| Non-optimized builds (SRT_QUAL = default) | Optimized builds (SRT_QUAL = maxopt) | |||
| 5x5 Matrix | 6x6 Matrix | 5x5 Matrix | 6x6 Matrix | |
| LinearAlgebra
Matrix * Matrix (*) Matrix not specialized Matrix not declared symmetric [tsnmul10] |
Min: 4.39 secs (4.39 µsecs)
Max: 4.48 secs (4.48 µsecs) |
Min: 5.51 secs (5.51 µsecs)
Max: 5.68 secs (5.68 µsecs) |
Min: 4.3 secs (4.3 µsecs)
Max: 4.4 secs (4.4 µsecs) |
Min: 5.5 secs (5.5 µsecs)
Max: 5.65 secs (5.65 µsecs) |
| LinearAlgebra
Matrix * Matrix (*) Matrix not specialized Matrix declared symmetric [tsnmul10] |
Min: 6.6 secs (6.6 µsecs)
Max: 6.8 secs (6.8 µsecs) |
Min: 8.28 secs (8.28 µsecs)
Max: 8.51 secs (8.51 µsecs) |
Min: 6.65 secs (6.65 µsecs)
Max: 6.97 secs (6.97 µsecs) |
Min: 8.35 secs (8.35 µsecs)
Max: 8.59 secs (8.59 µsecs) |
| LinearAlgebra
Matrix * Matrix (*) Matrix specialized Matrix not declared symmetric [tsnmul10] |
Min: 3.26 secs (3.26 µsecs)
Max: 3.34 secs (3.34 µsecs) |
Min: 3.68 secs (3.68 µsecs)
Max: 3.77 secs (3.77 µsecs) |
Min: 3.29 secs (3.29 µsecs)
Max: 3.6 secs (3.6 µsecs) |
Min: 3.67 secs (3.67 µsecs)
Max: 3.9 secs (3.9 µsecs) |
| LinearAlgebra
Matrix * Matrix (*) Matrix specialized Matrix declared symmetric [tsnmul10] |
Min: 6.72 secs (6.72 µsecs)
Max: 6.85 secs (6.85 µsecs) |
Min: 8.82 secs (8.82 µsecs)
Max: 8.95 secs (8.95 µsecs) |
Min: 6.7 secs (6.7 µsecs)
Max: 7.05 secs (7.05 µsecs) |
Min: 8.82 secs (8.82 µsecs)
Max: 9.26 secs (9.26 µsecs) |
| CLHEP
HepMatrix * HepMatrix (**) [tsnmul11] |
Min: 3.67 secs (3.67 µsecs)
Max: 3.75 secs (3.75 µsecs) |
Min: 4.61 secs (4.61 µsecs)
Max: 4.66 secs (4.66 µsecs) |
Min: 3.45 secs (3.45 µsecs)
Max: 3.58 secs (3.58 µsecs) |
Min: 4.54 secs (4.54 µsecs)
Max: 4.6 secs (4.6 µsecs) |
| CLHEP
HepSymMatrix * HepMatrix (**) [tsnmul12] |
Min: 5.6 secs (5.6 µsecs)
Max: 5.74 secs (5.74 µsecs) |
Min: 7.15 secs (7.15 µsecs)
Max: 7.34 secs (7.34 µsecs) |
Min: 5.53 secs (5.53 µsecs)
Max: 5.71 secs (5.71 µsecs) |
Min: 7.2 secs (7.2 µsecs)
Max: 7.38 secs (7.38 µsecs) |
| CovMatrices (***)
CovMatrixX * BiVectorX [tsnmul13] |
Min: 0.52 secs (0.52 µsecs)
Max: 0.54 secs (0.54 µsecs) |
Min: 0.67 secs (0.67 µsecs)
Max: 0.69 secs (0.69 µsecs) |
Min: 0.48 secs (0.48 µsecs)
Max: 0.54 secs (0.54 µsecs) |
Min: 0.66 secs (0.66 µsecs)
Max: 0.71 secs (0.71 µsecs) |
All timings are based on 1,000,000 multiplications.
