Volume level at recording

ManInMotion711

ManInMotion711

New member
Ive been told to record all my tracks with the volume level at -10.0 Db. So thats what i do. Welp i forgot to do it and i was wondering does it matter if i just turn the things down to that point now or is it crucial that it be recorded that way? Dumb question but ya never really know.:guitar:
 
ManInMotion711 said:
Ive been told to record all my tracks with the volume level at -10.0 Db.
Click to expand...

Assuming its digital yes, did you record at 0db? If the tracks don't sound distorted and there is no clipping you could apply -10db gain reduction to the tracks (if you really cant, or don't want to re-record them) I think turning the faders down accomplishes the same thing but if you have more recording to do (and mixing) it may be better to have them at -10db.

Good Luck
 
Yeah, ideally it's suggested to record anywhere from -12db to -18db. But the main thing is to not clip. So, if you're not clipping, don't worry about it, just turn down the tracks according to how loud/low you think they should be in the mix.
 
PDP said:
Assuming its digital yes, did you record at 0db? If the tracks don't sound distorted and there is no clipping you could apply -10db gain reduction to the tracks (if you really cant, or don't want to re-record them) I think turning the faders down accomplishes the same thing but if you have more recording to do (and mixing) it may be better to have them at -10db.

Good Luck
Click to expand...

ManInMotion711 said:
Alright Thanks RAMI
Click to expand...
What is he? Chopped liver?????


:D
 
A good little experiment to help ease how little it matters in 24bit
Record one at -30 (or what ever..) and one at ' -2, or six'
Then match the gains for a playback comparison.
 
The whole point is to capture the cleanest signal possible -- Running the gear around where it's spec'd is usually the easiest way to get there. -10dBFS is the absolute top of what I'd be shooting for personally (only for the most transient sources - usually percussion).

Sure - you can turn the tracks down -- But if the damage is done, it's done. If the circuitry was saturating and distorting because the voltage was too high, turning it down isn't going to repair it.

Blah, blah, blah, here -- Proper Audio Recording Levels | Rants, Articles
 
Massive Master said:
The whole point is to capture the cleanest signal possible -- Running the gear around where it's spec'd is usually the easiest way to get there. -10dBFS is the absolute top of what I'd be shooting for personally (only for the most transient sources - usually percussion).

Sure - you can turn the tracks down -- But if the damage is done, it's done. If the circuitry was saturating and distorting because the voltage was too high, turning it down isn't going to repair it.

Blah, blah, blah, here -- Proper Audio Recording Levels | Rants, Articles
Click to expand...
Absolutely. I figured that since the tracks are already recorded, if they sound good and not distorted (in a abd way), probably no harm in keeping them.

But for future reference, refer to Massive's post. It has more to do with the analog side of your chain.
 
RAMI said:
if you're not clipping, don't worry about it
Click to expand...

Yes, the notion that digital audio "sounds better" when recorded at levels well below clipping is simply false. Proof is here:

The Truth About Record Levels

Now, it's possible to have distortion at levels below 0 dBFS, but that's due to improper level setup between an external preamp and sound card. Otherwise, digital recording is perfectly clean right up to the point of hard clipping. Further, the peak reading meters in DAW software are much more accurate than the old-style VU meters we used with analog tape. So to the OP, if the meters don't hit zero, you're all set.

--Ethan
 
The noise floor on tape was so high that recording hot masked the noise better, plus you got a pleasing compression effect.

But with digital you have a huge amount of dynamic range, so you can safely back off the levels. Digital clipping is very ugly, and can't be fixed in the mix.

Drums especially have such fast transients that its easy to clip and not realize it until later. Assuming you are recording at 24 bits, -10 or so for peaks should be a safe range.
 
Digital clipping is incredibly difficult, if not impossible to do in a 32-bit floating point system which is what the majority of DAWs are these days. My point is that almost all clipping you'll experience is from slammed levels being output at the D/A which then, in turn, clips the analogue section. Visa versa for the A/D end.

Try it. I dare anyone here to try and accomplish digital clipping in any 32-bit FP DAW.

And for the record, digital clipping is a lot easier to fix than analogue clipping combined with saturation distortion. If you've ever used the pencil tool in Wavelab, you can actually draw a smoother waveform to lessen its effects. I have successfully "rounded out" clipped waveforms and the result was unnoticeable.

Just saying.

Cheers :)
 
Ethan Winer said:
Yes, the notion that digital audio "sounds better" when recorded at levels well below clipping is simply false. Proof is here:

The Truth About Record Levels

Now, it's possible to have distortion at levels below 0 dBFS, but that's due to improper level setup between an external preamp and sound card. Otherwise, digital recording is perfectly clean right up to the point of hard clipping. Further, the peak reading meters in DAW software are much more accurate than the old-style VU meters we used with analog tape. So to the OP, if the meters don't hit zero, you're all set.

