[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: MEX/Mkoctfile questions about matrices algorithms...why slower?
From: |
David Bateman |
Subject: |
Re: MEX/Mkoctfile questions about matrices algorithms...why slower? |
Date: |
Wed, 15 Nov 2006 20:52:14 +0100 |
User-agent: |
Thunderbird 1.5.0.7 (X11/20060921) |
钟 伟生 wrote:
> Hi,
>
> I really appreciate your reply. ATLAS project maybe an efficient way to
> do the improvement on matrix math. But I am still wondering to know the
> reason why the run time increased.
>
> Theoretically speaking, while using the interface functions
> (MEX/DEDUN-DLD) combining matlab/octave with C/C++ language, the matrix
> math should be faster than before. So, what is the key problem which
> makes it slower? The algorithms or the overhead generating by the MEX?
> If the front one, I will try to solve it. Thank you!
>
>
Because when you use the "*" operator in octave/matlab there is probably
a large chance that it is using ATLAS internally. Therefore, rewriting
it in MEX/OCT can only have the effect of slowing things up..
D.