I’ve mentioned before that I much prefer to work with WAV or AIFF files rather than MP3, but some people aren’t clear as to why – so let me explain further.

When songs are mixed, they’re usually mixed down to a WAV or AIFF file. Then, if the artist wants an MP3 copy, that mixed file is converted to MP3. However, quality is lost during the conversion process compared to the original mix. If I can work with the original mix, then I’ll be able to master while working with the highest quality version of the song.

There’s an old computer saying “garbage in, garbage out,” which basically means if you feed it bad data, you’ll end up with bad data. It’s the same kind of situation here – mastering a high-quality file will give a better end result than mastering one that has been data-compressed.

Some people think that converting an MP3 file to the WAV or AIFF format will restore the quality lost during the conversion to the MP3 format, but this is not true. The MP3 format actually deletes sound (that’s why the file size is smaller), and once that sound is deleted, you can never get it back. So while you can convert MP3 to WAV, all you’ll have is a faithful reproduction of the MP3 format file – not the same level of quality as the original.

One of the best things about doing mastering for Broadjam is I get to hear some really great music I probably wouldn’t have heard otherwise. I recently got into a conversation with one client who has had several songs mastered (like many of you, it just takes hearing one mastered version before coming back for more), and he wondered why none of his songs were ever chosen for song placement. After all, I’m not the only person who’s told him he writes great songs (and he has a superb voice, too), so wouldn’t that allow him to make the cut?

Not necessarily, and it’s not at all about “People just don’t recognize good music.” Many song placement opportunities revolve around video that needs audio, and I’ve done a ton of audio-for-video work in my life, both for movies and industrial/promotional videos. If there’s one thing I’ve learned, it’s that the music has to be, in most cases, subservient to the video or at the most, complementary to it. If the music is too interesting, it won’t work because it will take attention away from the video.

There are exceptions, of course; there’s that scene in “Chariots of Fire” when the music carries the visuals – there’s no dialog or other audio at all. And how many times on TV has the music swelled to a crescendo toward the end of a show, as it draws to a conclusion? But if you listen to the music leading up to that crescendo, odds are it was mixed pretty far back while the characters set up the scene, and built up only at the very end.

The problem with the singer I mentioned at the beginning is that he has a very distinctive voice – almost like “David Bowie meets Bob Marley,” if you can imagine such a thing (and even that only gives you a rough idea). His songs are also quite sophisticated, with nice little “ear candy” flourishes and a solid, attention-getting mix. And that’s the problem with placing his songs: They don’t really leave any room for the video, and what makes them such a delight to listen to makes them unsuitable for a lot of video (unless the producer and director would be willing to write a scene around the song, which does happen sometimes – but rarely).

I’m not saying this to discourage anyone, it’s just that we’re talking about a case of “the right tool for the right job.” When you get a notification like “Instrumental about partying needed for teen movie,” remember that the party will be the focus, not the music, and you need to come up with an instrumental bed that can sit in the background and not draw too much attention to itself, while nonetheless forming a perfect, seamless complement to the visuals. Is that hard to do? You bet it is! But once you learn what the video world wants, it makes it a lot easier to give it to them.

Now let’s look at how your brain functions during the mixing process, because in actuality, the human brain is a dual processing system—and that impacts artistic activities. The left hemisphere is involved in more analytical tasks, such as math, decoding directions, reading, and so on. The right hemisphere is more involved with creative tasks and emotional responses; it’s the part that “feels” rather than “thinks.” This is not some weird new age philosophy—it’s possible to hook up electrodes to people’s heads, and see which hemisphere of the brain is working during a particular task.

So what does this have to do with mixing? Everything, and here’s why.

In general, it’s difficult for people to switch back and forth between the two hemispheres. Every musician knows what I’m talking about: suppose you’re in a right-brain groove, generating an idea a minute, when all of a sudden there’s a technical glitch. Now you have to switch over to left-brain mode and begin the troubleshooting process. When you start playing music again, the groove is gone, because your brain became stuck in left-brain mode.

