Fix it in the Mix

PLEASE NOTE: This article has been archived. It first appeared on in August 1998, contributed by then Editor-in-Chief Rip Rowan and Media and Mastering Editor Lionel Dumond. We will not be making any updates to the article. Please visit the home page for our latest content. Thank you!

How Not to Get a Good Recording

When I was first starting out as a musician, I was blown away by even small recording studios.

The studio where my band was cutting its teeth was a little garage operation. The owner had converted the utility room and garage into a control room and a studio. This was pre-ADAT, with a 16-track digital recorder, a 24×16 mixer, a couple of DATs, some standard rack effects (DBX compressors, Eventide Harmonizer, Yamaha reverbs, etc.), and a few decent mics.

The problem with the studio was that the recording space was dead. Dead-dead-deadsky. Over a foot of fiberglass insulation and sandbags had eaten the life right out of the room. The Yamaha piano, normally a bright instrument, sounded limp. You couldn’t get a guitar to feed back. The drum booth, about the size of a phone booth, was too small to allow the full tone of the kick drum to develop, and the cymbals would just kill you.

Of course the point of all that insulation and sandbagging was to keep the sound out of the neighborhood. It worked. Metallica could have loaded their entire PA into that room and be completely inaudible outside. Hell, it would have been almost inaudible inside!

The reason the room sounded dead was because – even with a foot of fiberglass insulation – bass frequencies below 200 Hz were not being absorbed. Everything else was being very well absorbed. The result was as if you plugged in a room reverb and filtered off the mids and highs from the reverb. A heaping spoonful of low frequency mud. Early, low frequency reflections created in a small, dead room cause the bass frequencies to be overly loud and unclear.

Moreover, the sound of some instruments like snare drum, piano and electric guitar are created in large part by the interaction of the instrument with the room. All instruments interact with the room, of course, but some instruments are particularly affected. And in this room, these critical instruments were lame. You simply cannot record a quality rock song with a dead snare and dead guitars.

Dead sound in the studio means dead sound on tape. No matter what outboard effect you like, you can’t polish a turd into a diamond. Garbage in = garbage out.

No matter what we would say about the sound we were getting, the monkey running that studio had a single, irritating reply:
“We’ll fix it in the mix.”
And no matter how long and hard we would mix that puppy, we’d get crap. To which there was the second, even more irritating bullshit response:
“We’ll fix that in mastering.”
Later, I heard the comeback to these irritating statements that I wish I had known then. This quote came not from a music guru but from the Father of Business Process Reengineering – Dr. Michael Hammer – who said:
“‘FIX’ is something you do to a CAT!!!”
He is so right. Some things can’t be fixed, but need to be blown up and started over. The bad sounds created in this studio’s rooms were not surmountable with any downstream processing. They were just bad. All the wrong harmonics had been captured. All the worst resonances were present. The mixes were mud. The instruments sounded small and distant. The room was just plain bad and it sounded that way on tape.

Bad mics can do the same thing as a bad room. Take the ubiquitous Shure SM-57. One of the most easily spotted mics in any rock studio. Not a bad mic at all. Sometimes, in certain applications it’s the mic that sounds the very best. But, the SM-57 can be a little harsh in the upper-mids (~4KHz) and a little “covered up” (poor transient response, and poor response above 8K).

If you’re not quite happy with the sound of that 57 on that snare or guitar, and you have a choice, then try a Sennheiser 421 or an AT-4050. Both of these are great mics on a guitar or snare, and though they add a similar character as a 57, they have more body and brightness. It’s a better idea than going with the 57 and saying “we’ll fix it in the mix.”

Mic preamps are another culprit. Preamps can make a warm sound brittle, and can soften or slur transients. This effect is most apparent on sizzly-clear instruments like cymbals. If you’re using a small-diaphragm condenser to record hi-hats, and you find your hats are lacking in high-end definition, question your mic preamp.

The advent of low-cost home studios has brought the issue to a head. Home and project studios do not have a plethora of great mics, and usually do not have a great-sounding large room. Often, everything is close-miked. The sound is very “direct” with no space, no room for the instruments to breathe.

In the end there are few rules, but there are some helpful principles:

1. Make use of the space you have. Use the living room, bathroom, bedroom and the kitchen. Definitely use the garage when applicable. Move the furniture out of the biggest room to make it more reflective. Get some sounds happening in that space.

2. If you’re building or customizing a space for a project studio, DON’T OVERDAMP!! Pay attention to bass frequencies and flutter echoes. For the bass, you need bass traps. Foam won’t do it. For flutter echoes, put up diffusers, not absorbers. A bookcase randomly stacked with books will do the trick nicely. A tall ceiling does wonders. In the end your room should still have a “sound”, and not be totally dead. Remember that absorbers don’t soundproof. All the Sonex foam in the world won’t stop the sound from going right through the walls.

