What is your mixing process?

  • Thread starter Thread starter ibleedburgundy
  • Start date Start date
ibleedburgundy

ibleedburgundy

The Anti-Lambo
In my quest for mixing advice, I am finding anecdotal pieces that are helpful but really I would like to know what is your process from start to finish? Reason being, there must be a more efficient way than simply trial and error.

I'll start with what I generally do and please share what you do and why I should do things differently if you feel that way.

After I get all my tracks recorded I generally start with drums. EQ individual drum tracks (this usually involves taking the mids out of toms and bass drum). Add effects to drums. Get levels for drum tracks and pan them simultansously.

Move on to bass. Get the right sound with effects and eq, then bring to the right level relative to drums.

Move on to guitars, EQ, effects, levels.

I fucking hate singers 99% of the time so my shite almost never has vocals.

Then I move on to the overall mix and tweak levels and pan stuff if I feel like it.

Sounds pretty straight forward but I am curious how everyone else does it.
 
Singers.....ours just bailed on us...this morning!!! Ding, turn the page.

Same mix approach here.
 
Well....when you have no vocals, IMO it's much easier to mix things in most cases.

I usually start with drums too...but I never EQ at the beginning, rather, I focus on getting basic levels set between all the tracks, along with their rough pan positions using basic starting points.
That would be the first phase.

Then once I have a good balance of levels and my basic panning set...I will then apply EQ adjustments while monitoring how those EQ adjustments affect the overall balance setup I did in phase one, and I then make spot-adjustments to the levels and panning if needed as I apply the EQ adjustments to all the individual tracks.
That would phase two.

Since I work mostly WITH vocals...I then turn my attention specifically to the vocals and other lead elements, and fine-tune them and then go back and make even more finer adjustments to all the other tracks AFA their levels, panning and EQ. I might make several passes doing little tweaks.
That's phase three.

It's really a constant back-n-forth process...you tweak one thing and then readjust everything else to fit. There is no set-n-forget step where you EQ something at the outset and just move on from there, never touching it again.
Every adjustment has an impact on all the other elements...so you keep fine-tuning in ever-smaller steps until you reach a point where everything gels and you don’t feel the need to touch the knobs/faders anymore.
I'll run several stereo mixdown…listen…and go back and fine tunes some more during the final phase. :)
 
Well....when you have no vocals, IMO it's much easier to mix things in most cases.

I usually start with drums too...but I never EQ at the beginning, rather, I focus on getting basic levels set between all the tracks, along with their rough pan positions using basic starting points.
That would be the first phase.

Then once I have a good balance of levels and my basic panning set...I will then apply EQ adjustments while monitoring how those EQ adjustments affect the overall balance setup I did in phase one, and I then make spot-adjustments to the levels and panning if needed as I apply the EQ adjustments to all the individual tracks.
That would phase two.

Thanks for the reply dude.

The reason why I generally EQ and add effects before I set levels is because EQing and adding effects changes levels anyway. That is not to say I never go back and adjust EQ/effects because I definitely do.
 
The reason why I generally EQ and add effects before I set levels is because EQing and adding effects changes levels anyway.
The other side of that coin is the fact that you should EQ instruments for how they sound in the mix, not on their own. How can you know what needs to be EQ'd until you listen to everything?

As far as levels changing, it's not drastic. EQ shouldn't make that big a difference in levels. And if it does, 95% of the time, you should be cutting, not boosting. So everything will go down in volume a little bit.
 
My general routine

1 Drums
2 Bass
3 Lead vocal

if the song sounds okay with just kit, bass and vocal, then I feel I am on the right track.

4 Everything else
 
I tried something different from an article I read about mixing. I start with basic drums and vocals. Get the vocals right, good level, eq, comp, effects, verb etc. Then i start to bring everything else in. Sounds strange, but it helped me to get the vocals to sit properly in the mix. Usually, they are either too present, or not present enough.
 
The other side of that coin is the fact that you should EQ instruments for how they sound in the mix, not on their own. How can you know what needs to be EQ'd until you listen to everything?

I know I should probably get the ideal bass drum sound without EQing, but for some reason I don't. So on every song I've ever recorded, I take the mids out of the bass drum.

I will try your method though. Right now I'm working on something and for the first time ever it sounds good to me without changing anything at all. That didn't stop me from going in there and adding EQs, reverbs, compressors, limiters, gates, etc. But after I did all that, I wasn't even sure if it was better that way.
 
When recording a project, I'll spend 2x as much time setting up mics and dialing in the sound I want before I even think about rolling tape (or bits:D).

My drum mix is done before I'm even tracking....usually because I've listened to the material before I track it. I'm lucky wherein the projects I do mix, I'm usually sent a .wav of the tune first. Actually, I pretty much insist on that.:p
I know, I know...we don't often have that luxury and I only do because I limit how much work I take on, which frankly, aint all that much.

