Master volume when gain staging?

[irpacynot]

I feel like I may have already asked this, so please forgive me. But I'm curious how most of you handle this when gain staging. Since Waveform defaults the master volume to -3 db, are you leaving that at its default, bringing it up to 0, and in either case, are you compensating the gain db when doing so? Or, does the master volume have no reflection of the actual gain db when viewing the track meter?

Thanks

[Watchful]

I bring mine up to 0dB (it's set via a template), as I can create my own headroom via gain staging and balancing. I don't think keeping it at -3dB provides me any benefit by the time I'm finished.

[dysjoint]

Master volume only really controls the level leaving the daw and hitting your interface or output device (which can still clip as it's converting digital back to analogue) So I just use it to make sure my Presonus isn't hitting the red. You can still mix to -X lufs regardless of the Master fader level.

[irpacynot]

I see. Thanks, guys.

So, it seems the default master setting is a kind of fail-safe of sorts.

[Watchful]

I think it's a relic to the days when you could send your Waveform (then just called Tracktion) project to a mastering service from the menu in the lower left corner. I think it should be your call what your default level is. Everyone masters differently.

[Maarid]

Due to the organized folder rendering structure, I stopped using the master fader completely a long time ago, the plug-in master bus is used exclusively for displaying the overall output signal, spectrogram, reference, and so on.
If we shift our attention to the folder that will be rendered and its fader is perceived as the master, then its position is always 0. The limiter and/or Clipper at the end of signal processing is always required, even if you manually control each track for volume, a parameter such as lufs or "felt loudness" is measured separately from peak values the signal. The recommended lufs usually ranges from -12 to -9 by standards, modern tracks try to increase the volume as much as possible, I personally prefer to focus on -6. The higher the lufs value, the louder the music sounds by ear, but .. the more it gets clamped down and loses in dynamics and volume range, the maximum lufs should be measured at the loudest place in the track in your opinion and set it almost at the beginning of work, so you can control that your quieter parts of the track can be listened to by quieter ones, adjusting the volume in different places, you will be able to adjust /restore the dynamic range of the track, taking into account the volume of the output signal you need.
Another interesting technique, which I consider positive, is to reduce the output signal on the limiter itself by -0.2 at the very end, so we eliminate the possibility of physical clipping on the equipment if someone decides to listen to our products too loudly. I'm not sure if it works, but I prefer to stick with it.
Well, either I didn't understand the question, it's possible.

[irpacynot]

...The recommended lufs usually ranges from -12 to -9 by standards, modern tracks try to increase the volume as much as possible, I personally prefer to focus on -6. The higher the lufs value, the louder the music sounds by ear, but ..

But wouldn't that drive the peaks way up? I'm bringing in acoustic guitar tracks from an old Roland VS machine, and even gain staging the tracks to -18 peaks I have to spend a good hour denoising each track using the Bertom plugin. Which is fine, but I imagine driving those gains way up there would make it insane for me to denoise.

[Maarid]

Which is fine, but I imagine driving those gains way up there would make it insane for me to denoise.

One hour? It's unbearably small - you can't make music perfectly, you just need to be able to stop yourself
The mechanics of raising the rms level are based on increasing the overall sound density, where the quiet parts become louder and the peak bursts are cut off, turning the audio recording into a kind of sausage of uniform volume, which is done by the clipper, as I said above.
The mechanics of record clearing have a wide range of different methods that are selected and used depending on the specific situation. The first and most obvious place to start is if you are satisfied with the reduced noise level using bertom, but in the end you get the noise back on the clipper, try two things that you need to use intuitively based on your hearing.
As soon as you raise the rms and get noise, use nova on a noisy track and try to find the frequency where the noise is with one bundle, just removing the frequency with noise in most cases will change the sound of the instrument itself, so use a trick, if necessary, mask the noise, try experimenting and looking for your instruments that also they are playing at this moment and recorded well, try to find the one that plays closest to the noisy frequency in the scale, add volume to it in this place. The main thing is not to forget that the noise from the equipment can be extremely insidious and can sound simultaneously in several places, working with such material is comparable in complexity to the complexity of translating technical literature using a pocket dictionary of foreign words.
If you don't have enough time, I would recommend searching for clean recordings or perhaps simulating a live instrument. If you play yourself personally, I sincerely advise you to become a guru of sculptural sound production from personal experience.

[irpacynot]

One hour? It's unbearably small - you can't make music perfectly, you just need to be able to stop yourself
..
[/quote]

Sorry, I think I worded that poorly. I didn't mean an hour for each track, I meant about 45-60 minutes to denoise about 20 tracks.

But yeah, I've also started including Nova in the denoising process for noise that I still hear after using Bertom. I try not to go below (-73.5) on that main meter on the denoiser, at least with guitars. And then I might bring Nova in to repair very specific frequencies, putting the Q all the way to 6, and not lowering the frequency past a single decibel.

[Maarid]

putting the Q all the way to 6, and not lowering the frequency past a single decibel.

Bertom, it's still a hair styling gel, and if you're not combed, you still need to use a hair comb. However, start by installing the limiter on the master bus and adjust the volume first, then start scalping the main lead track and then descending masking the frequencies with each other.
To make it more interesting for you, I suggest you use this set of plugins. - https://archive.org/details/BlueLab_audio_plugins
Specifically, I advise you to use GHOST on the master bus, it is a plug-in for real-time spectral analysis, where you can visually monitor the overall frequency pattern of the entire composition. The fact is that every time you try to get rid of noise, you involuntarily concentrate on high and medium-high frequencies, it seems to you that the track has stopped making noise, and the track is in an airless chamber and just hums, and it also does not stop making noise, because you cannot find noise in the range of 200-300 Hrz.
To let the air back in and restore the vitality of the sound, try to monitor the spectrogram so that the volume in the whole picture adheres to +\- equal values, and more specifically, there are no obvious and obvious voids.