1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
|
Default matrix/vector min size to enable Blas/Lapack optimization: 0
Used matrix/vector min size to enable Blas/Lapack optimization: 0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
perfMatrixMultiplication is a Catch v2.9.2 host application.
Run with -? for options
-------------------------------------------------------------------------------
Benchmark matrix-matrix multiplication
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:217
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3)x(3x3) - Naive code 100 227 3.5866 ms
160 ns 159 ns 161 ns
3 ns 1 ns 6 ns
(3x3)x(3x3) - ViSP 100 251 3.5893 ms
146 ns 143 ns 155 ns
24 ns 2 ns 44 ns
(6x6)x(6x6) - Naive code 100 128 3.5968 ms
283 ns 282 ns 285 ns
6 ns 5 ns 8 ns
(6x6)x(6x6) - ViSP 100 134 3.5778 ms
261 ns 260 ns 262 ns
2 ns 0 ns 4 ns
(8x8)x(8x8) - Naive code 100 78 3.6036 ms
459 ns 458 ns 460 ns
4 ns 0 ns 8 ns
(8x8)x(8x8) - ViSP 100 82 3.5998 ms
428 ns 427 ns 429 ns
3 ns 0 ns 7 ns
(10x10)x(10x10) - Naive code 100 47 3.5861 ms
745 ns 745 ns 748 ns
6 ns 1 ns 16 ns
(10x10)x(10x10) - ViSP 100 50 3.62 ms
694 ns 694 ns 696 ns
4 ns 0 ns 11 ns
(20x20)x(20x20) - Naive code 100 7 3.885 ms
5.523 us 5.518 us 5.538 us
40 ns 3 ns 78 ns
(20x20)x(20x20) - ViSP 100 7 3.7891 ms
5.292 us 5.287 us 5.305 us
35 ns 2 ns 66 ns
(6x200)x(200x6) - Naive code 100 5 3.6485 ms
7.258 us 7.254 us 7.272 us
33 ns 3 ns 75 ns
(6x200)x(200x6) - ViSP 100 5 3.599 ms
7.182 us 7.177 us 7.196 us
38 ns 3 ns 81 ns
(200x6)x(6x200) - Naive code 100 1 16.0614 ms
152.954 us 151.946 us 156.023 us
8.084 us 3.034 us 17.626 us
(200x6)x(6x200) - ViSP 100 1 15.6444 ms
146.265 us 145.838 us 147.249 us
3.144 us 1.339 us 5.364 us
(207x119)x(119x207) - Naive code 100 1 485.185 ms
4.84537 ms 4.84266 ms 4.84732 ms
11.657 us 8.522 us 15.112 us
(207x119)x(119x207) - ViSP 100 1 488.668 ms
4.7803 ms 4.77946 ms 4.7812 ms
4.431 us 3.977 us 5.209 us
(83x201)x(201x83) - Naive code 100 1 135.486 ms
1.43162 ms 1.38005 ms 1.56063 ms
385.738 us 181.381 us 751.986 us
(83x201)x(201x83) - ViSP 100 1 134.647 ms
1.37954 ms 1.34677 ms 1.53771 ms
314.991 us 2.099 us 751.305 us
(600x400)x(400x600) - Naive code 100 1 15.2179 s
150.73 ms 150.528 ms 150.998 ms
1.19996 ms 976.732 us 1.40187 ms
(600x400)x(400x600) - ViSP 100 1 14.9137 s
148.665 ms 148.449 ms 148.938 ms
1.2353 ms 1.00954 ms 1.44016 ms
(400x600)x(600x400) - Naive code 100 1 10.1837 s
102.167 ms 101.995 ms 102.413 ms
1.04125 ms 791.732 us 1.29251 ms
(400x600)x(600x400) - ViSP 100 1 10.1381 s
101.378 ms 101.211 ms 101.618 ms
1.01171 ms 759.658 us 1.25666 ms
-------------------------------------------------------------------------------
Benchmark matrix-rotation matrix multiplication
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:294
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3)x(3x3) - Naive code 100 143 3.575 ms
