Figure of merit

Discussion of chess software programming and technical issues.

Moderators: hgm, Dann Corbit, Harvey Williamson

lauriet
Posts: 199
Joined: Sun Nov 03, 2013 9:32 am

Figure of merit

Post by lauriet »

Hey all,
It seems to me that a 'figure of merit' of a chess program would be how long a n+1 ply search takes compared to a n ply search. eg does it take 3 times longer to search 10 ply as it does to search 9 ply, or 10 times longer.
Does this make sense and how do chess programs compare in this regard. Does it follow that a better program follows this rule ?


Laurie
User avatar
stegemma
Posts: 859
Joined: Mon Aug 10, 2009 10:05 pm
Location: Italy
Full name: Stefano Gemma

Re: Figure of merit

Post by stegemma »

lauriet wrote:Hey all,
It seems to me that a 'figure of merit' of a chess program would be how long a n+1 ply search takes compared to a n ply search. eg does it take 3 times longer to search 10 ply as it does to search 9 ply, or 10 times longer.
Does this make sense and how do chess programs compare in this regard. Does it follow that a better program follows this rule ?


Laurie
This is called "branching factor" and in effect it could give an idea of the strength of an engine.

For more information see this link:

https://chessprogramming.wikispaces.com ... ing+Factor
Author of Drago, Raffaela, Freccia, Satana, Sabrina.
http://www.linformatica.com