Monday, August 19, 2013

Parallel Compression explained (Part 2)

In my last post I gave you guys a run down on the function of downward compression, now let's get into the real fun and complicated stuff.  Upward compression.  Phonetically, linguistically, and logically the words make sense together, upward as in compressed from the bottom instead of the top.  Imagine the bottom as the bottom of our input-output plot, and the top as, well, the top.  The bottom of the graph correlates to the softer parts of the signal, while the top is the louder parts.  So upward compression is basically taking all sounds UNDER the threshold and BOOSTING them by an amount determined by ratio.

Upward compression plot showing a threshold of -17.5dB.  Notice how the louder sounds are unaffected while the softer ones are boosted.
Now, the question is, what makes this any better than regular old downward compression?  I mean, they're essentially accomplishing the same thing with the dynamic range of the signal, right?  Yes, they are, but louder portions of signals contain very important transient and sound envelope information that are ultimately affected by downward compression.  Thus, upward compression gives us the best of both worlds; unaffected transients and a more malleable dynamic range that we can play with.  Unfortunately, machines that can do this effectively are few and far between and that's without mentioning the fact that with this operation your noise floor rises from the grave, rearing it's horrific head.  For example, think about a noise floor of -80dbFS and a threshold set at -20dBFS.  Now even with a gentle 2:1 ratio your noise floor will be raised to a very intrusive -50dBFS.  So now you understand why the practical application of such a device is unrealistic.

This is where parallel compression comes into play.  By mixing a dry, uncompressed signal with a compressed signal you can mimic the activity of upward compression, even though it isn't true upward compression.

You'll get a plot that looks more like this with one parallel compression track

Now, for those of you who want, or care to have, a real expert explanation of this go here.  Otherwise keep experimenting with it on your own and be sure to check out my video tutorial on how to set it up in Pro Tools.  Happy mixing!

2 comments:

  1. Kevin,

    When I say I am not a fan of compression, I am really saying that I am not very good at using compression. I have used compression as an effect before that turned out pretty good, but for the most part my compression turns out pumping. Thanks again for the parallel compression tutorial, it is definitely something I plan to try in the near future.

    Kyle Schroeder
    Tubes and Transistors

    ReplyDelete
  2. Hey Kevin, awesome explanation of parallel compression. I would like to say that I learned something from this and I am thankful for the information. The tutorial was great as well. Thanks.

    ReplyDelete

Comments