09Apr
By: John Shelton On: April 9, 2013 In: Audio Engineering Comments: 0

You know how it goes. You’ve recorded a song digitally. After spending hours mixing, adjusting, listening, taking it out the car, re-adjusting, re-listening, your digital recording just doesn’t sound as good as a comparable professionally recorded song. Your recording sounds “more harsh” or doesn’t have the “air” or “space” or “warmth” that the pro recording does. Sure, a big part of this is that the professional recording was made in a professionally designed space by a professional engineer using expensive professional equipment, but, as my dad used to say, “A poor craftsman blames his tools.”

You may be surprised that you may actually be able to remove some of that harshness and add back some of that air, warmth, or space on your own. Is it a new plugin or equalizer setting, a complex frequency adjustment or processing technique? Nope. The answer may be that your recording levels are too high. Simply turning down the input gain on your preamp can actually make your digital recordings sound better. There are reasons for this and I will explain how it works and what to do in more detail, but first, some examples from my past.

Analog v. Digital Recording

My first experience with audio recording was at home. In high school, I worked at a kennel all summer so that I could purchase my first audio workstation; a Tascam 414mkii. This 4-track cassette recorder got me hooked on recording. Because it was a simple analog device, the recording process was fairly basic: plug in a microphone, arm the track for recording, turn up the trim on the channel until the average recording level was in the center of the vu meter (0dB), hit record and play at the same time, and the recording would start. The vu meter on the recorder went up to +6dB. Per the manual, my average recording level should stay around 0dB, but if it went up to +6dB occasionally, as long as it didn’t go over, that was “acceptable”.

Several years later, when I first started experimenting with digital recording, it was also at home. Instead of the old Tascam Portastudio and cassette tapes, I had moved on to an original Mbox and emagic Logic 5. As I had no formal training up to that point, I utilized the same recording process: plug in a microphone, arm the track for recording, turn up the gain on the Mbox until the average recording level was as close to the top of the meter as possible (0dB), hit record and then play, and the recording would start. If the level on the recording went over 0dB, it would clip, and I would stop recording, turn down the gain slightly and re-record the track.

Only one of the two methods utilized for setting recording levels above is good. Surprisingly enough, the high school audio engineer got it right. Why? To understand why, we need to talk about what a decibel is, what line level is, and what headroom is.

For what I feel is the definitive guide to this concept, I strongly encourage you to check out the following article. This article was prepared by a mastering engineer and focuses the discussion on preparing for mastering, but is very insightful and worth reviewing:

http://www.massivemastering.com/blog/index_files/Proper_Audio_Recording_Levels.php

If you want to understand how I conceptualize these topics and what I do when tracking, then please read on.

The Decibel (dB)

A decibel (dB) is a unit used to measure the intensity of a sound or the power level of an electrical signal by comparing it with a reference level. To put it another way, it’s a measurement that is a comparison between two levels: the level being measured and a reference point. This can get very confusing because there are many different reference points used, resulting in various ways to define a decibel. For example, on the vu (volume unit) meters on an analog console, 0dB, or 0dBVU, is referred to as line level. Line level for professional audio equipment is +4dBU. Line level for consumer audio equipment is -10dBV. Like 1 mile is also 5,280 feet or 1,760 yards, the distance itself is the same; the measurement is simply using a different reference point (miles, feet, or yards). The various reference points for dB work in a similar fashion.

Headroom

Analog gear allows for headroom. Headroom is distance between line level and the point in which the analog circuit begins to clip. On my Tascam 4-track, the vu meter went all of the way up to +6dBVU. After +6dBVU, the signal would start clipping. The distance between 0dBVU (line level) and +6dBVU (where the signal began to clip) is headroom. This is why the Tascam manual said it was “acceptable” if the signal went up to +6dBVU occasionally.

dBFS (Full Scale)

Digital gear uses a completely different reference point called dBFS (full scale). Unlike analog gear that allows for headroom over 0dBVU, 0dBFS is the absolute highest level that a digital signal can achieve. All measurements expressed in dBFS are negative numbers; less than 0dBFS. You’ll also hear 0dBFS referred to as “all 1’s” because all of the bits in a sample at 0dBFS will be all 1’s instead of a mix of 1’s and 0’s. 0dBFS is where the signal clips in your recording software.

So, if line level is 0dBVU, +4dBU, and -10dBV, what is it in dBFS?

Good question. The answer is a bit disappointing: it varies . . . based upon the interface/converter being utilized. Avid sets the +4dBU inputs on the HD I/O to give you 18 dBFS of headroom (-18dBFS). That means that, if you were to pass a sustained signal from your analog device at 0dBVU into the +4dBU inputs of the Avid HD I/O, it would show in Pro Tools as -18dBFS. Many manufacturers use -18dBFS as the factory standard, but once again, it varies, so check your manual.

Despite the variation, the below chart from the Gearslutz forum, which is based upon 0dBVU/+4dBU being equal to -16dBFS, is incredibly helpful in putting things into perspective:

Analogue - Digital scale

How Does This Relate to Improving Recording Quality?

Line level (0dBVU, +4dBU, -10dBV) is the level for which your analog gear was designed to function; where you will have the highest quality signal; least amount of noise and distortion. That preamp on my old Mbox was an analog device. Although attached to a digital recording interface, the signal didn’t become digital until after the preamp when the signal passed through the analog-to-digital converter. As it is an analog device, the preamp was designed to function best at line level.

When I cranked the gain knob on my preamp to right below 0dBFS, let’s say peaking at -2dBFS as an example, working backwards in the chart above, that would have been +18dBU. This is WAY over +4dBU. That means that the preamp was not operating in the way it was intended; it would introduce noise, distortion, and eliminate my headroom. It’s also important to note that, once that signal is recorded, it’s recorded. All of that noise and distortion was already introduced and captured in the recording, so turning it down after the fact just reduces the level of a noisy, distorted signal, it doesn’t actually reduce the noise or distortion in the recording. Let’s say I recorded a song with 24 tracks of audio in this same fashion. That means that I’m piling noise and distortion on top of noise and distortion. In the end, I’ll likely end up with a mix that, despite my best efforts, sounds extremely harsh.

If this sounds like something you face every time you record, the solution is simple: set the input gain or recording level lower when tracking.

How Do I Set My Recording Levels?

I treat the green part of the meter in Pro Tools like I would the 0dBVU light on my old Tascam recorder. I try to keep my input signals well into the green around -18dBFS, but if it occasionally peaks up and hits the yellow part of the meter (which is -12dBFS), that’s okay, but I don’t want it to go much over that, nor do I want signals to be in the yellow for extended periods of time.

Sonic Benefits

For me, the benefits of this approach are far reaching. I find that each track, even after equalization and dynamics processing, doesn’t clip. Rarely am I doing extreme things with the Trim plug-in in Pro Tools to turn stuff down. Additionally, as I use a lot of outboard analog gear during the mix process, that analog gear, which is designed to receive +4dBU source signal, actually receives that level from Pro Tools as intended. Ultimately, I’m much more satisfied with my recordings because they sound less harsh and more open.

Turning down the recording levels works very well for me. The only way to see if it works for you is to try it.

I’ll be back again soon with mixing tips that also involves turning it down.