Single Comment

Reading back through the dicusion about this variant prompts two remarks:

1) I was a bit surprised by Ralph's Betza remark that this is 'serious, heavy-duty Chess'. Not that I doubt this is true, but I think it holds for mny other variants as well. The armies do not have to be different to make all the strategic issues of FIDE Chess carry over to it (and indeed, FIDE Chess uses 'identical armies'. Many other variants have the 'feel' of FIDE Chess. This is mainly caused by having the same Pawn, which is known to be the 'soul' of the game since Philidor.

So games like Berolina Chess, or Heian Shogi definitely have a very different feel then Chess. But Knightmate already looks very familiar to a Chess player (despite the funny 'King' moves). Capablanca Chess also strikes me as very similar. I guess it is mainly a matter of how seriously you take the game.

2) There was a very interesting remark over how long it took the FIDE army to subdue an inferior army. This might be an artifact of the evaluation function of the program handling the FIDE pieces. The effective value of strong pieces is depressed by the presence of weaker pieces of the opponent. This effect (for which Reinhard Scharnagl coined the term 'Elephantiasis correction') is tantamount to having evaluation terms that are proportional to the product of the number of (selected) white and black pieces.

Example: in normal Chess, a Queen is worth more than Rook + Bishop. But if each side would have 3 Queens, trading your first Queen for Rook + Bishop is actually a good trade! This because the removal of 33% of the light pieces of the opponent increases the effective value of your remaining 2 Queens by more than the intrinsic deficit of the trade. This is not a positional effect: it does not matter where the pieces are on the board. It is purely dependent on the material composition.

So even if the pieces of the opponent are inferior to yours in a one-on-one comparison, it could be wrong to avoid trading them. Because trading the first few creates the freedom for your remaining superior pieces to affirm their superiority, which they would not have when they have to run for their life to avoid 'bad trades'. 

A numeric example: Suppose you have 4 pieces worth 600 (centiPawn), and the opponent has 4 pieces worth 525. But suppose that the presence of one such opponent piece would depress the effective value of each of your pieces by 25, because it can interdict you access to the part of the board covered by the lighter opponent piece. With 4 against 4, this means each of your pieces is depressed by 100, down to 500, so in fact you are 4x25=100 behind, in stead of 300 ahead!

Not accepting this, and handling your pieces like they were equal to their inferior opponents, will allow us to unleash their unrestrained power of 600 against the opponent Pawns or King, so that to avoid losing Pawns or being checkmated, the opponent will be forced at some point to trade. (The opponent can certainly not avoid trading; this would depress the effective value of his army by about 400 on top of the intrinsic 300 deficit.) After the first trade it is 3 against 3, and in the trade-avoiding strategy we would see out pieces depressed by 3x25=75, for 525 each, so now we are about equal, in stead of 100 behind. But we will continue to ignore threats of further trading, so that our effective piece value remains 600. When the opponent under pressure trades us down to 2 vs 2, each of our pieces is suppressed by 2x25=50 to 550 in a trade-avoiding strategy, and even under trade-avoidance we would already be ahead by 50. Trade-ignoring we would lead by 150. After trading the fore-last piece the effective value of our remaining piece would be 600-25=575, and we would still be 50 ahead. So even now, trading would not have to overly worry us, athough we should not imagine that we can keep up the 150 advantage we have in a trade-ignoring strategy forever.

Trading the last piece would truly equalize us, and as that is not better than the +50 advantage we could reap in a 1-1 or 2-2 situation by employing the trade-avoiding strategy, we would switch to trade-avoidance only after the first two trades, where for the third piece we would still be prepared to take some risk of trading if that could increse our attack on other material. Only the last piece would have to be jealously protected.

Moral lesson: By being afraid of an inferior opponent, you give him the advantage. Treating your pieces as if they are worth more might in practice make them worth less. Don't do it, then!

I would be surprised if Zillions would know this. And that explains why it would take so long to beat Seperate Realms with FIDE. It should simply seek a few quick, nominally bad trades, and it would still be left with ennough advantage to get it over with quickly. In stead it cowers away, avoiding trading, until out of pure need it is forced to make the trades of lose other material, only to discover that after it has done so, the pressure is lifted and it can lash back with unrestrained power.