(Continued from part 1, part 2.)
I decided to voice the melody using the vibes opcode. Here's where I met my first major problem ... and made a major discovery.
First the problem. No matter how I try to voice the vibes -- filtering, enveloping, and so on -- I get a lot of transient noise in the initial strike. I don't mean natural vibe 'clank', I mean the sound of a digital waveform leaping point-to-point. Some call it the 'zipper' noise. It's also an annoying crackle, like someone at the theatre carelessly wadding up a cellophane candy wrapper.
But then, the discovery. I found out about being able to include necessary files within the unified Csound file, as Base64 attachments!
That's one of the coolest features I've seen. No more files that stop working when they are moved from their home directory, copied, or sent to someone else.
So, here it is, version 3 of 'Mini-LaVie'. I've also corrected a number of minor boo boos, for instance, having the reverb based entirely on the left channel :-)
| Attachment | Size |
|---|---|
| mini_lavie3.csd | 7.83 KB |
Including Files as Base64
So happy to be able to run your example and be reminded about the ability to
include the necessary files in the .csd. That is a great and often overlooked feature.
I'll share it with my students for sure.
-dB
Base64 inclusion
Base64 inclusion is probably the coolest feature I've seen recently, especially for audio snippets. I think the story of Csound is that you can always find something new and unexpected if you keep your eyes open.
I sidelined the final version of La vie while I learned a few granular synthesis tricks and ran some microtonal experiments. I had an idea to reharmonize this song in 19-TET, but I got sidetracked by trying to turn the ocean into an instrument. I'm finding it's almost easier to create an entirely new song than reinterpret an old one -- especially one that has a definite voice and texture.