Input fader v track fader

When tracking what fader should I be looking at for the recommended -18dbfs.

I’m assuming the track the audio is getting wrotten to right?

Do both input and track faders give different readings altogether?

Tah

With recording, I usually adjust the input levels on my sound card to get a distortion free recording. When mixing, I use the output fader of the channel to get the desired levels.

This being said, maybe a more experienced tracker can chime in?

I agree leave the input fader at 0 or don’t touch it. Only adjust your interface / sound card input to get a smooth signal. Always below 0… Watch your track signal level.

I add compression and gate to the input to get the best recording :slight_smile:

The input fader (and its pre-gain if used) controls the level of the signal being recorded.

The track fader controls the level of the playback or monitor signal coming from the track.

Say you have both faders set to 0dB, and a constant sound source (like a test tone) connected to the input and it is putting out a signal at -10dB. Both faders should show -10dB. If you record this the recording will be at -10dB independent of the track fader’s setting while recording. You can set the track fader to +5dB or -40dB and the only thing that changes is the monitor level you hear - the recorded signal will still be -10dB.

interesting.

So I’m safe if I leave both the input and track faders at 0 and set my interface gain so I’m getting readings under -0dbfs? i know a lot of folks shoot for -18dbfs to give plenty of headroom in a dense mix. My style of music is alt/folk and is fairly stripped back (no drums) can I get away with tracking slightly hotter? I usually end up with 10 tracks tops, guitar, vocals, bass, french horn, violin, keys etc… the videos I’ve watched on gain staging/leaving headroom in mix have all been mixes that are fairly busy.

tah

As long as you don’t clip the signal you are fine. The headroom is important if you use plugins that emulate hardware including the gain levels.
Personally I aim for a signal around -12db, and have set the meters to green up to -12db yellow to -6 then it goes to red anything above -6 db. That makes it easy to spot anything approaching a clipped signal.

The busyness of the mix only matters in the context of having more sound sources contributing to the final output. It is really irrelevant when deciding what level to record any individual track at. If you just consider the actual recording (ignoring for the moment peakae’s accurate comments about plug-ins) there are 2 main considerations in determining a recording level - minimizing noise & avoiding distortion.

In the analog tape days folks tended to record pretty hot. First the noise floor on tape is fairly high so you wanted to have your recorded signal as loud as possible to reduce the noise in the final mix. Second when tape is slightly overloaded the distortion can sound like it is adding warmth. So it wasn’t necessarily seen as a bad thing.

With modern digital recording all that changes. The noise floor is very low so having the recording level substantially below 0dB doesn’t really contribute noise in the mix. Second digital clipping at 0dB sounds really nasty so you really want to avoid it. So the modern practice is to record at a lower level. But there is nothing magical about the actual level as long as it stays below 0dB. Folks tend to record double-digits down as that keeps them a long way from the risk of clipping and even at those lower levels there is no need to worry about noise.

That’s the long way to say that neither the style of the music nor the number of tracks and busyness of the mix matter much when setting the level.

If I record with peaks at say -12dbfs most of the average recorded audio ends up being well below that, roughly -20 dbfs (especially my vocals) is this not perhaps too low!? The peaks are usually due to poor mic technique (nothing automation and compression can’t solve)

Probably not, but you can determine that for yourself by listening. Put a compressor (something modern and clean that doesn’t add noise as part of its emulation) on the recorded track to even out the vocal performance. Solo the vocal and adjust the fader (and pre-gain if needed) on the track so you are peaking just below 0dB. Now listening at normal levels do you hear any additional noise in the vocal signal (i.e. noise added from the track not just noise in your setup like from a power amp, etc.)? If so then you could boost the recorded level, if not you are fine but could still record hotter if you want. Keep in mind that a small amount of noise will get masked by the mix.

I fall into the camp that almost never put effects in the record chain. But if you have a singer you know is going to produce wide volume swings by moving around or just the way they sing, that’s a good reason to add some light compression before recording.

Is it common practice to boost the fader/pregain/compression after recording to get nearer to 0dbfs. …?

On individual tracks it isn’t. Just set them relative to each other by how prominent you want them to be in the mix. But you do in general want the output mix to peak a bit below 0dB (this is an oversimplification because you might want one song to sound soft compared to another song etc.). Use the meters to tell you when you are close to clipping & your ears for everything else.