3. If you’re doing a sparse mix (a small number of instruments and vocals) then record instruments in stereo to give each one as much size as possible. Don’t close-mic. Put a mic in the other room and leave the door open. “Open a box, make up an effect.” If you’re doing a dense mix (lots of layered guitars, keyboards, and vocals) then each instrument needs to be pared down to allow other instruments to peek through. Don’t immediately reach for EQ. If a guitar is supposed to sound thin, then get a thin-sounding mic and back it away from the source. The track might sound like a EQ was used, but the mix will benefit from NOT using the EQ knob.

4. Consider a mono drum miking technique. Drums occupy a lot of sonic space, and, depending on the type of music, a lot of that space can be wasted. And before you write in, realize that I’m a drummer. Next month I’m going to focus on drum miking – we’ll get into greater depth then.

5. Focus on mics and miking techniques to achieve the sound you need. The problem with EQs is that most of us only have, at best, only one sweepable mid and a high and low EQ. We are likely to be boosting or cutting the same frequencies in various tracks. This leaves the unmistakeable fingerprint of our EQ on the recording. How much better it would be to, instead, apply the complex filter of a wierdass mic. For that reason I like to use any strange mic if it has a unique sound (and isn’t too noisy). And, if you have to EQ, then try hard to CUT instead of BOOSTING. It’s often easy to hear what the sound is missing. Instead, try to hear what’s covering up the sound you want to hear. Need treble? Try cutting back the bass or mids.

6. Remember that SOUND quality should preempt NOISELESSNESS. If you DO have a unique mic, piece of gear or room that has some noise, don’t be afraid of recording that noise. You know, most listeners aren’t even conscious of low-level noise. But they sure get excited when – through the noise – a cool sound pops out. So if that old Vox happens to hum really loud – but it still sounds great – plug ‘er in and let ‘er rip.

7. Don’t try to fix it in the mix! You’ll spend more time turning knobs (plastic or virtual) and getting frustrated than you would spend by moving or changing the mic, setting up the sound source differently, or changing rooms. Getting it right the first time really pays off.

Don’t Forget the Omni

Close miking often results in a “direct” sound, lacking in air, warmth, and depth.

One reason that close-miking became popular was the lack (in most studios) of a truly “great” sounding room in which to record in the first place. And, even in very good sounding rooms, close miking is often used either to achieve acoustical separation and minimize bleed; to boost low frequencies via the proximity effect; or simply out of habit. The use of cardioid mics in this traditional manner can result in the “direct”, space-less sound you’ve described, even in the best-sounding of rooms!

Unfortunately, many recording engineers have a fear of omni mics borne of inexperience with them. Can you get good stereo imaging with omnis? Can you achieve separation in a multi-mic setup? How do you “aim” an omni?

The fact is, if you’re trying to grab some of the ambient sound of a room along with the true, direct sound of an instrument, omni (or multi-pattern mics used in their omni position) can be your best friend. Because of their superior off-axis response, their lack of proximity effect, and their “forgivingness” in relation to precise aiming, it can be really easy to “mix in” not only the sound of the room, but to precisely control the exact amount of room sound you want.

The Inverse Square Law states that the loudness of a source varies as the square of the distance from that source. If you double your distance from a sound source, the amplitude you hear will be decreased by a factor of four; if you triple your distance from a sound source, the amplitude will be decreased by a factor of nine, and so on. The same works for moving closer to a sound source.

Here is the practical application. Let’s say you’re miking a relatively loud guitar amp with an omni mic at 8 inches. You’re getting mostly direct sound, with only a very small hint (probably imperceptible) of room ambience. By doubling the distance to 16 inches while keeping recording levels constant, you’ve altered the “mix” between the “direct” sound and the “room” sound by 12 dB. At 24 inches, the relative levels between direct and room sound have changed by about 19 dB. If you had only wanted to alter the relative direct-room mix by only 6 dB, you would only have had to move the mic out to about 11.3 inches.

Omnis are a good choice for not only accurately capturing ambience, but for precisely controlling it. This won’t work with directional mics, because of their roll-off and/or rejection of off-axis sounds, as well as their changing on-axis response with respect to distance from the direct source. By using an omni, you’ll also avoid the well-known “pinched” quality directional mics can exhibit, especially on acoustical instruments (of course, you may *want* that pinched sound…)

I guess the bottom line is this — don’t be afraid to try the omni pattern on your mic!