Looper's Delight Archive Top (Search)
Date Index
Thread Index
Author Index
Looper's Delight Home
Mailing List Info

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

Re: Mobius scripts audio examples



Simeon, you are much farther along on scripting than I am. Can you actually specify a non-semitone value?
Also, I'm thinking the math is not quite on -that the rate factor should actually be the reciprocal of the values in the /cycle (1/ratio) column eg. half triplet = 1/.33333=3
k

On Thu, Mar 3, 2011 at 12:26 PM, Simeon Harris <simeonharris40@googlemail.com> wrote:
well i tried it with some simple rhythms and the results were pretty unusable using my exisiting rate shifts. i guess with more thought as you've laid out, you could possible do something pretty cool.

one thing i'd really like is the ability to predict how a rate shift will affect the length of a loop and therefore calculate multiple rate shifts so that the overall loop length is unaffected. i tried to work it out (the math isn't particularly hard), but the results didn't work in practice...

sim


On Thu, Mar 3, 2011 at 6:00 PM, Keith Smith <kahsmith@gmail.com> wrote:
LOL again - The things that happen to loopers in bed:

I started thinking about the purely rhythmic possibilities of this. Let's use a percussion sound instead. Same thing; record it on 3 tracks. Now rate-shift them to generate polyrhythmic patterns.

So, I made a spreadsheet to calculate the frequency ratios over 2 octaves. The reciprocals of those ratios give you the time values of the rate-shifted tracks, so if you consider 1 cycle to be a whole note (4 beats) in 4/4, then rate-shifting +12 gives a 1/2 note, +24 gives 1/4. OK. Now shifting +7 gives you very nearly a 1/4 triplet, +17 very nearly a 16th, +19 a 1/8 triplet.

The last 3 would drift, of course, because they're not exact, but they're pretty close, so they might be useful for a short time, or let 'em drift and deal with it. It does make me wonder: If you could script to the exact rate-shifts for various note *time* values, then all kinds of neat mayhem could be created very quickly by using Simeon's method of recording a single sound (or pattern) on three tracks at once and then rate-shifting the other 2 tracks. I'm thinking 3 tracks for chord manipulation and 3 more for percussion sounds, leaves 2 for melodic looping. Of course you could use different sounds on each track.

The spreadsheet isn't exhaustive (pdf at http://www.samplesmith.com/LD/Rate-shiftingForPolyrythms.pdf ). It's just a germ of an idea that needs a lot more thought. -if anyone wants the .xls file, just email me -or let me know if I'm re-inventing the wheel ;)
k