Topic: Boids and midi update

[BlinkenLights]

agentC, 10 seconds should be plenty.

if i can get em into laserboy.. that would be sweet.

[Agent C]

I think I'll do that.  I can just switch color mode from RGB-1.0 to HSV-1.0, 1.0 denoting the max range of a value.  Then I'll just compute the heading from the two component vectors.  I'll kick out about 10 seconds tonight or tomorrow as an unformatted wav.  Laserboy can work with those to some extent, right?  From there I'll work on getting a formatted wav.

[BlinkenLights]

wait.. LaserAgent, LaserWave and LaserBoy.. all working witH LB formated wav files...

thats 3, 3's the magic number. least thats what schoolhouse rock taught me.

[James]

Laserboy can work with those to some extent, right?  From there I'll work on getting a formatted wav.

Yes. LaserBoy is getting pretty good at making the most out of imported "foreign" waves.

Actually, if you turn off any optimization that would change the number of points per frame, then LaserBoy could import it perfectly.

To import foreign wave data into LaserBoy you set a points-per-frame number. If that is a constant in the source wave then the import will be perfect.

James. 

PS. Hey Blinken, dralva knows how to write to LaserBoy formatted wave too!

[Agent C]

Ok, here are a few seconds of a laser boids flock with heading based color modulation.  There are 20 boids, each a single point, no dwell time, no optimization.  The wav file is 16 bit, 6 channels, channel 6 should be silent, 30000 samples per seconds.  I tried to import into Laserboy but didn't have much luck.  I haven't really tested the disk based wav writer so the file could be garbage.  Tomorrow evening I will try outputting an optimized multichannel wav and playing it back via VLC.  For your experimental purposes: http://meoem.com/boids.wav

[BlinkenLights]

Agent FYI i found this cool page online...

http://mindprod.com/products2.html

[James]

Excellent!

This is yet another great example of a "wave-from-outer-space"!

My code is all busted up right now, but I tried this in an earlier version and I don't get what I should.

So that's a good thing!

I've got more to fix and a great example of wave that I need to figure out how to make use of!

Thanks!

Once I get through this rebuild, I want to start posting all the detales about the LaserBoy wave header extensions and data sub-code markers. I'd like to get some input from anyone who might also be interested in working with this format.

James. 

[BlinkenLights]

unless the wav is crap..
he said he had not tested yet.. play it on an o-scope first.

[James]

It looks fine in Audacity.

[BlinkenLights]

does that do 3d view? maybe the math is screwed..


or maybe laserboy is screwed

[James]

It looks like a track assignment issue.

This can happen if a wave header reader starts to read the formatted data at the wrong location.

If I invert the wave in LaserBoy the channels get screwed up.

James. 

[BlinkenLights]

so the pointers in the header are pointing to the wrong places..

[James]

Not nec.......

I'm not ruling out that LaserBoy is goofing up the header read.

Like I said, it looks OK in Audacity.

I'll have to take a look at it in a hex editor.

James. 

[Agent C]

There won't be any 3d data unless I'm missing something.  I'm not totally clear on writing multichannel waves.  The data is interleaved right?  So 1 frame of data should be short X, short Y, short R, short G, short B, 0

[James]

Yes. That's right.

signed short int

{X, Y, R, G, B, 0}

What does your wave header look like?

James.