| efficiency of eigenvalue calculations [message #55199] |
Tue, 07 August 2007 13:20  |
R.G.Stockwell
Messages: 163
Registered: October 2004
|
Senior Member |
|
|
any have experience with solving eigenvalue problems
in IDL, as compared to how fast it is in Fortran?
I have some legacy code that uses eigenvector/value
decompositions, on arrays of the size 200 by 50.
However, it has to do thouasands of them.
Is IDL (trired and triql for instance) nearly as fast as Fortran for this
stuff, or should
I stay with fortran?
Cheers,
bob
|
|
|
|
| Re: efficiency of eigenvalue calculations [message #55286 is a reply to message #55199] |
Tue, 07 August 2007 19:02  |
Kenneth P. Bowman
Messages: 585
Registered: May 2000
|
Senior Member |
|
|
In article <46b8d3fb$0$495$815e3792@news.qwest.net>,
"R.G. Stockwell" <noemail@please.com> wrote:
> any have experience with solving eigenvalue problems
> in IDL, as compared to how fast it is in Fortran?
>
> I have some legacy code that uses eigenvector/value
> decompositions, on arrays of the size 200 by 50.
> However, it has to do thouasands of them.
>
> Is IDL (trired and triql for instance) nearly as fast as Fortran for this
> stuff, or should
> I stay with fortran?
>
>
> Cheers,
> bob
I would try LA_TRIRED and LA_TRIQL. They probably *are* Fortran
(or optimized C).
Ken Bowman
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
