if you use buses you also have the advantage of bouncing that stereo buss as a stem, which is something I usually do near the end, you can then control all the guitars, drums, bass, keys e.t.c on one single fader and for example sidechain the drums to the lead vocals, for me it makes it easier, like painting with broad strokes.
What higher-level mixing techniques are a MUST for professional sounding stuff?
- Thread starter Divinejames
- Start date
ashcat_lt
Well-known member
Gain staging is something that causes a lot of people a lot more confusion than necessary. It doesn't matter at all if you're mixing in a modern DAW. Don't distort anything unless you want to. That's about it. A lot of the things we use in the box won't distort even if you want them to. In analog it's complicated a little by the noise floor, but honestly, the general rule is the same. Don't distort anything unless you want to. About everything in analog will distort at some point.
I feel like you might be leaving something on the table there. Not that you really always have to add a bunch of plugins on every bus, but there are some good things you can get out of your bus chain that can't get other ways very easily if at all. There may in fact be some things you're doing at a track level that could actually be better handled at the bus level. A lot of little things at various stages - track, group bus, master - can often be better than one big thing at just one stage.
jkuehlin
New member
Good point Drew. Would you mind elaborating on what you mean by 'thinking about gain staging' a little more so others who are following along here can discuss as well?
I'm of the opinion that is does very much matter, regardless of weather a signal is in the DAW yet or not. Perhaps if anyone else has a differing opinion than Ash they could chime in on how and why.
ashcat_lt
Well-known member
Why don't you start? How and why do you believe it "does very much matter"?
jkuehlin
New member
Here's a couple hints:
Why do plugins such as the Opticom XLA, Neve 1073, and Decapitator have output trim?
Why to plugins like the API 2500, J37, Voxbox, and Distressor have input trims? (We are not talking about compression threshold. Input trim is a dedicated gain staging feature, independent of the compression controls).
The Slate Trimmer is a plugin with a unique algorithm that manages trim. Its not simply a volume knob.
The stock Avid single band EQ has both input trim and gain makeup. Why?
Not sure how there's anything wrong with discussing techniques and tool that are relevant to people who have genuine questions about them. Is this not what a forum is for?
Last edited:
DrewPeterson7
Sage of the Order
In reverse order... It's not that I don't use bus processing, and in situations where I'm looking at the bus as a single instrument - say, a bass guitar broken into low end and high end parts, or a bus combining two seperate mics on a stereo-mic'd guitar - then I absolutely treat the bus as if it was a single instrumental track and go to work on it as if it were. But, for example, if I'm bussing all the various parts of a drum kit into a single stereo drum submix, or if I'm taking a whole bunch of different (and not all doubling each other) rhythm guitar tracks into a single submix, then I'm way more likely to prcess at the track level than the bus level. Never say never, and I could see doing some light compression or maybe a mild EQ tweak at the bus level or something... But I feel like for the most part if I'm EQing rhythm guitars I'd rather go at it on the indivudual tracks.
Re: gain staging... Well... I suppose I could have said this more clearly, and to be fair I think it's probably initially more of a concern in tracking vs mixing (if I had a dime for every time I heard someone say you need to track as close to -0db as possible to "use up all the bits" but then have every track in the project turned down 6-12db...). But, when it comes to mixing, I think gain staging, or at least being AWARE of how the mix decisions you're making are impacting gain, definitely matters.
Like, if you're running a track through an EQ and boosting at 2.8khz by 3db... that's impacting track gain. Right off the bat when auditioning you're going to want to adjust output gain to try to balance the perceived volume so you're not preferring the EQ'd version just because it's louder, but far more importantly, that plugin is going to impact every single plugin downstream from it that has any sensitivity at all to gain. If that's it, then no harm no foul, the DAW won't internally clip so you're not really hurting anything... But, if you're running a track through an EQ and then into a compressor, then by selectively boosting a particular frequency band you're in turn changing the amplitude of the output signal, and making it look differently when it hits the front of the compressor... And, by clamping down on parts of the signal over a certain threshold, potentially changing how the EQ tweak is perceived as it passes through the rest of your FX chain, etc.
With only one or two plugins you're probably not getting yourself into too much trouble, but if on your bass guitar you've got an EQ, a saturation plug, and two compressors, then unless you're really careful in setting up the gain staging and keep an eye on what exactly your FX settings are doing, then its possible to accidently start to overload the front of a compressor a lot harder than you meant to when, say, dialing in a little more saturation and some extra energy around 250hz. And, while at the end of the day it doesn't matter whats happening as long as it SOUNDS good, I'm kind of of the mindset that you want to make as much of this stuff as possible intentional choice and not happy accidents, and that unless you know for a fact hitting that first compressor harder because of the saturation and low end you added in is making the bass track gel a little more because you also went in and auditioned it with more "neutral" gain staging with the compressor hitting around the same way as it did before the new tweaks, then you're really rolling the dice.