In a conventional recording studio situation, the engineer lives in left-brain mode, the artist stays in the right brain (e.g., doesn’t have to worry about level-setting and such because the engineer takes care of that), while the producer has the difficult job of trying to integrate the two. If you’re trying to perform all these functions at once by yourself, you’ll find it’s not all that easy. This is why it’s always great if you can have associates to help during the mixing process.

However, if you’re flying solo, there are still ways to reconcile the right brain/left brain dichotomy. The most important goal is to make sure you don’t have to think about left-brain activities, so you can stay in right-brain mode. If working with your multitrack recording system of choice becomes second nature, it will be that much easier to stay in right-brain mode. Here are some tips on how to do this.

• Learn the keyboard equivalents for various operations. Once memorized, it takes less effort to just hit a couple of keys than to locate a specific area on the screen, move your mouse to it, go down a menu, select an item, etc.
• Use Layouts to organize specific combinations of windows for certain tasks, like mixing, overdubbing, different types of editing, etc. This requires less effort than opening windows and dragging them around.
• The use of color and graphics goes well with right-brain thinking, as your brain can decode colors more easily than words. This is why it can be helpful to color-code tracks, especially if you’re consistent about it—e.g., always making the bass tracks or guitar tracks the same color.

A lot of you have asked for tips on how to make your mixes better, so let’s give a few tips.

An important point is to realize that the idea of a single person writing, recording, producing, and mixing is a relatively new concept. Until the advent of the home studio, the musician was traditionally part of a team of (hopefully) experienced and musically intelligent people. Two of the people who play an important role on this team are the producer and engineer. In a home or project studio environment, the musician doesn’t necessarily have access to these high-powered talents, and has to perform those roles from within. Although this may seem difficult at first, this experience is probably one of the greatest teachers you can have in learning how to be objective about your playing, your style, and your sounds.

During the mixing process, it helps to be aware of the ideal role of each of the three participants (musician, producer, engineer) so that you can assume those roles at will.

The producer oversees the process, rides herd on the arrangement, gauges the overall emotional impact, and makes artistic judgements about what does and does not work. To fulfill the function of a producer, you need to see each piece as part of a whole, and each track as part of a final composition. If you know where you are going, it’s a lot easier to get there; the job of the producer is to figure out where you are going.

The musician participates in the mix on any one of several levels, from simply observing the producer to making sure the production remains true to the original intent of the music.

The engineer is the one at the session who doesn’t drink, smoke, talk much, or complain, and is responsible for translating the producer’s needs into a technological solution. If the producer says the vocals need more “presence,” it’s up to the engineer to decide which tweaks will result in that particular effect. Of course, this is a stereotype and no stereotype is accurate, but every engineer I ever worked with respected the job and took it seriously. It can be helpful to adopt an engineer’s attitude when mixing; forget about whether you could have done a better solo, and simply work with what you have.

By becoming familiar with these roles, you can apply their differing outlooks to your music and obtain a more balanced perspective. Above all, don’t just mix the music; produce it. (But don’t overproduce it—sometimes tracks are best left unprocessed, and sometimes parts should be removed to create space for other parts. Don’t fall in love with the elements that make up a particular piece of music; keep your focus on the final result.) Always keep in mind that the whole point of mixing is to turn a collection of tracks into a cohesive musical statement.

A lot of the people I master ask if there’s anything they should do when mixing that would make for a more “mastering-friendly” piece of music. I’ve already mentioned avoiding distortion, leaving some headroom, and not using signal processing on the stereo output, and as those elements are so important, I’ll say it one more time:

“Hot” mixes are NOT good mixes for a mastering engineer.

Actually, I lied. I’m going to say it yet again:

“Hot” mixes are NOT good mixes for a mastering engineer.

Do I need to mention it again? No? Okay, next topic.

