Re: Can octave scripts be compiled into standalone programs?

[AlexM]

>>not really. I saw matlab can do this and wonder if/how much the
>>performance can be improved by compiling
But in large matlab programs the gain is not that much since they have a
highly optimised interpreter that skims the line of automated C code. If
doing it by hand, the gains are big.

Octave compiling to C is sometimes a humungous gain, I had a 50x gian in
coding the Fwrd/Bwrd alg into C++.

Ciao,
Am




Evan-4 wrote:
> 
> On 8/30/07, Jordi Gutiérrez Hermoso <address@hidden> wrote:
>> On 29/08/2007, Evan <address@hidden> wrote:
>> > to improve performance?
>>
>> Is the interpreter slowing you down? Can't vectorise any further? Do
>> you have any idea where your bottleneck is?
> 
> not really. I saw matlab can do this and wonder if/how much the
> performance can be improved by compiling
> 
>> > if yes, how?
>>
>> There have been previous requests and attempts to get this done, but
>> nothing definitive. You can use the liboctave classes in your own C++
>> programs, which is almost as good as taking an Octave script and
>> compiling it into native C++.
>>
>> FWIW, here's the previous attempt I know of to create an Octave compiler:
>>
>>     http://www.program-transformation.org/Octave/WebHome
>>
>> although it looks like you can't reach the actual servers where the
>> compiler is found. :-(
>>
> 
> Thanks any way
> 
>> - Jordi G. H.
>>
> 
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://www.cae.wisc.edu/mailman/listinfo/help-octave
> 
> 

-- 
View this message in context: 
http://www.nabble.com/Can-octave-scripts-be-compiled-into-standalone-programs--tf4346038.html#a12404035
Sent from the Octave - General mailing list archive at Nabble.com.