Like, does that make sense? Gain within a DAW is totally clean even well over 0db internally... But plugins, especially ones trying to add some "analog" elements or that DO impact dynamics, absolutely need to be used with an eye on gain staging. At least, I think they do.
Gtoboy
Well-known member
IMHO gain staging when tracking is more a "don't get unwanted distortion" deal but mixing , not only in the box plugins but routing to hardware units gain staging (or Trim) is something to keep an eye on NOT for distortion but to make whatever i'm sending program material through works right in it's "sweet spot" to get the result I am listening for. I'm in the box so no i'm not worried too much about clipping, I'm trying to get a particular sound and with many(most?) plug ins/devices that means hitting them just right in the input section. No different from matching input impedance to the microphone-not all mic's like the same input.
jkuehlin
New member
Very well said!! That's an excellent answer to ashcat and altarman's response a from couple days ago.
In short, as Drew has illustrated, it is about how someone manages their signal as its passed from plugin to plugin down the insert chain. And the behavior of some plugins changes depending on how you are hitting them. The ability to change and manage levels effectively as adjustments are made throughout the session requires an ideal starting point. Again, great explanation Drew!
This is a good point also. The reason there even IS a 'sweetspot' in some plugins (take digital gates for example) is because the hysteresis and release curves are NOT linear. This is what we call 'program dependent'. And as Gtoboy has realized, distortion/saturation is not the only type of program dependent plugin. Another excellent example of why gain staging practices 'very much matter' when ITB.
Do you find there are people here attempting 'talk' themselves into becoming a pro? Very interesting observation.
Last edited:
miroslav
Cosmic Cowboy
ashcat_lt
Well-known member
I'm not sure how this is any more difficult than "turn the knobs til it sounds good and if that doesn't work, try a different knob." It is nice to have some idea of how a given knob might help or hurt, and sure even how different levels might affect different stages in the chain, but it's not really the super-critical, ultra-mysterious, life or death situation that causes people to post multiple "What's the right level" threads before they even feel like they can do anything. All that actually matters - especially in the box - is whether it sounds right or not. It's not at all unusual that the wrong levels are what make a given thing sound right.
But I guess I was actually trying to head off that whole thing where people get all caught up trying to make sure everything starts and stays just right nuts on -18dbFS. There are plenty of reasons why that might be wrong at any given step of the process. Even if there is a "sweet spot" in a given process, there's usually wiggle room, there are shades in between, and frankly it probably depends more on the dynamic range of the signal more than the absolute levels. We have plenty of ways to turn things up and down before, after, and between any plugins we might be using, and we should not be afraid to use them.
Now the thing about changing EQ before compression (or anything else, especially other non-linear processes) in order to shape the way that process responds IS something of an intermediate - if not advanced - technique that is worth playing with and learning about. I do a whole lot of "pre-emphasis/de-emphasis" - sandwiching like a compressor between a pair of EQs that are more or less complimentary to one another. Maybe you boost the bass going into the compressor to get it pumping, but then when you get the compression you want, it's too much bass, so you knock it back down on the other side. Some of that can be done by way of sidechaining or using other techniques, but it's not exactly the same thing.
Wanted to mention too, since it came up, that I generally would not boost a frequency on an EQ except that I want that frequency to be louder. If I don't want that frequency to get louder, but I do want the sound to have a greater proportion of that frequency, I'll usually cut the stuff I don't want. That is, Subtractive EQ is the answer to that particular "problem". I don't bother level matching. I change the EQ and if that means I then need to adjust the overall level, then I do that. It's not true that things always "sound better" just because they're louder in a mix. It's actually almost as likely that they will just sound wrong, because they're now too loud for the mix. I don't know if my point is really coming across here, maybe it's just a fine philosophical distinction, but in a lot of ways it's like that whole thing with Pan Law. It literally only matters when you're moving the knob. If you pan your shit and it gets louder or softer or whatever, you move the fader and it's fixed.
DM60
Well-known member
I have to agree with much of what is said here. The main point of affect on a plug in, EQ before a compressor will very likely have an impact on your compressor. Just use threshold adjustment for example, if I reduce a sound that would reduce the overall input, that would definitely have an impact.
Most other things I can' say how or if I can even hear the difference, but there are some that are very obvious and noticeable.