Quite a few tunes I’ve been sent have the voice mixed too low. The vocal is the focus of most songs, and you want to make sure it’s clear and distinct. After mastering, some people have even listened back and agreed the voice needed to be louder…so they remixed the song, and sent it back for a second pass at the mastering. In every case, they liked the version with the louder vocal more.

Another reason why this is important is because some of the music I get has a sort of “blanket” over the sound, where the music lacks clarity. I can fix that, but in the process of doing so, making all the other instruments stand out more can make the vocal sound less prominent by comparison. This is especially true with a “collapsed” soundstage, where the mix sounds more mono than stereo (another common problem).

One “Broadjammer” sent me a file and before I mastered it, I suggested he mix the vocal up. He didn’t think that was a good idea, so I asked if his mixes used automation. He said yes, so I said okay, do me a favor…send me a mix the way you want it, then run off another mix with the vocal up 1.5dB, and another with the vocal up 3dB. I mastered all of them and guess which one the artist chose? Right, the one with the vocal up 3dB.

So why don’t they get it right during the mix? I think it’s all about the difference between mixing and mastering. When mixing, they’re listening to each individual track, and if they hear the vocal, they figure it’s okay. But when the mastered version comes in, it’s usually at least a week or two after the mix was done, and they get to hear it with fresh ears – and more importantly, they may hear it for the first time as a complete, cohesive piece of music, not a collection of tracks. That’s when they realize that yes, vocals need to be prominent.

One last thing: I think another reason why people mix their vocals too low is because they’re self-conscious. Don’t be. Your vocal is what it is, and if it’s not a great vocal, making it lower in level won’t make it any better – in fact it will be worse, because people won’t be able to hear it! Whenever you think your voice isn’t good enough to spotlight in the mix, think of two vocalists: Bob Dylan and John Lennon. By any classic standards, Bob Dylan’s anything but a great vocalist. Yet he found a distinctive style, didn’t run away from it, and it became his trademark. Lennon reportedly hated the sound of his voice, but it seems millions of Beatles fan respectfully disagreed. So don’t be afraid of your voice; shine a spotlight on it.

Every now and then, I get asked to master an MP3, AAC, WMA, or other data-compressed file. This is different from audio compression, where someone puts a compressor or limiter across the output and squashes the living daylights out of the dynamics; data compression actually removes bits of the file so that it takes up less memory. All the formats above are called lossy compression formats, because when you convert a WAV or AIF file to one of them, part of the music is gone forever.

How much is gone depends on the bit rate. Higher bit rates take up more memory, but sound better. The “internet standard” bit rate is 128kbps (kilobits per second), which is not exactly high fidelity. 320kbps is much better, and is the maximum bit rate attainable with current-day MP3s. Still, you’re losing something, and mastering a data compressed file is always discouraging because I know I’ll hit a wall where no matter what I do, I can’t take it past a certain point because of limitations in the file itself.

I always ask the client if I can possibly get a higher quality version, but sometimes a data-compressed file is all someone has. (By the way, converting a data-compressed file to a WAV or AIF file doesn’t solve the problem, because remember, we’re dealing with a lossy format and some of the music simply isn’t there. So, what you end up with is an accurate reproduction of the data-compressed file, not an improvement on it.) In that case, I can either reject the job, or see if I can improve matters—while warning the client that this is a less-than-ideal situation.

With data-compressed files, the success of the mastering job depends more than ever on the recording quality and the type of material. As I’ve mentioned before, mastering puts music “under the microscope” so if the material was well-recorded, that’s a huge help. Also, loud, thrashing guitars will be more forgiving than, say, a solo harp recording.

Tonight I received a mastering job in m4v format (Apple’s data-compressed AAC format). I wasn’t very hopeful, but I really liked the music and wanted to make it sound as good as possible. Much to my surprise, although it took a lot of effort, I was able to make it sound much better than the original. Of course, if I’d had a higher-fidelity file in the first place, it would have sounded considerably better. But, if you want to define mastering as “taking a mix and making it sound a whole lot better,” this qualified.

