summaryrefslogtreecommitdiff
path: root/gistanalysis/notes
blob: af9084a279a8d5c807b545c506a3939713974cf8 (plain)
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
ls srcpre reqs

audio fro input or playback
wave + fft + other



create objects from the audio wavs or fft
manipulate them with forces - the audio itself + transforms + fractals
artifically create a low quality transform algorithm
give them a lifetime - maybe auto to preserve frame rate
animate them - line changing length, or growing into a geometic shape


control the number and persistence of the (dots/sparks/chips)

1st step: collect live audio data into 2 point arrays and draw it
how do we deal with the screen size??

we draw to the output windows size: points are in output window coords
not resolution independednt:
will probably need to fix this some time


1st example: make a colourpolyline of centre, 256 points
frame budget= 2k points, so make it have a life of 8 frames max
ways to extend this- less lines, just points, 

a connected series of lines
a constellation of dots
a series of shapes that grow or change or interact

are these necessarily different things?
what do they have in common?
- audio data that is stored over a few frames

start with the dots one.

how to make sh!t look good.

how to draw the dots.. will this ever really look good, are there realistically gonna be many


we like drawing lines.

we could draw waveforms in loops rather than across?

we could draw fft bars

we could graduate the colours

make a realistic breakdown between technicalities (ie audio, mapping) and content

a better way to plan scanning area not pixel based? guess warp scale is ok for now
subtlety not really the thing?
drawing dots would be good.
warping dots in interesting ways
wraparound
other kinds of audio analysis



1- mirror (L/R) 
easy fading
transform the origin ie vertical hold effects
segmenter effects
actual 3D?
2- an FFT thing with bars, similar, what's the ideal interface
3- sample waveforms randomly
4- more transforms, abstract transforms
5- set size of points
6- transition between colours
7- show or blank on onset
8- transform on onset
n- scriptable transforms

- audio input
- mapping
- fix length overlap
- some kind of svg thing for ambulance with audio animation

scale isn't scaling from centre - move evrything to centre and use centre feature of ofxHelios
X add a master scale
X rotate isn't working at all
set modifiers for transformations to animate
need to hook up midi


1. get full control of placement and sizing - set up mapping framework
2. get a full, smooth waveform working
3. apply segmenter
4. animate origin and output seperately - tv hold effects
5. hook up to midi - get controller 

restore from old project? 
don't need concept of output pos but could use concept of output scale
make new project with bones of that project? np - add in positioning interface?

-> audioplot mirror
onsets turning things on
get some content for ambulance
colours in audio
midi hookup to chaos
-> get audio input with fft


seems to be a bug when closing while the mic is live

mirror
set more sensible limits for points and plots