Peak vs RMS Compression

[AlecBeretz]

Do you guys have a preference? Use both?

Just curious. I just sorta figured out how to use each and I've found I love using peak-compressors on vocal tracks to sort of "automate" the volume for me. I put the threshold high enough so I'm only killing the biggest peaks, and I use like a 5:1 ratio. I get a lot more volume out of the vocals and it still sounds very natural, not too squashed.

 

[Dags]

Hi Alec
I have always liked the smoothness of the default RMS compression sensing found in the Logic compressor plugin, especially for vocals (can be switched to peak for catching more of the transients of source material if required)
ProTools, by default, uses a peak sensing compressor on the bog-standard compressor plugin. I worked that out by listening to the same track put through both software compressors with very similar settings to achieve around 4-6 dB reduction. Protools made the compression much more audible and by switching Logic's compressor to peak I got the same sound.
Anyone know if PT has an RMS sensing option switch somewhere or is it fixed at peak sensing?

Dags

 

[mixsit]

Maybe a diff angle than you're asking but I generally go through my best 'first guesses as to the comps I think might sound nice or be a good fit- and most (many..) have attack options.
To get to your point though there's at least a few of my faves do have the full feature sets and let you sort of roll your own response styles. BlueCat for one. The other (I use way more often) is the old red Timeworks', and I think 'Peak style pretty much stays on that one.

While we're at it, for most work where it's an option I almost always tilt towards hard knee' (and maybe lower ratios-) over soft knee'.

 

[GONZO-X]

if i want dynamics in my music, i allow for less RMS, which allows me higher peaks.

it's all in how you set the compressor, and it really is extremely variable.

if you make your RMS high, it means you are flattening your dynamics.

which also means, when you mix a bunch of tracks like that, it will sound like one monotonous wall of sound.

 

[jaynm26]

Do you guys have a preference? Use both?

Just curious. I just sorta figured out how to use each and I've found I love using peak-compressors on vocal tracks to sort of "automate" the volume for me. I put the threshold high enough so I'm only killing the biggest peaks, and I use like a 5:1 ratio. I get a lot more volume out of the vocals and it still sounds very natural, not too squashed.

No such things as preference really...It would be like asking whats your preference Reverb or Delay. Their two different meters. Only time RMS can truly be helpful is if your looking at bus level of RMS or RMS of the entire mix or a master. So in the mix as far as individual instruments it will be peak. I could be wrong some other more seasons veterans might find some unique uses for RMS. If you do please shine the light.

 

[mixsit]

No such things as preference really...It would be like asking whats your preference Reverb or Delay. Their two different meters. Only time RMS can truly be helpful is if your looking at bus level of RMS or RMS of the entire mix or a master. So in the mix as far as individual instruments it will be peak. I could be wrong some other more seasons veterans might find some unique uses for RMS. If you do please shine the light.

Gonzo's talking about mix density/levels and loudness, you're looking at metering types..
OP's taking about compressor attack speeds and detector peak or averaging styles.

 

[GONZO-X]

right, but it's all related.

 

[bouldersoundguy]

Peak and RMS just mean faster and slower reaction. I think in terms of attack and release rather than peak or RMS.

 

[mixsit]

if i want dynamics in my music, i allow for more RMS, which allows me higher peaks..

Fix that for ya then

 

[GONZO-X]

heh

you know, it's real hard to 'teach' compression.

it's a real touchy-feely thing, that some people just never get.


and they have pre-conceived notions of what a 'compressor' is supposed to do, and sound like.


less rms


mean, the Peaks of the RMS is low......

which means, hardly any compression is occuring, allowing the peaks thru, which drives the RMS down.
i'm talking AVERAGE RMS here, and always have been, always will be.

 

[bouldersoundguy]

heh

you know, it's real hard to 'teach' compression.

it's a real touchy-feely thing, that some people just never get.


and they have pre-conceived notions of what a 'compressor' is supposed to do, and sound like.


less rms


mean, the Peaks of the RMS is low......

which means, hardly any compression is occuring, allowing the peaks thru, which drives the RMS down.
i'm talking AVERAGE RMS here, and always have been, always will be.

RMS basically means average level over time. Peak means the short term, instantaneous level. "Peaks of the RMS" is contradictory.

An RMS compressor with affect the level on a longer time scale, evening out the average level while leaving the short time scale dynamics (peaks) less affected. A peak compressor can bring down the peaks without substantially affecting the RMS level.

 

[GONZO-X]

that why i wanted to make clear i was referring to 'average' rms

 

[bouldersoundguy]

that why i wanted to make clear i was referring to 'average' rms

But RMS means average. It can't be anything else.

 

[Mo Facta]

Basics.

RMS means Root Mean Square which is defined as the square root of the squares of the means (average) of the incoming discrete peak values. In other words, RMS is calculated by the taking each discrete peak level, squaring it, calculating an average, and then calculating the square root of that average sum. A compressor set to RMS will monitor this mathematical function of the incoming signal and react accordingly.

Obviously, this will have a rise and fall time independent of, yet related to, the instantaneous peak level and will be determined by its RMS mathematical function.

But like was said earlier, a RMS compressor will generally react slower because the mean peak calculation is averaged. This makes it a good choice for compressing sources with relatively narrow margins of RMS swing. I like using RMS compression on vocals set to a low ratio and a low threshold, which can help make the results more transparent. Peak compression is good when you want to grab things.

However, as always, YMMV.

Cheers

 

[mixsit]

..
But like was said earlier, a RMS compressor will generally react slower because the mean peak calculation is averaged. This makes it a good choice for compressing sources with relatively narrow margins of RMS swing. I like using RMS compression on vocals set to a low ratio and a low threshold, which can help make the results more transparent. Peak compression is good when you want to grab things.

However, as always, YMMV.

Cheers

Interesting angle 'rms for narrow rms swing, need to digest that

I've always considered attack time in general to be where a huge portion of a comp's reaction lies. Consider a waveform's front end shape, speed and height. Look at that (or 'compare) it to your attack time, and how much effective ratio in change is going on in those first milliseconds.

.now ..where else could I have gotten into the usable notion of Low Ratio Peak Limters

 

[GONZO-X]

This makes it a good choice for compressing sources with relatively narrow margins of RMS swing.

thank you, mo facta.