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 100 216 3.6936 ms
172 ns 165 ns 184 ns
46 ns 31 ns 66 ns
(3x3)x(3x3) - ViSP 100 219 3.7011 ms
167 ns 165 ns 172 ns
15 ns 3 ns 31 ns
(6x6)x(6x6) - Naive code 100 129 3.7023 ms
308 ns 295 ns 328 ns
79 ns 56 ns 112 ns
(6x6)x(6x6) - ViSP 100 127 3.683 ms
280 ns 277 ns 289 ns
26 ns 10 ns 51 ns
(8x8)x(8x8) - Naive code 100 80 3.696 ms
462 ns 461 ns 463 ns
4 ns 0 ns 8 ns
(8x8)x(8x8) - ViSP 100 89 3.6935 ms
413 ns 413 ns 415 ns
5 ns 0 ns 11 ns
(10x10)x(10x10) - Naive code 100 48 3.7008 ms
849 ns 810 ns 909 ns
245 ns 182 ns 326 ns
(10x10)x(10x10) - ViSP 100 58 3.7004 ms
645 ns 637 ns 666 ns
60 ns 17 ns 119 ns
(20x20)x(20x20) - Naive code 100 7 3.8731 ms
5.538 us 5.53 us 5.562 us
59 ns 2 ns 132 ns
(20x20)x(20x20) - ViSP 100 10 4.043 ms
3.914 us 3.909 us 3.924 us
31 ns 8 ns 61 ns
(6x200)x(200x6) - Naive code 100 6 4.3518 ms
7.242 us 7.234 us 7.264 us
62 ns 10 ns 119 ns
(6x200)x(200x6) - ViSP 100 11 4.0194 ms
3.629 us 3.621 us 3.653 us
60 ns 2 ns 127 ns
(200x6)x(6x200) - Naive code 100 1 16.7183 ms
155.724 us 153.807 us 159.392 us
13.075 us 8.221 us 21.689 us
(200x6)x(6x200) - ViSP 100 1 12.8256 ms
127.576 us 126.209 us 130.558 us
9.866 us 5.164 us 16.692 us
(207x119)x(119x207) - Naive code 100 1 494.329 ms
4.9908 ms 4.98164 ms 5.00157 ms
50.402 us 43.99 us 66.873 us
(207x119)x(119x207) - ViSP 100 1 225.965 ms
2.26113 ms 2.25987 ms 2.26424 ms
9.499 us 4.867 us 19.559 us
(83x201)x(201x83) - Naive code 100 1 136.936 ms
1.3683 ms 1.36736 ms 1.36923 ms
4.735 us 4.183 us 5.626 us
(83x201)x(201x83) - ViSP 100 1 69.1109 ms
703.049 us 693.708 us 720.098 us
62.567 us 38.643 us 104.346 us
(600x400)x(400x600) - Naive code 100 1 15.2181 s
151.581 ms 151.194 ms 152.09 ms
2.24227 ms 1.81167 ms 2.92544 ms
(600x400)x(400x600) - ViSP 100 1 6.67635 s
64.9744 ms 64.8007 ms 65.1767 ms
956.587 us 821.225 us 1.12209 ms
(400x600)x(600x400) - Naive code 100 1 10.1197 s
103.109 ms 102.587 ms 103.793 ms
3.0202 ms 2.4384 ms 3.73497 ms
(400x600)x(600x400) - ViSP 100 1 4.72905 s
45.6781 ms 45.2911 ms 46.1563 ms
2.20046 ms 1.87937 ms 2.85897 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 100 154 3.696 ms
232 ns 231 ns 233 ns
3 ns 0 ns 8 ns
(3x3)x(3x3) - ViSP 100 236 3.6816 ms
150 ns 146 ns 157 ns
27 ns 17 ns 41 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 100 135 3.7125 ms
255 ns 255 ns 257 ns