The first number is the time for all the inversions while the time
in parenthesis is the time for 1 multiplication.
The string in the bracket is the name of the test module build for
the particular test.
When the results say "Non-Optimized builds," then all the libraries
and the test module were build with the "SRT_QUAL" variable having the
"default" value.
When the results say "Optimized builds," then all the libraries and
the test module were build with the "SRT_QUAL" variable having the "maxopt"
value.
(*) Since we can't do Matrix * BiVector in LinearAlgebra , we did
Matrix * Nx2 Matrix.
(**) Since we can't do Matrix * BiVector in CLHEP, we did HepMatrix
* Nx2 HepMatrix.
(***) Covariant Matrices are inherently symmetric.
Matrices and Vectors used in the tests:
LinearAlgebra:
5x5 matrix
1e+03
4
6
8
10
4 1.1e+03
16
18
20
6
16 1.2e+03
28
30
8
18
28 1.3e+03
40
10
20
30
40 1.4e+03
10-element "bivector":
2
4
4
6
6
8
8
10
10
12
6x6 matrix:
1e+03
4
6
8
10
12
4 1.1e+03
18
20
22
24
6
18 1.2e+03
32
34
36
8
20
32 1.3e+03
46
48
10
22
34
46 1.4e+03
60
12
24
36
48
60 1.5e+03
12-element "bivector":
2
4
4
6
6
8
8
10
10
12
12
14
CLHEP:
5x5 matrix:
1000
4
6
8
10
4 1100
16
18
20
6
16 1200
28
30
8
18
28 1300
40
10
20
30
40 1400
10-element "bivector":
2
4
4
6
6
8
8
10
10
12
6x6 matrix:
1000
4
6
8
10
12
4 1100
18
20
22
24
6
18 1200
32
34
36
8
20
32 1300
46
48
10
22
34
46 1400
60
12
24
36
48
60 1500
12-element "bivector":
2
4
4
6
6
8
8
10
10
12
12
14
CovMatrices:
5x5 matrix:
+1.0000000000e+03
+4.0000000000e+00 +1.1000000000e+03
+6.0000000000e+00 +1.6000000000e+01
+1.2000000000e+03
+8.0000000000e+00 +1.8000000000e+01
+2.8000000000e+01 +1.3000000000e+03
+1.0000000000e+01 +2.0000000000e+01
+3.0000000000e+01 +4.0000000000e+01 +1.4000000000e+03
10-element bivector:
+2.0000000000e+00 +4.0000000000e+00
+6.0000000000e+00 +8.0000000000e+00 +1.0000000000e+01
+2.0000000000e+00 +4.0000000000e+00
+6.0000000000e+00 +8.0000000000e+00 +1.0000000000e+01
6x6 matrix:
+1.0000000000e+03
+4.0000000000e+00 +1.1000000000e+03
+6.0000000000e+00 +1.8000000000e+01
+1.2000000000e+03
+8.0000000000e+00 +2.0000000000e+01
+3.2000000000e+01 +1.3000000000e+03
+1.0000000000e+01 +2.2000000000e+01
+3.4000000000e+01 +4.6000000000e+01 +1.4000000000e+03
+1.2000000000e+01 +2.4000000000e+01
+3.6000000000e+01 +4.8000000000e+01 +6.0000000000e+01
+1.5000000000e+03
12-element bivector:
+2.0000000000e+00 +4.0000000000e+00
+6.0000000000e+00 +8.0000000000e+00 +1.0000000000e+01
+1.2000000000e+01
+2.0000000000e+00 +4.0000000000e+00
+6.0000000000e+00 +8.0000000000e+00 +1.0000000000e+01
+1.2000000000e+01
Testing system specs:
CPU: 500 MHz PentiumIII (Katmai)
Memory: 128MB SDRAM
Cache: 512KB
Hard Disk: IBM 13677MB
Video Card: 16MB Diamond Viper TNT 1.95E
OS: Fermi Linux Release 6.1.1 (Strange)
Kernel: 2.2.16-3 on an i686
X-Server: XFree86 Version 11.0 Release 3360
Tests were compiled and build using GCC 3.0.1