well first of all Ian Shepard is a real mastering engineer and he is not using freeware except that tt meter.
and no you don't want to listen to a lot of these people saying push push push. look use compression and eq to shape the way its going to feel if you are in a style that have a lot of over compressed music than you will have to do that to match levels. And I never put a limiter on my mixes. the best way to mix is at a good low level that how CLA does it. I do it that way and so far no problems everything comes out clean and clear. and mastering should be its own process. mix the music in the mixing stage and master it with maybe multiple compressors doing little bits of compression each it helps keep the music dynamic and gets it loud at the same time.

Wow, I am not really sure about this.

There is a real danger of hitting the drum bus and the master faders too hard.

Even when I am tracking sometimes the overall levels of all the tracks recorded at a reasonable level get too hot and I have to select "all" faders and pull them down 2 or 3 dBs as I am starting to hear some distortion.

Pushing everything individually from the very start seems pretty dangerous to me, and I believe that keeping the levels under control will be your key to it sounding good when it is mastered and loud.

As I said before the important thing for me is to not waste my time doing "subtle" things in the mix only to hear them get pushed up to the front in level afterward. Again I put a brick-wall limiter in the insert of my master so I have an idea what is going to happen, before it happens.

ducter does have a good point if you have a limiter there you can at least work to help make it sound great even after limiting. but yea mixing is all about subtle edits you want the best edit you never heard haha. just small things here and there add up but I also use a K-meter mix at low levels and have my speakers set up for k-meter specification. look if you want things to sound good make sure to no do everything in solo a bass or kick is solo can sound amazing but will not work in the mix.

Use subtractive EQ for the most part and only additive for the main instruments and transient peaks of each element. Also feel free to squeeze as much extra level through the best limiter you can before your dynamics get squashed. Dont forget fader riding too becuase using a compressor on smooth parts will unintentionally start the waves to pump and spoil the moment, especially on a master. The most important factor is a good mix, even frequency spread with no clutter or harsh EQ, then its just a job of getting towards the level of a commercially mastered similar track. (Dont forget Lo cut, producers swear by it.)

PS_ Concerning the above, NEVER overrun every channel into overdrive, some yes, very delicately for a gentle summing glue. Unless your looking for next to no headroom and a certainly harsh Mix with extra distortion then feel free to run your tracks into the red. Same goes with plugins, try and keep any clipping to an extreme limit as this is where all the shit comes from.Only try this at home kids with analog

ha I was about to add pretty much the same thing but +1 on all of this but if you have to to alot of pushing try using a compressor doing tiny tiny reduction and stack a few it works and does not do the pumping

+1 Gentle summing for instruments, over driven guitars, and super snappy kick drums and snares. Then strap on a powerful master buss with a really powerful DSP fx to fatten things up and pull your master Fader back 2-3 db before mixdown. Then use sonograms and multiband eq and dynamics or character shaping fx to give it whatever artistic flair it might require and normalize to taste after checking for stereo phase, lo freq energy center and the high freq impact. overall width and where you have used good seperation when panning to instruments that collide one pan left the other pan right, and not to forget, freq problem solving such as 2nd and 3rd partials

Best way to test it (if EDM) is to mix it in a DJ set, and see if anything sticks out.

As for actual tools, the TT Dynamic Range Meter (offline) is great, as it gives tracks an official DR Rating.

There's a danger of hitting anything too hard, not really a reason to put anything off.

Putting loudness off until the master is a terrible idea - when the compression smashes down a peak, it doesn't just lower the level of whatever caused the peak, it lowers the level of the entire mix at that point - so you'll have a rogue snare hit that peaks the entire mix, and then your master limiter momentarily lower the volume of not just that snare but the guitars/synths/vocals/bass/et cetera taking place at the time. It's far better to control that snare's dynamics and then work your way upward.

We are talking about two different things.

I am saying that making every track as loud as possible puts you in danger of overloading buses and the master and I stick by that. It is a road leading to distortion and digital harshness that is really unpleasant.

I treat some of the more dynamic elements in a mix, eg lead vocals, bass, some guitars, before they hit the master, so that they don't interfere with the overall "feel" of the dynamics of the project.

It determines how they "sit" in the mix for me. They are controlled to not "pop" out or suddenly disappear. It might have the added benefit of lowering the peaks, but that is not the purpose, to gain in loudness, it is the control of the blend of the various elements I'm after. Mastering for the mastering engineer, who has fresh ears.

My master fader has a insert with limiter/maximizer on it right from the start of a project so there are no surprises when the tracks mixed low in the track, backing vocals, pads are mastered. They don't suddenly become louder and spoil the balance of the mix as I wanted it to be.

Things stay pretty well the same as what I wanted them to be in the first place.

If you want to compare several tracks (e.g. RMS, Spectrum) Toscanalyzer is your friend

br
-tom

Listen and compare.

Tools like Sonnox Oxford Inflator give more 'apparent' loudness.

This is BS. Care to explain?

How do I know when my track is loud enough?


What sort of question is this ?