3 ns 0 ns 8 ns
(4x4)x(4x4) - ViSP 100 215 3.698 ms
169 ns 169 ns 169 ns
1 ns 0 ns 2 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 100 223 3.6795 ms
153 ns 150 ns 162 ns
28 ns 16 ns 46 ns
(3x3)x(3x1) - ViSP 100 239 3.6806 ms
155 ns 155 ns 156 ns
1 ns 0 ns 4 ns
(6x6)x(6x1) - Naive code 100 219 3.6792 ms
165 ns 164 ns 167 ns
3 ns 0 ns 8 ns
(6x6)x(6x1) - ViSP 100 204 3.6924 ms
197 ns 189 ns 210 ns
51 ns 37 ns 72 ns
(8x8)x(8x1) - Naive code 100 190 3.686 ms
192 ns 186 ns 207 ns
45 ns 21 ns 81 ns
(8x8)x(8x1) - ViSP 100 183 3.6966 ms
195 ns 195 ns 196 ns
3 ns 0 ns 6 ns
(10x10)x(10x1) - Naive code 100 170 3.706 ms
210 ns 207 ns 216 ns
18 ns 4 ns 33 ns
(10x10)x(10x1) - ViSP 100 159 3.7047 ms
218 ns 218 ns 219 ns
3 ns 0 ns 7 ns
(20x20)x(20x1) - Naive code 100 88 3.7312 ms
399 ns 391 ns 414 ns
51 ns 26 ns 80 ns
(20x20)x(20x1) - ViSP 100 89 3.7202 ms
404 ns 404 ns 405 ns
2 ns 0 ns 4 ns
(6x200)x(200x1) - Naive code 100 37 3.7074 ms
919 ns 918 ns 921 ns
5 ns 0 ns 10 ns
(6x200)x(200x1) - ViSP 100 30 3.753 ms
1.269 us 1.258 us 1.291 us
75 ns 45 ns 125 ns
(200x6)x(6x1) - Naive code 100 27 3.7071 ms
1.262 us 1.255 us 1.296 us
67 ns 0 ns 160 ns
(200x6)x(6x1) - ViSP 100 26 3.7466 ms
1.413 us 1.38 us 1.526 us
273 ns 79 ns 615 ns
(207x119)x(119x1) - Naive code 100 3 5.0646 ms
15.572 us 15.304 us 16.159 us
1.906 us 1.008 us 3.257 us
(207x119)x(119x1) - ViSP 100 2 4.4566 ms
21.737 us 21.277 us 22.537 us
3.009 us 1.938 us 4.32 us
(83x201)x(201x1) - Naive code 100 4 4.6444 ms
10.619 us 10.553 us 10.923 us
610 ns 59 ns 1.447 us
(83x201)x(201x1) - ViSP 100 3 4.6578 ms
15.677 us 15.524 us 16.078 us
1.131 us 391 ns 2.242 us
(600x400)x(400x1) - Naive code 100 1 22.2641 ms
208.053 us 205.12 us 214.11 us
20.595 us 12.289 us 36.479 us
(600x400)x(400x1) - ViSP 100 1 23.3768 ms
233.743 us 233.577 us 234.127 us
1.218 us 650 ns 2.335 us
(400x600)x(600x1) - Naive code 100 1 17.7946 ms
163.147 us 162.437 us 164.908 us
5.262 us 1.598 us 9.387 us
(400x600)x(600x1) - ViSP 100 1 24.0175 ms
238.468 us 238.319 us 238.799 us
1.073 us 590 ns 2.069 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 100 242 3.6784 ms
152 ns 152 ns 152 ns
1 ns 0 ns 2 ns
(3x3) - ViSP 100 215 3.698 ms
167 ns 166 ns 167 ns
1 ns 0 ns 3 ns
(6x6) - Naive code 100 160 3.68 ms
228 ns 227 ns 228 ns
2 ns 0 ns 4 ns
(6x6) - ViSP 100 129 3.7023 ms
279 ns 278 ns 279 ns
2 ns 0 ns 4 ns
