Gain staging is only important in the analog domain and for A/D and D/A converters to preserve good audio specifications for the signal chain. Because the modern digital audio engines use a floating point arithmetic, no degradation can occur except if you shift the correct gain by plus or minus 96 dB or more. In a normal mix situation, keeping the peak meters active in the green region, few dBs away from the red zone is enough to get the best possible results.
For the A/D and D/A converters, just keep the modulation away from the last dBs on the meters.
For the analog part of the signal path, just refer to the equipments specifications. This link can help you to understand how to manage the different dB values found in the manufacturers documentations:
404 Not Found
Keep the technical stuff simple and focus on your music
Please note that gain staging during mixing and mastering, whether done with software or with hardware, is to a great degree about "reading" the information density of the content and with gain tuning the mix towards where its information density peaks according to how the brain perceives the signal at various voltage levels. Fundamentally this process is intended to maximize the amount of resonance potential present in the mix. The better the signal capacity of the D/A, the more resonance potential you are able to output. Gain staging on setup A and setup B varies depending on the setup when their voltage capacity varies. The engineer with the more available headroom in his audio interface has better visibility into the information density and can drive the hardware towards much higher resonance potentials than what the software counterparts are able to mimic, hence also a better sound quality. So essentially gain staging is critical both when using software and hardware.
A major reason for the great sounding masters out there has to do with the quality of the gain staging produced from great monitoring (in terms of flatness) and within a great voltage window. This is a major reason for the "keep it simple" idea, that yes in deed you get far simply by having lots of experience of moving the faders in a high quality context. But there is more to gain staging besides that also in these contexts. Essentially it is about having great visibility into and understanding about the level of resonance potential and information density present in the audio, since that will determine what type of emotional experiences the music can produce, in a way how emotionally "alive"/"vital" it can be perceived.
In a low voltage monitoring context, you cannot as easily notice that you have too much low end and too little high end, you cannot as easily and with as high quality correct it either, well it is possible but in practice it is much much harder. And if the monitoring is not flat in the high end these frequencies are going to be unbalanced too, this might in combination with high frequency overlaps you might not hear as easily be causing harsh upper mids and high frequencies when brickwall peak limited, so that when you push the mix to a commercial level it sounds harsh in comparison.
What you can do when you gain stage, no matter the quality of your monitoring context, is to apply temporary EQ filters on the MID and SIDE components so that when you perceive the mix to be at full balance and at the ideal integrated LUFS as a result of the gain staging, then when you undo these EQ filters, the mix jumps very close to your optimum fundamental frequency and integrated LUFS targets, both in stereo and on MID and SIDE respectively, so not only do you get a more optimal information density this way, but also a better stereo image. When you do this, you can monitor the integrated LUFS of the MID and SIDE components by excluding/bypassing them from the temporary EQ filters by having them placed before the EQ filters. This means that when you undo the EQ filters, the fundamental frequency shifts upwards to the near optimal places on the MID and SIDE components, but the LUFS figures stay at the target levels on the MID and SIDE components. Also ensure the RMS levels are near their targets. This combination instantly places your master to the pro level when the recording has enough information density and when you have not done something very wrong. Having the right "loudness" characteristics in the center and on the side contributes in a major way to the perception of a pro master. Similarly when you have this balance totally off, it quickly starts sounding pretty bad. This technique you can optimize further by instead measuring the same values for low, mid, high respectively, because when those are at the targets, the sum will be too. And you can take it even further. Find out what works the best for you. What one can say though is that two similar sounding mixes in the same genre with the same loudness characteristics will sound roughly equally well balanced, but not necessarily equally good, because that depends on other things such as the respective information density at these loudness characteristics, the song, the amount of resonance potential, the amount of frequency overlaps and so on. A few tracks out of tune in one of the mixes can be enough to create a big difference in the perceived quality. In terms of the production, having vocals in one of the versions can be enough for a major difference in perceived quality between the two. And so on. This will make you quickly realize that balance and dynamics those are only contributing to a certain perception of a mix, more than anything it is about recording a great song/production with high information density. And at this point it becomes more about working towards establishing great tones on the sound sources in their arrangement/production context and having them play the right things at the right time. A good room for instance will balance the captured frequency response of every single sound source in the mix with a much higher information density because more of the sound source "gets through the room". A room with bad acoustics will create tons of resonances at specific frequencies into each sound source, that makes a huge difference.
By doing it this way you are also able to stay a bit more gentle with your EQing. Just don't A/B while you have those temporary EQ filters enabled. Establish the correct gain structure as early as possible, then maintain it until the final master is ready.
The good thing about creating the optimum gain structure very early, is that you will very early be aware of the information density present in your mix and can figure out whether you should go back to production and recording or allow the mixing to go past the point of the rough mix. Because sometimes that might be tricky to tell, but at the optimum gain structure it is obvious.
The tricky part about all of this, is that you need to both have the right gain structure and at the same time have the frequencies work in their context, meaning having the frequencies pulled together to create the desired perceived sound of each sound source in the mix. This means a very important thing in gain staging - when you pull together the right perception of say two sound sources let's say in the mid component, let's say bass guitar and kick, you must lock/group the faders before you gain stage, meaning you must apply the same amount of gain increase or decrease on the two tracks when you gain stage towards the target, else you will create additional frequency overlaps during the gain staging. So the art kind of becomes to have a high information density, high resonance potential, low amount of frequency overlaps in the context, the right dynamics and the right loudness characteristics. Those are 5 highly critical dimensions where we need high quality.
