why does cubing my audio signal not do as expected

this is a simple question that probably comes from a fundamental misunderstanding on my part of how reaktor works

i try to cube an audio signal to shape its envelope (x^3). i try this with a sine osc and get no audio out. tried in primary and core - why doesn't this work?

 

I don't know what you expect it to do, but think about the math.

Start with a sine of amplitude 2, i.e. varying between -1 and 1.
Multiply it by itself and (-1 * -1) = 1 and (1 * 1) = 1. Values in between 0 and one get smaller, e.g. (-.5 * -.5) = .25, (-.25 * -.25) = .0625. The frequency is double the original signal.

The square ends up being a signal of amplitude 1, with double the frequency of the original, varying between 0 and 1. Multiply by your original signal again, and you get a signal varying between -.5 and 5, with the same fundamental frequency as the original, plus an overtone.

Now start with a signal of amplitude 1, varying between -.5 and .5. You end up with a signal of amplitude .25 varying between 0 and .25. multiply it by the original again, and you get a signal with an amplitude of .125, with the same spectrum as our (-1,1) signal when it's cubed.

So cubing a signal produces a new signal with an amplitude that is a fraction of the original.

I suspect you're starting out with low amplitude signals and ending up with a new signal with a much lower amplitude.

 

yeah i expected it to reduce the volume and introduce some distortion/shaping but how come the frequency changes? i thought since the zeros would be in the same spot the freq would be the same.

i probably ****ed up with the connections which is why i couldnt hear anything when i tried (not at my reaktor pc when posted)

 

I tried running a sine oscillator, amplitude 1 into a multiply module with 3 connections, and then the output of the mult into the output of the instruments. I got plenty of output volume.

Check to make sure you use an audio voice combiner before going to the output of your instrument. Or set the polyphony of the instrument to 1. The 'silence' problem usually comes from trying to push polyphonic signals into monophonic inputs.

 

Maybe it's not doing what you expect because you are expecting the wrong thing.

You don't cube audio to make it louder, you multiply it by a constant value - a positive value. Think about squaring. Squaring audio is the same as ring modulation - it's multiply two signals together. Ring modulation changes the harmonic content - regardless of whether the two signals being multiplied are the same or not. Cubing is multiplying then multiplying - that's like ring modulating twice.

Have you looked at the signals on an oscilloscope in reaktor? That's always the best way to learn about the math of audio signals.

- CList

 

...after re-reading your post, I think what you want to do is cube the output of an envelope *FIRST* and then multiply the result of that operation with your audio signal.

- CList