The One-Man Multitrack Band
Part 3: Miking, Editing, Mixing, and Tweaking
This part is the second part on recording, and the third part of this
series. This section includes a discussion of recording and
The Recording Process
A lot of people wonder why in the world I'm doing my own recording and
editing. The answer comes down to economics. You can easily spend
ten grand or more to record an album at someone's overpriced house
For 20 grand, you can build a very, very nice recording studio, assuming you
have a spare room to use for it. This isn't right for every band, but it
often makes a lot more sense than renting a studio for every album. For
those of us who can't afford to blow that much money (or don't have the
space), a fraction of that will buy you a decent home studio.
For example, a dual-2 GHz PowerMac G5, Bias Deck or Logic, and an M-Audio
Delta 1010LT will run you around 4 grand. Add half a terabyte of disk
space for a few hundred, and you're good to go.
A good engineer helps, and a good band is pretty much a must, but what makes
many CDs from less well-known bands bad is a lack of time spent tweaking,
re-EQing, sweetening, and so on. They're in such a hurry to get it done that
they don't spend the amount of time that's needed to get everything "just
right". This is not a weekend activity, or even a good way to spend a couple
of weeks. To do it right should consume you for months. If it doesn't, you
probably did something wrong.
I'll discuss my thoughts on editing more later, though. First, I should
write about the recording process itself. For the purposes of this
section, I am assuming a modern studio with computer-based editing, thus
I will ignore issues like bounce mixing to a new tracks, deciding whether
to record tracks wet or dry, and other limitations of tape-based studios).
There are four main concerns to keep in mind when recording: hardware,
engineering, microphone placement, and the performance itself.
The first part is the easy part. You either have decent hardware or you
don't. If you don't, you try to work around the problems as best you can.
That's one place where good engineering kicks in....
The second part takes years of practice to get right, but good hardware
and software can largely make up for it. For example, if you have
hardware and software capable of 24-bit audio, it gives you enough
dynamic range that getting perfect levels the first time is of less
importance than with traditional 16-bit systems.
There are a few other areas where engineering is vital, though. Checking
each recording session for hum and noise while you're recording can save
hours of time later. Catching bad cables before you end up missing the
perfect solo also helps. Having spare cables, of course, is an absolute
The third part, microphone placement, could literally be a book unto itself,
but it's worth covering briefly here for completeness.
Microphone Placement for Piano
Microphone placement for pianos is a bit of a black art. No two engineers
will exactly agree on placement for any given piano. A good rule,
though, at least for grand pianos, is to place the microphone under the lid
above where the bass strings cross the treble strings in a grand piano. If
there is a sound hole there, you should probably place it directly
above that. Otherwise, just take a guess.
I like to mic grand pianos with a stereo microphone pointing towards the
player. This has two purposes. First, it gets a little more of the
characteristic pedal thump, hammer impacts, and damper noise that makes
it obvious that you're dealing with a real piano and not a keyboard.
Second, it gives you a very different sound in both channels, dominated
by low and high sounds, respectively. This means that you can adjust
equalization by adjusting the level and pan of the two channels. This
is VERY helpful if you ever end up with your microphone in a
slightly different place, as it allows you to easily recolor the sound to
match the earlier recording.
For an upright piano, I generally place a mono microphone (like a Shure 555SB)
nestled into the gap between the vertical columns on the back. It doesn't
matter much how you point it, as the sound is pretty generalized by the
time it gets to that side of the soundboard anyway.
Microphone Placement for Clarinet
Proper microphone placement for clarinet depends largely on the sound
you're trying to achieve, but in my opinion, the most "true" sound comes
from placing the microphone a few inches above the left hand fingers at
about the second or third tone hole from the bottom.
Microphone Placement for Brass
Brass placement also depends on the desired sound. If you place a
microphone at the same height as the horn and point it directly
into the bell, you will get a very crisp sound, which may be appropriate
if you're doing a fanfare for trumpet, but sounds rather thin for trombone.
I find that a good warm tone can be achieved by aiming the bell above
the microphone by a few inches. That way, you get more reflected sound
relative to the direct sound, and by being out of the direct path of the
sound, you also end up with more of the lower harmonics, which also adds
to the overall richness of the sound.
