Re: execution speed in *oct files

[Joao Cardoso]

John W. Eaton wrote:
> 
> On  3-Jun-1999, A+A Adler <address@hidden> wrote:
> 
> | I just submitted an oct file for 2-D convolutions
> | to octave-sources.
> |
> | While I was writing it, I noticed that the standard octave
> | syntax seems to be very inefficient.

...

> 
> |            Why are pointers so much more efficient than
> |              matrix refernces?
> 
> Because the indexing operators invoke some functions to do the
> indexing calculation.  I'd guess that either the functions are not all
> inlined, or the compiler is not able to convert the addition and
> multiplication of the indexing operation to a simple increment (as you
> have), or both.

I also found Octave Matrix class very slow, if compared with an
equivalent *C* program using pointers (of course). Up to ten times
slower in same cases, usually three to four times slower.

Are you using gcc or egcs, I presume? Has anybody compiled Octave with
other compiler? Any hints?

Joao

-- 
Joao Cardoso                |   e-mail: address@hidden
INESC, R. Jose Falcao 110   |   tel:    + 351 2 2094322
4050 Porto, Portugal        |   fax:    + 351 2 2008487



---------------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.  To ensure
that development continues, see www.che.wisc.edu/octave/giftform.html
Instructions for unsubscribing: www.che.wisc.edu/octave/archive.html
---------------------------------------------------------------------