Soft-knee compression

So, you think you’re pretty familiar with compressors. You know how they work, what they do. You know what the basic controls do – attack, release, ratio and threshold. But maybe you’re not sure about ‘knee’. Something about being soft? What is knee, and – more importantly – what is it useful for?

Normal compressors behave depending on the level of the audio. When the audio is below the threshold, nothing happens. When the audio is above the threshold, the level is reduced by an amount related to the compressor ratio. This is what’s called a ‘hard knee’ – there is an instant transition between unity ratio (1:1) and maximum ratio (whatever the ratio knob is set to). Soft knee, by comparison, causes there to be a gradual transition from unity ratio to maximum ratio. This means that some audio that is only just below the threshold may be compressed, but at a lower ratio than audio above the threshold.

It’s useful to think about soft knee as a kind of variable ratio. It’s especially useful when you want low-level signals to be compressed a little, but high-level signals to be compressed more aggressively. A good example of this is a vocal performance where most of the audio is fairly even but there are some syllables that are quite loud. If you use hard knee compression, you’ll have three options:

  1. High threshold, high ratio. This will control the peaks, but the rest of the performance will be uncompressed. You might need a second compressor to apply some more gentle leveling.
  2. Low threshold, high ratio. This will bring all the levels into check, but may sound too aggressive or choked – especially if you’re aiming for a more natural sound.
  3. Low threshold, low ratio. This will give you the gentle levelling, but the loud syllables will still pop out due to the low ratio.

Using a soft knee, you can apply some gentle levelling compression, while still controlling the loudest peaks. Set the ratio to the maximum you’ll need, and gradually lower the threshold until the lower-level audio is being gently compressed. That way you’ll have the gentle levelling with a low ratio, but the louder parts will be compressed at a higher ratio. It’s a little like having two compressors in series (one reducing the peaks, the other just gently levelling the audio), but in a single compressor. It’s also easier to control because it’s less complicated, and it’s a little more sophisiticated because the transition between low ratio and high ratio is smooth.


    • Jacaré
    • July 30th, 2009

    Hello Kim
    Thank you for explaining this. Never thought of the soft knee use like that, thou, I always used the hard knee aproach to percussive sounds. Always check your blog, since you write in very explicative way. Thank you for giving away some of your knowledge.

    • MOK
    • November 10th, 2009

    Nice, that’s definitely something I was always a little less clear on. I knew the basic idea, but what really helped was the way you illustrated the three typical categories of compressor use, and where the soft knee tends to fits in relative to those. Kudos again.

  1. No trackbacks yet.

Comments are closed.
%d bloggers like this: