ok i see i kind of tested it i see what u mean although i still try to understand....so about lowering down specific frequencies as i was explaining above that can only be done with dynamic eq (or eq)

 

ok now i understand that all frequencies are lowered when compressed

but that makes me think that volume knob does the same now....is it the same after all

volume knob turns down all frequencies same way?

 

It's not the same. A waveform, as the name states, consists of waves. If you zoom way in you'll see.
Let's take many sinus waves in one long chain. The difference between those waves top and bottoms in db (that's what the "dynamic" is) gets reduced as long as their peaks are above the threshold. So they get squeezed (or compressed) because as stated, the difference gets reduced. The bottoms do not move down. (Dynamic compression)

If you reduce db overall, then the bottoms will move down as well. (Volume knob)

Thats why it gets louder if you raise the db after compression to the same amount, because the bottoms now are higher in relation.

 

Last edited: Apr 26, 2022

He can also add density to the waveform out of thin air, but it has to be with a flying fader.

 

You could say a compressor can react (and its reaction is programmable with attach release and ratio) much faster then you and therefore you need it when that is needed and also because taming the attack of 1000 snare hits by hand can be a little tyring. A rider will do it the way you would do it (at leat it tries and might even do better and be more patient and not get fatigue, but has no feeling), but is absolutelt to slow for percussive sounds.
But compressors are way more versatile; Tame peaks, bring out low level detail, even out levels, bring out the room sound, change the transients, tame them and increase them etc etc. Most of these things are impossible by hand or automation, to much and to fast.

 

Would you take the Swiss Army Knife or the little white plastic toothpick they put in it?

 

Both. My typical workflow fader ride/volume automate prefader. Than compress...

 

no samsone, thats not true!
step 1: everything above the threshold triggers the compressor, okay? high, low, wtf frequencies dont matter.
step 2: the triggered compressor lowers the WHOLE signal, the frequencies dont matter again too.
(of course, w/o attack, release, ratio, multiband, sidechain eq blafasel...)
if youve got a quiet 400 Hz signal it will pass through. but when a big fat loud 2 kHz tone appears, the compressor lowers everything, the 2 kHz, AND the 400 Hz!!!
in a special sense the auto vocal rider and the compressor work similarly, only there triggered differently
but the compressor, limiter, vocal rider (and de-esser btw) lower the whole signal level, once triggered!!!

edit: too late, way too late...

 

Last edited: Apr 28, 2022

Can't vote cuz is only one option. My workflow: Vocal Rider then Compression.

 

Over the years Ive gotten the best results from minimizing the amount of compression on vocals and using compression only as a finalizer to create an even gloss of distortion throughout the performance. My personal process is going through the clip gaining the vocal by hand to get things in a generally even range. This prevents you from overloading the compressor to get things balanced or dealing with certain words getting pushed down too far following a particularly loud syllable as the compressor hasnt fully recovered. My process is clip gain / fix clicks / de-plosive / timing. If its full dry, gentle limiting. Pro L or Goodhertz Faraday. Then some broad dynamic EQ / De-essing if needed. Then I run out to a UTA670 or a BG2. If youre fully ITB I like the UA 1176A and Softube CL1B 2. Goodhertz Vulf can sometimes hit as well.

EDIT: to clarify I level by hand over a tool like vocal rider. Doing things by ear takes much longer but it translates to maintaining the emotion in the performance as opposed to automated normalization of sorts. Sometimes you want certain lines to hit the compression harder to drive home the intensity.

 

It's been quite a while since I tested a Vocal Rider plugin, but at least back then I wasn't really impressed. We humans react differently than electronic circuits or software. We don't react to bit or voltage values, but to pressure differences in the air.

It's kind of funny how some people try to explain things they don't understand themselves. Don't get me wrong, I'm not making fun of you, it just puts a big smile on my face. Reminds me of my former self.

Anyway, compressors don't work the way they are described here. For simplicity's sake, let's assume a VCA compressor:
The voltage signal coming into the compressor is split into two paths. To the signal circuit (the audio path), and to the control circuit (the sidechain). Both the audio path and the sidechain now lead to different inputs of the VCA (Voltage Controlled Amplifier). The VCA is a component that basically does nothing but boost or attenuate voltage. Well, the sidechain is connected to the VCA as a control voltage. But only the voltage in the audio path is processed by the VCA. But now we have a problem, because the voltage in the sidechain is basically the same as the voltage in the audio path. This is not what we want. We need a detector circuit for our time constants, a threshold and a ratio. We achieve all this by inserting a gate in our sidechain path. First a peak or rms detector that we can use to control attack and release. Then a gate that serves as a threshold. That means it only lets through voltage that exceeds X volts. And finally a ratio circuit. All this is now connected to the control input of the VCA. If you want, you can add a filter before this in the SC circuit. All the VCA does now is, it reduces the signal in the audio path accordingly, when signal is present at the control input. The VCA is not interested in which frequencies are present. The VCA doesn't know that at all. It simply regulates the voltage in the audio path according to what arrives at its control input.

I have built this as an example and splitted a stereo signal to 4 channels. Channels 1&2 pass the gate unprocessed and are connected to the input of the volume plugin. The gate processes the channels 3&4. I use the output of the gate as control signal for the volume plugin. Voilà, a compressor!

 

Honestly mentioning developers advertising is not really an interesting answer.
Better mention your experience.

 

I remember renowned mixing engineer Alex Tumay saying to never use any of these Waves "Rider" plugins because "You don't know what it's doing." since it kind of just automatically adjusts volume levels and you have limited input.
But then on the other hand I've seen Rob Kinelski using it on a Billie Eilish mix, and I'm pretty sure I've seen it being used in some other big productions/mixes too.

As for your question in the OP, there's literally no reason why you should ever have to choose between one or the other since they don't do the same things. Ideally you'd "ride" the channel to adjust any obvious volume differences by yourself and then use a compressor afterwards.

 

There's also plugins like Waves MV2 that offer both downward and upward compression in one plugin, I've heard mixing engineers say that it's their plugin of choice for vocals, apparently it greatly reduces the time you'll need to manually automate volume and such.

 

Did anyone here realize that you give advice to a guy that never in his life worked with a vocalist and vocals - and generally with an entire production? Funny to watch these dick measurement contests (“I know more” vs “no!, I know morer!!”) but really, is good to know who you work so hard to teach