requirements for success in battle have clearly changed. The antiquated idea that God is on our side cannot compete with the modern paradigm of resource management. An effective general must consider his resources—men, equipment, supplies and land—as having some real value that yields to a common rate of exchange. He must then optimize the utility of his army within the confines of his limited resources. To my knowledge, no game engine has ever address to this creative aspect of warfare.

Street Values

An apparent weakness of such a game would be that the price of resources must be accurately set by the initial programmers. Theoretical prices will not do. The cost of a particular power must reflect its ‘real-world’ contribution to winning a game. Otherwise, a savvy player will soon learn that certain powers are overpriced and others are underrated. This would eventually lead to a kind of ‘Sim Army elite’—experienced players who, through collective experience, had divined the street value of various powers. These players would repeatedly win, though not through skillful play. They would win through a kind of insider trading.

One creative solution to this problem is to let the price of powers float. For each consecutive game, the cost of a power would be adjusted according to the relative number of units purchased in the last round. Eventually, a price-to-value stasis would be reached.

In fact, economically inclined players might wish to compete on a totally economic level. They would test their skills by outfitting an army within the confines of their budget. The computer would then determine the winner by playing itself as the actual contestants look on. Data from this outcome would then be used to determine power prices for the next round.

The ultimate solution would be an evolutionary program that establishes the real value of each power through repeated self-play. These values would then be "hard wired" into the program prior to distribution. But this clever fix would destroy a rather interesting feature of the game. I would prefer a game wherein all powers are initially given course values and their real values are determined through repeated play. It is interesting to note that these values would not be the same for each set of players.


I found David’s terminology to be quite clear and in a very positive way reminiscent of LISP. There are two terms, however, that I would recommend changing. They are orthag and stealth.

I interpret orthag to be an abbreviation of ‘orthogonal-angular’. I think a better term would be orthan. My reasoning here is that the CV community is just becoming familiar with the term orthogonal. In fact I have heard a number of neophytes pronounce it "or-THAG-onal," which conveniently rhymes with diagonal. This being a potential problem, I would change the term at the onset.

The term stealth should rightly be theft. Stealth implies invisibility, whereas the piece enjoying this power is not invisible but is clearly a thief. A serendipitous benefit of this change would be that the word stealth would become available for other applications. Consider, for example, a stealth piece that is invisible only on its home turf. Or a stealth piece that is invisible every other move? Moreover, consider a stealth thief !


David has truly described a "metagame" here. He has defined his primitives so skillfully that the ensuing arguments may be perceived as tautological. In the spirit of Euclid, he has created a theoretical foundation for an application that I find particularly interesting. I would be very disappointed if this idea did not find some medium of expression.

Written by John William Brown.
WWW page created: March 26, 1999.