Re: Default precision in Octave

[Hans Hegeman]

>J.B.J.W.Hegeman skryf:
>>
>> Is there an option in octave to set the default precision?
>> Currently it seems that octave is using doubles as default precision
>> Is is possible to set this to real variables?
>>
>This will at a minimum require extensive source code modification and
>rebuilding.  It is not a job for a non-expert.  In fact, doing any
>numerical work in single precision is also not for the non-expert.

>If I have not made myself plain yet: don't even try.  Go and do
>something safe, like driving a motorcycle on an icy road.

>Dirk
The only reason I was asking is:
We run a 'simple simulation' with very large matrices, the final precision
is (not yet) very important.
Matlab runs the program perfectly, but we only have a licence for Windows
and not for linux.
Therefore it would be nice to use octave, to me it seems to be a powerfull
program.
Running the the simulation using octave gives a Memory exhausted error,
although the same program is only using 750 Mb using Windows NT 4.0 (on a 1
GB system).
I know I can change the source code of my program, store a tmp-file and what
so ever,
I only want to have a fast solution to run the simulation using linux.

Thank you for the suggestions you made, but I already was aware of that.

Look at the mailing-list for 16-3 and 17-3, some-one has the same
allocation problems as I do have using linux (Redhat 6.1)

Best regards,
Hans



-----------------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.che.wisc.edu/octave/octave.html
How to fund new projects:  http://www.che.wisc.edu/octave/funding.html
Subscription information:  http://www.che.wisc.edu/octave/archive.html
-----------------------------------------------------------------------