Re: [Scid-users] More on SCID UCI implementation vs others and affects on engine output

Hi, I played a few minutes with Polyglot and configured Toga 1.2.1a to be handled as a xboard engine. In this position r7/p1p2pk1/1p6/4qpp1/2PP4/P3R3/2P3PP/6K1 w KQkq - 0 1 where the queen must be recaptured (!) I get this output : Depth: 12 Nodes: 9106K (765 kn/s) Score: -99.88

Re: [Scid-users] More on SCID UCI implementation vs others and affects on engine output

Hi, I checked engines' behavior : - changing hash size works (controlled by the memory usage of processes) - I tried to change one parameter in Toga (King Safety) and put traces in Toga's code : it is indeed taken into account. So I don't know what is going on here. Engines parameters are saved

[Scid-users] More on SCID UCI implementation vs others and affects on engine output

Continuing a previous email discussion where I checked Glaurung under SCID vs Glaurung under ChessBase GUI... I can make Rybka 2.3.2a behave EXACTLY the same under SCID as I can under ChessBase : just use polyglot instead of using the "built-in" means within SCID of configuring a UCI engine. W