Sound Levels and Recording (Part.2)

Forewords : to fully understand this part, the reader would benefit of reading the Part 1 in which some important informations are detailed.

OPTIMAL LEVELS IN DIGITAL RECORDING

As already mentioned in the first part, the ideal level in analog recording is quite simple as the VU Meters indicate it.

During the tracking process the Sound Engineer needs to adjust the gain to reach the 0db VU knowing that there is some space for analog distortion.

Ok but…what is the equivalent level Digital Audio ? When carefully looking the Logic Pro Peak Meter i.e. it is obvious that there is NO recommended level !

However, if we consider digital equipment as a sound card, for example (or rather A/D converters that transform the analog signal into digital one), this digital device should also have an ideal operating level. Unfortunately, this level is not officially defined by any standard.

Therefore, there are slight variations between brands or even countries with a margin being between – 20 dB FS and - 16 dB FS. However, in most of the case the levels is set at – 18 dB FS.

Accordingly, when recording you should pay attention to be as close as possible knowing that a little variation of 2-3 dB in both sides won’t be an issue.

When proceeding this way, you can ensure a comfortable margin of safety of 18 dB FS for the peaks since the maximal value is set at 0 dB FS.

NB :
This margin of safety is also called Headroom. Such Headroom is very important in the context of “ITB” or “In the Box” recording when everything is made with a computer and a DAW, which represents the majority of users.

Regarding the sound itself, more Headroom implies more space where your sound can breath. At the opposite, closer you are from 0 dB FS less dynamic occurs because of the small difference between the RMS and Peak Levels.

Ok but… This implies a new question regarding how to measure this, because the terminology RMS is mentioned as it is for analog recording, while the DAW don’t usually include VU Meters. Consequently, how would the Sound Engineer know the level of – 18 dB FS is reached ?

The solution is to download an external plugin that may be free of charge as the FreeG made by Sonalksis

NB :
In the FreeG, the red lines indicate the RMS level while the green ones indicate the Peak level. A quick look at this example show that there is about 20 dB difference between the 2 meters in the left channel and a little bit less in the right one…which is a lot !

Now, you would be able to make your recordings with the indications about RMS levels, great tool isn’t it ?

WHAT ARE THE ADVANTAGES TO WORK WITH RMS ?

• You can be reassured to make distortion-free recordings because it is rare to record an instrument that has more than 18 dB of dynamic.

• You minimize or avoid master bus saturation issues. When everything is recorded too loudly in a session, the master bus will be saturated before you start to Mix and you’re going to spend a lot of time lowering every track’s levels.

• If some Hardware is on the recording chain, tracks will already be at the right level and will not produce distortion.

• The same applies for Plugins that emulate analog hardware such as compressors, EQ and others that can be subject to distortion too.

• Your sessions will benefit in homogeneity, tracks recorded with RMS levels of similar range will be tracks be of a similar levels without major deviations. The fact to have all faders at level 0 will immediately result by a mix generally almost balanced. (This point will be treated in a future post).

• To conclude the part devoted to recording, a last general comment : trying to record close to the 0 dB FS is methodological mistake. Closer you will be to 0 dB FS results in less space for the “Peaks”. As a consequence, they will have more and more distortion. Saturated peaks totally distort a recording that will seem uninterested and lifeless.

This error is partly due to the confusion between the 0 VU of the analog and the 0 dBFS which, as you now know, have nothing to relate. Recording as loud as possible in analog was necessary because it was the only way to “cover” the background noise generated by the recorders and the different devices used.

In digital, it’s absolutely useless since the 24-bit recording allows having 144 dB of dynamic.

This gain in dynamis does not occur upwards, you can’t record louder and louder. The gain has to be done downwards, which pushes the background noise generated by any audio interface (background noise that is already far lower than in analog and the preconceived idea that ones has to record as loud as possible to mask this noise is totally obsolete even with interfaces at very modest prices because the background noise is nowadays derisory.

IN SUMMARY, TO PERFORM A QUALITY RECORDING REQUIRES

• The use of a plugin allowing you to read RMS levels
• Record to -18 dB FS RMS or as close as possible to such value
• Always take care not to exceed 0 dB FS

NB :
Would you exceed 0 dB FS while at -18 dB FS RMS, then the solution would be to lower the recording level accordingly or, even better, process to a reduction of the dynamic range by using a compressor when possible.

That’s all for today guys, I really hope having been clear enough in my explanations, and that all these information will be contributing you to progress in processing your mixes. Thank a lot for being readers of my posts.

The 3rd part in which I’m going to provide additional detailed information (with examples) regarding why I don’t recommend recording with the use of peak levels. Some information about levels in mastering will also be covered.