How to choose Ratio on a compressor?

[heavymetalmixer]

It doesn't seem to have been mentioned but one should use the ratio and threshold controls in tandem, ie the more of the one the less of the other (unless you're going special fx, of course ). So if you have a very low threshold (ie compression kicks in sooner) you would have a low ration ratio (2:1, for example). Conversely, if you had a high ratio (eg 8:1) you'd have a high threshold (compression would kick in later).
You shouldn't twiddle the one in isolation.

Thank you for your attention.
TGIF!

Exactly, this is why I made my question in OP, like, if you compare all the ratios they would need a different threshold to achieve the same amount of gain reduction, but they obviously have a different sound, that I still don't know where to use on.

[Unaspected]

It doesn't seem to have been mentioned but one should use the ratio and threshold controls in tandem, ie the more of the one the less of the other (unless you're going special fx, of course ). So if you have a very low threshold (ie compression kicks in sooner) you would have a low ration ratio (2:1, for example). Conversely, if you had a high ratio (eg 8:1) you'd have a high threshold (compression would kick in later).
You shouldn't twiddle the one in isolation.

Thank you for your attention.
TGIF!

Exactly, this is why I made my question in OP, like, if you compare all the ratios they would need a different threshold to achieve the same amount of gain reduction, but they obviously have a different sound, that I still don't know where to use on.

Low threshold and ratio are going to sound more natural in a musical sense, compared to an abrupt taming of peaks.

[chk071]

This may sound weird, but, the Dune 3 manual actually has a pretty good description of what a compressor's parameters do (check page 76 and following): https://www.synapse-audio.com/DUNE-3-Manual.pdf

[heavymetalmixer]

It doesn't seem to have been mentioned but one should use the ratio and threshold controls in tandem, ie the more of the one the less of the other (unless you're going special fx, of course ). So if you have a very low threshold (ie compression kicks in sooner) you would have a low ration ratio (2:1, for example). Conversely, if you had a high ratio (eg 8:1) you'd have a high threshold (compression would kick in later).
You shouldn't twiddle the one in isolation.

Thank you for your attention.
TGIF!

Exactly, this is why I made my question in OP, like, if you compare all the ratios they would need a different threshold to achieve the same amount of gain reduction, but they obviously have a different sound, that I still don't know where to use on.

Low threshold and ratio are going to sound more natural in a musical sense, compared to an abrupt taming of peaks.

So, ratios like 4:1 and higher are for sounds or songs that need a more obviously-compressed and distorted sound, right? I think I'm getting where this is going to.

[Unaspected]

It doesn't seem to have been mentioned but one should use the ratio and threshold controls in tandem, ie the more of the one the less of the other (unless you're going special fx, of course ). So if you have a very low threshold (ie compression kicks in sooner) you would have a low ration ratio (2:1, for example). Conversely, if you had a high ratio (eg 8:1) you'd have a high threshold (compression would kick in later).
You shouldn't twiddle the one in isolation.

Thank you for your attention.
TGIF!

Exactly, this is why I made my question in OP, like, if you compare all the ratios they would need a different threshold to achieve the same amount of gain reduction, but they obviously have a different sound, that I still don't know where to use on.

Low threshold and ratio are going to sound more natural in a musical sense, compared to an abrupt taming of peaks.

So, ratios like 4:1 and higher are for sounds or songs that need a more obviously-compressed and distorted sound, right? I think I'm getting where this is going to.

It's difficult to generalise when compression can be used for so many tasks. For example, you can also use higher ratios to push sounds back in the mix.

[VELLTONE MUSIC]

yep it is not necessarily distortion or serious change of the sound,but to make it sounds right in the mix or to express something more obviously - i did a lot of experiments and sometimes best ratio was 5,5 with no obvious reason that logic tell me to follow but i guess if the rest of instruments are bright,only certain compression level could bring back specific freq spectrum of instrument in the mix ,other option is to change type until find right sound,it is all fantasy...

[jochicago]

Compressors are tough regardless. Everything depends on the application and the source material. It also depends on the compressor itself.

You can get away with high ratios with something very rounded like an opto. But it depends on the material as well. Say you have a vocal track that is very dynamic and your typical compressor approach, say an 1176, is struggling to tame it. So you can start with a transparent compressor (say Kotelnikov) with a high-ish ratio of 6:1 and aim the threshold to start around 2/3s of the working range, just to squash the louder dynamics but not add much character. Then you can chase that with your regular compressor of choice, say the 1176 at 4:1, and now that the material is more even it will sound like a regular treatment on a regular track.

As a rule of thumb, for me:
Typical track compression 2 to 5 ratio. I often do more than one compressor on a track so I try to get only a few db GR per stage. My vocals sound more pristine and present than a lot of the stuff I see online, and this is one of the reasons. I would never take off 12-15db with an 1176 on a vocal unless we are talking about some meaty rock application. For most things you are better off taking off less and doing 2 stages. I also thicken with parallel tracks, so that also takes away the pressure of high GR on the lead track.

