should i be caring about peaks?

if so what i should be doing about them?

and how will that help?

my question is more for edm based

thanks

 

Last edited: Jul 28, 2022

At mixing level, a subtle soft clipping on "offending tracks" will do the trick.
You will gain dynamic overhead, without loosing anything important.
You may even get pleasing harmonic content in "presence" domain (3.5-5 kHz)

If they are too "obvious" to be clipped, another solution is "transient shaper" (there are free ones) :
https://www.whippedcreamsounds.com/free-transient-shapers/

At master bus/mastering level ... it is a LONG story
So forget about it

It will end in war, like any mastering/stereo bus thread

 

With a "brickwall limiter" blunt the transients until you have an acceptable result.

 

Some people might say avoid them like the plague. It may seem that one or two peaks might not affect the sound of your mix. However, those things are cumulative, and they might affect the sound of your mix. As @PifPafPif and @BEAT16 suggested, you might wanna soft clip them or add a limiter. Nevermind the style, whatever contributes to clear mixing is welcome. I'd say avoid them. There's the whole debate on 32-bit floating point and how it is impossible to actually clip anything until it reaches the two-buss, but why bother?

 

so how should i understand exactly if something needs clipping or not......what the difference need to be between Peak and Rms?

 

A "healthy", dynamic audio track should have most of its reading at about half the meter bar, with peaks about two thirds of it. However, a lot of tracks sound great when they are pushed harder, or you might need to make them stick out in your mix. That's when you often see the meter go into red, and you need to do something about it. There are several options: add a compressor, a clipper, or a limiter. All of them will reduce the peaks somehow differently. Compressors reduce the difference between the peaks and the body of your audio file. Limiters and clippers will basically chop them off before they hit 0dBfs. Each one has a different flavor and will make your track sound different. Peak metering basically reads the peaks and is useful to detect clipping, and RMS reads the amplitude more the way the ear perceives it.

EDIT: typos and a heads up to @No Avenger for the correction. Thank you, buddy.

 

Last edited: Jul 28, 2022

is it like a haircut? sometimes not taking a haircut for a while is not that bad

nevermind most times its bad

 

Last edited: Jul 28, 2022

nah, not important at all. It's propaganda for Compressor hardware/software makers to sell their products. Don't believe the internet and youtube. Catch all them peaks and slay them down. -4LUFS is what important, everything else is just lame attempt to make music.

 

Peak is the bigest peak, in other words your loudest point.

RMS is the average peak, it stands for root mean squared and is a average used for sin wave forms to calculate the average opeak.

you want to know both for differant reasons.

If you want to know how loud your track is compaired to something else then use RMS.

Peak is the loudest point, so in the real world context it is the point where you speaker will be fully pulled in or pushed out.

Since your EDM based what you want is to get your peak and RMS quite close together. This can be done with limiters and compressors.

Don't jst turn the volume up and let it clip unless your using analogue equipment and digital clipping is brutal, to your audence's ears and your speakers over time.

 

Peaks are transients, which are quite important - even in EDM. If you take too much of it out (to the point where the difference between peak and average RMS/loudness is too small) it loses its "impact", "energy", "excitement", etc. It would be like speaking with your mouth in a open and fixed position, like at the dentist.

 

Last edited: Jul 28, 2022

Nailed it. Words of true wisdom...

 

Yes, everytime peaks are growing you need to cut them.

This again is amazing.

 

@Baxter Like your avatar ?

Yes, peaks/transients/whatever NEED to be present. Clarity, punch, definition ... call it like you want.
That's a big part of "loudness war" cadavers.

Some devs "accentuate" them in a trashy manner ... like Slate FG-X (sounding like transient butchery to me ...).
Or kill them by psycho-acoustic tricks. Because delay lines (used in psycho acoustic tricks) kills transients by "smearing" them.
Like Tone-2 Akustix (i hate Tone2 coding AND psycho acoustic sh*t on ALL their plugins, killing bass phase AND transients).
Even worse at mastering stage ... like FG-X or Akustix are intended to

Keep the transients/peaks ALIVE and UNDER CONTROL

 

Agreed with almost everything except FG-X. I love it and it possibly is my oldest paid-for plugin. When used properly, it can sound amazing.

 

Just don't push "transient" section

My fav "character" limiter is Elephant.
But it is NOT "transparent" at all.

 

Not all peaks. I mean I just come from pornhub with an ongoing true peak... can't really recommend to cut it,,,

 

Last edited: Jul 29, 2022

Produce and mix a LOT of songs. Pay attention to the results you’re getting - what do you like, what don’t you like?

For the things you don’t like, figure out WHY you don’t like them, then experiment with trying to get them to a point where you do like them.

When you do that, you’ll have a better idea of how to ask questions where the answers will actually be useful to you - I’ve noticed a lot of your threads are posting very kinda arbitrary questions, as if you’re looking to try and collect a bunch of blanket rules/statements that will suddenly all fall into place, and every mix you do will magically sound incredible.

You question about peak vs rms level is both arbitrary and pointless. I’ve noticed in a couple of your other threads that you seem to sorta have the idea that a production/mix is soemthing you can “solve” like a mathematical equation.

Sorry to tell you, but if there were shortcuts like that, then everyone would be an incredible producer/mixer, and let’s face it, that’s just not the case is it?

The way you get good at this is by doing it. A LOT.

For sure, it can be useful to chat to other folks, see how others’ approaches differ from yours, and pick up the odd tip and bit of inspiration, but the only way you’re truly gonna learn, understand and improve is by actually sitting down, getting off the forums, opening up your daw and putting the work in.

I'm not trying to rain on your parade or anything here - you're obviously very enthusiastic and keen to learn/improve, which is awesome! But I feel that you're kinda falling into a trap that many people often do, in searching for a one-size-fits-all 'instruction manual' on how to get good.

There isn't one set way to make a mix sound fantastic - tbh every single other producer, mixer, and ME I work with has a unique approach that is specific from them, and while we may well have some degree of overlap, we're constantly working to figure out what works best for us individually.

People don't hire you for your skills - when you reach a certain level, skills are assumed - they're hiring you for the uniqueness you can bring to a project. If you can't offer something that's uniquely you, then why would anyone want to hire/collab with you?

Get very comfortable with sucking - we all sucked for a long time, then we figure some shit out, and we gradually suck less.



(That's probably a somewhat more rambling - and possible OT - answer than you were hoping for. So if you feel that I'm talking total horseshit with all of the above, I'll give you the straight answer to the peak/rms question:
Make all your peaks 12, and your rms 5, those are the industry standards).

 

Last edited: Jul 30, 2022

It depends on how loud do you want your song to be. If you want your song as loud as most top chart bangers, you should be caring about peaks.

 

LUFS is only important in the future cuz the future is fucking stupid and useless, other than that, LUFS doesnt make a good song, it just makes shit songs louder and cleaner to make up for the shitty song