I'll start mixing before I've finished tracking if I have the luxury of multiple takes on backing tracks, if not, I'll start moving things around during a dry run of the tune. When the tracking is finished, I usually have very little to do other than add whatever effects (I've already selected during the dry run and set up) and whatever little touches the mix demands. I spend waaay more time on the front end than I do on the backend.:cool:
 
Thanks for the reply dude.

The reason why I generally EQ and add effects before I set levels is because EQing and adding effects changes levels anyway. That is not to say I never go back and adjust EQ/effects because I definitely do.

Well...it's not like you shouldn't.
I mean, if I hear the Kick has way too much LF energy, I may roll it off a bit right at the outset....BUT...when I have all my other tracks up and somewhat level/pan balanced...
…I will take *out* the EQ from the Kick channel just to see if that initial roll-off was even needed.
And same goes for any other tracks or any other EQ/FX I might have added during the initial rough mix.

If you apply too much EQ/FX per initial track...you can end up tailoring your subsequent track decisions to THAT track...and so on, and so on. I like to first hear the raw mix with basic/ballpark settings...that way I can see if there are any issues...rather than mix in “pin-ball” or “domino” fashion from track, to track, to track. I will always “pull back” and try tracks with EQ/FX in & out all throughout the mixing process to make sure I even need those things.
Does that make any sense?

All in all...it's a give-n-take, back-n-forth process until things start to gel...
 
In my quest for mixing advice, I am finding anecdotal pieces that are helpful but really I would like to know what is your process from start to finish? Reason being, there must be a more efficient way than simply trial and error.
I eliminate much trial and error by setting up an initial game plan before I start mixing. If you are the one doing the tracking, much of this you can think out before you even start mixing, especially if you are the composer or arranger of the piece. If you're coming into the song for the first time at mixing, like is often the case for me, then you can start with a rough "faders up" listen and just give the song a run through to give it an overall critical listen.

This is where you have to play producer/engineer and not just engineer, and listen to the various elements of the song and decide how you want to *arrange* the mix in order to highlight the creative elements you want to and throttle the ones that aren't so important.

What are the main elements and hooks of the song? Are they especially strong or important lyrics? Is it a lead instrument that needs to be featured? Is there a call/response dynamic that needs highlighting? Maybe it's just a really cool riff? Are there melodic hooks that really sell the song? How much spice to the fills add (or not) to the sound? Maybe the beat or simply the rhythm is the groove that rally sets the rest of the song?

And so forth, in any combination as necessary for the song. By dissecting the song in such a way and IDing the musical and creative elements you want to hear, and what the song itself is really meant to sound like in that respect, instead of just the cold, almost arbitrary method of just building it instrument by instrument, you already start building a mix and a sound design for each track at the same time without just blindly poking around.

I still pretty much roughly follow (or at least start with) a "rhythm/bass tracks first, accompaniment instruments and vocals second, and lead instruments/vocals third" approach - though that may easily change if my creative analysis of the song takes me in another direction instead (e.g. "I have GOT to get those keyboards right first because they really are the center of gravity", and so on).

But even when following the standard track order, I'm alreading listening to the tracks in a slightly different way, within the context of how I want the mix to bring the song out instead of just "lets get this track to sound OK and see how we want to mix it later." I find this approach to wind up not only in a fresher-sounding mix at the end (instead of just another cookie-cutter mix), but eliminates much of the trial-and-error process by actually creatively listening first.

G.
 
...and where would those faders be panned for this run through???



:D:D:D
Everything hard right!!! ;) :D



Seriously, I think that's optional, and a lot depends on how solid your game plan is going in. If I've never heard the tracks before, I tend to just throw the faders up with the pans wherever they happen to be. Of course, if it's in software, the new project will probably have default pannning with everything centered. I'f I've already heard something and have a basic scheme in my head already, I'll start with that panning scheme.

I personally tend to get panning out of the way pretty early. Wherever the pans may be, I'll tend to move them to position first. I've found from this board that many prefer to leave everything centered and perform the ol' mono check first, and save panning until everything else is done.

For me, I like to set pans pretty early because it helps me to listen for the things I mentioned earlier. Determining levels, automation (which I personally use heavily) and EQ comes easier for me when I hear how things sound in place an in spacial relation to each other. But that's me; I think that's one of those personal decisions, and if one can hear and work better by saving their panning for later, I figure that's cool too.

G.
 
For me, I like to set pans pretty early because it helps me to listen for the things I mentioned earlier. Determining levels, automation (which I personally use heavily) and EQ comes easier for me when I hear how things sound in place an in spacial relation to each other. But that's me; I think that's one of those personal decisions, and if one can hear and work better by saving their panning for later, I figure that's cool too.

