Marcin Kasperski schrieb: Hi!
>> You don't imagine the tricky stuff I had to do to make >> Scid analyse a whole game, recursively taking into >> account all variants and sub-variants ! I had to make >> some stacks to keep track of score changes, etc. So the >> idea is certainly good, but would mean big rewrite of >> code. Not today ! ;-) > > Who said, that it is to happen today ;-) One could imagine for the time beeing to do it for a "plain game" first and do it the other way round if variations exist. Probably it is of most interest in the case that I do not have an already analysed game. I wonder wether one would not have to go to the end of each variaton first and then step backwards for the mentioned to work. Sounds like a requirement for clever bookkeeping, indeed. (I admit that I did not have this complex scenario in mind with my intial suggestion.) > But there is a reason why most (all?) chessbase programs > analyse back. Exactly thats why I asked Stefan as he change it in Shredder 11/Linux. So I wondered whats the reason. > To give you even more things to consider, some tools (like > CA) use multipass schemes. Details vary, but usually you > first quickly scan through the game (using 1-5 sec/move), > then analyse it normally, but give some more time in > places where evaluation significantly raised/dropped, and > sometimes makes even deeper analysis in crucial points of > the game... This sounds a resonable approach, but also not to simple to implement. I think the stepping is a crucial part here. Sounds a bit like adaptive step algorithms for differential equation solvers. To add an idea from the Shredder GUI: they use for analysis several engines that analyse the position and a "master brain" that then selects the best move out of that analysis. Don't know how much they gain through this "triple brain" as they call it. -- Kind regards, Alexander Wagner Universitaetsbibliothek Ilmenau Langewiesener Str. 37 98693 Ilmenau Tel.: 03677/69-4521 , Fax.: 03677/69-4617 ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Scid-users mailing list Scid-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scid-users
