I recently did a small talk at Bathcamp on Cymatics and computer visualizations of audio and @Z303 - who was in the audience - sent me a link to this physical spectrum analyzer. I’d shown a Processing sketch that did something very similar but in the virtual world so seeing it working in the real world was great fun.
And my processing version…
During the recent stint I did making audio reactive visuals for the recent Peter Gabriel tour we had to create a track for Apres Moi. For part of the track Marc Bessant suggested creating a 3d Battlezone style landscape from the audio loosely based on the Unknown Pleasures album artwork created by Peter Saville for Joy Division. This is something we’d seen flight404 doing before and it seemed to fit with the pace and tone of the track.
I figured it would only be a short step away to use a variation of the code to recreate the cover itself as an audio reactive piece. Turns out I’m not the first to have thought this but hey, originality is overrated.
The song used is the first track from the album, Disorder.
For reference here is a fan shot video of the version created for Apres Moi…
Noticed this via Peter Gabriel’s twitter feed (@itspetergabriel) An LA Times review of the Peter Gabriel Hollywood Bowl concert with some really great comments about the visuals I made. Very chuffed indeed!!!
An expansive video screen that spanned about two-thirds of the Bowl’s stage, enough to obscure the orchestra behind it, displayed a sequence of horizontal lines moving up as the screen itself raised. Soon a batch of vertical white lines joined, moving gracefully in tempo with the music, only Gabriel’s voice indicating his presence on stage until the screen was high enough to reveal the humans — all wearing black — behind it.
Reed’s “The Power of the Heart” nodded to “Ol’ Man River” as he sang “You and me we sweat and strain” in that wonderfully gritty, soulful voice of his. The accompanying imagery on an additional triptych video screen on the wall behind the orchestra was at its core supremely simple: two columns of swirling white dots resembling active molecules, each with a red dot, presumably the nuclei, at the center. As the song progressed, the nuclei moved from their respective outer screens toward one another in the center screen, an extraordinarily moving image in conjunction with Gabriel’s song about the difficulty of two hearts finding, and staying, together. It’s going to make for one killer concert DVD.
Seriously clever Photoshop goodness coming in CS5.
Recently spent a fantastic 6 days on a soundstage at Shepperton Studios preparing sound reactive tour visuals (with Processing) for the Peter Gabriel New Blood Tour with a great team of talented people. I had a great time and produced some work I’m very proud of. Some videos of the concerts have started to appear on YouTube including the work I did for the rear vertical screens on “My Body Is A Cage”.
And the visor screen from “Mirrorball”
Paul Graham published this article about maker and manager timescales which defined the main differences in perception and schedule between those that make and those that manage and the problems that can cause.
At about the same time that article was published I started a transition from a maker to a manager that is still ongoing and surprisingly hard to navigate. The usual expected problems of being promoted from within a team were minimized by virtue of the fact I was primarily the designer and front end guy working with a team of pure developers and so was always slightly apart from the crowd anyway. That and the startup I work for is small and has a pretty flat hierarchy.
Surprisingly my main problem has been the redefinition what I expect of myself. My expectation used to be that at the end of the day there would be a small pile of code or images that are production ready and in the git repo. And suddenly my output is my thoughts, opinion or influence sometimes in wireframe form and sometimes verbal so how can I as an employee measure my own performance when I have nothing to show for it? We are a tiny company so of course I still cut code and I still photoshop production images from time to time but that only appears to confuse my poor addled maker brain and skew the teams agile velocity.
Part of navigating this transition will affect this journal as well. Not sure how as yet, it may increase the frequency of code and imagery as I’m no longer getting my ‘maker’ fix at work but the more text based output may well increase as I practice the art of expressing an opinion with words rather than through a medium such as music or design.
Using RWMidi library I quickly converted the Game of Life example that comes with Processing to work with the Novation Launchpad.
import rwmidi.*;
MidiInput input;
MidiOutput output;
int sx, sy;
float density = 0.5;
int[][][] world;
int note;
void setup()
{
size(8, 8, P2D);
frameRate(5);
sx = width;
sy = height;
world = new int[sx][sy][2];
input = RWMidi.getInputDevices()[2].createInput(this);
output = RWMidi.getOutputDevices()[2].createOutput();
// Set random cells to 'on'
for (int i = 0; i < sx * sy * density; i++) {
world[(int)random(sx)][(int)random(sy)][1] = 1;
}
}
void draw()
{
background(0);
// Drawing and update cycle
for (int x = 0; x < sx; x=x+1) {
for (int y = 0; y < sy; y=y+1) {
note = y * 16 + x;
//if (world[x][y][1] == 1)
if ((world[x][y][1] == 1) || (world[x][y][1] == 0 && world[x][y][0] == 1))
{
world[x][y][0] = 1;
set(x, y, #FFFFFF);
output.sendNoteOn(0, note, 127);
} else {
output.sendNoteOn(0, note, 0);
}
if (world[x][y][1] == -1)
{
world[x][y][0] = 0;
}
world[x][y][1] = 0;
}
}
// Birth and death cycle
for (int x = 0; x < sx; x=x+1) {
for (int y = 0; y < sy; y=y+1) {
int count = neighbors(x, y);
if (count == 3 && world[x][y][0] == 0)
{
world[x][y][1] = 1;
}
if ((count < 2 || count > 3) && world[x][y][0] == 1)
{
world[x][y][1] = -1;
}
}
}
}
void noteOnReceived(Note note) {
println("note on " + note.getPitch() + ":" + note.getVelocity());
if (note.getVelocity() > 0) {
int nx = note.getPitch() % 16;
int ny = (note.getPitch() - nx) / 16;
output.sendNoteOn(0, note.getPitch(), 100);
println(nx);
println(ny);
world[nx][ny][1] = 1;
}
}
// Count the number of adjacent cells 'on'
int neighbors(int x, int y)
{
return world[(x + 1) % sx][y][0] +
world[x][(y + 1) % sy][0] +
world[(x + sx - 1) % sx][y][0] +
world[x][(y + sy - 1) % sy][0] +
world[(x + 1) % sx][(y + 1) % sy][0] +
world[(x + sx - 1) % sx][(y + 1) % sy][0] +
world[(x + sx - 1) % sx][(y + sy - 1) % sy][0] +
world[(x + 1) % sx][(y + sy - 1) % sy][0];
}
First thing I did with my new Launchpad? Nope, I didn’t hook it up to Live and jam. I hooked it up, fired up Processing and worked out how to light up the pads and recieve the notes so I could hack it to do anything I wanted to.
One thing I noticed straight off the bat is the note numbers are not incremental. Each row starts 16 higher than the last - I assume to allow for the same firmware to work in a 16 x 16 layout model - and there is a very simple way that reminds me of the technique of bitshifting I used back in the bad old days of DVD authoring to convert from one to the other.
// To go from XY to the note number noteNumber = y * 16 + x; // To go from the note number to the XY x = noteNumber % 16; y = (noteNumber - x) / 16;
Hopefully that will save someone ten minutes of head scratching.
Followed the great step-by-step from here.
But there’s one bit missing to get it working if you’re getting the 403 errors.
You need to run the following command to tell Plesk to use your spanking new vhost.conf
/usr/local/psa/admin/sbin/websrvmng -u --vhost-name=<domain>