Microphone Placement for Guitar
Back when I was in junior high, my father taught a recording class at
the local university. One of the many tricks I picked up from sitting
in on that class was to mic guitars with two pickups, one near the
tone hole and one above the frets. This gets you the ability to
decide in post-production how much fret noise to include relative to
Placing the mic slightly off from the tone hole has a similar effect to
pointing brass instruments slightly away from the microphone, creating
a richer, more vibrant tone. Thus, I tend to put the mic slightly towards
the frets from the tone hole, and about a foot out to keep it from
getting in the way. A second mic can be added over the frets if desired;
I sometimes do that, depending on my mood.
Microphone Placement for Percussion
Percussion instruments can be tricky to mic, mainly because it isn't always
clear where the interesting harmonics are being generated. In many cases,
they are generated by the shell of a drum, but in some cases, they mainly
come from the head. If you're trying to get the same sound that you would
get when listening from a few feet away, you may find it easiest to simply
put your microphone a few feet away.
That having been said, for most drums, you will probably get the best sound
my carefully mixing two microphones---one microphone from above the top head,
one from below the bottom head or open bottom. For shakers and other hand
instruments, keep the microphone as close as possible without putting it
so close that you accidentally hit it.
I don't use drum sets generally, so my advice on that is largely from
memory, but a good rule is a directional (cardioid) mic pointing straight
down over the snare, one under (to get more of the snare sound and to
pick up the toms), two over the cymbals (assuming a setup where one cymbal
is on either side, and where one is near the high hat) and a bass drum mic
for the bass drum. An omnidirectional mic over the drummer's head can
also add useful depth to the sound.
As always, drum set configurations vary, so you may have to experiment a
bit to get the sound you want. Just remember that it is always better to
have more microphones than less if you plan to experiment in
Microphone Placement for Voice
As with any instrument, microphone placement for voice determines the
color of the sound you will obtain. While this can be further adjusted
to some extent with effects, one key point sticks in my mind. Make sure
your placement is consistent between takes. There's nothing more annoying
than having three takes of a tricky section and not being able to use
one of them because it sounds different from the others.
The distance between the microphone and the singer has a tremendous impact
on the tone that the microphone picks up---more so than is typical
with instruments---and thus this aspect is of the greatest importance
as far as consistency is concerned. The closer the microphone is to the
singer, the richer and warmer the tone will sound. That's why many singers
naturally tend to swallow the microphone---they're trying to get a warmer
tone because it is more pleasing to the ear. (And all this time, the
singers thought they were just trying to make the sound louder....)
A good distance is between 4-6" from the singer's mouth, assuming a
typical vocal microphone. However, this does vary from microphone to
microphone, so some experimentation is needed to find that magic point
where the sound is neither muddy (from having the microphone too close)
nor thin (from having the microphone too far away).
Another critical issue in recording vocals is the microphone height. This
is the area that is probably most thoroughly neglected by musicians, but
a poorly placed microphone can make Luciano Pavorotti sound like Fred
MacMurray (remember the original absent-minded professor).
The reason that most people don't place microphones correctly is that they
have a fundamentally incorrect notion about how sound is produced.
The sound of the human voice does not entirely come from the mouth and
throat. The entire human body acts as a resonance chamber, adding color
to the sound. If your microphone is pointing directly at the performer's
mouth from right in front of it, much of this resonance will be lost,
buried by the fundamental tone coming from the direct sound.
For the male voice to sound believable, much of this tonal depth must
be present. For the female voice, this is of less importance, as the
harmonics are closer together and are harder to hear anyway. (A few
harmonics up from the fundamental, and you're outside the range of
human hearing.) This is one of the main reasons why the female voice is so
much easier to record well than the male voice---you can literally place the
microphone almost anywhere and get an acceptable recording of the female
voice. Not so for male voices.
A good microphone placement for the male voice, in my experience, is
slightly below the jaw and a few inches out, pointing horizontally or
slightly upwards. This gets most of the direct sound, but balances it
nicely with the additional harmonics generated in the chest, resulting
in a rich, full tone.
The absolute worst place to put a microphone for male vocals is above
the vocalist's head. This gives you none of the chest reverberation,
very little of the direct sound, and lots and lots of upper harmonics
bouncing off of hard surfaces nearby. The result is a very distant sound,
tinny and lifeless.
Such placement is bad even for female voices, but because the lower
harmonics of their voices are the ones being reflected rather than the
upper harmonics, the result, while still hollow and weak, is at least
This placement is also the reason why so many choral recordings sound so
incredibly bad, with way too much soprano sound and with no real warmth to
the basses and tenors. Often it has little to do with the singers and
much to do with the overuse of overhead microphones. These are fine for
adding brightness to reduce the natural loss you might get in a church
sanctuary or theater with cloth-covered seats or pews, but that is the
limit to their usefulness, in my opinion.
