Ito,
I understand that your library optimizes
vector1 = matrix1 * vector2 + vector3
(Stroustrup's example, right? Double loop, no temporaries)
How general can the expressions allowed by MET be when one achieves
optimal performace?
As an example, how about
matrix1 = matrix2.transpose() = matrix3;
vector1 = vector2 + vector3 + vector4 + vector5;
vector1 = vector2 + matrix1 * vector3 + vector4;
vector1 = matrix1.transpose() * vector2 + vector3;
matrix1 = scalar1 * matrix2.transpose();
?
Regards,
Oleg Trott http://www.columbia.edu/~ot14/
email: ot14@columbia.edu
Ph.D. Candidate
Columbia University phone: (212) 305-0257 (work)
Palmer Lab (NMR) fax : (212) 305-6949 (work)
--------------------- Object Oriented Numerics List --------------------------
* To subscribe/unsubscribe: use the handy web form at
http://oonumerics.org/oon/
* If this doesn't work, please send a note to owner-oon-list@oonumerics.org
This archive was generated by hypermail 2b29 : Wed Feb 20 2002 - 03:20:14 EST