Mixing Technique 1

http://pullpud.tripod.com/mixing.html

There are as many different ways to mix a recording as there are engineers; however, certain methods
can never steer you wrong.  This article will discuss one basic approach to a successful mix.
 
First, it's always a good idea to make a final mix on a different day than the tracking was done.  This
is true for a number of reasons - namely, after recording all day, most people are not only tired but
their ears are saturated.  Start on a fresh day for the mix if at all possible.
 
A WORD ON EQ:  Use EQ to replace missing bass or treble (by using the high and low shelving controls),
reduce excessive bass or treble, boost room ambience (high frequency shelf), improve tone quality (using all
the controls), and help a track stand out in the mix (by using the parametrics).
An instrument's sound is made up of a fundamental frequency (the musical note) and harmonics, even when
playing only a single note, and it is these harmonics that give the note its unique character.  If you use EQ to
boost the fundamental frequency, you simply make the instrument louder, and don't bring it out in the mix.
It should be noted that a particular frequency on the EQ (say 440 Hz) corresponds directly to a musical note on
the scale (in the case of 440 Hz, to the A above middle C - hence the expression A-440 tuning reference).
Boosting the harmonic frequencies, on the other hand, boosts the instrument's tone qualities, and can therefore
give it its own space in the mix.  Below are listed useful frequencies for several instruments:
Voice: presence (5 kHz), sibilance (7.5 - 10 kHz), boominess (200 - 240 kHz), fullness (120 Hz)
Electric Guitar: fullness (240 Hz), bite (2.5 kHz), air / sizzle (8 kHz)
Bass Guitar: bottom (60 - 80 Hz), attack (700 - 1000 Hz), string noise (2.5 kHz)
Snare Drum: fatness (240 Hz), crispness (5 kHz)
Kick Drum: bottom (60 - 80 Hz), slap (4 kHz)
Hi Hat & Cymbals: sizzle (7.5 - 10 kHz), clank (200 Hz)
Toms: attack (5 kHz), fullness (120 - 240 Hz)
Acoustic Guitar: harshness / bite (2 kHz), boominess (120 - 200 Hz), cut (7 - 10 kHz)
 
The thing to remember about EQ is not to get carried away - be specific and use it only when you need it, where
you need it.  If you get the mic placement correct and use good preamps on a good sounding instrument, you
shouldn't need much.
 

I

O.K.  Start with the lead vocal by itself.  If you used a good mic and a good preamp, all you should
need is a little compression and reverb, and perhaps a bit of delay depending on the song and the
production style (I personally like vocals rather dry).  Patch in the compressor and set the ratio to
around 10:1 or so.  Then set the threshold so that the compressor is grabbing the loudest peaks; this
will allow you to set the overall gain of the track higher.
 
Now add just enough reverb to sweeten the track (or to your taste if you want a cathedral . . .)  If the
voice sounds overly boomy, roll back the bass a bit; if it sounds muddy or muffled, boost the high
end a little.  Again, if it went to tape correctly, you shouldn't need much EQ.
 
Next, go to the kick drum.  You shouldn't need much if any reverb or any other effect here (I don't
usually use any effects at all on kick).  If the rest of the kit is spilling over badly, you may want to
gate the track; I rarely find this necessary.  Find a recording (by somebody else) that you like and 
reference it (A-B) back and forth with yours, paying special attention to the sound of the kick drum.
Do you have basically the same amount of slap, thump, etc. This is not to say you should attempt to
copy someone else's production, but rather that you need a frame of reference that you know is an
acceptable industry-standard mix.  Apply EQ if necessary.  Pan it straight up the middle.
 
Go to the snare.  You will probably need to EQ the snare a bit.  I've found that a little cut around 800hz
and another fairly wide boost around 5K usually does the trick if I printed the track flat and I'm going for that general "radio" sounding snare drum sound.  Select an appropriate reverb for the
snare - plates are usually nice.  Set the levels of the kick and snare so that they bear the proper
relation to each other; you can also watch the peak meters on the mixer to see that the kick and snare
register about the same.  I usually pan the snare right up the middle with the kick.
 
Bring in the toms.  You may need to find a passage in the song that contains a good roll so that you
can set the proper levels.  Volumes could be written on reverbs, EQ, gates, etc. used to obtain
different drum sounds; for example, Phil Collins uses all of the above and then some on each individual
drum to achieve his trademark sound.  It just depends on what you're going for, how much time
you have to spend tweaking, and how much gear you have to work with.  If your gear is limited,
you can't go wrong by going for the most natural sound possible.  You will probably want to pan
the toms to the left and right of center (about 11 o'clock - rack tom - and 2:30 - floor tom).
 
