Author: Gerd Isenberg
Date: 12:19:22 01/19/06
Go up one level in this thread
>>>assert(expression == 0 || expression == 1); >>>foo[expression](); >> >>No - avoid that for simple expressions! > >You might be surprised: >http://chessprogramming.org/cccsearch/ccc.php?art_id=178068 > >Depends on the compiler, of course. Ok, a few years old - it was quite ok on P3 iirc. To my knowledge on recent processors (amd64) indirect branches with (not necessarily random) changed target addresses implies a missprediction. Also i would never replace a simple array access by indrect branch. a = x[boolexpression]; a = foo[boolexpression](); // returns either x[false] or x[true]
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.