X32 : optimal gain for studio recording ?

User avatar
DaveH
Posts: 44
Joined: Sun Oct 10, 2021 11:01 pm
Location: France

X32 : optimal gain for studio recording ?

Post by DaveH »

Hi there,

reading here and there, I often notice that it is recommanded to adjust the gain to hit around -18dB to -12dB, and lighting only the first orange light or the gain meter (that offers

When I do this, I find it strange not to "profit" of the fullscale, and the maximum number of bits available.
And when I record in my DAW, the waveforms are ridiculously small, and hard to read.
More than that, using compressors on such signal need to lower the threshold at very small values.

I really wonder why we should do this in studio.
(in live situation, I can understand that it's to keep some confortable headroom in case a musician changes his volume, or did not play at "full expression" at soundcheck)
Newbie in this vast X32 world. :geek:
User avatar
pvannatto
Posts: 1330
Joined: Wed Apr 14, 2021 3:48 pm
Location: Ontario, Canada

Re: X32 : optimal gain for studio recording ?

Post by pvannatto »

Hey @DaveH , I know others may disagree with me. But I tend to run my gains hotter than the "theoretical" optimum level of -18 db for live events (as well as the occasional recording setup). I prefer to have the signals bouncing around in the yellow range. Occasionally vocals do go hot enough to light up the clip (red) LED. But I've never noticed any distortion during a live event. My advice is do to what you have to do to get a good waveform in your recording (without clipping).
Paul Vannatto
Global Moderator
User avatar
MJKlein
Site Admin
Posts: 1020
Joined: Wed Apr 14, 2021 12:08 pm
Location: Hukou, Taiwan
Contact:

Re: X32 : optimal gain for studio recording ?

Post by MJKlein »

Each DAW has an optimum recording level. You should use that as a reference. That being said, I normally record at -12. I also mix at around -12 so I have some headroom to work with for Mastering.
湖口鄉新竹縣台灣 Hukou Township, Hsinchu County, Taiwan
Behringer X32/Dante + Soundcraft Ghost + X32ReaperAutoMate
English/中文
This site is hosted at Dathorn: https://dathorn.com/index.php?r=489
https://www.youtube.com/@phoenixmediaforge
User avatar
DaveH
Posts: 44
Joined: Sun Oct 10, 2021 11:01 pm
Location: France

Re: X32 : optimal gain for studio recording ?

Post by DaveH »

Ok @pvannatto , for the moment I work at -12, with exceptionnal peaks at -6 and I do not hear any problem, so I probably go on with this.

@MJKlein I wonder why DAW would have an "optimal" reference (I already read this too, so it's not for polemic, but by curiosity and need to understanding). Once the analog signal is digitalized, all the rest is numerical. Sums and multiplications. Why would these calculations run "better" at -18 or -6 ? For me, logically, in digital world, using the wider number of bits (out of 16 or 24) is the best, to increase the resolution of the sound. I must miss something...

And what about plugins ? Those who simulate "saturation" should work on the signal level. They should be "hotter" when you get close to max level, no ? So if you stay at low level, they won't get any warmth ?
Newbie in this vast X32 world. :geek:
User avatar
MJKlein
Site Admin
Posts: 1020
Joined: Wed Apr 14, 2021 12:08 pm
Location: Hukou, Taiwan
Contact:

Re: X32 : optimal gain for studio recording ?

Post by MJKlein »

DaveH wrote: Mon Oct 18, 2021 6:15 pm
@MJKlein I wonder why DAW would have an "optimal" reference (I already read this too, so it's not for polemic, but by curiosity and need to understanding). Once the analog signal is digitalized, all the rest is numerical. Sums and multiplications. Why would these calculations run "better" at -18 or -6 ? For me, logically, in digital world, using the wider number of bits (out of 16 or 24) is the best, to increase the resolution of the sound. I must miss something...
It's all about gear designs:


You've spoken of "the wider number of bits" as if the bit depth is affected by the recording level. Resolution is an entirely separate issue with regard to recording level.

The concept of louder signals being better is an analog mindset, where the S/N ratio is much different. Earlier you mentioned waveform size. Unless you're editing, careful visual monitoring of the waveform isn't necessary (or even possible?). When editing, zoom in on the section you're working with. As long as you have good metering when recording, looking at the waveform is only good for a quick check to make sure that audio is being recorded. Waveform size alone is not a good reason to increase the recording level. Every DAW has some way of scaling the zoom for viewing tracks.
DaveH wrote: Mon Oct 18, 2021 6:15 pm And what about plugins ? Those who simulate "saturation" should work on the signal level. They should be "hotter" when you get close to max level, no ? So if you stay at low level, they won't get any warmth ?
Not exactly. Clipping can be accomplished in many ways, with the same results. You can overdrive a guitar amp by actually running the waveform into clipping, or you can use a clipper (a circuit designed to do the same damage to the waveform as actual clipping) like in modern guitar pedals. We've been using hardware clippers in the broadcast field for decades.

