[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gcc-patches
Subject:    Re: [PATCH, libstdc++, complex] complex multiplication algorithm improved
From:       Robert Dewar <dewar () adacore ! com>
Date:       2010-09-15 14:39:04
Message-ID: 4C90DA88.5060406 () adacore ! com
[Download RAW message or body]

On 9/15/2010 4:58 AM, Basile Starynkevitch wrote:

> I am not a processor expert, but common folklore says that floating
> point addition is slower than floating point multiplication.

That's odd folklore, it's not true on most processors, and there
exist processors where the opposite is true.

> And there might perhaps be some rounding or precision issues (IEEE
> floating point is a tricky beast in the ugly details).

Not really, it is clearly and well defined, what ugly details did
you have in mind?
[prev in list] [next in list] [prev in thread] [next in thread]