Skip to main content

Several years ago, A friend of mine who I had previously mixed a record for decided for his next record to buy some gear and do it at home. A familiar scenario huh? Anyway, he called to ask what kind or gear to get, etc. A few months went by and then he called one day and said I'm done recording and I think it's pretty good, let me ask you a question, how do you mix? I tried not to laugh cause he was asking seriously. I know what thought processes I use when confronted with this fun but sometimes scary situation, but I'd like to know how you guys go about it. I'll chime in after a few posts and give ya'll my take on it.

Topic Tags

Comments

audiokid Sat, 02/04/2017 - 20:30

bouldersound, post: 447199, member: 38959 wrote: Simple solution, poverty. I have no access to real LA2As and all the preamps and eqs (in that studio) are M-2600 MkII channels. If the preamps in that board aren't especially interesting at least they're forgiving. Anything in the vicinity of 0dBVU sounds about the same and it doesn't clip until +24dBVU. Honestly, in the rush of tracking (hobby band, limited time to devote) it's the best compromise. If I had to sort through preamps and find a small sweet spot we'd never get anything done, and/or I'd screw it up.

Well, there's a Solo 610 that I break out when appropriate, and then what you say hits pretty hard. You can't just fix that after the fact because it is so responsive to levels.

Same thing at the "other" studio that uses a variety of preamps (Daking, Grace 101, V72 etc.) into an Apollo 16. In fact a recent vocal session spanned two days, and the solution was to leave the mic, goboes and Daking preamp/eq untouched. With that kind of gear I do gravitate toward your method. I guess that means, yes, I hear what you're describing.

So, really, I'm flexible. Depending on the gear and the musical style I may go either way, mix while tracking (and start over when tracking is done) or leave it untouched (other than level and pan).

Excellent. Thanks for the explanation and coolness.

JWGT Wed, 02/22/2017 - 15:26

I'm more leveling while tracking so the performers can monitor while they record. In fact, I do more headphone mixes than I would like. Once tracking is "mostly" done then I start mixing from the bottom up - specifically bass and kick. If I can make that sound groovy then I start moving up from there. I also prefer to fix timing and pitch issues as I go on each track. Once I've been through every track that way, I start to focus on the flow - verse to chorus and back, then beginnings and ends. Then I start tone shaping, adding effects etc. When I think things are getting balanced I add some combination of parallel compression and master bus limiting and go through the levels again a few times, focusing on different tracks. Once I'm getting near the end I like to focus on the bass and the overall sound signature. At this point i switch from Behringer truths (ribbon model) to Shure 840s and in ears to hear the bass. If I can hear decent bass with the Shures and its not swamped in bass with the in ears I am in the right range. I listen on magnepans for the overall balance, then on my laptop, my phone, etc. I try to inject a few reference tracks hear and there throughout,

Lots of good ideas here - I like the one about mixing in mono. So often I am doubling guitar or vocal harmony tracks and immediately separating into stereo mixes. I think working in mono could be a great way to start.

DonnyThompson Thu, 02/23/2017 - 05:21

I guess I've mainly been a "start with the vocal track first" kinda cooker... that is, of course, if the song has a vocal.
Whenever possible, I like to start with the LV and build the mix around that.

There are many engineers whom I respect who start with drums and bass, and it works well for them - I can totally respect that - but it always seemed to me that getting the vocals sounding good and in place first allowed me to bring up the other tracks in support of the vocal track(s).
( My instructor used to say, "The vocal is the focal!" ). For the most part, it's a method that has served my mixes ( and my clients) fairly well over the years.

Although, I can't say that there haven't been plenty of times where I haven't started with the rhythm section - especially if it's an ongoing project, where perhaps the vocals haven't yet been recorded with a keeper take...
So I certainly can't say "I never start with drums and bass."

FWIW

Nosirrah yrag Thu, 06/08/2017 - 18:43

My mixing is not that good due to lack of experience but, this is how I was taught. After tracking and editing, it's all faders up for a quick static mix. Then I switch over to mono. Go to the loudest/climatic part of the song. I start there, tweaking out the drums, bring in the bass, then lead vocals. I think start bringing in all other parts. I'm still mono, and I'm not working on anything other than volume balance and eq if needed, I then go through the other parts of the song the same way, paying attention to volume (making notes as I go about what I think will be needed when I start into automation). After I get the song at the overall levels I want. I'll start really tweaking things down. Compression, vocal bus treatment, drum bus treatmentBass, I started doing something I was shown, about getting the bass to sit in the mix with small speakers. Bus out/send the bass track to reamp the bass, with a slightly distorted, mid, to mid highs, pushed up quite a bit, low end left alone , or rolled off a tiny bit, blend back into the bass buss to just sit underneath the main bass track.
I try to keep a focus on what the song is giving me and mix in the fx to maintain that image. I pop out of mono start panning, and working with automation to bring elements in, and out of the mix giving it some interest, energy, and working towards the build up( I reference back to Mono often to make sure I'm not washing something out in the mix). I always use reference tracks to help me out since I don't have years and years of experience. Again I almost always start out the climatic/loudest part of the song and mix down from there. This makes volume leveling easier for me for some reason.

