[rescue] SMP on intel wasteful?

[Joshua D Boyd]

On Thu, Jun 27, 2002 at 12:54:25AM -0400, George Adkins wrote:
> > If this sounds kinda hazy, it is mainly because I don't have handy my
> > definition of what a strict vector is.  I know that [x y z w] doesn't
> > count, mainly because of the w element.  And I know that packing two
> > sound samples together like [right-chanel1 left-chanel1 right-chanel2
> > left-chanel2] doesn't count because the right will usually mean
> > something different.  [right-chanel1 right-chanel2 right-chanel3
> > right-chanel4] would count though.
> 
> as I understand it, this would mean that things like fft's on large data sets 
> should be much faster...  
> Because instead of :
> 
> select operation 1
> fetch value (element 1)
> perform operation
> store result
> fetch value (element 2)
> perform operation 
> store result
> (and so on...)
> 
> you can instead do:
> 
> fetch Vector (elements 1 to n)
> select operation 1
> perform operation (on all elements in Vector)
> select operation 2
> perform operation (on all...)
> select operation 3
> perform operation (on all...)
> store fully processed Vector (elements 1to n)

I'm not sure that 1D FFTs are the best example.  2+D FFTs might be a
good example though.



-- 
Joshua D. Boyd