On a parallel bus I'll easily go to 6:1 ratio if the goal is to make a harsh bus that I'll then mix in to taste using the volume fader. This is the thickener that allows the main tracks to stay more dynamic and elegant.

On a regular bus: 1.5-3 ratio. Even with a transparent compressor, I'm trying to avoid any pumping or loss of charisma that comes from the compressor taking some chunks off at the wrong spots. Buses usually carry more than one track, so it is best to touch that more gently. Bonus points for using a compressor that lets you decouple L/R on a bus.

Master bus: 1.2 - 2.5 ratio. If the master needs more compression then probably something's off at a previous stage. The comp on the master is there to glue the mix, and too much GR will definitely add pump and/or suck the life out of the performance. If I want more than that, I may use a second compressor to touch up with a different character. For instance I may have a transparent compressor followed by a Fairchild-style or an LA2A-style for a touch of personality.

Having said all of that, keep working on letting your ears set the knobs. You are listening for presence, energy, charisma, pump, cohesiveness. If you are dialing it wrong you'll get flatness, dullness, pumping at the wrong time, squashed, uninteresting. This is why it is so important that you know your compressor. It is a very custom artistic tool, no 2 are the same. The same numbers do different things on 2 different compressors. You have to spend time with your choice compressor and learn it like a weapon.

When I dial in the master compressor I'm not even looking at the knob numbers. I'm hearing for that sweet spot. The music starts a bit unruly, overly energetic, then I start to dial the compressor in and I start hearing cohesiveness, parts coming together to work as a team, then as I keep dialing it in I hear that energy start to squash, the vibrance becomes subdued, the instruments and vocals start sounding more distant. That's when I went too far. Then I start dialing it back to find that sweet spot. Usually ends up being around 1-3 GR at 1.7 ratio (if the compressor has a ratio, because I'll use a classic opto too).

I hope some of that helps. Sometimes I feel this must read like ramblings.

[heavymetalmixer]

Compressors are tough regardless. Everything depends on the application and the source material. It also depends on the compressor itself.

You can get away with high ratios with something very rounded like an opto. But it depends on the material as well. Say you have a vocal track that is very dynamic and your typical compressor approach, say an 1176, is struggling to tame it. So you can start with a transparent compressor (say Kotelnikov) with a high-ish ratio of 6:1 and aim the threshold to start around 2/3s of the working range, just to squash the louder dynamics but not add much character. Then you can chase that with your regular compressor of choice, say the 1176 at 4:1, and now that the material is more even it will sound like a regular treatment on a regular track.

As a rule of thumb, for me:
Typical track compression 2 to 5 ratio. I often do more than one compressor on a track so I try to get only a few db GR per stage. My vocals sound more pristine and present than a lot of the stuff I see online, and this is one of the reasons. I would never take off 12-15db with an 1176 on a vocal unless we are talking about some meaty rock application. For most things you are better off taking off less and doing 2 stages. I also thicken with parallel tracks, so that also takes away the pressure of high GR on the lead track.

On a parallel bus I'll easily go to 6:1 ratio if the goal is to make a harsh bus that I'll then mix in to taste using the volume fader. This is the thickener that allows the main tracks to stay more dynamic and elegant.

On a regular bus: 1.5-3 ratio. Even with a transparent compressor, I'm trying to avoid any pumping or loss of charisma that comes from the compressor taking some chunks off at the wrong spots. Buses usually carry more than one track, so it is best to touch that more gently. Bonus points for using a compressor that lets you decouple L/R on a bus.

Master bus: 1.2 - 2.5 ratio. If the master needs more compression then probably something's off at a previous stage. The comp on the master is there to glue the mix, and too much GR will definitely add pump and/or suck the life out of the performance. If I want more than that, I may use a second compressor to touch up with a different character. For instance I may have a transparent compressor followed by a Fairchild-style or an LA2A-style for a touch of personality.

Having said all of that, keep working on letting your ears set the knobs. You are listening for presence, energy, charisma, pump, cohesiveness. If you are dialing it wrong you'll get flatness, dullness, pumping at the wrong time, squashed, uninteresting. This is why it is so important that you know your compressor. It is a very custom artistic tool, no 2 are the same. The same numbers do different things on 2 different compressors. You have to spend time with your choice compressor and learn it like a weapon.

When I dial in the master compressor I'm not even looking at the knob numbers. I'm hearing for that sweet spot. The music starts a bit unruly, overly energetic, then I start to dial the compressor in and I start hearing cohesiveness, parts coming together to work as a team, then as I keep dialing it in I hear that energy start to squash, the vibrance becomes subdued, the instruments and vocals start sounding more distant. That's when I went too far. Then I start dialing it back to find that sweet spot. Usually ends up being around 1-3 GR at 1.7 ratio (if the compressor has a ratio, because I'll use a classic opto too).

I hope some of that helps. Sometimes I feel this must read like ramblings.

yep it is not necessarily distortion or serious change of the sound,but to make it sounds right in the mix or to express something more obviously - i did a lot of experiments and sometimes best ratio was 5,5 with no obvious reason that logic tell me to follow but i guess if the rest of instruments are bright,only certain compression level could bring back specific freq spectrum of instrument in the mix ,other option is to change type until find right sound,it is all fantasy...