Saturation plugins are software clippers and they perform their function at any level as long as the settings such as drive and threshold, are considered. That's the only situation where the level matters. If you set up your clipping with a certain amount of drive and threshold at a respective input level, but you drop the input level to the plugin, it will react differently. You can make additional adjustments to the drive and threshold to compensate for the level drop and then it will function as before. But the sound of the plugin will be the same at either input levels (within reason of course) as long as you adjust the controls to the same degree of effect.

If you aren't getting any warmth from a saturation plugin it's because it's not being driven hard enough. Of course, there is a limit. If (for some strange reason) the guitar was recorded at -50 LUFS there might not be enough signal to drive the plugin. As long as the input level falls withing the operating range of the plugin, you can achieve whatever saturation the plugin is capable of generating.
湖口鄉新竹縣台灣 Hukou Township, Hsinchu County, Taiwan
Behringer X32/Dante + Soundcraft Ghost + X32ReaperAutoMate
English/中文
This site is hosted at Dathorn: https://dathorn.com/index.php?r=489
https://www.youtube.com/@phoenixmediaforge
User avatar
DaveH
Posts: 44
Joined: Sun Oct 10, 2021 11:01 pm
Location: France

Re: X32 : optimal gain for studio recording ?

Post by DaveH »

@MJKlein

> The concept of louder signals being better is an analog mindset, where the S/N ratio is much different.

Really (at first sight) I do not understand that.
Let say you record with a ridiculous low level : your waveforms will be very short. It means that you are only using the very last first bits of the 24 bits integers to digitalize and store your data : so your digital dynamic range will be very short. You will not have 2^24 values to discretize your sound, but maybe 2^6. And then, to hear the sound, you will have to raise the fader, ie. to multiply all these values by 2^18 if you want to normalize them. Then you imagine that there will be a lot of artifacts, due to aliasing (like in small pictures you zoom in and pixels become squares).

> Earlier you mentioned waveform size. Unless you're editing, careful visual monitoring of the waveform isn't necessary (or even possible?).

That's precisely what I do : as a musician, in my DAW, when I make arrangements, I often edit visually, to find the note or the drum hit where I want to cut, etc... Big waveforms are important for me. And BTW in the video you sent me, Kenny Gioia says that he normalized all his waveforms after recording to make them bigger (= he extends the "useful" bits to their maximum, but a bit too late in my mind).

BUT he makes a statement : he says that digital gear (and converter) are designed to work best around -18dB (as were analog gear in the past). So he makes a difference between recording levels that should be around -18dB for a best A/N conversion, and playback levels where, once recorded, he normalizes his waveform at 0dB.

That's interesting. And I understand better now. Because I know that A/N converters are a lot of complex maths and electronics. :geek:

Anyway thanks for your explanations. As I'm not pro, and I don't want to waste my time normalizing my takes, I will lower a little bit (from my -6dB peaks), like @pvannatto and aim -12dB peaks (like Kenny says, because infact he aims an average -18, but allows peaks at -12).

For plugins that works with level (saturation, limiter & compression), then I suppose that we have to tweak up the input level as you say. But I always find it frustating on the compression cuves of the X32 to see the audio spot moving only in the bottom part... :D
Newbie in this vast X32 world. :geek:
User avatar
MJKlein
Site Admin
Posts: 1020
Joined: Wed Apr 14, 2021 12:08 pm
Location: Hukou, Taiwan
Contact:

Re: X32 : optimal gain for studio recording ?

Post by MJKlein »

I say, do what you want, however you want to do it.
湖口鄉新竹縣台灣 Hukou Township, Hsinchu County, Taiwan
Behringer X32/Dante + Soundcraft Ghost + X32ReaperAutoMate
English/中文
This site is hosted at Dathorn: https://dathorn.com/index.php?r=489
https://www.youtube.com/@phoenixmediaforge
User avatar
DaveH
Posts: 44
Joined: Sun Oct 10, 2021 11:01 pm
Location: France

Re: X32 : optimal gain for studio recording ?

Post by DaveH »

MJKlein wrote: Tue Oct 19, 2021 10:18 am I say, do what you want, however you want to do it.
Oh, do not misunderstand : I take all your advices into account. They are very valuable to me. :)
Newbie in this vast X32 world. :geek:
User avatar
pvannatto
Posts: 1330
Joined: Wed Apr 14, 2021 3:48 pm
Location: Ontario, Canada

Re: X32 : optimal gain for studio recording ?

Post by pvannatto »

Hey @DaveH , I would suggest you follow @MJKlein suggestions, since he is far more experienced with a studio setup than I am. Most of my experience is with live sound, with a smattering of recording. Btw, it was MJ that suggested aiming for -12 db, not me.

For me, I mix primarily with my ears, as well as watching the meters during any live performance.
Paul Vannatto
Global Moderator
User avatar
DaveH
Posts: 44
Joined: Sun Oct 10, 2021 11:01 pm
Location: France

Re: X32 : optimal gain for studio recording ?

Post by DaveH »

Yes, yes, don't worry, I already took that into account in my recordings of today !
Average at -18 to optimize the digital chain, and very rare transients at -12, no more -6 hitting ! Now that I understand the underlying reason, it's more easy to accept.
And by the way, these vumeters deliberatly show a limit green-orange at -18dB, so there must be a good reason afterall :D
Newbie in this vast X32 world. :geek:
Post Reply

Return to “Recording Engineering”