Re: Matrix Multiplication

On 2008-01-07 15:54:04 -0400, user923005 <dcorbit@xxxxxxxxx> said:

On Jan 6, 1:25 pm, Evgenii Rudnyi <use...@xxxxxxxxx> wrote:


You asked for comments on your Matrix Multiply. You "forgot" to mention
that F90 has that as an intrinsic which will be as efficient or not as
any other system that has a run time support for such an operation.

I think it is a good idea to let the Fortran 9x programmers give it a
try.
If they can come up with a marvelous solution, I would like to see it.

To the Fortran camp, this article:
http://matrixprogramming.com/MatrixMultiply/

Discusses matrix multiplication. Can you provide an optimal Fortran
9x solution for comparison purposes?

Why would anyone develop a matrix multiply when there is a MATMUL intrinsic?

Mostly you can expect to be told to read the fine manual.

Are you still under the impression that Rudnyi is doing anything other
than a three loop N^3 in his various examples. He has figured out that
there are various languages, including F90, that share the attribute
of having matrix multiply built-in and that all of them will be equally
efficient, or not, depending upon how dilligent their implementors are.
No effort required on the part of the end programmer.




.

Relevant Pages

  • Re: Matrix Multiplication
    ... that F90 has that as an intrinsic which will be as efficient or not as ... If they can come up with a marvelous solution, I would like to see it. ... Are you still under the impression that Rudnyi is doing anything other ... Then if a F90 or F95 benchmark does not exceed the speed of the Atlas ...
    (sci.math.num-analysis)
  • Re: Fortran vs C++ for scientific computing
    ... I know you can operate very easily of arrays in F90, ... A and B really corresponded to a matrix multiplication. ...
    (comp.lang.fortran)