armadillo
Armadillo
C++ library for linear algebra & scientific computing



Code extract



Add two matrices


Matrix size: 4x4

Approximate speed-up relative to
IT++:
15 times
Newmat:
10 times


Matrix size: 100x100

Approximate speed-up relative to
IT++: 3.5 times
Newmat: same speed



Add four matrices



Matrix size: 4x4

Approximate speed-up relative to
IT++: 15 times
Newmat: 10 times


Matrix size: 100x100

Approximate speed-up relative to
IT++: 6 times
Newmat: 1.5 times



Multiply four matrices



A: 100x80
B:  80x60
C:  60x40
D:  40x20

Approximate speed-up relative to:
IT++:
2.5 times
Newmat:
10 times


A: 1000x800
B:  800x600
C:  600x400
D:  400x200

Approximate speed-up relative to:
IT++: 2.5 times
Newmat: 20 times



Submatrix manipulation


Matrix size: 4x4

Approximate speed-up relative to:
IT++:
16 times
Newmat:
44 times


Matrix size: 100x100

Approximate speed-up relative to:
IT++:
2 times
Newmat:
4.5 times



Multi-operation expression


Matrix size: 4x4


Approximate speed-up relative to:
IT++:
77 times
Newmat:
23 times


Matrix size: 100x100
(y-axis is non-linear)

Approximate speed-up relative to:
IT++:
1086 times (!)
Newmat:
5 times


sourceforge