Overall, I’m being dragged into becoming more tolerant about dealing with data-compressed material. Whereas before I used to pretty much reject it outright, these days I take a very close listen and make a first pass at mastering before deciding whether to keep going or not. Some songs just won’t fly no matter what I do, but others are indeed “masterable.”

Bottom line: If you’re going to send something in for mastering, please—please!!—send a high-quality WAV or AIF file. But if all you have is something lik an MP3, don’t give up; the more I work with them, the more I’m finding ways to get around the format’s limitations.

By now, many of you have discovered the difference that mastering can make to your music. And of course, as soon as you receive the mastered file, you want to listen to it – but you need to take a few issues into account.

The paramount concern is if your listening room is acoustically untreated, and has bad acoustics. Supposed you mixed in that room, and the bass tends to cancel, thus making the bass sound less prominent than it really is. You then boost the bass as you mix so that it sounds right, and hand off the results for mastering. The mastering engineer will (hopefully) have a properly treated room, recognize there’s too much bass, and reduce the bass somewhat to compensate.

Now you get the mastered file, listen to it in your room, and…that’s right, it sounds like there’s not enough bass because of the acoustical problems in the room that caused you to mix too much bass in the first place! So when listening to the mastered version, A/B it with well-mastered commercial recordings. If they all sound a bit bass-light, and your master sounds equally bass-light, then you know the bass is okay (or at least in the same general range as commercial releases).

If possible, it’s always a good idea to listen to the mastered file on an accurate system at some point (maybe you have a friend with a better listening environment, or can book a few minutes inexpensively during down time at a local studio). It’s also a good idea to to listen over a variety of systems so you know how it will sound on boomboxes, in car stereos, with earbuds, etc.

I also suggest investing in a high-quality set of headphones, and learning their response by comparing your mastered file to commercially-released music. For a given amount of money, you’ll usually get more accurate response with headphones than with speakers. Not only will this help you hear the nuances of the mastered file, but also, it’s a good idea to listen to your mixes at least a few times on headphones in order to catch any little artifacts that might get lost over speakers. I’ve received several files with little clicks or pops that I’ve removed, and the artist didn’t realize they were there – until I told them I’d removed them, and when they went back and listened carefully to the mix, realized that there were indeed “baked into” the mix.

In the past two blog posts, we looked at two techniques you can use when mixing to make the mastering process flow more smoothly—and here’s one more.

Getting Rid of Subsonics. Digital audio technology can—and sometimes does—record and reproduce energy well below 20Hz. This subsonic energy has two main sources: Downward transposition/pitch-shifting, and extensive DSP operations that allow control signals, such as fades, to superimpose their spectra onto the audio spectrum.

I ran into this problem recently when doing a remix of a soundtrack tune. I was adding some limiting to the finished mix, and in some sections, the level went way down, as if some hugely powerful signal was overloading the limiter’s control signal. Yet I couldn’t hear anything out of the ordinary.

Looking at the two-track mix showed a massive DC offset. After a bit of research, I noticed that these dips corresponded to places in the song where there was a long, rising tone. I had transposed the tone down by several octaves so it sounded like it was coming up from nowhere, but that transposition had moved it down so far into the subsonic region it created a varying DC offset. That’s the signal to which the limiter was responding. Adding a filter to remove as much as possible below 20Hz solved the problem.

You can also use a sharp low-cut filter with already mastered material to cut out subsonic frequencies, but it’s much better to do this type of processing before the files are mixed together, as this can lead to a cleaner mix. In fact, some engineers routinely cut off frequencies below an instrument’s lowest note. For example, guitar doesn’t have much energy below 90Hz, but there can be hum, thumps from hitting the neck, and the like. Cutting the response below 90Hz can remove these non-musical signals, resulting in a cleaner guitar part that won’t cause problems during the mastering process.