DrewPeterson7
Sage of the Order
Bolded two points for emphasis. First, I think the first bit I bolded, that it's nice to have some idea of what a knob does and why it might help or hurt before you start twisting it, is exactly what I'm getting at. In a modern DAW, there are a LOT of different knobs to turn, and you could try to get a killer mix going just by throwing a whole bunch of random EQ and compression tweaks at a mix, and a million monkeys with a million copies of ProTools will eventually produce Pet Sounds... But you can do it a heck of a lot faster if you know which particular knob to turn to make that high end fizz on your rhythm guitars get pulled back into the mix a little more.
Second, just to clarify, that's NOT what I'm saying, that there's some theoretical level that we should be running everything at, and if you happen to hit an EQ at -17.6dfFS, then the mix is ruined... So much as, you're going to get a lot better results, a lot faster, if you're at least AWARE of what's going on as the signal runs through a FX chain, and what's happening to the output at each stage.
So, heck... Maybe you could even say that what I'm talking about isn't even gain staging in the purest sense, so much as just a really good awareness of each step in the virtual signal routing as audio moves through the DAW. Part of that - a large part - is definitely gain, but it's less about specific levels and more about just making sure everything is operating in a range where it can be effective. Like, that the compressor you carefully dialed in to knock just the peaks down by 4:1... When you start EQing the track in front of the compressor and everything you try sounds like garbage, at least having the awareness that you might want to check your compression threshold again and maybe the issue isn't the EQ cut itself, it's that everything sounds weirdly thin because now your peaks are nowhere near that compressor threshold anymore. That kind of stuff.
And yes, to alterman's point, that's definitely something I've learned the hard way. "Ohh.... No wonder that sounds like crap, I'm overloading the %$@!#@#$ out of that thing!"
ashcat_lt
Well-known member
I think we're mostly saying the same thing. I think my main point is that in digital for sure and in analog with a coupe caveats, absolute level doesn't much matter. It's all relative to the specific piece of gear - software or hardware.
But yeah, I don't consider paying attention to what your doing to the sound and learning which tools do what to be really advanced level stuff. To me that's like basic. Maybe that's because I did start in analog where you're generally trying to keep a bus with four flat tires going 200MPH down a two lane tunnel with no shoulder space. But that's the basics. Learn your tools.
I don't want to make it out like pre-emph/de-emph is some secret thing I invented, but at this point it's another one of those things that kind of just intuitive to me. Like, if I had the compressor doing what I wanted to do, but I wanted to pull a little bass out of it, I'd put the damn EQ after the compressor because duh.
I figured it out like the first time I tried it. It made sense to me because I understood what these things were doing to the signal. And I probably did actually say "Duh" out loud. It happens to be the real secret of how and why all those boutique guitar pedals sound the way they do. Like at this point, any real innovation in overdrive/distortion/fuzz is in how you filter the signal before the nonlinearity - which frequencies will dominate and what harmonics will be generated, and after - what part of the mess that came out of the distortion stage do we actually want to hear.
But yeah, I don't consider paying attention to what your doing to the sound and learning which tools do what to be really advanced level stuff. To me that's like basic. Maybe that's because I did start in analog where you're generally trying to keep a bus with four flat tires going 200MPH down a two lane tunnel with no shoulder space. But that's the basics. Learn your tools.
I don't want to make it out like pre-emph/de-emph is some secret thing I invented, but at this point it's another one of those things that kind of just intuitive to me. Like, if I had the compressor doing what I wanted to do, but I wanted to pull a little bass out of it, I'd put the damn EQ after the compressor because duh.
I figured it out like the first time I tried it. It made sense to me because I understood what these things were doing to the signal. And I probably did actually say "Duh" out loud. It happens to be the real secret of how and why all those boutique guitar pedals sound the way they do. Like at this point, any real innovation in overdrive/distortion/fuzz is in how you filter the signal before the nonlinearity - which frequencies will dominate and what harmonics will be generated, and after - what part of the mess that came out of the distortion stage do we actually want to hear.I have a general guideline I use considering if an eq might or might not want to be before of after a compressor; If it's not going to make significant changes in level, or is part of track cleanup (i.e. filtering excess or unwanted low end -eq before the comp is fine.
And consider to the compressor's decetor circuit, most other tone adjustment's level changes appear rather small in dB compared to the changes in track volume you're typically trying to tame.
DrewPeterson7
Sage of the Order
Bingo - here's where we differ, then. I'm just on the cusp, at 37, where I started in digital, as a hobbyist, not working on tape, and I think that's one of the most fundamental differences between guys who cut their teeth on a DAW vs guys who cut their teeth tracking to tape and mixing on a console - all this stuff about keeping a careful eye on your levels is second nature for you because you HAD to think this way simply to not fuck anything up while you worked, and from that perspective, digital is probably kind of liberating because it's like the gloves coming off. For someone who started digital and never really had to worry about gain staging beyond the tracking process and ensuring you weren't clipping on input, though, it's entirely possible that you could get reasonably good at the basics of mixing without really thinking all that much about how the output of one track or plugin is impacting the impact of the next. And, from that perspective, especially as increasingly I've found "saturation" effects that II actually really like (I use the Sonimus Burnley73 on damned near everything these days) the level that the track is hitting the plugin really begins to matter.
