Larsson Endgame test, Rybka-Houdini, 90+30, i7, 4CPU

[Rubinus]

Deep Rybka 4.1 SSE42 x64 - Houdini 2.0c Pro x64 -> 23:17 (9:3=28, +-53ELO)

90m+30s, 4cpu, 345tb+several 6man, 1GB hash, no pondering.
i7-860, 2.8GHz, 8GB RAM, OS Windows 7 Home x64
replay games
download games (any *.pgn + uci options + Larsson test)[/url]

[MM]

Deep Rybka 4.1 SSE42 x64 - Houdini 2.0c Pro x64 -> 23:17 (9:3=28, +-53ELO)

90m+30s, 4cpu, 345tb+several 6man, 1GB hash, no pondering.
i7-860, 2.8GHz, 8GB RAM, OS Windows 7 Home x64
replay games
download games (any *.pgn + uci options + Larsson test)[/url]

Hi,

thank you for sharing your results. They seem to confirm what i ever thought, so Rybka is very strong in the endgame and Houdini is strong only in easy endgames but weak in strategical endgames.

Regards

[Jouni]

Note, that in this famous position

[d]8/5pk1/r5pp/P7/3R3P/6P1/5PK1/8 w - - 0 1

both played Rd5 and won! Probably it's also winning (sooner or later black is in zugzwang).

Jouni

[Rubinus]

Black rook is bad, needed to a1 ...