High Mixing Level

Help Help

Member
Hello! This is a beginner question that is probably common sense around here. I have a song I recorded and mixed in Cubase. I added some stuff and now it clips on the chorus', up to 2 dbs or so. The rest of the song sits under 0 though. Anyway, I know I need to get a grasp on gain staging as this kind of thing happens to me more than it should.

Given all that, would it be advisable or would there be any drawback to putting a limiter on the mixbus? I'm not looking to squash anything, but it seems like the dynamics would be preserved in doing so, just some occasional peaks on the loudest parts of the song being turned down. I've never put anything on the mix bus before as I'm wary of such things, but it would make things a lot easier.

Otherwise, there's turning the volume down on the master fader, but I've always been wary of doing that too(I'm so bloody paranoid lol). Does doing so make any negative difference in a DAW? The drawback is that I'm turning the song down a maximum of 3db due to the loudest peaks, though I guess that's what mastering is for? Thirdly, I could pull the faders down on every track I suppose, which I'm also wary of as I got the mix where I want it.

Usually I just do the latter and pull all the faders down, but I've always wondered about either a limiter for such scenarios, or else turning the master down. Would be a lot more efficient anyway.

Any responses and advice would be greatly appreciated! Thanks!
 
The simple (sorry, but it really is simple) answer is to always leave some headroom.
I mix so that I am never above -6db and often at around -10 to 15db.
If you start out like that, with track faders pulled below zero, it has two main advantages:
1) That you have to room to boost sounds in the final mix without going into positive dbs.
2) That you avoid the problem you faced.

I sometimes find myself having to pull all the track faders down half way through a mix because I don't have enough room to manouvre.
Hope that helps.
Better brains on here will explain better or differently.
Don't know why I bother posting on here. No-one ever acknowledges me or says thanks..
 
Last edited:
I'd say for this one just pull the master fader down and don't worry too much about it. In future leave yourself a few more dB of headroom when starting the mix. Maybe do a rough mix with no plugins, see where it lands then make an across the board adjustment to guarantee yourself headroom before moving on to the in depth mix.
 
If it’s going to be mastered separately, and you’re confident it actually sounds the way you want, then don’t worry about the overs. Just render to a floating point (32 bit is fine) format and deal with it down the road. Floating point doesn’t clip.

The problem is you can’t be sure what it sounds like because it’s being clipped on the way out to your monitors. Turning down the master a bit will let you actually hear the unclipped mix. Then you’ll probably still just leave it for mastering. :)

Course floating point also has no practical bottom limit, so if you render after turning down the master, the result will fine too. Whether your peaks are -18 or +18 literally means nothing as long as youre rendering to FP, and unless you’re rendering the real final distribution master, you should be rendering to FP.
 
Thank you guys so much for the replies! I really appreciate it :)


If it’s going to be mastered separately, and you’re confident it actually sounds the way you want, then don’t worry about the overs. Just render to a floating point (32 bit is fine) format and deal with it down the road. Floating point doesn’t clip.

The problem is you can’t be sure what it sounds like because it’s being clipped on the way out to your monitors. Turning down the master a bit will let you actually hear the unclipped mix. Then you’ll probably still just leave it for mastering. :)

Course floating point also has no practical bottom limit, so if you render after turning down the master, the result will fine too. Whether your peaks are -18 or +18 literally means nothing as long as youre rendering to FP, and unless you’re rendering the real final distribution master, you should be rendering to FP.

Thanks ashcat! I usually mixdown to 24 bit. By render to 32 bit I assume you mean in mixdown to send to mastering. Please excuse my ignorance! No issues with transferring such a large file online to an engineer? Is it normal to mixdown to 32 bit/Floating Point? I've always assumed 24 bit was the standard.

If I were to mixdown to 24 bit, is there anything at all lost in turning down the master fader? I feel like this is a stupid question in a DAW and it makes no difference, but I used to record on tape where it does matter, and my OCD needs assurances ha. Plus I read conflicting opinions on this once on some forum with less knowledgeable people than this one, and so a doubt has lingered.

If I were to turn down all the faders the exact same amount, this is essentially identical to turning the master fader down right? Is there any way this can effect the mix? Again, I'm pretty sure this is obvious but I'd appreciate some quick reassurance haha.

Thank you for your patience with me!
 
Right now 24 bit is the standard, but 32 float is also becoming more common. Most DAWs have been using 32 float internally anyway and some are going to 64 float. Pro Tools switched from 40 bit fixed sometime around version 9 or 10. So delivering in floating point makes a certain amount of sense. That said, unless you really do something wrong 24 bit fixed is fine.

I don't think you lose anything by turning faders down. Any analog noise captured at the ADC will go down with the faders. If you're using buses or aux sends with any level dependent plugins you'll have to consider the effect turning faders down will have on those.
 
It depends on how many tracks you have and how much fader automation you’ve already got going on whether what I’m going to suggest is feasible, but as I like to use the idea of getting a static mix in place first, I will use “clip gain” first to get things balanced so I’ve got track and master faders at zero with the master peaks allowing appropriate headroom. This can help take care of many gain staging problems - whether self-induced or caused by tracks you’ve been sent.
 
I guess my thing is that yes, you should try to gain stage things and watch your master headroom and all that - among other things like I said so you're not monitoring a clipped version of what's actually happening. But like if you've already got it all mixed and it's clipping a little, and turning down the Master still sounds good, then you really don't have to turn down the master, but it doesn't matter if you do or not as long as you render to floating point.

I'm actually really bad about this. I'm not the least bit worried about red lights. I know about gain staging and I use it to my advantage, but if things start adding up at the master fader so that my converters are clipping, I just turn down the master. Then at a certain point I end up putting something like a mastering chain on the master. I adjust that chain (one way or another) to work with the level that's coming in, but to avoid 0dbFS at the other end with the master fader back at unity. Then I finish the mix and when I'm ready to "send it to mastering", I bypass that master chain and render to 32bit FP. It sometimes end up peaking at +6dbFS or more. That's fine. I knew what I was doing. It wasn't causing anything unexpected in my master chain, and I'll do the same kind of adjustment to make it hit the real mastering chain later. For now, I just render it and go, and then post the pictures in certain forums just to troll people. :)
 
Just throwing in late - Although it's arguably "bad form" (and you should check your gain staging at the start), 32-bit floating-point files would then be the way to go otherwise. Happens frequently. "Either drop ALL your faders by (a lot) or drop your master fader by 12-15dB and we should be okay."
 
I mix into a limiter and deliver to mastering both with and without a limiter so they have a choice as to which is prefered to best handle the job. Probably not the popular answer sure but effective and helpful as you constantly have a check of what squashing your record will; sound like and will help you avoid those pitfalls early on during the mix stage.
 
Back
Top