So, from a pure digital perspective, actually starting to consciously think about gain staging within your DAW actually CAN be a pretty powerful way of looking at a mix... But, I totally get that for someone who started on tape, it's like I just burst into a room of long distance runners and said, "you guys! I've got this super innovative technique I just thought up! I call it 'breathing' and you do it by inhaling and pulling oxygen into your lungs, and then exhaling to expell carbon dioxide, and if you keep doing it long enough, you won't die!"
ashcat_lt
Well-known member
IDK though. You're still talking about what seems to me like the very basic step of learning how various effects actually work. I seriously can't imagine just putting a plugin on a track without having a pretty good idea of what it's going to do. Like, is it just because some dude on the internet told you that you should? Drop it on there "Hey that works! I don't know why, but it sounds good." And I mean, I'm all for experimentation too. Drop it on there and find out what happens. Turn the knobs and see what they do. Change things before this and see how it responds to different inputs. Change things after it and see how it impacts things downstream. Fine. You don't have to be able to draw the circuit schematic and explain what every part does. You don't have to be able to explain it at all. You do (I think) need to come to your type of understanding. But again, this is like basic competency.
It really freaks me out when I see threads about the order of guitar pedals. Filter before or after distortion? Distortion before or after chorus? Where should I put the reverb? If you know what each one does, you can make educated assumptions about what to expect if you cascade them, and then if you know what you want, you put things in the order that is most likely to get you there. Flange before delay is completely different from delay before flange and if you don't immediately understand why, I really don't know what to tell you except maybe link to a page that explains the basics of the effects. Or like, "Plug it in and find out."
It really freaks me out when I see threads about the order of guitar pedals. Filter before or after distortion? Distortion before or after chorus? Where should I put the reverb? If you know what each one does, you can make educated assumptions about what to expect if you cascade them, and then if you know what you want, you put things in the order that is most likely to get you there. Flange before delay is completely different from delay before flange and if you don't immediately understand why, I really don't know what to tell you except maybe link to a page that explains the basics of the effects. Or like, "Plug it in and find out."
DrewPeterson7
Sage of the Order
Welcome to the internet. :lol:
ashcat_lt
Well-known member
See I'm not really down with that though either. I have my experience, you have yours. I don't honestly believe that experience with legacy systems is necessarily important to somebody who is just never going to use them. Yes, some of what I did in analog informs my usage of digital tools, but the important principles really are the same either way, and somebody who really understands things in the box should be able to parley that understanding to analog workflow. They might need help with patchbay and a qualified tape operator, but they'll know what needs compression and what needs EQ and what seems to be distorting at some stage and all that basic stuff.
And you know if what you're actually trying to do is work as an assembly line worker churning out industry standard consumable products, then you probably are served best to learn the it by rote so that you can do it exactly like the people who set the industry standards. Buy the right shit, put it in the right order, pull up the preset or set the knobs like your hero does. Rock on. Get paid. Doesn't bother me.
Gtoboy
Well-known member
I have to say I completely agree with you that (for me) the best way to learn is to get in there an tweak stuff and see what it does, though I prefer to read about the concepts and some applications I can compare to/try to emulate for understanding purposes. Where I diverge is I have found when giving exactly that advice to newbs(especially millennials expecting instant gratification and black and white answers eg what's the best way to do x?(!!!) that it's almost necessary to add the caveat about gain staging, sweet spots, etc otherwise I hear back "o this plug in sucks you don't need this, it doesn't work, I don't get how you get that sound, etc. You don't need to ask someone "Does this sound good?", we have listened and practiced and learned what we like , etc. Not so much for many just starting out, and though I am always going on about just try it and see , even crappy sound can have a use and IF you pay attention you will always learn something but let's face it , that's not what some newbs want to hear. So IMHO throwing out the hint that gain staging can change the sound, sometimes drastically, can be an "Aha!" moment for those starting out.
ashcat_lt
Well-known member
I haven't actually been arguing that.
In fact, this is pretty much exactly my argument. It's important when you're starting out. It's not a "higher-level" technique. It's basic to the whole damn thing. But also I'm definitely saying it's not as hard and weird and worrisome and critical to have it exactly perfect as so many people - especially new people - end up thinking. You give a gentle hint about gain staging and the next thing you know they're normalizing every file at several points in the process and doing horrible things to keep it at -18 rather than -17 or -19. They end up mixing by meter, second-guessing their aesthetic decisions, and very often not getting much actually done.