Single Comment

Silver Elephant Chess. Missing description (10x8, Cells: 80) [All Comments] [Add Comment or Rating]

H. G. Muller wrote on Sat, Oct 3, 2009 03:04 PM UTC:

I must agree with David. I did lots of empirical piece-value determinations on the computer, with setups like the array for this game (putting two exo-pieces in a context of orthodox pieces on a 10x8 board). I did not test the Silver Elephant, but the (FA) piece was already very close in value to a Knight. (In pairs, that is; being color bound there likely is a pair-bonus effect involved.)

My empirical formula for 8-fold symmetric short-range pieces was (5/8*N+30)*N centi-Pawn, where N is the number of target squares. For 10 squares this would give 362. (On a scale where N=280.) But in other tests forward moves turned out to contribute about twice as much to value as sideway or backward moves. If I count forward moves as 4/3 and backward moves as 2/3, I get N=32/3 and a value of 391. And this piece seems to have mating portential as well, which should probably be good for a bonus.

For comparison: a single Bishop on 10x8 is worth about 330 on the same scale, with a bonus of 50 for a pair.