This is the second installment of a sound a day. It is a bit longer. I have now implemented a continuous randomization of the note values and durations in the sequencer, as well as the delay times in the K-S algorithm. It has an intriguing quality now. It never repeats and yet is quite rhythmic which makes your mind try to assign patterns. There are fleeting bits of pattern that are tantalizing without actually going anywhere and yet I find it quite fascinating to listen to. I have created two copies with independent timings and sent one to the left channel and one the right. The polyrhythms and occasional harmony add to the nagging feeling that something is happening that makes some sense, and yet it is driven by random number generators and filtered noise generators. It feels a bit fractal to me although there is nothing really fractal about it. It will be interesting to see if adding a fractal nature to the “patterns” sounds more interesting. By this I mean creating nested rhythmical and harmonic blocks. I’ll give it a go and post the results here soon.
This is just a short sample of a very basic (and mostly uninteresting to listen to) sequence. What IS interesting about it is the way it was created. I used Pure Data to generate the sound and create the sequence. There are no samples used here, and no “soft-synths” although strictly speaking I created a simple “soft-synth” here. There is only one sound source here, a sine wave oscillator being fed back into a 2nd order lowpass filter and a delay line. This is in fact the Karplus-Strong method of physically modeling a string. I have left the burst of sine wave very short here so the attack is a bit stronger than it would be with a normal string I think. But it’s quite remarkable how nice it sounds, with so little going into it. There are no additional effects. This is the dry sound coming out of Pure Data. I have attached the patch for anyone who is interested.