Very good info here both of you!

Btw @jochicago " . . . Bonus points for using a compressor that lets you decouple L/R on a bus . . ." at least for me unlinked/dual mono compression isn't a good idea, it messes up the width and produces some weird modulation effects.

[No_Use]

I like the A.R.R.T. (attack, release, ratio, threshold) approach (from 'Mixing with your mind'), it's roughly described here (post #4):
viewtopic.php?t=389365

Note the starting point, you basically crank up everything to max (so you hear easily what the comp does) then dial back til it fits.

[heavymetalmixer]

I like the A.R.R.T. approach (from 'Mixing with your mind'), it's roughly described here (post #4):
viewtopic.php?t=389365

Note the starting point, you basically crank up everything to max (so you hear easily what the comp does) then dial back til it fits.

I've been using that approach to anything in mixing for a few months and it's really useful, many Youtubers started using it since like 2017.

[Unaspected]

I like the A.R.R.T. approach (from 'Mixing with your mind'), it's roughly described here (post #4):
viewtopic.php?t=389365

Note the starting point, you basically crank up everything to max (so you hear easily what the comp does) then dial back til it fits.

I've been using that approach to anything in mixing for a few months and it's really useful, many Youtubers started using it since like 2017.

That's essentially the technique I wrote about here: https://unaspectedstudios.wordpress.com ... ompressor/ - though without any acronyms. Attack and release settings are definitely easiest to hear with extreme settings so it's logical that you would force those conditions first.

I concur with Jochicago; in that it's really down to the program material and where you want to get it to. The best thing is to simply understand what the compressor is doing to the audio and then just listen as you make adjustments.

[VELLTONE MUSIC]

i want to ask something - in real physical world higher freqs are faster,lower are slower so by default i set bass with slower attack, leads and pluck with faster ...is there any other basic rules like these when work with frequencies,for example is it stereo and reverb under 100hz absolutely forbidden or just need more attention cuz some of my fav kicks have reverb or minimum distortion of any kind in upper spectrum,from my experience i'd say lower freqs accept more saturation than higher ones,in other words is there forbidden stuff and what man must b careful mostly?

[Unaspected]

i want to ask something - in real physical world higher freqs are faster,lower are slower so by default i set bass with slower attack, leads and pluck with faster ...is there any other basic rules like these when work with frequencies,for example is it stereo and reverb under 100hz absolutely forbidden or just need more attention cuz some of my fav kicks have reverb or minimum distortion of any kind in upper spectrum,from my experience i'd say lower freqs accept more saturation than higher ones,in other words is there forbidden stuff and what man must b careful mostly?

All audible frequencies travel at the same speed in a consistent medium.

[heavymetalmixer]

i want to ask something - in real physical world higher freqs are faster,lower are slower so by default i set bass with slower attack, leads and pluck with faster ...is there any other basic rules like these when work with frequencies,for example is it stereo and reverb under 100hz absolutely forbidden or just need more attention cuz some of my fav kicks have reverb or minimum distortion of any kind in upper spectrum,from my experience i'd say lower freqs accept more saturation than higher ones,in other words is there forbidden stuff and what man must b careful mostly?

A known technique for changing the frequency range where the compressor focus on is the "side-chain HPF". As you may already know low frequencies have the most energy but they're also the most unstable because of room reflections. Because of this using a compressor results in a non-musical track that jumps and distorts like crazy.

The side-chain HPF cuts low frequencies from the compressor detector (but not the original signal, nor the output), so the compressor doesn't have to deal with so many wild low frequencies and focuses on more on the mid-range. Some compressors even allow you to set the side-chain HPF really high.

[cron]

i want to ask something - in real physical world higher freqs are faster,lower are slower so by default i set bass with slower attack, leads and pluck with faster ...is there any other basic rules like these when work with frequencies,for example is it stereo and reverb under 100hz absolutely forbidden or just need more attention cuz some of my fav kicks have reverb or minimum distortion of any kind in upper spectrum,from my experience i'd say lower freqs accept more saturation than higher ones,in other words is there forbidden stuff and what man must b careful mostly?

Distortion from very fast attacks and releases tends to be more apparent on lower frequencies. Lower frequency = longer wavelength. If the attack and/or release of a peak compressor is shorter than one cycle, it can really 'cut into' the cycle and significantly change its shape. Meanwhile up at the high end of the spectrum, the same attack and release settings are likely to span multiple cycles and will thus produce less harmonic distortion.

Stereo and reverb under 100Hz really comes down to whether you're aiming for the club or not. Club systems tend to be mono (though tbh most of the dives I've played seemed to have stereo systems), and if you're aiming for vinyl you'll be able to get a louder cut with a centred bass end. I've noticed a lot of EDMish tunes are extremely bass-light these days incidentally. Optimising for mobile phone/earbud listening and cutting bass to achieve more volume on those systems, perhaps?