Now add the drum overheads.  I usually pan them about 9 and 3 o'clock for a good stereo spread
that stays tight and focused.  If you have reverb on the snare and toms, you shouldn't need to add
much more here as too much reverb = mud.  Try to set the levels so that everything sounds as close
as possible to the kit being played in the room.  Reference against that other recording again.
 
Set the vocal level relative to the level of the drums.
 
If you have extra compressors, you should insert a compressor each set to around a 4:1 ratio on the
kick and snare channels; peaks from these 2 instruments are the biggest reason many people have a
difficult time getting their CDs to play as loud as commercial ones.  Although, let me say this: a good
many CDs these days are ridiculously overcompressed, so don't get carried away.  Let me also say that on most of the records we hear on the radio, the sound of the drums is not exactly what a real drum set sounds like if you're setting in the room with it.  When, in the real world, for example, would you hear a 2 second reverb on the snare and yet the kick is still dry as a bone - not in any real room I've ever been in.  Yet this sort of thing is the standard on records these days.  So don't be afraid to experiment.  It all depends on whether you're a "purist," or a "producer" (or a little of both, somewhere in between).
 
Next, bring in the bass.  Without changing the level of the kick drum, you may want to solo just the
kick and bass tracks and adjust the bass level such that there is a good "lock" between the two.  When
you hit the right spot, you'll know it; the two should complement one another almost as though they
were one instrument.  Some compression is almost always necessary on bass.  I use about a 4:1 ratio.  Just don't set the attack too fast, or you'll squash all the life out of it and it'll just sound dead.  
A little chorus or flange (just a little) can sometimes sound nice on certain songs as well.
 
Now listen to what you have so far: vocal, drums, bass.  Contrast it against your reference and make
minor adjustments if necessary.  What you have so far is the bedrock of the song and the focus of the
recording.  Everything else will be added to (1) support the vocal and (2) complement the rhythm.
If you discover a track along the way as you are building the mix that either takes away from the
vocal or clashes with the rhythm, either scrap it or do a last minute fix.  Once you have the drums,
bass and vocal sounding right with one another, change them under no circumstances. 
 

II

Now we approach the other rhythm instruments (guitar, piano, etc.).  If there is both rhythm guitar
and piano, pan them opposite one another (around 3 and 9 o'clock).  Bring them in one by one and
check your reference after each.  You may have to thin a strumming acoustic guitar by rolling off some
of the bass if it seems to be mushing up the rhythm section.  Also, avoid using too much reverb
on rhythm guitars as this will lead to muddy sound.
 
One by one, add the lead instruments where they occur.  You've probably heard the whole track
through 5 - 10 times by this point.  You can adjust their panning as suits you; suffice it to say that
each lead instrument should have its own spot in the stereo image.  Here is where you can go nuts
with all those fancy effects processors should you so desire.  As you're bring in a lead, pay special
attention to whether or not it clashes with the vocal.  Guitars tracked with tube amps don't usually
require much compression, as one of the effects of tube saturation (distortion) is a natural limiting.
Violins often need quite a bit of limit, depending upon the player.  
 

III

Things should be sounding pretty good by this point.  The last thing to do now is blend in the harmony
singing and percussion / string sections / sounds of exploding bombs, etc. if they occur.  One trick
I've found that helps blend the harmony vocals is to add a bit of chorus to them along with the reverb.
I really like the chorus in the Eventide Harmonizer for this application (I'm also fond of the
Lexicon PCM-80 for chorus).  
 
Now, listen to a few seconds of your reference recording and a few seconds of your mix.  You should
be very much in the same ballpark.  Don't touch anything.  Go get a coke, make a cup of coffee, take
5 minutes - call your mother and say "Hi."  The point is to let your ears freshen up just a bit.
 
Come back and listen top to bottom for anything weird; you can be setting the level on your mixdown
deck (Hard Drive, 2-track reel-to-reel, DAT whatever) at the same time.  If everything still sounds
good then Bombs Away - print it!
 
One final note: a well arranged and well tracked song shouldn't require a great deal of fader-riding;
you may have to bring up a lead instrument here and there, ride a vocal on soft passages or at the ends of lines, but that should be about it.