G.[/QUOTE]

I too like to do my panning pretty early in the game... It helps me with eq when everything is already sitting where I want them spacially as well.Then I usually start with setting all my tracks input levels with the trim adjust. Just seems to help me to get a more even and balanced mix in the end easier. Then I listen to the tracks faders up to see what might be wrong and needs fixing , jot it down on paper. Then down faders and Start with getting a decent rough kick sound and level, work my way through the drums quickly from kick , snare ,ovr's (I usually only use 4 mics on my drums). Then listen to the kit as a whole and see if I think im on par somewhat. Then bring the bass into the mix and listen all the while while adjusting levels abit, work into rythym guitar then vocals. all the time going back and forth between soloing the instruments and the mix . At this point i usually add some verb to Snare , gtr , vox and such . Once I have a decent rough I decide what may need some Eq and maybe some comp. burn my rough to cd and see how i like it and maybe get some input rom the client to see if im on par with where they want to go with the mix .Then I usually get tired Fall asleep and have a dream that I went pee in my sleep...
 
Probably not a model you should follow, but finish tracking, set levels and pan, pour a drink, and then just listen for shit that sounds wrong. :lol:

Though, I'm kind of from the school where if it doesn't sound "pretty much" like a mix when you're done tracking, then you're not done tracking.
 
I wrote out my whole process about a week ago... So I guess a copy paste is in order. But I hate a straight copy/paste so I'll write some new stuff in there.


Before the paste, I will say that adding instruments one at a time might not be the best idea. I see a lot of people saying they start with drums, move to bass, move to guitar, etc. A typical (whatever that is) song puts huge importance on drums. Probably second to only vocal. Doesn't make sense to me to mix something as important as drums without any context from the rest of the song already present.

And this is a topic for another day, but if you are mixing your own stuff you have to learn the difference between a mix problem and a tracking problem. 90% of mix questions (and mastering questions) I see on this board are really tracking problems. If it doesn't sound damn near ready for the radio before you do a single bit of mixing...That's a tracking problem.

Anyway, onto the (augmented) paste:

How I mix:
Zero the board. Meaning put all faders at the same level somewhere in the middle, take out all eq and all effects. Pan everything center. Just listen for a bit. Everything is playing. Nothing is altered.

I like a "starting point" so I don't end up running out of head room. I'm usually safe if I solo the unaltered bass drum, move it to where the master buss hits -14, and then pretty much leave the bass drum fader there and move the other faders around it. But we're not moving of those other faders yet until we turn the rest of the tracks back on. So turn everything back on. Nothing has moved yet aside from the bass drum fader.

On a side note, if you do run out of headroom down the road, don't just pull down the master fader. Most likely you are cliping imediately after summing, and then the master fader pulls down the already clipped sum. Pull down your individual tracks instead.

But back to the mix: Turn the monitor volume low (this is personal preference). Start moving faders up and down. Get your track balance correct. For anything tracked with multiple mics (drums for instance), start flipping phase back and forth on tracks to make sure everything is "in".

Listen for anything that could use some EQ. Until you get some real experiance under your belt, don't try to "improve" a track that already sounds good. Just hunt out the stuff that clearly needs help.

Shoot, even after you have a ton of experience under your belt you probably don't want to mess with any track that already sounds good. But by then, your experience will tell you if that "messing" is a good idea or not.

If you are confident with compressors, start thinking about them. Again, listen for things that need help instead of trying to improve good tracks. Learn the difference between something that needs EQ and something that needs compression. Something that needs a fast attack compressor can sound an awful lot like something that needs added low mids. Learn how the release time of your compressor can "eq" the sound. Often, quicker releases sound "brighter". Depends on what you got. Play around with extreme compressor settings on a mix you don't intend to use to get a feel for the sound.

The most important button/checkbox in your EQ/Compressor section is "on/off" or bypass or however your system lables it. As you make EQ or compressor adjustments, constantly do A/B comparisons with the bypass. Sometimes I spend 10 minutes EQing an instrument, think I nailed it, and then hit the bypass to find I actually made it sound worse. Most EQs and dynamics have makeup gain. Set this correctly so you do not have any volume change when hitting the bypass. It makes the "did I make this better or worse" judgement much easier.

Anyway, move some faders. Change some eq. Move some faders. Change some EQ. Every now and then bring the volume up loud. See if that exposes a problem that you didn't notice down low. This is also a good time (after hearing the whole mix for a significant time period) to solo the kick drum and snare to see what you've got there. Don't make any tone changes solo. Just...learn what's there. Then go back to full mix. Pull the volume low again. Move levels, EQ. Etc. When you're happy and it's time to bring it all back up loud again, maybe this time you really crank it up and leave the room. Or walk clear across the house/building 3 rooms away. This always uncovers fun surprises in the mix. You will find mix problems you didn't hear before.