All the bullshit to do with the loudness wars which are changing yet again. Someone once sent a demo to a company but was told his overall levels were not up to the standard they expect. Ie, you need to use the volume you are given for each Mix session very judicially and know how to compress dynamics properly so you get the best dynamic range from them before all the peaks disappear. Its all so easy to clip and Limit and use distortion to fill out the sound but it is the natural qualities and individual uses by each artist that are done uniquely while working through each project always deciding what is the best for the audio as you mix and decide where you want the music to go. These days its become where everyone wants results in 1-2 days, where I think for an average track to sound properly done it should take around 1-4 weeks if not more, the 1-4 weeks is the time I would spend on making sure that a track is started and finished and mastered and listened to on as many stereos and speakers as possible. At times in the industry there is a minimal level for each kind of musical genre where the music frequencies should Fit. The track in my sig was composed in 1 afternoon and then mixed and mastered quickly in the next afternoon with a fresh pair of ears but still turned out quite good. Practise will always make perfect,especially in this game.

Ouch, ouch, OUCH!

So much misguided information, then "post production" misinterpreted for "mastering". Sorry, but some correciton is due.




First and foremost, I really think about reporting you as spammer due to your link-pimpage. But the main "ouch factor" here is with the peak meter.

Granted, you should watch the digital peak of a signal, but what you describe is the current demise of audio productions "hit it as close as 0dBFS and then raise it a bit higher". You gave no relation to the actual loudness, which is not(!) measured by a digital peak meter, but by a loudness meter with at least 300ms raise/integration time. And depending on the desired standard, with or without weighting filter.

Please stop spreading that wrong information.






This is plain nonsense since we have to separate loudness with maximum peak.

Analog is in no way superior in terms of loudness than digital. Granted, you have a smoother editing (unless you use 16x oversampling in the digital realm, which can get close to analog behavior), but it doesn't add more dB in terms of loudness.

You can drive both into oblivion, even faster and more agressive with digital.






What you talk about, or should talk about (like Chase), is proper gain staging in the post production stage. I wrote tons of posts here on KVR regarding that topic. One is in my KVR marks.

Setting the correct levels right from the start (even with slight overdriving) compared to fixing everything while mastering should be the correct way to do things. And if you use a calibrated system, you do not(!) need a brickwall limiter on your master bus to mix through while post production.





Good post, but if you know that nowaday productions hover around -7dB RMS to -6dB RMS loudness wise, you know where to shoot for. If you want to shoot for that level (which I do not recommend).





I admire the effort, but "official" is nothing except for the EBU R-128 (broadcast) standard currently. The DR rating only gives you an indication of the crest factor (range between overall loudness to maximum digital peak) of a song, not the actual loudness. It was not considered an official value.

Unfortunately there is no standard set for music, even the K-System is only a recommendation rather than a strict rule. But this all can change in the near future with the "Music Loudness Foundation" or in other words their proposal for "Loudness Normalisation for Music with playback devices".






You should care, since it's also a representation of your production. On top of it, these so called "labels" still ignore suitable loudness limits. Of course the workload is off your shoulders (leave it to a mastering engineer, like you should in the first place if you are not one yourself), but on the long run - do you want to be known like Metallica with "Death Magnetic"?





Here's how I usually do it:

1) my studio is setup to the K-System v2(!) (in short, it's the same K-System, only with the ballistics of the EBU R-128 meter), though my worklevel is not(!) +89dBc (SPL).

2) I work on a production with a reference track (if desired), but the reference track will be at the same loudness as the producion I need to edit

3) I go by the rule "less is more", meaning: EQ, barely any compression (if ever), barely any stereo FX (though a touch here and there is okay), no agressive loudness raising, etc. I abandonned exciters years ago.

4) If I need to raise the loudness up to a certain point (example K-12 and higher), I take off the workload of the limiter with a clipper in front of the brickwall limiter.




Some notable things:

a) with the introduction of the EBU R-128 broadcasting standard, there were a lot of things set in motion. One of the things is implementing a loudness normalisation scheme while playing back music on standalone, portable or PC (Win/Mac/Linux) bound devices. You can still master all the way you want, but the loudness of all tracks will be normalised to a certain value (to my understanding, currently around -16LU). Squashed productions will sound less lively than dynamic productions.

b) learn your tools, especially what they measure. The Dynamic Range meter is (in my opinion) not a suitable loudness measurement tool. Especially if you use the outdated concept from the early days of that recommendation. Also learn the different weighting filters and what they are for.

c) If you want to work future-proof, grab an EBU R-128 loudness meter that can be freely configured. Then setup the reference level/offset of the Momentary Loudness (or ML in short) to 0LU = -16dB/-14dB to resemble K-16 or K-14. Then mix your production to that loudness while not exceeding a digital peak of -1dB True Peak. Advantage: barely any loudness raising/limiting needed, and you get a very dynamic and good sounding production. I wrote a white paper on that topic - you can find it on my technical blog. A good VST plugin for starters is EBU Loudness/Compact by ToneBoosters.

d) if this is all too much for you to handle, give your production unmastered(!, read: with suitable headroom) to an actualy mastering engineer.



But to get back to the OP's initial question:
You know when your production is loudn enough, if you do not(!) exceed a certain maximum loudness which is either reasonable or a standard. K-14 (either the original version one by Bob Katz, or the revised v2) is a great starting point to work against the loudness war.

If your listeners say "it's not loud enough/not competitive" - ignore them.