Davedog Thu, 06/08/2017 - 19:57

When I track, I usually know where everything is going to sit in the song. Overdubs are like little mini mixing camps for the artists benefits....I make the mix for them at this time as close to finished as it can get.

When I cook someone else's crap it can go so many different ways. Usually it's a long repair process followed by a burst of inspiration and then a phone call.........I digress

When there is a vocal, I have found that getting the bass, drums and the vocalist on the same relative level is all important to the quality of the finish. When your basic groove and the lyrical melody is bangin, everything else is cookies and cream.

JayTerrance Fri, 06/09/2017 - 09:37

How do I Mix Today? Pretty much like the old-school mix engineers...just that I'm attempting to add this Sound Design/Layering knowledge into the whole process. This whole post below is based on the "personal experimentations" that I've been working with lately...so while some of my thoughts might be "out of the box" for some of you; others may have already "been there - done that".

I've read lots of good information in this thread above about the typical "Mix Engineering" process. That is, the tried and true process used for many years (even in the days of tape). But with the plethora of technology tools that exist today and the experimenting I've been doing, I'm beginning to slowly grasp that their may be an additional level that nicely compliments the typical mix engineering phase. And it rarely gets discussed in here.

To compete (or even get close to) some of the professional productions I hear today, I wondered if delving into "Sound Design Engineering" in addition to the typical "Mix Engineering" would be worthwhile. It used to be that Sound Design was an area exclusive for the synth programmers and their layering techniques, but that has expanded to more than just synths today. I'm now working/experimenting with layering (and ADSR) on most everything (but mostly snares, bass, and guitar). In fact, some of the snare sounds I hear today, there is no way to accomplish without a decent understanding of Layering and ADSR. Layering with a frequency block to create a subtle change in tone or layering with a frequency block utilizing just the attack (A of ADSR) on an instrument. How can I get the attack of this instrument to "pop" by utilizing the attack sound of another instrument? or the sound of the same instrument mic'd at a separate angle/distance? Or how do I utilize layering the D;S;R elements as well. It seems that answers to these questions are becoming necessary to compete with the vibrancy, clarity and depth of mixes being produced today. After the "experimenting phase" that I'm going though, I see the line becoming more blurred between sound design engineering and mix engineering.

The more I delve into ADSR and frequency layering, the more I'm starting to look at basic eq'ing as a more drastic/crude way to accomplish something. That is, the old method of me reaching for an eq right away to accomplish some complimentary eqing is looking more and more like "creating a fine furniture with a chainsaw" type of mentality at times. It's still necessary in many cases, but I tend to now think first of how I can make eq corrections on an ADSR type of vision. So...Do I really need to totally eq out 4-6 db's in this frequency range? Or can I just shorten up the sustain(S) of this frequency range for a more subtle (and complimentary) effect? Or do I shorten up the sustain (or conversely reduce the attack) of this frequency range with a sidechain tool like a compressor? (for a little more aggressive cleaning but not yet as aggressive as eq'ing it alone). Or do I use a spectral tool in combination with ADSR? (still not as aggressive as eq'ing it alone).

There's a ton of info out there on layering with the vast majority of it geared towards "dance", but I'm finding it valuable even for the more natural music. It's really a trial and error type of path to go down because certain sounds that you think will layer nicely, just do not work once you put them together. And can drive a person crazy at times. But on the other hand, when I've gotten some layers that do work, it can really "pop". Then I'm spurred on to experiment even more with it. Here's a link to a simple article that is unfortuantely focused on dance but nonetheless, the fundamentals of layering are explained...
http://www.nyonyxx.com/production-help/2015/6/12/sound-design-guide-mixing

And another on Drums and Phase manipulation...
http://www.soundonsound.com/techniques/layers-complexity

DonnyThompson Sat, 06/10/2017 - 04:47

I find myself using parallel compression more these days. The thing is, the process isn't new to me, it's not like I just leatn d how to do it, I used to use it all he time, back when I was cooking on consoles; it was a frequent method for me for many years ... but I got away from it for awhile after I went digital... I can't even give a reason for why I did. But I've gotten back to using it as part of my DAW mixing method.
I don't really have a particular go-to compressor that I use - it depends on what I'm mixing and how I want certain tracks to sound. Sometimes I want an Opto character, other times FET, and sometimes the stock Compressor in Samplitude is perfect; other times, I also dig the Waves R channel's Compressor (probably the only Waves plug I really use anymore as my Waves plugs are older 32 bit, and I never upgraded) but I like it because it's very transparent, and I like to use it when I want to compress something without it sounding like it's being compressed. I'm sure there are others out there that are like that as well, but that's the one I use.
The parallel GR seems to give me more control; and because it resides on its own Aux fader/channel I can adjust EQ and levels more easily. It also allows me to automate more smoothly because I can just grab the Aux fader and pull or push the fader level during the automated mixing.
Although, this brings up another facet of my mixing approach - I really don't automate my mixes much anymore. I prefer to use Samplitude's Object Editing feature instead, which I find to be an awesome feature. I'd never worked with OE before I got Samp. At first it was a bit of a steep learning curve, but as I became more familiar with it and what it could do, I found myself using it a lot more than I do the automation feature.
Anyway, FWIW. :)