Go back. Lower the volume. Adjust. Raise volume. Repeat.

In the middle of all of this, you will find that you need to mute tracks every now and then to concentrate on something else. This is important: Be sure that you remember the sound of what you muted while you make adjustments. Mute as few tracks as possible to hear what you need to hear. Give it all a listen after unmuting before finishing your adjustments.

As an alternative to muting a track or two, if you have enough headroom you can just crank up the volume of the single track you are trying to hear. When you lower the volume of this track to put it back in the mix, don't assume that it goes back to the same place. In fact, listen to everything else as well. The tone changes you made might mean that some other track has to go up or down.

Once you are happy with all THAT, then maybe you start moving your pan knobs around. Be aware that after something pans, you might have to adjust the level to compensate. How do you know when you have to make a volume compensation after a pan?

This is just my brain, but I literally see the mix as a solid three dimensional object sitting on the desk right in front of me. I don't try to imagine it as some visual metaphor or anything. That is just how it appears to me, and I couldn't perceive a mix any other way if I tried. Anyway, that is how I know where to pan a sound and how loud it should be. If the "mix object" sitting in front of me has an obvious gap 1 foot back, 14 inches to the left, and 8 inches off the ground, well then I'm going to put an instrument there!

Your brain's millage may vary, but the point is to learn how you personally perceive a mix and use that to your advantage so you can build a mix that you know to be correct. I learned by years of listening to my favorite professional mixes. The listening doesn't count if you are driving or cooking or doing homework. Nothing but you and the music. Then you will know how you perceive a mix, and thus how to build a mix.

Anyway, when you are doing this panning and your sound field is coming together you can start thinking about digital effects. Think "short delay" for things that are wrapped in a bubble and don't seem to want to join the rest of the instruments. Think reverb for things that are too "close". Just as we did with the eq, don't look to improve any track that already sounds good. We are just adding effects to things that need help.

Then it's back to low volume/level tweaks/eq tweaks. Back to high volume. Repeat etc etc. Your wife is wondering if you're still alive by now and you are wondering what the heck there could possibly be left to learn about these tracks, but you're not done until you feel it. You will know when you're done.

If the CD sounds good in your car, think of that as confirmation that you are done.

And remember, you have to do this independently for parts of the song that sound different. But don't mix one section entirely before even starting on the other. Bring it all up together. And mix automation/saved mix scenes are your best freind when moving from section to section.
 
Last edited:
Probably not a model you should follow, but finish tracking, set levels and pan, pour a drink, and then just listen for shit that sounds wrong. :lol:

Though, I'm kind of from the school where if it doesn't sound "pretty much" like a mix when you're done tracking, then you're not done tracking.

Dude. That's a great model. You summed up my exact technique in like 4 lines of text. Took me an entire essay.
 
Dude. That's a great model. You summed up my exact technique in like 4 lines of text. Took me an entire essay.

Yeah, but if you're trying to figure out the process someone goes through while mixing, yours is approximately 6,128x more useful than mine. :p I state the spirit - you actually give a pretty good walk-through of the mechanics.

Besides, I didn't see "pour a drink" anywhere in yours. ;) Which, to be fair, probably (and past a point certainly) doesn't help the process, I'm sure, but probably puts you in the same mindframe as the listener.

I guess to try to expand a little bit, Generally when I start looking for things that don't sound right, I start with the interplay between the bass and drums. This is something I've always struggled with, trying to get a clear, articulate low end to my mixes, so it's something I tend to latch onto. I write instrumental guitar music, so I totally fixate on guitar sounds and whatnot, but partly because of that guitars come pretty easily to me (or, rather, I know what I want my guitars to sound like in advance almost perfectly, so it's much easier for me to get them right while tracking). It's the instruments I'm not as personally focused on (guitar and bass) where I may not have as clear an idea how I want everything to fit together in advance, so when I start mixing I generally start by thinking about how best to fit them together, and then how to get the guitars to gel on top of them.

Of course, after four scotches, this is generally a lost cause. :D

EDIT - and of course the other thing is I'm a total hack compared to a good 90% of this board, and not someone who should normally be giving out advice. :p
 
Probably not a model you should follow, but finish tracking, set levels and pan, pour a drink, and then just listen for shit that sounds wrong. :lol:

Though, I'm kind of from the school where if it doesn't sound "pretty much" like a mix when you're done tracking, then you're not done tracking.

Haha hell yeah, that's pretty much exactly what I do :D Same school too, I have my standard cubase project template with all my eq's pans fx and etc. set when I start a new proj, so I'm basically dialed in from the start. I loop it for an hour while I'm having a couple beers and playing on the web or whatever.

I'm sure it'd be a lot different if I was recording someone else..
 
Back
Top