Microphone Case Study and Conclusion
Quite possibly the worst example of microphone use I've ever seen or heard would
have to be a church I visited in Tennessee a few months ago. The church
was huge, but rather than allow its natural reflections to create a vibrant
sonic environment, they deadened it with so much acoustic foam that you could
package every iPod Apple ships for a century with it. That's not what
made it sound bad, though.
Their speakers were all horns. That didn't
help. That's still not what made it sound bad, though. Their podium
mics were the tinniest things I'd ever heard, but that still wasn't the
The biggest problem was that they had two concepts for microphones: border
mics and ceiling mics. Border mics are great for capturing the sounds of
tap dancing. They are terrible for quite possibly everything else. Just
say no. Ceiling mics... see my rant above.
The net result of all these problems combined was that the entire place
sounded like what would happen if you took a nice sound system, clipped the
wires leading to the woofers, clipped the midrange wires, and tuned to an
AM radio station. 'Nuff said.
Unless you have no other option for sound support, don't rely on
overhead mics except to add high frequencies to make up for the natural
decay of the room you're in (for sound reinforcement). It is quite impossible
to get a truly good sound out of any other use of overhead mics.
In short, microphones should generally be placed a few inches away from
wherever the sound is coming from, with care taken to make sure the sound
is really coming from where you think it does.
If you keep those things in mind, it should help you to come up with
ideal placement for almost any instrument or voice combination that
you might run into.
- A clarinet's sound comes mainly from the tone holes.
- A guitar's sound comes partially from the body, partially from the
tone hole, partially from the strings, and partially from the frets.
- The sound of a voice comes in part from the nose, throat, and mouth,
but also in part from the chest cavity.
- The sound of a brass instrument comes from the entire body of the
instrument, not just the opening in the middle of the bell.
- Direct sound almost always puts too much emphasis on the fundamental.
- Overly indirect sound puts too much emphasis on upper harmonics.
The Editing Process
In my opinion, a good mix takes at least five passes:
- First pass: draft recording
- Second pass: rerecording
- Third pass: editing
- Fourth pass: reediting
- Fifth pass: sweetening
Cut scratch vocals and instrumentation. Edit until you can stand it, and until it is rhythmically consistent.
In this stage, you should Replace instruments one at a time. Once you have rerecorded all of the instruments, you should record several cuts of the vocals. Edit everything until it sounds reasonable. Add draft effects (reverb, EQ, and so on).
Third pass: editing.
This is actually dozens of smaller passes. Repeatedly play through each song with a notebook and note anything you don't like (rhythmic errors, bad pitch, pops and noise, vocal plosives, timing variation between tracks, and so on).
It helps to burn a CD at the end of each editing pass. Listen to it everywhere---at home, in the car, at work---listen to it until you know every single spot that feels uncomfortable, then go back and edit. Repeat this process over and over until you can't hear anything wrong.
This is the stage where you may end up adding some compression in order to get sufficient control over gain, depending on your audio software (and depending on the style you're trying to achieve). Your mileage may vary.
Fourth pass: reediting.
Take a week off and don't think about the music. Listen to other people's music exclusively. After that, go back and do the same thing you did in step 3. This will probably take four or five passes before you're satisfied, and you'll feel like each pass gets you closer and closer to the desired quality.
By the time you get to the end, the music will sound technically perfect, but flat and lifeless. This is where most people stop the first time, hence the reason that many self-produced CDs sound much less polished than they should.
Fifth pass: sweetening.
In this stage, you should add equalization, consider stereo simulation if you recorded things like a piano monaurally (though a little goes a long way), and always shift most of your instruments off-center to add stereo spread.
Listen to it through as many different sets of speakers and headphones as you can. Each one will sound a little different. Try to find the right balance between instruments and the right equalization so that it sounds good everywhere you listen to it. Hopefully, you already did some of that in the third and fourth passes, but it helps to check it again every time you add effects, as they tend to change the character of the sound in ways you might not expect.
By the time you have done all this, you should be utterly exhausted and ready
to scream. Take a few weeks away, then go back and listen to it again, just
to make sure you're happy with it. If you are, ship it; if not, go back and
edit it until you are. If you aren't under a deadline, it is far better to
take the time needed to get a good recording than to cut corners to get
it out sooner.
The one-man-band series continues in the fourth and final part,
The Kitchen Sink. Also watch for other articles about related
subjects as Thirty-Six moves from production into duplication,
distribution, and promotion.