Re: [Scid-users] Re : 3.7 beta 4 - tree cache and game analysis

2009-02-28 Thread Zarkon Fischer
. Incidentally, if I set multipv to 2 or 3 then multipv=2 or 3 is used to annotate. But not multipv=1. Zarkon On 2/26/09, Pascal Georges wrote: > Did you change UCI engine settings via the configuration dialog ? > > Pascal > > 2009/2/26, Zarkon Fischer : >> Actually, now

Re: [Scid-users] 3.7 beta 4 - tree cache and game analysis

2009-02-25 Thread Zarkon Fischer
Actually, now game analysis is only using multipv=4 even though multipv=1 for infinite analysis. I'm not sure why it sometimes works and sometimes not, but I guess there is an issue here. Regards Zarkon On 2/20/09, Zarkon Fischer wrote: > Could you give the number of nodes per sec

Re: [Scid-users] 3.7 beta 4 - tree cache and game analysis

2009-02-20 Thread Zarkon Fischer
Could you give the number of nodes per second with various multipv and the engine you use ? I know well the code you used by many engines for multipv, so I know what impact it has. But maybe other engines behave differently (which I don't understand : sorting scores and lines before an output is m

Re: [Scid-users] 3.7 beta 4 - tree cache and game analysis

2009-02-18 Thread Zarkon Fischer
> > A couple of things I've come across: > > 1.When using the annotate button in the analysis pane to annotate a > game, the analysis is done in multipv mode (4 lines) which I guess > would be 4 times slower than if it were done with just a single line. Not at all. The difference is bearly notice

[Scid-users] 3.7 beta 4 - tree cache and game analysis

2009-02-16 Thread Zarkon Fischer
Pascal SCID's new interface is looking excellent. With the option to remove the game info from beneath the board, it even looks pretty good on a 1024x600 netbook! A couple of things I've come across: 1.When using the annotate button in the analysis pane to annotate a game, the analysis is done i