Why you can’t fool the dentist.

Only flossing your teeth the night before your appointment doesn’t work. It’s the same with your mastering engineer

(Before we get down to nitty gritty here, this article is aimed at the mixing engineer who is an intermediate or beginner, or an artist who hasn’t had their tracks mastered before or they used a cheap “mastering engineer” who just took their money and didn’t try to help them. The rest of you already know better or should! - Dave)

Each of the 3 WAV images above look like they have drastically different dynamic ranges. They don’t. Each of those images is the same exact record shown at different output levels. The first is the original. The second is the level dropped 18dB and the third is dropped 10dB.

Because I still come across this issue every-so-often, I thought showing it this way might help better demonstrate why smashing the dynamic range on your record in the mix stage, cripples your record in the mastering stage. Especially if the master is significantly clipped like the example shown here.

To understand this, think of dynamic range as the difference between the lowest (quietest) signal in the audio, to the highest (loudest) signal. If 0 is the lowest and 10 is the highest for example, that ratio is 10:1. It could be the slightest tail of a reverb at the end of your song that represents 0 and the 808 slapping through the mix representing 10.

Normally, to get a good master, your mix should be done with the master fader (stereo bus fader) at 0 and the peaks just below clipping. This is the result of paying attention to your individual track levels. The old “leave 6dB of headroom” on your mix comes from the analog tape days where you used a VU meter to set your levels. With digital you can run that peak signal all the way up to zero as long as you don’t clip - quote me on this!

Now here’s where it gets tricky. Louder sounds are perceived to sound “better” to the human ear in comparison to their original level. Put a limiter on an 808 and turn up the make up gain and instantly you have a “better” sounding 808, “better” of course being subjective. If your 808 sample is already limited, adding more limiting will eventually work against you. So naturally, you do the same thing on your master bus right? - slap a maximizer or Ozone or L2 on your master bus and instantly it sounds “amazing”. Don’t get me wrong, in most cases it does sound better if your mix is done well. But in a lot of cases it gets pushed right to the threshold of clipping in the plugin itself, even when the master fader level on your DAW shows the signal is below zero and not being clipped. And for sure, it is reducing the dynamic range which can quickly become a slippery slope.

So at this point, you’ve set the dynamic range, whether that’s -14 LUFS, -10 LUFS or something extremely heavy like -4LUFS. That 10:1 ratio is SET (baked in) when you bounce the mix.

When the mastering engineer gets a hold of your mix, they will know INSTANTLY if you’ve used a maximizer/Ozone/L2/etc in the same way your dentist will know instantly whether you regularly floss or not (the bleeding gums is always a dead giveaway when they pick and floss your teeth). You can pretty much tell just by looking at the waveform if it’s been processed with heavy gain reduction. The peaks will look shaved off and will all be uniform in their maximum level like a straight line across the top and bottom as shown in the image below (which happens to be the same mix as the 3 images at the top).

The mid section of the waveform will also look more “dense” across the mix. Below, we have a mix that has been very lightly limited (or not at all). You can see the peaks are just below zero in the loudest parts of the song so its taking advantage of all the available dynamic range and is ready to master.

So really, Dave. What is the problem here?

I’m glad you asked. Let’s talk about clipping. Clipping is a nice word for DISTORTION. If you send a mix that has distortion to a mastering engineer, your project can become a rescue mission (or is rejected for not being ready to master). As good as specialized modern plugins are at trying to remove clipping, they can only do so much. And even when you remove some of the clipping, the dynamic range still doesn’t change. If you’ve used heavy gain reduction on your mix, your mastering engineer has minimal control in what they can do to that mix to reach its full sonic potential. Often, your mix will end up quieter than the original you sent to be mastered and getting things like your vocal to come forward and shine in the mix will be at the expense of your snare getting quieter in the mix. It becomes a trade off and deciding what has to be sacrificed. Mastering should be about taking your mix to its fullest quality level, not making compromises that cause it to fail to reach it’s full potential.

My mastering engineer sent my mix back asking me to take off the heavy bus processing that was used, or to reduce it a bit. What do I do?

Usually this just means to mute whatever plugin you were using to make it loud or reduce the amount that is being processed. If you take off the processing though, and your mix is clipping the master level meter, you have a couple of options. The best is to reduce the levels of all of your individual tracks equally so that there is no clipping and then re-bounce/export. The second, which is not considered “best practice” is to lower your master fader level until the clipping stops. What you don’t want to do is leave the gain reduction plugin as it is and just lower the volume of the master fader, which defeats the purpose and doesn’t change the reduced dynamic range (ratio) issue we’ve been discussing (this is like only flossing the night before). And finally DO NOT, I repeat, DO NOT open the mixed WAV file, lower the volume, and then bounce/export it like that. You’re not creating headroom or making it passable to master. Mastering engineers are quite capable of turning down the level like this one their end - what we need is to have you turn down the gain reduction processing instead. In the end, it comes down to this: Do you want to master the record yourself or do you want a professional mastering engineer to do it? If it’s the latter, send a great sounding mix and they will almost always take it further than you can on your own.

Let’s talk about 2 track beats for a minute.

I get it. Money can be tight sometimes. You’re trying to manage a budget for your project and paying for trackouts from a beat maker sometimes just isn’t possible. I’m not going to talk about the mix engineer’s job that is limited because they can’t make adjustments to individual elements in the beat. The same issue with you sending a mix to a mastering engineer that has been heavily processed with gain reduction ALSO APPLIES TO YOUR MIX that you are monitoring and tracking vocals with. When you hear a prospective beat, most producers have used a maximizer/limiter so that it sounds loud and is slapping extra hard to get you excited about buying and using it. This is just good business sense. But what you need to ask for is either a version that doesn’t have that processing on the master bus, or to include both versions in the purchase. When you go to track and mix your record with this 2-track beat, use the quieter version and then use your maximizer on the stereo bus so that you still get all that hype and excitement while you’re performing your vocals. Then, when you’re done, simply take off the maximizer, bounce it and send it to the mastering engineer. If you don‘t do this, and you track and mix your record with a heavily processed 2-track beat, you’ll be in the same boat as before where a mastering engineer won’t be able to do anything to your record and they’ll ask you to take off the processing and you’ll be like “I can’t. The beat came that way.”. Get the unmastered beat if not the trackouts.

Hopefully this sheds some light on managing dynamic range, clipping and getting a mix that a good mastering engineer can hit out of the park for you. Send me a message with your comments here! - Dave