Re: [boost] BOOST::SIMD - handling double : precision vs speed

Subject: Re: [boost] BOOST::SIMD - handling double : precision vs speed
From: joel falcou (joel.falcou_at_[hidden])
Date: 2009-04-03 01:19:26

Le Jeu 2 avril 2009 22:13, Christian Henning a écrit :
>
double is almost 4-times slower than float. Seems to me that one should
> favor float over double.
This x4 already contaisn a x2 from
the difference of vector size. Rest is due to the internal algorithm to
achieve correct precision.

> I still would like to see
boost.simd appear and do some more testing.
This won't prevent
boost.simd to appear at all, I was just checking soem points ;)