Limited Time Only: Vocal Plugin Sale | Shop Now »

10 Mastering Myths BUSTED | Audio Myth Busting

Apr 08, 2021

Let’s BUST 10 popular mastering myths—we’ll filter out the facts, avoid limiting our viewpoints, compress a whole bunch of information in one article and maximize our understanding.

By Craig Anderton

10 Mastering Myths BUSTED | Audio Myth Busting

If you search the web for myths about mastering, you’ll find pretty much the same ones, recycled over and over again, often from mastering house blog posts. Some of them are self-serving, some are genuinely helpful, and based on experience, but I wanted to find some new myths we can bust (as well as some older ones that need a refresh).

So, get out your sledgehammer, put on your safety goggles, and let’s start smashing those myths!

1. Don’t go over 0 when mastering for CDs, and your levels will be fine.

No, because your highest peaks need to be below 0, even if only by a tiny amount. A CD pressing plant that receives a master with a string of samples at 0 will probably reject it because there’s no way they can be sure the audio stopped at 0 and didn’t go over 0 during the mastering process.

2. All the streaming services use different LUFS levels—so now I have to make a master at -16 for Apple Music, -13 for YouTube, -14 for Spotify, -11 for Spotify Loud, etc. LUFS sucks!

Not at all! You can master to any LUFS level that sounds the way you want the music to sound. Limiting the dynamic range to -9 LUFS or so is common with rock music—it gives a certain sonic character. The streaming companies will just turn that track down to their standard playback level (Fig. 1). Your track won’t sound louder than other tracks, but it will have the character you wanted in your master.

Waves PAZ Analyzer

Figure 1: Based on readings from Waves PAZ Analyzer, this song’s LUFS is -11.0. Uploading it to Apple Music won’t blow anything up—they’ll just turn the level down to -16 LUFS.

3. True Peak readings are technical BS; it’s stupid to worry about true peaks going over 0. (A “professional” mastering engineer actually said this to me.)

The reasoning is that because true peak indicates signal levels over 0 after going through D/A conversion, not in the digital media itself, it will be noticed only by consumers on playback…and they listen to MP3s on earbuds, so screw ’em. Still, if preventing distortion at the consumer’s end isn’t enough incentive to pay attention to true peak readings, you’ll hear this distortion while mastering because you’re listening to the signal after D/A conversion. So, you might mistakenly think the distortion is in the master itself.

4. My masters for CD and streaming sound really good, so now I’m ready to master for my peeps who want to do vinyl releases. I’ll just make sure the bass is centered so the needle doesn’t jump out of the groove.

Mastering for vinyl is why mastering engineers acquired the reputation of being able to do something no mere mortal could ever understand. The process involves a deep, physics-based understanding of the tradeoffs involved with running time, bass response, distortion, surface noise, high frequencies, and even album sequencing (softer songs are best at the end of a side, due to the phenomenon of inner groove distortion.) Whoever cuts your vinyl may be able to compensate for some mistakes made while mastering, but for the sake of your clients, have an engineer with extensive vinyl experience do the mastering.

5. Mastering can never fix a bad mix.

Well, as the old saying goes... “never say never.”

Case in point: I was asked if I could do anything to salvage a song from the early ‘90s that had never been released because although the song was great (and had a gorgeous vocal), the drum machine kick was mixed so loud that it ruined the song. The original multitrack files were long gone, and no EQ or dynamics could fix the kick without wrecking some other aspect of the mix.

However, I found an isolated kick sound in the intro, brought the stereo mix into a DAW, and created a track with a copy of the isolated kick in all the same places as it appeared in the stereo mix. Then I flipped the track with the isolated kick out of phase, and amazingly, it canceled the kick in the master almost completely. Changing the level on the kick track dialed in just the right amount of cancelation for the perfect kick level—mission accomplished! Although I'd say it’s true that mastering can rarely fix a bad mix, I’d never say “never.”

6. Only analog gear is good enough for truly professional mastering.

When audio engines were 16 bit and running on Motorola 68000 chips, this was true. But today, it’s a myth. In blind, A/B testing, even the best audio engineers can’t hear a difference between analog processors and state-of-the-art digital processors—thanks to more powerful digital technology, better knowledge about what contributes to the “analog” sound, and more evolved algorithms to quantify analog behavior (Fig. 2). Of course, some analog gear is great…but so is some digital gear. There’s no reason not to use one technology, the other, or both.

Abbey Road TG Mastering Chain

Figure 2: Abbey Road TG Mastering Chain. Digital emulations have evolved to re-create not only individual effects but complete analog mastering chains.

7. Musicians can’t do mastering; it’s too demanding and too different a skill.

Anyone who’s learned how to play a musical instrument knows a thing or two about learning a skill, and there’s no reason why musicians can’t master their own music. The real question is whether they should.

Sure, you can master your own music, but without experience, ears, proper monitoring and objectivity, you probably won’t master it well until you’ve learned the mastering skill set. So, we’ll classify this myth as “partially true.” However, you’ll learn a lot if you try your hand at mastering. Master a mix, then have a pro mastering engineer master it. You’ll learn much as you compare the two. Reality check: Every world-class mastering engineer started off not knowing how to master.

8. Normalizing files for an album project is just plain wrong.

This is partially true because some people blindly trust that normalization will produce the same volume level for different material. And that would be true if all the songs in an album had the same peak levels and average levels—but they almost certainly won’t.

However, normalization does have its uses. I normalize all cuts to some level under 0 and then look at the LUFS readings. That provides a good indication of which songs have louder perceived levels than others. After adjusting them for the same LUFS readings, I then listen to them analytically because ears are the ultimate judge of level. Some songs that have the same LUFS as another song may need to be a bit softer or louder, based on the nature of the music. Although depending on normalization isn’t a good idea, using it as a starting point can be.

9. Mastering is best done on the master bus when mixing.

This is a myth because it’s not the best way. It’s one way. Personal bias alert: I still consider mixing and mastering as two different processes. Mixing is about getting the best balance and tone among all the tracks, while mastering is about optimizing the final mix.

What’s more, I listen differently when mixing vs. mastering. At least for me, mixing is about listening to what multiple, separate elements contribute to a unified whole, while mastering is about listening to that unified whole and seeking out every little element that needs tweaking. I prefer to treat mixing and mastering as separate disciplines and do them at separate times.

10. Two-for-one myths! EQ should go before dynamics / dynamics should go before EQ.

Some people think one or the other is the way to go—but both are myths, and both are true because the correct answer is situational. Compressing or limiting after EQ “undoes” some of the boosting or cutting that EQ does, while EQ after compressing or limiting undoes some of the dynamic range consistency introduced by compressing or limiting. I usually favor EQ before limiting because I try to avoid adding too much limiting anyway, and it will catch the occasional peak introduced by EQ. But when mastering a classical solo harpsichord album by Kathleen Mcintosh, I added a tiny bit of limiting (shhh! don’t tell anyone!), and then a little EQ afterward, of course making sure to stay at the proper peak levels.

Always remember, the music will tell you what it wants…not your gear, and not the internet.

Musician/author Craig Anderton is an internationally recognized authority on music and technology. He has played on, produced, or mastered over 20 major label recordings and hundreds of tracks, authored 45 books, toured extensively during the 60s, played Carnegie Hall, worked as a studio musician in the 70s, written over a thousand articles, lectured on technology and the arts (in 10 countries, 38 U.S. states, and three languages), and done sound design and consulting work for numerous music industry companies. He is the current President of the MIDI Association.

Check out more audio myth busting: 11 Compression Myths BUSTED.

Want to get more tips straight to your inbox? Subscribe to our newsletter here.