233 ns 232 ns 235 ns
3 ns 0 ns 9 ns
(3x3)x(3x3) - ViSP 100 237 3.5787 ms
144 ns 144 ns 146 ns
3 ns 0 ns 8 ns
-------------------------------------------------------------------------------
Benchmark matrix-homogeneous matrix multiplication
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:379
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(4x4)x(4x4) - Naive code 100 132 3.6036 ms
262 ns 262 ns 263 ns
3 ns 1 ns 6 ns
(4x4)x(4x4) - ViSP 100 198 3.5838 ms
169 ns 169 ns 169 ns
1 ns 0 ns 3 ns
-------------------------------------------------------------------------------
Benchmark matrix-vector multiplication
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:467
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3)x(3x1) - Naive code 100 234 3.5802 ms
147 ns 147 ns 147 ns
0 ns 0 ns 1 ns
(3x3)x(3x1) - ViSP 100 264 3.5904 ms
133 ns 133 ns 134 ns
0 ns 0 ns 1 ns
(6x6)x(6x1) - Naive code 100 205 3.5875 ms
166 ns 166 ns 167 ns
2 ns 0 ns 5 ns
(6x6)x(6x1) - ViSP 100 219 3.5916 ms
155 ns 154 ns 155 ns
1 ns 0 ns 3 ns
(8x8)x(8x1) - Naive code 100 183 3.5868 ms
184 ns 183 ns 187 ns
7 ns 2 ns 16 ns
(8x8)x(8x1) - ViSP 100 193 3.5705 ms
172 ns 172 ns 173 ns
1 ns 0 ns 4 ns
(10x10)x(10x1) - Naive code 100 159 3.5934 ms
207 ns 207 ns 209 ns
2 ns 0 ns 6 ns
(10x10)x(10x1) - ViSP 100 171 3.5739 ms
194 ns 194 ns 195 ns
2 ns 0 ns 5 ns
(20x20)x(20x1) - Naive code 100 83 3.5856 ms
387 ns 387 ns 388 ns
2 ns 0 ns 5 ns
(20x20)x(20x1) - ViSP 100 87 3.6105 ms
383 ns 382 ns 384 ns
6 ns 6 ns 8 ns
(6x200)x(200x1) - Naive code 100 35 3.668 ms
921 ns 921 ns 923 ns
5 ns 0 ns 10 ns
(6x200)x(200x1) - ViSP 100 37 3.6038 ms
895 ns 886 ns 906 ns
50 ns 46 ns 52 ns
(200x6)x(6x1) - Naive code 100 26 3.6192 ms
1.242 us 1.24 us 1.247 us
13 ns 1 ns 29 ns
(200x6)x(6x1) - ViSP 100 28 3.6708 ms
1.187 us 1.185 us 1.19 us
11 ns 4 ns 19 ns
(207x119)x(119x1) - Naive code 100 3 5.3244 ms
15.86 us 15.697 us 16.471 us
1.39 us 283 ns 3.146 us
(207x119)x(119x1) - ViSP 100 3 5.0892 ms
21.083 us 20.009 us 22.267 us
5.731 us 5.18 us 7.049 us
(83x201)x(201x1) - Naive code 100 3 3.636 ms
10.72 us 10.705 us 10.756 us
112 ns 41 ns 209 ns
(83x201)x(201x1) - ViSP 100 4 4.5908 ms
10.302 us 10.195 us 10.599 us
795 ns 98 ns 1.636 us
(600x400)x(400x1) - Naive code 100 1 22.1507 ms
201.556 us 200.63 us 203.234 us
6.211 us 4.095 us 10.957 us
(600x400)x(400x1) - ViSP 100 1 21.8398 ms
203.591 us 202.66 us 206.086 us
7.081 us 1.993 us 14.139 us
(400x600)x(600x1) - Naive code 100 1 18.5221 ms
168.297 us 166.807 us 170.544 us
9.235 us 6.836 us 13.256 us
(400x600)x(600x1) - ViSP 100 1 18.2562 ms
165.329 us 164.449 us 166.724 us
5.566 us 4.082 us 8.912 us
-------------------------------------------------------------------------------
Benchmark AtA
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:548
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3) - Naive code 100 220 3.586 ms
