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
|
I have done all the raw rendering and now I just need to use ImageMagick
to do motion-blur, cross-fades, and add the nanosecond/rotation display.
Given a clear plan, that is relatively quick work, so I hope to get the
DVD in the mail to John some time in the next week.
Here is the last version of the video I did before we brought in John:
http://video.google.com/videoplay?docid=-199611355948259531
Here are the various animation speeds. This is really a note to myself
about how to do correct motion blur.0
Fast: fastpng/fast_cpk_png, use raw frames directly
0.15 ps/sec (5 fs/frame)
Medium: fastpng/fast_cpk_png, average 4 frames
0.6 ps/sec (20 fs/frame)
Slow: slowpng/slow_cpk_png, average 10 frames
6 ps/sec (200 fs/frame)
Slow jumpy: slowpng/slow_cpk_png, use every 10th frame
6 ps/sec (200 fs/frame)
All animations are done at 30 FPS.
CPK = van der Waals atom display, fat spherical atoms
B+S = ball-and-stick atom display ("tubes" display)
The sequence:
(1) Titles 1 & 2: Medium speed, CPK, 23 seconds
I didn't render enough raw frames for this entire piece, so I
ran for 8 seconds with title 1, then started over and re-ran the
animation for 15 seconds for title 2. I have enough raw frames
for 18 seconds of medium-speed animation.
(2) Cross fade: 5 secs medium CPK, half-second fade, 5 secs medium B+S,
(3) Title 3: 18 seconds fast B+S (8 secs with title, 10 secs without)
(4) Title 4: 18 seconds medium B+S (8 secs with title, 10 secs without)
(5) Title 5: 18 seconds fast B+S (8 secs with title, 10 secs without)
(6) Titles 6 - 8: 25 seconds, fast jumpy B+S
(7) Cross fade: 5 secs slow jumpy B+S, half-second fade, 5 secs slow jumpy CPK
(8) Titles 9 & 10: 27 secs slow CPK (not jumpy)
Whenever there is no title on the screen, the nanoseconds and rotations
show in the upper left, in fixed-width font, with three digits after the
decimal point. Since adding these is trivial, I will produce all sequences
both with and without them, and you can decide when to cut them in or out.
My plan is to make a DVD with two top-level directories (with nsecs/rots,
and without) and in each of those, put eight subdirectories with the
640x480 frames for the eight pieces shown above.
==============================
I won't be doing anything with titles, but just for reference, here is
what was previously used for titles. I've used HTML tags to
notate line breaks, peragraphs, boldface, a horizontal dividing line (hr),
and indentation (blockqoute). Titles 7 and 8 appear along the bottom of
the screen.
1: Molecular dynamics of a small bearing at 300K, Computational modeling
by NanoEngineer-1, <Nanorex logo>, (c) 2006 Nanorex, Inc.
2: The device shown is a low-friction bearing with a tight ring that holds a
rotating shaft. <p> All dynamics are at room temperature (300K) with
a shaft frequency of 5 GHz. Apparent motions will be compared at
several display rates and with different imaging methods. <hr>
Initial display rate: 0.6 ps per second <br>
Imaging: multiple position samples per frame
3: Base rate: <b>0.15 ps per second</b> <p>
At this speed, the thermal motion of individual atoms is clearly
visible. Shaft rotation is almost imperceptible.
4: 4 times base rate: <b>0.6 ps per second</b> <p>
At this speed, the thermal motion of larger structures becomes
obvious. Shaft rotation remains inconspicuous.
5: 40 times base rate: <b>6.0 ps per second</b> <p>
The 5 GHz rotation is conspicuous. Atomic vibrations are now a
blur of motion.
<blockquote><b>Each animation frame shows merged images from
many position samples.</b></blockquote>
6: <b>Same rate, 6.0 ps per second, but imaging only one sample per frame.
This omits realistic motion blur.</b><p>
Imaging a single position sample per frame creates the false
appearance that atoms oscillate at the frame rate. This illusion
is common in molecular animations.
7: A stroboscopic illusion: motion speeds appear wrong
8: Realism requires multiple samples per frame
9: Covalent structures of the kind shown in this video can be realistically
simulated using current molecular dynamics techniques.<p>
Producing structures of this kind, however, will require molecular
nanofabrication techniques that have yet to be developed.
10: Simulation and animation by Will Ware <p>
Molecular dynamics by NanoEngineer-1 <p>
Device design from <i>Nanosystems</i> <p>
(K. E. Drexler, Wiley Interscience, 1992) <p>
<Nanorex logo> <p>
(c) 2006 Nanorex, Inc. <p>
<i>Rendering powered by POVray</i>
I think that last one should say "POV-Ray" rather than "POVray". That seems
to be the spelling preferred by its developers.
|
