Hi, I've been producing for about 6 or 7 months now and I've been desperately trying to grasp compression in it's entirety but I just cannot seem to get a grip on how the threshold knob does it's job. The confusing part is the way dB works. I've noticed that 0 isn't the minimum value and this makes such a headache for me. Also, with the compressors I'm using, I seem to be unable to set the threshold to a positive value.
I've tried fiddling around with the controls and learning by ear, but that's a little challenging with the equipment I'm using. I guess my question is sort of centered around understanding dB levels. Can anyone make this less confusing for me? Thanks in advance!
Help with compression (threshold)?
Re: Help with compression (threshold)?
Just open up your daw, play a constant tone, look at the meter and start changing its volume. Even a simple look at the meter will tell you that 0db is the maximum volume.
Exilium wrote:distorted square
Re: Help with compression (threshold)?
If you are using a dynamic compressor, the compressor will affect the sound spectrum above the db selected in the threshold,
- reducing that section proportionally to the chosen radio (4:1, 2:1, etc)
- with an attack (delay in the activation of the compressor)
- and a release (delay in the desactivation of the compressor)
- reducing that section proportionally to the chosen radio (4:1, 2:1, etc)
- with an attack (delay in the activation of the compressor)
- and a release (delay in the desactivation of the compressor)
Re: Help with compression (threshold)?
1. It doesn't affect the spectrum.K4OS wrote:If you are using a dynamic compressor, the compressor will affect the sound spectrum above the db selected in the threshold,
- reducing that section proportionally to the chosen radio (4:1, 2:1, etc)
- with an attack (delay in the activation of the compressor)
- and a release (delay in the desactivation of the compressor)
2. Attack/release time doesn't delay the activation of a compressor. It means how quickly it will act. And this isn't the same as delay.
Exilium wrote:distorted square
- Turnipish_Thoughts
- Posts: 684
- Joined: Wed Apr 20, 2011 12:34 pm
Re: Help with compression (threshold)?
sounds like you don't know what a decibel is. it isn't an absolute value, but a difference between two values, specifically a logarithmic scale that represents orders of magnitude changes in sound pressure level micropascals. (hamfisted explanation but I hope you get me)
0dB is your highest value because that's your clipping threshold in digital audio (the point at which the wave sample's value would exceed the bit width, values higher than this can't physically exist so the audio engine 'clips' the sample to the maximum applicable value), the clipping threshold is therefor a handy reference amplitude to calculate differences in volume from. so in digital audio, something being '0dB' means there's no difference between the amplitude of your audio and the maximum amplitude capable within the digital domain (things will never get louder than this, increasing the fader will just add clipping distortion to the waveform). Lowering the amplitude to -12dB would be lowering the signal to the point there's a 12 decibel difference between the signal and the clipping threshold.
That's why it seems backwards, it isn't. It's just the nature of what a decibel is, and that it's arbitrarily representing differences in loudness. It can represent lots of other things, it's just a logarithmic scale
I don't understand why you'd want to set the threshold on a compressor to a positive value, why would you want the threshold to be up in clipping territory when you can just lower the volume pre compressor and have a sensible threshold to exactly the same effect? The threshold is arbitrary, it's compared to the level of the incoming audio. Specifically it is the at which any audio going above, is compressed.
So say you have some audio going in at -12dB, you have the compressor threshold set to -24 and a ratio of 2:1. -24 is double the range of -12 so you can say the audio is going into the compressor at "12dB over the threshold". a ratio of 2:1 will essentially halve that amplitude (for every 2dB over the threshold, the audio will be lowered by 1dB, so 4dB over will be lowered to only 2dB over e.t.c.). Meaning the output audio will be -18dB instead of -12dB (-12dB + -6db attenuation). In other words, you've quietened the audio by 6dB.
It is a bit weird to get your head around and seems topsy turvy, but it does actually make sense when you get a grasp on what a decibel 'is' and that you're working in negative numbers.
0dB is your highest value because that's your clipping threshold in digital audio (the point at which the wave sample's value would exceed the bit width, values higher than this can't physically exist so the audio engine 'clips' the sample to the maximum applicable value), the clipping threshold is therefor a handy reference amplitude to calculate differences in volume from. so in digital audio, something being '0dB' means there's no difference between the amplitude of your audio and the maximum amplitude capable within the digital domain (things will never get louder than this, increasing the fader will just add clipping distortion to the waveform). Lowering the amplitude to -12dB would be lowering the signal to the point there's a 12 decibel difference between the signal and the clipping threshold.
That's why it seems backwards, it isn't. It's just the nature of what a decibel is, and that it's arbitrarily representing differences in loudness. It can represent lots of other things, it's just a logarithmic scale
I don't understand why you'd want to set the threshold on a compressor to a positive value, why would you want the threshold to be up in clipping territory when you can just lower the volume pre compressor and have a sensible threshold to exactly the same effect? The threshold is arbitrary, it's compared to the level of the incoming audio. Specifically it is the at which any audio going above, is compressed.
So say you have some audio going in at -12dB, you have the compressor threshold set to -24 and a ratio of 2:1. -24 is double the range of -12 so you can say the audio is going into the compressor at "12dB over the threshold". a ratio of 2:1 will essentially halve that amplitude (for every 2dB over the threshold, the audio will be lowered by 1dB, so 4dB over will be lowered to only 2dB over e.t.c.). Meaning the output audio will be -18dB instead of -12dB (-12dB + -6db attenuation). In other words, you've quietened the audio by 6dB.
It is a bit weird to get your head around and seems topsy turvy, but it does actually make sense when you get a grasp on what a decibel 'is' and that you're working in negative numbers.
Soundcloud

Serious shit^Altron wrote:The big part is just getting your arrangement down.
Brothulhu wrote:...EQing with the subtlety of a drunk viking lumberjack

Who is online
Users browsing this forum: No registered users and 0 guests