josh kopeček

Two days and our performance is in a building in the town square. I’m not altogether absolutely happy with the improvisation yet. I would like it to be a bit more impromptu, but it’s starting to move in the right direction. I have to learn how to use a new tool, come to terms with performance using it and create all the materials. In a week of working with it I think I’m starting to get the hang, but I need to think about the structuring of the materials in terms of shape. At the moment I’m controlling textures/atmospheres using the USB controller, but it would be nice to have something more manipulatable, like my coming controller/interface. A range of envelopes/gestures/morphologies would be nice, rather than variations on a slow fade. Also, there is very little I’m controlling in terms of timbre, at the moment it’s done my simply changing to a different sound and combining layers, a very flexible and fertile procedure, but limited in terms of the amount of time it takes to switch sounds. This could be done automatically, but it would be nice (and very possible) to do this in real-time.

I’m not sure we should ever move away from the sense of a computer being a servant, in fact it has been proven that it is very difficult for it to be any more. It is very effective as a servant, and has the possibility to extend our actions and interpret our behaviour, but not act reliably of its own volition. It is only as useful as the information we put into it.

So. Just a few notes again from Sicily. Angela and I are working today towards our performance on Friday. The Mexicans arrived yesterday, and we’ve been exchanging a few dialogues. They are thinking of doing another festival with Cadmium next year in Mexico City, and would like to do a concert at Chop of alternative musics, as part of it. Personally I’d love to go to Mexico and perform, and I really like their style. It’s a kind of dancefloor mixup, made with traditional instruments – flutes, ocarinas, saxophones, etc.

I have finally got a good improvisation going with Live and some control routed through Max. Angela is happy with it too – there are some good sounds coming out, but I have the feeling that it sounds like ‘Live’. There’s a particular sound that the Session time stretcher has – both the granulator and the fft, which doesn’t make it ideal for live performance, but then Max is so convoluted that to do some of the most basic things that Live does would take months to program in Max, and even then it would be very unstable. I think a lot of people just end up rewriting some of the basic features of Live in Max without realising it, but then there is this element of direct control over sound that Max/MSP has over Live. If one could get all the interwoven signal processing features of Max, i.e. let every element be able to control another then it would be perfect. The problem is making a system which is not too difficult for the average user, yet powerful enough to satisfy the needs of the complex musician. There is a compromise between the two, that one can pass audio and MIDI between them, make plugins for DSP in Max and control what you will, but as always there are barriers to overcome.

One cannot always create all the sounds they would like to use in real time, a lot of preparation in terms of sound making must be done before the process starts. Preprocessed sounds are infinitely better to listen to that real-time processed ones, especially in the live instrument processing domain. As soon as you introduce a computer into the equation, the sound quality dramatically reduces.  Like João Pedro was saying, he has to create all his sounds in the studio before trying to do it live. He has not heard a solution for live processing that works, and I trust his opinion. However, in that domain one is comparing studio created work performed on stage to live work created on stage, and there is a fundamental difference there.

Evening:
It’s very loud and busy here at the moment. Kukultech (the Mexicans) are working on some material, practising and performing. Andres has discovered his camera, the one he uses for everything, is completely broken and is going to cost a lot of money to repair. It’s extremely hot and humid, difficult to work in circumstances like these, and my computer is complaining a lot – not overheating but struggling to keep cool.

I start to think that it would be easier to live somewhere that’s not England in order to do workshops, but then there are so many barriers in the way that it would make the difficulties in Britain seem trivial. But, saying that, Tullis is moving to Barcelona and leaving (almost) everything he has established in Britain behind.

So, the first of the entries from Sicily. We’re working on a new performance, Textures, and Angela is having difficulty working both on Ninnananna and this too. I am having problems getting SuperCollider to work with basic MIDI protocols, which seems crazy, but I’m going to start and see if there is a way of doing it with RubyCocoa instead. I think I’d be spending less time doing really basic things instead of wasting my time on such a convoluted and non-intuitive programming language.

Some thoughts: visual feedback (and tactile, I’ll get to that) seems such a basic thing that many of the programs for advanced musicians working with computers are incapable of supporting in a simple and stable way. SuperCollider has crashed a number of times when I’ve been here, doing simple things, it is buggy and prone to disaster. The problem is that ixiQuarks would be a great tool to use live with MIDI control, if only it worked. Live seems to be much better (more stable), but is too fixed around the quantised loop paradigm. If I could control external AUs from within, and liberate the loops from the ‘warp’ function, it would be perfect, especially by routing all the audio through SoundFlower, from ixi, Max etc. With Max/MSP again, there are problems with control, even though it is quicker to program than SC.

The solution at the moment is:

Route all audio through SoundFlower – Jack destabilises programs and cannot remember setups.
Send ixiQuarks audio through MaxMSP, in order to control levels etc using MIDI.
Use loops and sequencing in Live in order to maintain a structure in performance.

The solution in the future is:

Have a standalone control program – Max, SC, RubyCocoa – that sends data (MIDI/OSC) to the others.
Have multiple sound producing programs – Max, ixiQuarks, SC, MetaSynth – that are controlled by the above program and routed through soundflower.

That way we can have an external, controllable fx rack and keep everything as independent as possible, and finally integrate external interfaces such as the Arduino.

* OSC and Ruby – already sorted
* MIDI and Ruby
* Audio and Ruby