[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug-gnubg] address@hidden: Skill Threshold and 2ply question]

From: Øystein Johansen
Subject: [Bug-gnubg] address@hidden: Skill Threshold and 2ply question]
Date: Tue, 22 Oct 2002 19:35:15 +0000
User-agent: Mutt/1.2.5i

----- Forwarded message from address@hidden -----

From: address@hidden
Date: Tue, 22 Oct 2002 09:15:22 EDT
Subject: Skill Threshold and 2ply question
To: address@hidden
X-Mailer: AOL 7.0 for Windows UK sub 10502

After I have analysed a match I want to step through and see any errors even
if it it just 0.001 worse than the best move. 
I tried setting doubtful threshold to 0.00 but it says it must be positive.
I set it to 0.001 which it accepted but when I step through the match it 
go to errors above this level.
I set it to 0.01 and still it doesn't work.It seems to only go to errors 
above 0.04.
My second question. Analzyed at world class sometimes GNU only evaluates a 
single move at 2ply. Would it not be better to have a minimum of 2 moves?
Otherwise there is no point as there is nothing to compare against and in 
fact the second best move (1ply) could be the best move if analysed at 2 ply?

----- End forwarded message -----

Later he writes me this message:

> I now realise I have to rerun the analysis if I want new thresholds!
> Could you maybe give us an option not to rerun the analysis.
> many thanks


reply via email to

[Prev in Thread] Current Thread [Next in Thread]