future mobility evaluation term

[smcracraft]

Anyone tried this?

http://mysite.verizon.net/vzesz4a6/current/id80.html

[PK]

this topic pops up from time to time - but now it is the first time when author promises empirical tests. IMHO this isn't going to be competitive within a traditional chess program framework, but might be good for a heuristic searcher (Steven Edwards, where art thou?)

[Gerd Isenberg]

Anyone tried this?

http://mysite.verizon.net/vzesz4a6/current/id80.html

Progressive Mobility is on my todo list...

Some implementation hints from cpw:
Multiple Sliding Pieces ...

For bitsets with multiple sliding pieces one can apply a fill-algorithm in each of the eight ray-directions to determine attacks. Since we keep disjoint directions, there is still a unique 1:1 relation from each bit of the target set to it's source square. This approach is great to determine target sets we may reach in two or more moves.

and Flood Fill Techniques with multiple sliders.

Keeping distinct ray-directions has some advantages, for instance in determining pinned pieces or discovered checks. Fill-Algorithms are great to feed in line-wise attack sets of single sliders, to determine a progressive mobility on the orthogonal rays, i.e. move targets in two moves, or trajectories against important squares or areas.

Code: Select all

 . . . . . . . .
 . . . . \ . . .
 . . . \ . 1 . .
 . . \ . 1 . \ .
 . \ . 1 . \ . .
 . . 1 . \ . . .
 . B . \ . . . .
 . . . . . . . .

[smcracraft]

Isn't it theorized that Rybka uses future mobility?

[bob]

Isn't it theorized that Rybka uses future mobility?

Wild-ass guess, just like most everything "assumed" about Rybka...

[Michael Sherwin]

Isn't it theorized that Rybka uses future mobility?

Somebody, I forget who, decided by some logic that I can not remember that when the knights were removed from the starting position, Rybka's search slowed way down and therefore Rybka must use future mobility. However, I posted in reply that with the knights on the board the best move (NB3) was more obvious to the search and therefore due to the nature of alpha/beta the trees were smaller and the search faster.

[Zach Wegner]

Isn't it theorized that Rybka uses future mobility?

I don't know about later versions (probably not), but Rybka 1.0 definitely does not.