(8x8) - Naive code 100 106 3.6888 ms
343 ns 342 ns 344 ns
2 ns 0 ns 5 ns
(8x8) - ViSP 100 84 3.7128 ms
420 ns 419 ns 424 ns
9 ns 2 ns 20 ns
(10x10) - Naive code 100 69 3.7329 ms
526 ns 526 ns 528 ns
3 ns 0 ns 7 ns
(10x10) - ViSP 100 55 3.718 ms
636 ns 636 ns 638 ns
5 ns 1 ns 9 ns
(20x20) - Naive code 100 12 3.8448 ms
3.132 us 3.129 us 3.141 us
23 ns 2 ns 46 ns
(20x20) - ViSP 100 9 3.6936 ms
3.899 us 3.895 us 3.909 us
28 ns 2 ns 51 ns
(6x200) - Naive code 100 1 9.8439 ms
95.755 us 95.706 us 95.871 us
359 ns 129 ns 627 ns
(6x200) - ViSP 100 1 13.1222 ms
125.529 us 125.438 us 125.681 us
587 ns 396 ns 820 ns
(200x6) - Naive code 100 9 3.8115 ms
4.238 us 4.234 us 4.253 us
33 ns 1 ns 76 ns
(200x6) - ViSP 100 10 3.971 ms
3.701 us 3.692 us 3.723 us
68 ns 36 ns 127 ns
(207x119) - Naive code 100 1 151.902 ms
1.48603 ms 1.47566 ms 1.50025 ms
61.785 us 48.536 us 79.359 us
(207x119) - ViSP 100 1 145.366 ms
1.42862 ms 1.41744 ms 1.44475 ms
67.892 us 51.549 us 101.145 us
(83x201) - Naive code 100 1 150.939 ms
1.5154 ms 1.5095 ms 1.52265 ms
33.262 us 28.107 us 42.342 us
(83x201) - ViSP 100 1 154.46 ms
1.53329 ms 1.52608 ms 1.5402 ms
35.96 us 31.586 us 40.965 us
(600x400) - Naive code 100 1 5.48345 s
54.0313 ms 53.9336 ms 54.1664 ms
578.342 us 460.41 us 870.919 us
(600x400) - ViSP 100 1 4.89592 s
49.1144 ms 48.8925 ms 49.4029 ms
1.28257 ms 1.03966 ms 1.69331 ms
(400x600) - Naive code 100 1 8.06838 s
79.9083 ms 79.6242 ms 80.2447 ms
1.56944 ms 1.35167 ms 2.03858 ms
(400x600) - ViSP 100 1 7.62138 s
77.7236 ms 77.1346 ms 78.684 ms
3.75861 ms 2.618 ms 5.98312 ms
-------------------------------------------------------------------------------
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 100 218 3.6842 ms
157 ns 156 ns 159 ns
5 ns 3 ns 9 ns
(3x3) - ViSP 100 199 3.7014 ms
176 ns 173 ns 184 ns
25 ns 12 ns 43 ns
(6x6) - Naive code 100 87 3.7236 ms
471 ns 470 ns 474 ns
9 ns 2 ns 17 ns
(6x6) - ViSP 100 106 3.71 ms
323 ns 317 ns 340 ns
46 ns 8 ns 84 ns
(8x8) - Naive code 100 97 3.7248 ms
349 ns 342 ns 364 ns
49 ns 27 ns 80 ns
(8x8) - ViSP 100 75 3.7125 ms
498 ns 473 ns 533 ns
150 ns 116 ns 188 ns
(10x10) - Naive code 100 66 3.7092 ms
513 ns 505 ns 532 ns
60 ns 29 ns 103 ns
(10x10) - ViSP 100 44 3.718 ms
804 ns 773 ns 854 ns
196 ns 137 ns 271 ns
(20x20) - Naive code 100 12 3.6948 ms
2.814 us 2.758 us 2.962 us
407 ns 55 ns 745 ns