--Ethan
Click to expand...
Avoiding distortion in the analog path is the only reason to keep the recording level peaking 6-10db below zero. Depending on your setup and the type of signal, you could be running the analog side well above its nominal level, which some equipment moment handle very well.

It also gives you a small buffer zone, just in case a musician plays something a little louder at some point.

On the digital side, it doesn't matter at all what the level is as long as it doesn't clip.
 
Mo Facta said:
Digital clipping is incredibly difficult, if not impossible to do in a 32-bit floating point system which is what the majority of DAWs are these days. My point is that almost all clipping you'll experience is from slammed levels being output at the D/A which then, in turn, clips the analogue section. Visa versa for the A/D end.

Try it. I dare anyone here to try and accomplish digital clipping in any 32-bit FP DAW.

And for the record, digital clipping is a lot easier to fix than analogue clipping combined with saturation distortion. If you've ever used the pencil tool in Wavelab, you can actually draw a smoother waveform to lessen its effects. I have successfully "rounded out" clipped waveforms and the result was unnoticeable.

Just saying.

Cheers :)
Click to expand...
Correct, you really can't clip internally in the DAW. But you can certainly clip the converters on the way in. Since their is no interface that works on floating point math, you can't let the input or output hit zero.
 
Agreed but I still think the affects of clipping the analogue stage will be far more damaging than simply lopping off the top of the waveform, which is what digital clipping does. Analogue clipping comes with all sorts of harmonic artifacts and distortion as it's gradual instead of a "cliff effect" as you would see in digital.

Cheers :)
 
I personally don't see why there is a debate?

It has been a long, hard furrow to plough to get where we are with digital recording, that being a state where we can capture virtually any dynamic range we want without "fader riding".

A couple of weeks a go I visited a jam night at a club my son infests armed with a KA6, two SDCs set as CO-i pair and an HP i3 lappy. The band line up varied quite a bit. Drums always and bass G, some kbds, electric guitar, amped acoustic G, 1, sometimes 2 singers, loudish PA and at one point amped up cello!

I just set the level for -10dBFS in Samplitude SE8 for the loudest snare crack and left things to cook. I don't drink but I could have got totally mullered and still got a good recording!
Had I been running tape I should have been sweating over a mixer all night!

Dave.
 
Mo Facta said:
Agreed but I still think the affects of clipping the analogue stage will be far more damaging than simply lopping off the top of the waveform, which is what digital clipping does. Analogue clipping comes with all sorts of harmonic artifacts and distortion as it's gradual instead of a "cliff effect" as you would see in digital.

Cheers :)
Click to expand...

I suppose, if you want to draw all the peaks back in, it's easier to do digitally. But digital clipping has a bunch of harmonic goofyness since it essentially turns into a square wave for that moment.

the point is, there is no good reason to get that close to clipping in the first place. With peaks at -6dbfs, you are still "using all the bits".

If you are recording at 24 bit, you can record with peaks at -30 dbfs and still have more resolution than a 16 bit file normalized to full scale. So it is really very, very forgiving with regards to recording at lower levels. There just isn't any reason to get so close to zero that you run the risk of going over.
 
Farview said:
Avoiding distortion in the analog path is the only reason to keep the recording level peaking 6-10db below zero. Depending on your setup and the type of signal, you could be running the analog side well above its nominal level, which some equipment moment handle very well.

It also gives you a small buffer zone, just in case a musician plays something a little louder at some point.

On the digital side, it doesn't matter at all what the level is as long as it doesn't clip.
Click to expand...

Well, there's a couple of other reasons.

The first is convenience. Once you start mixing multiple tracks they pretty quickly add up to more than 0dB(FS) so you have to pull everything down anyway. You might as well start a bit lower.

Working in 32 bit floating point takes away a lot of this because you can push above 0dB and then just pull down levels at the end. However, even though I know this works, it still strains my OCD brain to see levels like this.

The second reason I've only heard of. I'm told that certain plug ins, particularly those emulating analogue working, also emulate analogue clipping if you get anywhere near 0dB. I have to say that I've never encountered a plug in that does this but, then, I don't use many plug ins.

Anyhow, back to the first question. As long as there's no clipping, your recording is fine. Just reduce the level while mixing.
 
It is true about the emulation plugins. If you get something like the UAD 1176 or something like that, it assumes line level -18dbfs as line level. So peaking at around zero would be like feeding a real 1176 a signal well above that. So, just like a really 1176, it will distort. That is sometimes the point of the emulation plugins, some of the equipment they are emulation will do cool things when they are pushed hard. But all of that falls under the heading of "distortion".
 
Mo Facta said:
digital clipping is a lot easier to fix than analogue clipping
Click to expand...

Not only that, the notion that "digital clipping" is terrible is a myth. I proved this with a file to accompany my Audio Expert book where I clipped a gentle acoustic guitar intentionally by running it 2 dB over digital zero. It sounds fine. Maybe a tad thicker than the unclipped version. People who claim that digital clipping is horrible should actually try it for themselves!

--Ethan
 
You must log in or register to reply here.
Back
Top