[Insight-users] speed of svd

[Miller, James V (Research)]

Aren't most the Lapack routines in vnl already.  Is it just a matter of
making the vnl_svd use the Lapack routine instead of the Linpack routine?



-----Original Message-----
From: William A. Hoffman [mailto:billlist at nycap . rr . com]
Sent: Thursday, November 20, 2003 11:23 AM
To: John Biddiscombe; Bing Jian; Insight-Users at itk . org
Subject: RE: [Insight-users] speed of svd


There is some talk about moving to Lapack in vxl.  However,
I expect this will be 6 months or so from now before it makes
it into ITK.

-Bill


At 06:05 AM 11/20/2003, John Biddiscombe wrote:
>I think you'll find that vxl/vnl code uses linpack rather than Lapack
>(for svd at least) whereas Matlab probably uses lapack.
>
>JB
>
>> -----Original Message-----
>> From: insight-users-admin at itk . org 
>> [mailto:insight-users-admin at itk . org] On Behalf Of Bing Jian
>> Sent: 20 November 2003 03:49
>> To: Insight-Users at itk . org
>> Subject: [Insight-users] speed of svd
>> 
>> 
>> 
>> Hi, everyone,
>> 
>>     I assumed that C/C++ code would be faster than Matlab.
>> But to my surprise, I found that svd routine in matlab is 
>> much faster than the one in vxl. I tried it on a 660*660 
>> matrix, it took matlab <5 seconds. While in using vnl_svd, it 
>> took about more than 20 seconds.
>>    Does someone has an idea about this?
>> 
>>    Thanks in advance!
>> 
>> -- 
>> Best wishes,
>> Bing Jian
>> bjian at cise . ufl . edu
>> 
>> 
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk . org 
>> http://www . itk . org/mailman/listinfo/i> nsight-users
>> 
>_______________________________________________
>Insight-users mailing list
>Insight-users at itk . org
>http://www . itk . org/mailman/listinfo/insight-users 

_______________________________________________
Insight-users mailing list
Insight-users at itk . org
http://www . itk . org/mailman/listinfo/insight-users