Ableton Forum

Despite having toyed around with it for a few years I'm a relative noob when it comes to all things Max so bear with me here.

Basically, I'm doing a music video for an ambient tune that is very long and slow-moving. Variations in tone, frequency and volume shift at the pace of a lava lamp and I'd like these glacial variations to be reflected by some of the visual characteristics of the track. The contours of the track as a whole are pretty interesting - the amplitude increases steadily over the duration, with a couple of small peaks and valleys, and the track is comprised of several competing harmonics that drift in and out in different configurations.

My initial idea is to have the amplitude mirrored by the brightness and saturation of the images, with higher luminance and saturation values for higher volumes. In addition, though it might be more difficult to implement, I'd like to have the frequency characteristics mirrored by slight variations in colour. I guess this would require devising a scheme by which the full frequency range covered in the song will relate to a colour scale.

My question is, what is the best method of achieving these results? What software will be of use? I realise that Max 4 Live probably won't be much help (although Max itself might) but thought this would be a good place to seek counsel on the matter!

Feel free to have a listen to the track here.

- Joey

I do something very similar during my live shows.

http://www.synnack.com/blog/post/25/0xf8-phase-2

I am using M4L devices to send float data to a second laptop using udpsend based on frequency and amplitude data of specific tracks. The floats are used to drive effects in Jitter.

Amplitude is determined by just sending the audio in to a live.meter object which kicks a float out the other side.

Frequency data is gathered by using an fffb~ object. You don't really need more than M4L. You don't need 2 laptops to do this. I just do that for less risk on stage.

Thanks Tempus3er. Starting from scratch, I've tried following your methods and have got part of the way. But there's a couple of things I still don't quite get.

First, I've got the audio track coming from Live going into two live.meter objects (one for each channel). I'm not quite sure what to do after that. Where exactly can I send the float data to translate it into something my video editing software could understand? Or is this the wrong workflow? Would my process be different to yours considering I'm hoping to get my NLE to interpret and apply the data to video before rendering and exporting?

Second, I've managed to create a fffb~ object but again am faced with the same issue - how to translate this into frequency data that my NLE can understand.

Many thanks for the help so far!

I was looking for some info on something else and came across this. To answer your question - you can send data to the other source using midi or OSC data. I do not know the video software you are using, but it would need to be able to accept and interpret data coming in using one of these formats.

I would imagine OSC would be the more likely, so to send this data, you would use a udpsend object in Max/M4L. Within this object, you specify the the IP address of the computer you are sending data to (IP address 127.0.0.1 if the receiving software is on the same computer), along with a port number and send your float in there. Then presuming the software you are using can accept OSC data, you choose the same port number to look for the data coming in. In Max, you would use a udpreceive object, specifying the port number. The same float should come out of this object.

There's some useful information on OSC, along with some pretty useful Max objects here:

http://cnmat.berkeley.edu/

Hope that helps.