In the last blog post, we covered the importance of matching timbres when using loops so that the mastering process doesn’t have to turn into a salvage job. This time around, let’s put peak and average levels under the microscope.

Bringing Peaks into Place. Music has two main kinds of levels, peak levels and average levels. A lot of engineers use mastering to increase a tune’s average level, thereby making it seem louder (regrettably, some engineers and artists take this to an extreme, essentially wiping out all of a song’s dynamics because they sacrifice peak levels that contribute dynamics). To understand the difference between peak and average levels, consider a drum hit. There’s an initial huge burst of energy (the peak) followed by a quick decay and reduction in amplitude. You will need to set the recording level fairly low to make sure the peak doesn’t cause an overload. As a result, there’s a relatively low average energy.

On the other hand, a sustained organ chord has a high average energy. There’s not much of a peak, so you can set the record level such that the sustain uses up the maximum available headroom.

Entire tunes also have moments of high peaks, and moments of high average energy. Suppose you’re using a hard disk recorder, and playing back a bunch of tracks. Of course, the stereo output meters will fluctuate, but you may notice that at some points, the meters briefly register much higher than for the rest of the tune. This can happen if, for example, several instruments with loud peaks hit at the same time, or if you’re using lots of filter resonance on a synth, and a note falls within that resonant peak. If you set levels to accommodate these peaks, then the rest of the song may sound too soft.

You can compensate for this while mastering by using limiting or compression, which brings the peaks down and raises the softer parts. However, if you instead reduce these peaks during the mixing process, you’ll end up with a more natural sound because you won’t need to use as much dynamics processing while mastering.

The easiest way to do this is as you mix, play through the song until you find a place where the meters peak at a significantly higher level than the rest of the tune. Loop the area around that peak, then one by one, mute individual tracks until you find the one that contributes the most amount of signal. For example, suppose a section peaks at 0dB. You mute one track, and the peak goes to 2. You mute another track, and the section peaks at 1. You now mute a track and the peak hits ¬ 7. Found it! That’s the track that’s putting out the most amount of energy.

Zoom in on the track, and use automation or audio processing to insert a small dip that brings the peak down by a few dB. Now play that section again, make sure it still sounds okay, and check the meters. In our example above, that 0dB peak may now hit at, say, 3dB. Proceed with this technique through the rest of the tune to bring down the biggest peaks. If peaks that were previously pushing the tune to 0 are brought down to 3dB, you can now raise the tune’s overall level by 3dB and still not go over 0. This creates a tune with an average level that’s 3dB hotter, without having to use any kind of compression or limiting.

Let’s look at some techniques you can use while mixing to make the mastering process go more smoothly, whether you do your own mastering or hand the project over to someone else.

Matching timbres. If you use loops in your music, be aware of loops whose characteristics are wildly different from other loops. For example, let’s suppose most of the loops were taken from a drum machine you use, but you also inserted a few commercially available drum loops. It’s likely that the latter were already “pre-mastered,” perhaps with some compression or treble boosting. As a result, they might sound brighter than the loops you created.

If you decide to boost the track’s overall brightness while mastering, the commercial loops will now seem “over the top” in terms of treble. I had this happen once when re-mastering a stereo track where everything needed a little extra brightness except for a high-hat loop. It took seemingly forever to use notch filtering to find just the high-hat frequencies and reduce those, while boosting everything else.

This kind of inconsistency can also happen if you use a lot of analog synths, which tend to have a darker sound, mixed with a few digital synths, which tend to be brighter. This will also give problems when mastering, because if you bring down the highs to tame the digital synths, the analog synths will sound much duller; if you bring up the highs, the digital synths may screech.

The solution is simple: To ensure that changes made during mastering will affect all sounds pretty much equally, before mixing, bring “minority” tracks into timbral alignment with the majority of the track’s timbres. However, don’t go overboard with this; some differences between tracks need to be respected (e.g., you might want a track to sound brighter or duller than others, regardless of any equalization done while mastering).