Channels level question

Hey people,
I Beginner and there are some things I don’t understand about gain staging and mix.

I create beat of Kick and Snare.
I noticed that in my case when the kick play, I go into the clipping in the stereo out.
So I lower the kick level and the problem solved, but now I hear the kick very low.
So I increase the headset sound to hear stronger, but this affect on all the project and all the mix.

What I’m trying to ask is when i start project, where do you start ?
should I first direct the headphones level for a specific volume, or first start from the Kick level and then direct all channels according to the kick ?

Thanks!

I don’t know if this can answer your question, but here’s what I think.

When you’re in a creative mode, if I can put it that way, the sound level doesn’t matter in a DAW as long as the output of the individual tracks and the main output don’t clip or exceed 0 dBFS and the ears can handle this.

The same thing could be said when it comes to mixing, however, the balance of the various instrument levels is essential before even adding EQs, Compressors, etc. It is also very important after adding all that.

Ultimately, the sum of the tracks on the master track when mixing should be between -23LUFS and -18LUFS (Integrated) in order to have enough space for the Masteirng. If this requires lowering the faders on the tracks to reach this level, this must be done while maintaining the instrumental balance. If you find that the mix volume is too low, turn the volume knob on the headphones or sound box amplifier. In no way will this affect the quality of the mix and the instrumental balance.

Oh boy, to a certain extend your question can be easily answered. To another extend we could get into the controversial area of gain staging.

Your first task is always to mix the levels of all channels relative to each other in a way that it sounds good. (Sounds good to you, to your friends, to your fans, to your client.) These channels can be thought of as the “sound sources” in a mix.
It doesn’t matter too much, which absolute level you chose on each channel. It matters that the levels RELATIVE to each other are nice. Only your ears and your experience can help you there.
Please note: It is practically impossible to clip channels at this stage. In theory you could mix everything to -20dBFS or to +20dBFS, there is no sonic difference.

The next area is the output of the mixed signal to your speakers/headphones. You need to make sure that the last point, where you can adjust volume for your overall mix (= the final gain stage), does not hit 0.0dBFS. It must stay below 0.0dBFS at all times. That point usually is the Stereo Out channel fader.
In order to be able to use a volume fader for mixing purposes and not only for gain staging I recommend to use such a construct:
grafik

The Stereo Out is usually created by Cubase by default. The PreMaster channel was created by me and is just a Group Track. All individual channels of the project (the sound sources) will be routed to the PreMaster. Only the PreMaster will then be routed to the Stereo Out. The PreMaster is used for artistic changes or mastering, the Stere Out is used for final gain staging only in order to prevent clipping (or to raise the level if the mix turns out a bit too quiet).

The final level for listening to is either adjusted at your amplifier/speakers or, in case you have Cubase Pro or Nuendo, in the Control Room.
grafik
Control Room’s listening volume knob. It does not affect the actual mix.

That was the nice part most people can agree on.
The controversial part is: Do you need to do gain staging at all other than adjusting all sound sources relative to each other and on Stereo Out? And if so, to which level? And just on some channels or on every channel?
The real question is: What exactly do you need gain staging for?
I have seen many videos on gain staging that address a specific situation only, even though it is claimed that the provided answer is universally true. I am at a point where I doubt that even all professionals understand this topic fully.

1 Like

Thank you very much for your answer.

Yes, the remaining questions is Lets say i arranged all the channels good together, and they are all bellow 0dbs, how much under 0dbs they need to be ?

I once read that -10dbs should be left at the end the mix, for mastering.

I can turn all the knobs down together and then go to -10dbs, but then I’ll hear everything low.

So at this part should the headphones be turned up more? or -10dbs is to much ?

Make it a habit right away to use the correct naming of things. Above I think you are talking about 0dBFS. Sometimes people get confused because they think one thing and read or write another.

To answer your question: If you are exporting a file that is fixed point then in order not to clip the file itself you just have to be below zero. So if you’re asking about sending to mastering there won’t be anything wrong with your file if it is -0.5dBFS. Now, a mastering engineer might prefer it to have peaks below a certain value, let’s say -5dBFS just as an example; well they can turn that down themselves.

The number “-10” (dBFS) is to some degree arbitrary as far as I know.

Part of setting up a workspace is figuring out how loud your monitoring needs to be. One way you can do this is to simply play back something that you know is “correct” and then set the monitor playback level to where you want that to be.

If you are mastering for example then you can simply take commercially mastered songs and import them into Cubase and play them back (with zero adjustment to the level of that track in the mixer). Adjust your speakers / monitoring so that the loud songs sound loud and the soft songs sound soft. Once you have done that you can take a mixed song and master it, and once it sounds loud it basically should be more or less loud, because you set your monitors up for that.

The difference for you if you are mixing and not mastering is that you can’t use a commercially mastered track to use as a reference because mastered tracks are usually louder than mixes. So you would need a good mix of a song and then you can do the same thing. If you do that then you will know that when your mix sounds loud it is loud.

You can adjust your monitoring levels as you go of course but I would recommend using Cubase’s “Control Room” and set the reference level the way I implied above. That way you can always get back to a good frame of reference whenever you want to check how loud your mix feels.

Should your headphones be turned up to compensate? Short answer is “yes”.

1 Like

When working inside a DAW there are basically 3 things that determin on which level your channels should be:

  1. Peak Meter
    Put the levels so that the peak meter can show you something about what is going on on that channel. That is if you put the level to +50dBFS the peak meter will be always fully lit, which is not helpful. If you put the levels to -50dBFS the peak meter will also not show much.
  2. Plugins that use an absolute volume level
    Examples for this are compressors with a threshold value, or limiters. These parameters usually don’t go beyond 0.0dBFS and if you mix at +50dBFS you will not get anything meanigful out of them.
  3. Plugins that emulate analog circuitry
    Some of these plugins emulate even the so called “sweetspot” of the analog models, ie. the plugin sounds best when the signal enters at a defined volume level. Personally I think this is one of the stupidiest ideas the plugin developers ever had as it imposes an artificial restriction on the user for no real sonic advantage. (The devs could just allow the good sound on all dynamic ranges.)

Those above limits apply to purely working in the box. If you implement hardware gear in your signal flow you must additionally adhere to that gear’s limitation.

There is no real “correct” level at which to mix. Many people like to go for -18dBFS as that corresponds to 0dB VU, which was widely used in the previous century and is still fairly well used nowadays.

That is old knowledge from the previous century. But we are well into the 21st century by now. As long as you create files, whose bit resolution is set to floating point (usually 32bit float) you don’t need to keep any headroom whatsoever on your files. Floating point formats ensure that the mastering engineer can turn down the volume by themselves to the level they like without any loss of quality whatsoever.

@Nivs719,

Sometimes it’s worth investing a little money in training. There are many courses offered regarding Cubase on the Net.

This guy knows the software very well and he does a good job. It also offers courses that are very well designed. Why not spend a little to have training that could make a difference instead of moving forward blindly…

1 Like