Tal’s Top 5
All types of audio productions end with a summing procedure, a mixing stage, an audio signals bottleneck, where all streams of audible events, spaced and stacked on a timeline, are being conditioned with final relative presence (volume), spectral location (pan), dynamic behavior (compression, automation), overall sonic character (EQ), and final “coloring” (effects). The mixing stage is THE moment of truth, where all creative and technical production elements are finally glued together, turning our work into a final, standalone, reproducible asset, a MASTER.
But in our never ending quest for the perfect master there are many creative and technical traps, which sometimes almost anyone, even the best, can fall into. Although modern day DAW’s and additional hardware and software tools can help us salvage almost any sonic disaster, there are still some elements that are extremely hard to fix-in-the-mix, if at all. Like….
Both mic and DI signals have to be electronically amplified prior to being sent to a recorder. Overloading an amplifier’s electronic circuit will cause it to generate squared envelope signals, which usually sound unpleasantly “fuzzy” or “fried”. Although certain styles of music and types of sound effects can take advantage of this phenomenon, distortion is for the most part a non-desirable, often non cure-able problem, one that’s hard to ignore and hard to hide. More importantly, it can always be added later on, so while recording, try being as observant as possible to this issue, and adjust the pre amp’s gain controls accordingly. An occasional over-saturated passage might not be bad, especially if it helps to emphasize a strong dynamic moment, but always reserve the option of having both clean and dirty signals. And remember that long, distorted passages eventually end up causing ear fatigue that turns any listening experience to an unpleasant one.
Every single electronic component that’s involved in sound generating, capturing, and processing, introduces a certain amount of noise to the final signal (hiss, hum). Dozens of such components can be found inside any recording chain, and the overall effect is cumulative. The audible relationship between the amount of signal relative to the amount of noise that’s being captured is called signal-to-noise Ratio. The rule of thumb is to always have much more signal captured than noise. Having too little signal in this cocktail will result in significantly louder noise level when turning the signal up in the final mix (adding gain). If there is close to the same amount of signal and noise in the source, turning the source twice as loud will also turn the noise twice as loud, and so forth.
Analog tape produces a lot of noise on top of the noise already introduced by the recording chain, therefore one has to be constantly aware of recording levels to overcome this issue. Digital audio does not suffer from this issue as much, but involves another factor – the sampling level. The “hotter” a signal is being sampled, the greater the binary resolution it generates, and ends up sounding better as a digital source. Always stop short of clipping the Analog-to-Digital converters, due to the nasty nature of digital distortion (unless desired as an effect).
Regardless of analog or digital capturing, try maintaining the lowest possible amount of noise in your recordings, mainly because the process that’s involved with trying to get rid of noise in the mix ends up hurting the recording’s quality quite a bit, especially when dealing with vocal tracks, quiet acoustic instruments, and other sources. There are several software and hardware tools available to deal with noisy sources (Digidesign DINR, Waves X-Noise, Cedar products), but the results can vary dramatically, based on the severity of the initial problem. The same conditions apply when having to deal with sources recorded with loud pops and clicks, created by bad drop-ins (punches), digital clocking problems, or unclean edits (no cross fades).
Mixing is an act of exposure, a sort of sonic and musical voyeurism. It shouldn’t be an act of hiding stuff because it sucks. Because if it sucks, it shouldn’t have been recorded to begin with.
Because of the track count limitations, pre-blending certain elements prior to the mixing stage is a normal thing to do. If 20 vocal tracks create a background stack, or three pairs of drum room microphones create the ambience sound, the session may become too heavy to play and monitor. It is then possible to take a pause in the production process, create premixes of these elements, and sum them down to a stereo mix, committing to a specific printed blend. While doing this is fine, losing the original individual components, or not including those components in further editing that takes place, will result in an inability to re-blend those elements during the mixing stage. Adding EQ and compression during the mix always changes the inner balances of any committed pre blend, and not having the source tracks for re-blending (now taking into account the current stage the production and mix are at) can cause a major problem. Try to always keep and include pre blended source tracks as an integral part of the session, and make them available for the mixer to re-blend upon need.
While out-of-tune monophonic signals like vocals, trumpet, and bass can be successfully re-pitched (using software tools like Antares Auto Tune), out-of-tune polyphonic signals like guitar, piano, and orchestra, cannot be re-pitched. There is no known cure for this illness, and it will irritate almost any listener. Keep a tuner handy all the time when recording guitars, make sure the piano has been tuned prior to the session, and make sure the Concert Meister visits the control room a few times during the session to recheck the tonal center of the recording. I can take a donkey and make it sound like Pavarotti, but when the guitar is out of tune or the piano is out of tune there is no cure for it.
Arrangement & Performance
A good musical arrangement usually mixes itself, while a bad one can result in days of non-successful fader moves and knob tweaking. Also, while it’s possible nowadays to fix almost any pitch or timing related musical performance problem, science has yet to develop a tool that can deal with lack of vibe, groove, inspiration, and soul, manifested as artistic and production values. Of all the five points this is my most important one. This is my message to producers out there: Don’t expect your productions to sound good if they’re not arranged and performed with passion.