vec_ipremultn16()

VEC_$IPREMULTN16                  Domain/OS                   VEC_$IPREMULTN16


NAME
     vec_$ipremultn16 - multiply a 16-bit integer vector by a matrix

SYNOPSIS (C)
     #include <apollo/base.h>
     #include <apollo/vec.h>

     void vec_$ipremultn16(
          short *start_vec,
          short *matrix,
          long int &m,
          long int &n,
          short *result_vec)

SYNOPSIS (Pascal)
     %include '/sys/ins/base.ins.pas';
     %include '/sys/ins/vec.ins.pas';

     procedure vec_$ipremultn16(
          in start_vec: univ vec_$integer16_vector;
          in matrix: univ vec_$integer16_matrix;
          in m: integer32;
          in n: integer32;
          out result_vec: univ vec_$integer16_vector);

SYNOPSIS (FORTRAN)
     %include '/sys/ins/base.ins.ftn'
     %include '/sys/ins/vec.ins.ftn'

           integer*4 m, n
           parameter (m = 3, n = 4)

           integer*2 start_vec(m), matrix(m, n), result_vec(n)

           call vec_$ipremultn16(start_vec, matrix, m, n, result_vec)

DESCRIPTION
     Vec_$ipremultn16 multiplies the m-element vector start_vec by the vari-
     ably dimensioned matrix matrix, and supplies the resulting n-element vec-
     tor in result_vec.

     In C, vec_$ipremultn16 applies the nxm matrix matrix as a left transform
     to the m-element column vector start_vec, and supplies the transformed
     n-element result in result_vec:

          for (i = 0; i < n; ++i) {
               result_vec[i] = 0;
               for (j = 0; i < m; ++i)
                    result_vec[i] += start_vec[j]
                                  * matrix[i][j];
          }

     In Pascal, vec_$ipremultn16 applies the nxm matrix matrix as a left
     transform to the m-element column vector start_vec, and supplies the
     transformed n-element result in result_vec:

          for i := 1 to n do
               begin
               result_vec[i] := 0;
               for j := 1 to m do
                    result_vec[i] := result_vec[i]
                                  + start_vec[j]
                                  * matrix[i,j];
               end

     In FORTRAN, vec_$ipremultn16 applies the mxn matrix matrix as a right
     transform to the m-element row vector start_vec, and supplies the
     transformed n-element result in result_vec:

           do 10 i = 1, n
               result_vec(i) = 0
               do 10 j = 1, m
                    result_vec(i) = result_vec(i)
          &                       + start_vec(j)
          &                       * matrix(j,i)
       10  continue

     start_vec
          An m-element vector to multiply by matrix.

     matrix
          A matrix to multiply start_vec by.

     m    The number of elements in start_vec.

     n    The number of elements in result_vec.

     result_vec
          An n-element vector that is the product of matrix and start_vec.

NOTES
     Vec_$ipostmultn16 transforms 16-bit integer vectors from the other side.

SEE ALSO
     vec_$dpremultn, vec_$ipremultn, vec_$premultn.
Museum
