The basic concept behind it is that most “classic” dance music is so formulaic it should be easy for a computer to generate. Generating all patterns and sequences from a small set of rules the unit is able to deliver new and original tracks in realtime.
The original implementation ran autonomously but the tracks lacked the (often criticized as predictable) build up/down structure so critical in forming a flow. Rather than trying to implement such a complex concept in such a simple and elegant piece of code I decided this task should be performed by a human.
Here’s what it sounds like using a trance sound set:[display_podcast]
Infinite Horizon features 5 sequencer channels:
- Chord This forms the base from which all the other sequencers derive their notes.
- Lead 1 The most complex of the channels, it is capable of generating different length patterns and variations on loops of the pattern.
- Lead 2 Based loosely on Roland’s TB-303 sequencer, ties and slides included.
- Bass 1 The most simple of the channels, it was the basis for and shares all of its code with Lead 1.
- Bass 2 Similar to Lead 2 but biased specifically for basslines.
As the picture shows, the controls are minimal. The big red button arms all channels to generate new patterns at the end of the 8 bars, effectively creating a new track. For each of the 5 channels, there are 3 controls:
- Mute Deferred latching switch with LED indicator.
- Forward/Back Generates a new pattern or restores the previous one.
- Clock Shift Slides the clock by a 16th for that channel.
This is not a commercial product at this time, but goes into the ‘Unobtanium’ category.
What do you think of the idea of generative sequencers? Do you see this as a crutch for creating more formulaic dance music or as a useful live performance and compositional tool?