Objects for AM
  • There are so many cool things demonstrated on this site but I'm stubbornly starting with basics and working my way slowly through. I am trying to demonstrate amplitude modulation as it transitions from creating rhythm to creating harmonic (or anharmonic) effects. Based on something I read a while ago it seems the there isn't a VCA function as such and the recommendation was to use the Level object.

    As the attached video demonstrates using an Osc output to modulate the Level object doesn't move the gain to a zero state while manually adjusting the control does.

    https://drive.google.com/open?id=1IGG3DoLwTPhitR4y5ui5n1tHIPCO5PZh

    What am I missing?
  • You are sending the level node a bipolar signal. The level node does not have a clamped input and the knob won't display a negative value, so you should either send it a unipolar signal or clamp the incoming signal. Probably the easiest way is to use the expression (x+1)/2.

    Screen Shot 2017-12-03 at 3.49.13 PM.png
    971 x 702 - 105K
    Level Node Unipolar Input.audulus
    25K
  • Robert, Thank you for the explanation. Problem resolved. Is it true that the Knob's response to a negative control value is to switch fully on while showing the Knob control position as fully off? Trying to make sense of what I see and what I hear. Taking the absolute value of the input perhaps?

    Ds
  • image

    So it turns out the level node is a different visualization of the multiply node. When a negative value is received by the knob's input, it multiplies the negative value and inverts the waveform. But because we can't aurally perceive phase inversion on it's own, it just sounds like a click or pop.

    The knob display can only show the min/max value it has been set to. If you right click on the knob you can see the range of the knob can be adjusted. I would recommend always leaving it set to the default 0-1. However if the signal it is connected to is outside that range the knob will appear to be fully on or fully off.

    If you wanted to model what a typical VCA does, like a Doepfer A-132-3, you would configure the patch something like this image

    You can see, the negative signal is clipped and anything exceeding the maximum parameter is also clipped.
    Screen Shot 2017-12-03 at 6.05.24 PM.png
    1878 x 1034 - 327K
    Screen Shot 2017-12-03 at 6.15.18 PM.png
    966 x 579 - 78K
    A-132-3 type VCA.audulus
    21K
    Screen Shot 2017-12-03 at 6.21.21 PM.png
    220 x 171 - 19K
  • Lots to absorb. Thank you.

    Ds
  • I don't mean to give you information out of a fire hose. So If there is something unclear or confusing, let me know. :)
  • The rate of information is just fine but I need to play around with it and see how it works.

    The problem as I understand it is that the Level Knob really wants a signal between 0 -1 and the modulating Osc provides a signal from -1 to 1.

    The Knob takes this negative portion of the control signal and inverts the phase of the input signal. The consequence of this is that if both the modulated and unmodulated signal are mixed together the original signal is canceled out during the negative part of the cycle. However if the modulated signal is listen to directly where there was silence there is the phase reversed original with a click.

    The first expression you showed takes the modulating signal and makes it all positive. The result is that the modulating wave is reduced in amplitude, the wave shape is undistorted and the gain of the amp is fully controlled.

    The second expression with the clamp formula which is designed emulate a Doepfer A-132 clips the modulating signal causing it to flatten somewhat. I tried the patch on my Doepher A-131 exp VCA and it didn't exhibit the nonlinear behavior but maybe that distortion on the A-132 is a feature :).

    Thanks for all your help. This is slowly beginning to make sense.


  • @robertsyrett thanks for explaining this so well!

    @dsu let us know if you need any more help! :)