My friend, Dan, asked some great questions in response to these posts on unity mixing, so let’s address them.
“When I run through a soundcheck I always start by setting every pre-amp level to 0db or close to it. This allows me to sit all my submixes for broadcast ect., to unity and then set them post, thus enabling my submixes to sound similar to my house mix with tweaks but without constant layer flipping. Doing this gives me enough head room to work up and down the fader and on the rare chance that I need to gives me some more room on the pre-amp should someone go emo on me and start whispering into the mic.
I was always taught that 0db was the happy place for signal to noise, broadcast, amps and gain structure the world, life and the ever after. =) Now working with pro’s every single day I find that even setting to 0db by faders are pretty close to unity most of the time. When I work with stupid people, I find that setting to 0 is crucial but it makes my faders look a lot like mountains and hills. So what does mixing at unity do to your input level on say an instrument or voice that is near the back of the mix? Wouldnt you be pushing a bunch of noise through the mains?
So give me some feedback. 0db = good or bad? what does it do to your submixs? I find when working with Vince, Amy, Toby, Smitty or any real pro’s that even when I set to 0db I am still pretty close to unity on the fader. However with idiots I have to work with sometimes my faders are all over the place.”
My perspective is that the answer is balance. Do you mix the whole show constantly changing gain? No. But every one of our younger engineers have mountains and hills on the faders, then wonder why the monitor mixes they’re creating for the artists are lacking (because all of our rigs have single console for FOH & monitors) and the mixes are inconsistent. I think the key is intentionality. If you decide that on a given channel which may be more buried that you’re going to ride the fader down intentionally, that’s cool. The problem is when you aren’t in control of the console and the summing bus is controlling you. Every channel is gained too hot, you have to burn it off somewhere because the summing bus can’t handle everything coming to it at 10, so faders are all over the place. One more thing: we’re talking about getting maximum return and a growing engineer can’t have everything, so what is the bigger trade off? I’d suggest on those buried inputs, the slight amount of extra noise you might get doesn’t matter as much as all the benefits to unity mixing below because the input is buried as it is. If you can’t give artists monitor mixes that they like and your house mix isn’t working, who cares if there’s a tad bit of extra noise on a channel? I’d take the noise but mixes that work any day.
Regarding submixes, how you gain the console probably isn’t as critical on POST mixes, because if you gain something too hot or too soft, the faders will compensate and the mix will still translate. Where unity mixing gives HUGE benefits is on PRE mixes, because just as you can put the faders near unity across the board and get a solid house mix, you can put aux mixes at unity (or based on some percentage) and know that the aux mix will also translate. If every wedge needs kick/snare/hat, put those three at the same level in the wedges that need them and you know the relationship between the inputs will translate there just like it does in the house. We have a stereo drum submix that feeds our ears mixes. I never have to adjust that mix week to week – all I have to do is push the faders up one at a time and build my house mix through channel gain so the drums are working with everything at unity and it translates nearly perfectly to the ears.
The better and more consistent the input, the easier things will be. So regarding major artists, if you have a band full of “A” inputs who give you great consistent input, its harder to screw it up. I think we’re basically talking about the same things – my addition to your 0 db s/n point is that riding in a good unity spot is just as important at the outputs of the console as it is at the front end. Often by running every channel at its own individual optimum gain can sacrifice the output bus or vice versa. Again, the answer is moderation.
One more addition to this whole thing that buttons up the topic well. Robert Scovill opened my mind to the use of subgroups AND vca’s and its become the last piece of this whole thing and might be the answer for your point. I submix my music to three stereo groups – rhythm (drums & bass), band (guitars & keys), and vocals. I’ve posted one of his articles recently about this topic here. Everything musically on my console fits into one of those three groups. This way, I can do the macro level balancing on the groups rather than having to do it on the fader or at the mic pre. I find that often times if I gain the guitars so they are hitting the mic pre’s hard where it sounds better on the console, the band subgroup can be 2-5 dB lower than rhythm to balance the whole mix with channel faders still at zero. Often times the rhythm & band groups are a couple dB below the vocals group for everything to balance as well. Sometimes I will pull all three of those groups back a few dB before it hits the left/right because its a great place to burn gain, keep mic pre’s hot, faders near unity, and not overload the summing bus.
UPDATE: Dave Stagl has an addition worthy of promotion to the main post.
“Another big thing I’m not sure you mentioned about keeping your faders near unity is the area around unity is where a fader has the highest resolution. For me, this is my main reason for mixing this way. Due to the logrhythmic nature of a fader’s throw, a millimeter of movement around unity could give you a .1 dB change, however if you’re only a 1/3 of the way up the fader’s throw that same millimeter could be a 5-10 dB change depending on a particular console. If your faders are riding low at the start of a show, you will have far less control of the mix than if your faders are at unity.”