-
Notifications
You must be signed in to change notification settings - Fork 10
/
profiling.txt
67 lines (49 loc) · 1.73 KB
/
profiling.txt
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
Slicing Test Case:
linux
nard@DeepSpace9:~/PhotonSlicer$ python3 PhotonSlicer.py -s STLs/legocog.stl -r 1 -l 0.5
Full: 11.6 sec
Excl Load: 10.6 sec
No Encode: 11.3 sec
No Fill: 10.9 sec
Cython triInSlice 8.2 sec (if we abort triInSlice in first line, time stays 8.2 sec)
Cython triInSlice 7.4 sec (after some optimization in Stl2Slices)
Cython triInSlice 7.0 sec (after making tri lists per slice)
Draw to a 24 bit img 7.4 sec
Draw to a 8 bit img 5.1 sec
No fills if normal is up/down 4.0 sec
Slicing Test Case:
linux
nard@DeepSpace9:~/PhotonSlicer$ python3 PhotonSlicer.py -s STLs/3DBenchy.stl -r 1 -l 0.1
Cython triInSlice 462 sec (after some optimization in Stl2Slices)
Cython triInSlice 74 sec (after making tri lists per slice)
24 bit img 74 sec
8 bit img 56 sec
No fills if normal is up/down 56 sec
Slice 62 sec:
if polypoints = 10 sec
ConvexFill = 14 sec
pfrom,pto,pmid = 8 sec
fillpoints.append(pfill) = 1 sec
fill = 22 sec
img[fillPoint[1], fillPoint[0]) = 12 sec
Floodfill = 10 sec
Slicing CPU (Stl2Slices) vs GPU (GL_Stl2Slices)
GL_Stl2Slices: 18 sec
Stl2Slices: 58 sec
==============================================================================================
Encoding Test Case:
linux
nard@DeepSpace9:~/PhotonSlicer$ python3 PhotonSlicer.py -s STLs/bunny.stl -r 0.5 -l 1
Full: 10 sec
No fill: 10 sec
No Rotate: 10 sec (in encode we rotate 90deg)
No Encode: 1.1 sec
Encode: numpy-code 0,14 sec
reps-code 0,02 sec
No Encode: 1.1 sec
Cython (no numpy): 7 sec
Cython (w/ numpy): 8.1 sec
Cython (nonumpy1D): 5.0 sec
Cython (nonumpy1DBlock): 4.6 sec
Cython (withnumpy2): 7.9 sec
Cython (numpy1DBlock): 1.4 sec