more generations

So I took the sound of knocking a piano and turned it into another layer of material for the piano piece I am working on. Turns out this piece is going to be a generative piece, as in it will be partially generated in real time.

Planning how the will all happen has kept me up way to late, but I think I have a plan, but I won't know until I check it all out in the light of the sun. Bring the sun to bear on the illusions of the deep darkness of night as it were.
pfNOCK


Now to generate some notation from this stuff and then sew it all together.

I have been thinking about how this is both a combination of things that I have made by hand and things that are being composed/compiled in real time. I like this mix control and lack there of, feels like improvising a bit.

pfTINK

I have been making some progress on a piece for piano and electronics. I spent some time making a large bank of sound files. These were all made by hand. I then created an automation system that reads through a set of 10 of the 75 files 20 times. There are three streams of audio, one begins in the 1-10 range, one in 11-20 and the other 21-30. These all advance to the next set of 10 after 20 files have been read. This continues until they have each reached the 70 category and run out of files. The choice as to which file is read is made by the sneak~ object (1994 by Karlheinz Essl), which operates thusly;

A bang in the left inlet makes a random selection of a supply sent to right inlet before. Note that there are no repetitions of the same element and that the chosen elements are always neighbors of the given supply.

Finally, the audio is intermittently run through a convolution reverb where the impulse response is that of an upright piano being struck with a ceramic spoon. This in effect makes the piano sounds sound more like piano sounds. This part was not automated as I simply wanted to do it by hand.

pfTINK


A much cleaner patch than the piece yesterday.

As a finished piece it lacks the contour and counterpoint of a bass line. Also, it is a bit consistent in volume. That all said, it is really a study and as such, I rather like it.

Envelope Piece

I was thinking about the creation and use of envelopes and made a little study piece. The breakpoint windows at the top of the patch are the piece in essence. I must admit that I was more interested in exploring the textures than any coherent compositional plan, but it sort of works. Right now it is very monophonic, but that would be another level of complexity and I didn't want to get to carries away.

EnvPiece


Most of the signal processing is pretty straight forward, using a modulo (% in Max land) and then multiplication of the signal in order to provide frequency control data for a saw tooth wave, rectangular wave as well control for amplitude. There is a lot of recycling, which makes the patch for messier than I ought to allow, but it is also fun to just make a piece in a few hours.

In the end, I now have a some techniques for generative music that I wanted for control of the Viola piece for Shannon and I have what turns out to be a pretty robust and flexible sketch pad for seeing and hearing ideas. One improvement that I should attempt is aesthetically interesting visualization of the data. But I grow fatigued.

Ok, less fatigued, but the video capture combined the 2 audio streams, there-by creating phase cancellation all over the place. It has an interesting sounds, but not really the right sound. Listening back and forth to the video and the audio is quite illustrative, phase cancellation can destroy stuff.

eMusic Technique 1

I got thinking that there are so many different little techniques that one can use in electronic music and that I might catalog some of them. I will make an effort to post more going forward. Sort of a cookbook of Electronic Music techniques.

Today I came accross a post for a conference on the duality of Binary. The first that entered my head is the fact that a great deal of electronic music is in two channels. Of course, it could be mono, or it could be n-channeled, but 2 channels is the predominant paradigm. One reason is that once you go beyond the duality of two, the gain is not as significant as from 1 to 2. (I am ignoring the fact that we all own stereo's and rather am considering why is that we own stereo's with 2 instead of 4 speakers, etc etc). There is also the fact that we have 2 ears, an obvious but profound statement. We do transduce a some auditory information through our bones and flesh and it is no surprise that the senatorial aspect of the sub woofer is a prominent feature in some music's.

But this is all an introduction to my thinking about the fact that when we produce electronic music, the resulting file is stereo. That said, when one commits global changes to 1/2 of a mix, it is a very radical and interesting thing. Hence, I made an experiment with altering the playback speed of 1/2 of a mix. The initial sound is something that I generated some time ago (not sure I remember how).

Source


I took this sound and split the stereo file into 2 mono files and then altered the playback speed of one file to slow and speed up. This is a shot of the max patch


I recorded the output and put this file along with the unaltered other channel to create a new version.

AlteredV1


This is a vary simple operation, but I think the results are rather striking.

Another means of accomplishing this is the delay of samples per channel. Here is a version I through together in Logic where the right channel has the a sample delay envelope. This does not alter the pitch as in the previous version and so the effect is different.

AlteredV2


Here is a screen shot of the logic session. The delay on the right channel is represented by the orange line in the second track.