152 ns 151 ns 152 ns
2 ns 1 ns 4 ns
(3x3) - ViSP 100 244 3.5868 ms
138 ns 138 ns 139 ns
2 ns 0 ns 5 ns
(6x6) - Naive code 100 144 3.5856 ms
230 ns 230 ns 231 ns
1 ns 0 ns 4 ns
(6x6) - ViSP 100 149 3.5909 ms
222 ns 218 ns 234 ns
31 ns 5 ns 58 ns
(8x8) - Naive code 100 95 3.61 ms
349 ns 344 ns 363 ns
41 ns 20 ns 71 ns
(8x8) - ViSP 100 98 3.5966 ms
336 ns 331 ns 348 ns
35 ns 7 ns 69 ns
(10x10) - Naive code 100 62 3.6394 ms
540 ns 531 ns 560 ns
63 ns 31 ns 109 ns
(10x10) - ViSP 100 63 3.6288 ms
524 ns 523 ns 527 ns
8 ns 3 ns 16 ns
(20x20) - Naive code 100 11 3.7554 ms
3.13 us 3.128 us 3.138 us
20 ns 1 ns 38 ns
(20x20) - ViSP 100 11 3.7114 ms
3.197 us 3.142 us 3.309 us
379 ns 209 ns 609 ns
(6x200) - Naive code 100 1 10.673 ms
96.248 us 95.999 us 96.85 us
1.86 us 968 ns 3.76 us
(6x200) - ViSP 100 1 10.5802 ms
96.96 us 96.408 us 98.916 us
4.629 us 1.213 us 10.424 us
(200x6) - Naive code 100 9 3.8205 ms
4.237 us 4.235 us 4.245 us
20 ns 3 ns 43 ns
(200x6) - ViSP 100 9 3.8025 ms
4.226 us 4.224 us 4.234 us
18 ns 1 ns 41 ns
(207x119) - Naive code 100 1 143.055 ms
1.44358 ms 1.44321 ms 1.44427 ms
2.515 us 1.638 us 4.646 us
(207x119) - ViSP 100 1 144.424 ms
1.44438 ms 1.44391 ms 1.44485 ms
2.404 us 2.211 us 2.608 us
(83x201) - Naive code 100 1 149.112 ms
1.49123 ms 1.49003 ms 1.49577 ms
10.666 us 2.255 us 24.942 us
(83x201) - ViSP 100 1 149.281 ms
1.52542 ms 1.49553 ms 1.66495 ms
278.385 us 19.581 us 658.686 us
(600x400) - Naive code 100 1 5.40658 s
54.1896 ms 54.1157 ms 54.3237 ms
491.847 us 321.487 us 805.342 us
(600x400) - ViSP 100 1 5.39861 s
54.0345 ms 54.0154 ms 54.0637 ms
118.763 us 87.188 us 172.826 us
(400x600) - Naive code 100 1 7.78527 s
77.9772 ms 77.9076 ms 78.0429 ms
344.104 us 261.422 us 486.234 us
(400x600) - ViSP 100 1 7.64336 s
77.7307 ms 77.617 ms 77.8411 ms
570.893 us 519.475 us 671.668 us
-------------------------------------------------------------------------------
Benchmark AAt
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:619
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(3x3) - Naive code 100 216 3.5856 ms
206 ns 206 ns 207 ns
2 ns 1 ns 3 ns
(3x3) - ViSP 100 237 3.5787 ms
142 ns 142 ns 143 ns
1 ns 0 ns 4 ns
(6x6) - Naive code 100 145 3.596 ms
229 ns 228 ns 230 ns
3 ns 1 ns 6 ns
(6x6) - ViSP 100 158 3.5866 ms
209 ns 209 ns 210 ns
3 ns 1 ns 5 ns
(8x8) - Naive code 100 98 3.6162 ms
338 ns 336 ns 344 ns
14 ns 4 ns 32 ns
(8x8) - ViSP 100 105 3.612 ms
314 ns 313 ns 316 ns
6 ns 3 ns 11 ns
(10x10) - Naive code 100 65 3.6075 ms
513 ns 506 ns 531 ns
49 ns 5 ns 103 ns
(10x10) - ViSP 100 72 3.6216 ms
454 ns 453 ns 456 ns
7 ns 3 ns 14 ns
(20x20) - Naive code 100 12 3.6348 ms
2.753 us 2.75 us 2.761 us
22 ns 1 ns 45 ns
(20x20) - ViSP 100 13 3.7817 ms
2.742 us 2.69 us 2.952 us
445 ns 49 ns 1.025 us
(6x200) - Naive code 100 9 3.6702 ms
4.065 us 4.048 us 4.127 us
150 ns 41 ns 344 ns