(20x20) - ViSP 100 8 4.1192 ms
4.853 us 4.715 us 5.154 us
988 ns 552 ns 1.8 us
(6x200) - Naive code 100 8 3.7096 ms
4.486 us 4.292 us 4.811 us
1.253 us 862 ns 1.847 us
(6x200) - ViSP 100 6 4.3122 ms
7.072 us 6.908 us 7.401 us
1.135 us 645 ns 1.816 us
(200x6) - Naive code 100 1 10.3833 ms
96.175 us 95.169 us 97.685 us
6.186 us 4.451 us 9.244 us
(200x6) - ViSP 100 1 16.2869 ms
148.771 us 146.851 us 151.947 us
12.325 us 8.601 us 21.818 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 91 3.6855 ms
375 ns 368 ns 388 ns
45 ns 28 ns 72 ns
(6x6)x(6x6) - ViSP 100 116 3.6888 ms
300 ns 290 ns 321 ns
70 ns 39 ns 125 ns
(20x6)x(6x6) - Naive code 100 48 3.7536 ms
686 ns 676 ns 711 ns
72 ns 22 ns 145 ns
(20x6)x(6x6) - ViSP 100 56 3.7128 ms
619 ns 602 ns 645 ns
105 ns 74 ns 142 ns
(207x6)x(6x6) - Naive code 100 6 3.918 ms
5.565 us 5.503 us 5.693 us
438 ns 203 ns 738 ns
(207x6)x(6x6) - ViSP 100 7 3.8073 ms
4.705 us 4.636 us 4.887 us
508 ns 97 ns 1.009 us
(600x6)x(6x6) - Naive code 100 3 5.3424 ms
16.683 us 16.404 us 17.068 us
1.66 us 1.262 us 2.143 us
(600x6)x(6x6) - ViSP 100 3 4.5849 ms
12.986 us 12.946 us 13.101 us
318 ns 139 ns 683 ns
(1201x6)x(6x6) - Naive code 100 2 7.3592 ms
32.231 us 31.831 us 32.749 us
2.309 us 1.756 us 3.13 us
(1201x6)x(6x6) - ViSP 100 2 6.0316 ms
26.636 us 26.112 us 27.754 us
3.736 us 2.116 us 6.882 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 95 3.7145 ms
356 ns 352 ns 366 ns
30 ns 6 ns 59 ns
(6x6)x(6x6) - ViSP 100 119 3.7128 ms
277 ns 276 ns 279 ns
4 ns 2 ns 9 ns
(20x6)x(6x6) - Naive code 100 49 3.7632 ms
675 ns 674 ns 679 ns
10 ns 0 ns 21 ns
(20x6)x(6x6) - ViSP 100 58 3.7468 ms
565 ns 564 ns 569 ns
8 ns 3 ns 18 ns
(207x6)x(6x6) - Naive code 100 6 3.714 ms
5.837 us 5.78 us 5.977 us
426 ns 140 ns 798 ns
(207x6)x(6x6) - ViSP 100 7 3.8605 ms
4.783 us 4.711 us 4.943 us
521 ns 273 ns 900 ns
(600x6)x(6x6) - Naive code 100 3 5.4993 ms
16.966 us 16.671 us 17.54 us
2.029 us 1.188 us 3.185 us
(600x6)x(6x6) - ViSP 100 3 4.6308 ms
13.465 us 13.289 us 13.859 us
1.275 us 657 ns 2.252 us
(1201x6)x(6x6) - Naive code 100 1 3.7059 ms
32.449 us 31.826 us 33.947 us
4.54 us 879 ns 8.32 us
(1201x6)x(6x6) - ViSP 100 2 6.0968 ms
26.739 us 26.353 us 27.648 us
2.837 us 1.475 us 5.519 us
===============================================================================
All tests passed (60 assertions in 8 test cases)
|