(6x200) - ViSP 100 9 3.6585 ms
4.074 us 4.071 us 4.079 us
19 ns 12 ns 30 ns
(200x6) - Naive code 100 1 9.9315 ms
89.743 us 89.437 us 90.33 us
2.081 us 1.156 us 3.126 us
(200x6) - ViSP 100 1 8.7793 ms
78.884 us 78.778 us 79.081 us
713 ns 435 ns 1.123 us
-------------------------------------------------------------------------------
Benchmark matrix-velocity twist multiplication
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:690
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(6x6)x(6x6) - Naive code 100 93 3.6177 ms
356 ns 355 ns 357 ns
3 ns 0 ns 8 ns
(6x6)x(6x6) - ViSP 100 143 3.5893 ms
230 ns 230 ns 232 ns
3 ns 0 ns 6 ns
(20x6)x(6x6) - Naive code 100 48 3.6528 ms
677 ns 676 ns 679 ns
5 ns 0 ns 10 ns
(20x6)x(6x6) - ViSP 100 94 3.5814 ms
349 ns 346 ns 354 ns
16 ns 7 ns 34 ns
(207x6)x(6x6) - Naive code 100 7 4.1615 ms
5.341 us 5.284 us 5.481 us
429 ns 143 ns 795 ns
(207x6)x(6x6) - ViSP 100 15 3.621 ms
2.243 us 2.21 us 2.326 us
246 ns 90 ns 459 ns
(600x6)x(6x6) - Naive code 100 3 5.1183 ms
16.476 us 16.299 us 16.819 us
1.216 us 726 ns 2.075 us
(600x6)x(6x6) - ViSP 100 6 4.0362 ms
6.152 us 6.071 us 6.361 us
594 ns 115 ns 1.088 us
(1201x6)x(6x6) - Naive code 100 1 3.5885 ms
32.537 us 32.155 us 33.553 us
2.92 us 870 ns 5.841 us
(1201x6)x(6x6) - ViSP 100 3 3.894 ms
11.938 us 11.743 us 12.388 us
1.434 us 724 ns 2.514 us
-------------------------------------------------------------------------------
Benchmark matrix-force twist multiplication
-------------------------------------------------------------------------------
/visp-ws/visp-fspindle/modules/core/test/math/perfMatrixMultiplication.cpp:775
...............................................................................
benchmark name samples iterations estimated
mean low mean high mean
std dev low std dev high std dev
-------------------------------------------------------------------------------
(6x6)x(6x6) - Naive code 100 90 3.6 ms
369 ns 367 ns 371 ns
9 ns 8 ns 12 ns
(6x6)x(6x6) - ViSP 100 141 3.5955 ms
235 ns 234 ns 237 ns
6 ns 3 ns 9 ns
(20x6)x(6x6) - Naive code 100 47 3.6096 ms
739 ns 731 ns 757 ns
54 ns 17 ns 106 ns
(20x6)x(6x6) - ViSP 100 93 3.6084 ms
354 ns 354 ns 358 ns
7 ns 0 ns 16 ns
(207x6)x(6x6) - Naive code 100 7 4.1657 ms
5.76 us 5.728 us 5.783 us
137 ns 89 ns 195 ns
(207x6)x(6x6) - ViSP 100 15 3.63 ms
2.212 us 2.207 us 2.23 us
41 ns 2 ns 95 ns
(600x6)x(6x6) - Naive code 100 3 5.1867 ms
16.291 us 16.157 us 16.781 us
1.187 us 250 ns 2.769 us
(600x6)x(6x6) - ViSP 100 6 3.9396 ms
6.07 us 6.042 us 6.142 us
206 ns 52 ns 411 ns
(1201x6)x(6x6) - Naive code 100 2 6.7678 ms
30.349 us 30.017 us 30.749 us
1.855 us 1.592 us 2.552 us
(1201x6)x(6x6) - ViSP 100 3 3.8268 ms
11.678 us 11.651 us 11.766 us
211 ns 8 ns 477 ns
===============================================================================
All tests passed (60 assertions in 8 test cases)
|
