The Flaws of Top Down Mixing! Why you should NOT do it!

In today's video I'm talking about the top down mixing method, and why I advocate against it. I know many people and mixing courses advocate for it, but unless you are a seasoned professional and work on similar songs all the time, it is extremely difficult to make it work without running into problems sooner or later.
Multiple levels of compression are very difficult to manage, and without EQs on your individual tracks you risk sending bad sounding audio into your reverbs and delays as well.
Not to mention revisions, which become very complicated as your mix is fully locked in.
#mixing #mixengineer #martinmixing
Visit www.martinmixing.com for masterclasses and collaborations!
00:00 Intro
00:20 Overview of top down mixing
02:35 the complexities of top down mixing
03:13 Top Down mixing forces you into a corner!
03:34 Logistical flaws of Top Down Mixing
04:17 Flaw 1, compression issues with Top Down Mixing
04:57 Your volume fader won't work properly!
05:40 Flaw 2, EQ issues, why it is better to EQ separately
07:34 Flaw 3, Your reverbs and delays will sound bad!
09:12 Flaw 4, Stems and Revisions to the mix will become very difficult to make
10:16 Just use Bottom Up mixing instead
10:40 Outro

[@JesusArmasOficial]

Interesting point of view for top-down mixing.
I am of the kind that thinks that we should use whatever works best for one.
I've been mixing using top-down mixing ever since I started mixing and discovered the method.
Today I work with a combination of the two methods.
Top-down mixing in the initial state of the mix and then work on the individual channels when more in depth while mixing.
I used to have processing on all my busses but not anymore.
My process now goes:
Gain staging session>Summing, bus compression, tape sound on mix bus>dealing with individual channels using channel strips and working on the groups only to do general i.e de-essing vocals, group limiting or saturation/compression but in small increments as it's been added beforehand on both the mix bus and the individual channels.
So all in all, the best is to combine the two.

[@perryyeldham6746]

I agree with your opinion of top down mixing. I don't understand why people say you shouldn't eq while solo'd. Of course you listen to, and eq, an instrument that needs it, both solo'd and not solo'd. Even more importantly, you listen solo'd because now you don't have other instruments masking certain frequencies and hiding problems. Finally, my humble opinion is that if each instrument sounds really good (aka sounds the way you want it to sound) by itself, then getting instruments to blend (which, after all, IS the definition of a mix) is very much easier. Let's face it, that's the way it was done in the old days when they didn't have tons of tracks, tons of channels, and tons of processors - you got the instrument to sound good, you put the right mic on it the right distance away, and besides some compression, that was most of the work done. Now you just have to balance a number of great sounding instruments. Let's not forget that mixing engineers were originally called Balance Engineers!;-)
Of course you can do some things top down and you adjust your mix bus periodically as sub-balances change, but I feel like the way we experience music emotionally should be paralleled by the way we mix - for pop music it's drums and bass, then get the guitars and keys to sit the way we want on top, then get the sweeteners to sit on top of that the way it feels best (horns, strings, percussion, fx), then get the vocals to sit on top of everything else, adjusting what's below as needed. Now we've got a good mix and we use the master bus to sweeten and massage it even more. or orchestral music, it isn't much different - bass and cellos and tympani first, then everything else moving up to the top.

[@BrunodeSouzaLino]

I think an ideal setup is a mix of the two, as both top down and bottom up mixing have the exact same issues, but they're opposites. Taking your EQ example, fixing stuff with EQ whilst your track is soloed means you just lost the context of how that track interacts with the other elements of the mix. Doing this on acoustic drums for example means you're treating it as a drum machine, rather than an instrument which happens to be comprised of several channels of audio that produce that sound when combined. Our ears are pretty good at isolating elements from the mix apart and all that trickery is built into our brains already, so soloing tracks should be reserved to special cases where you need to fix something, rather than a norm.
The way I mix goes as follows:
- Load all the tracks into your DAW of choice
- Rename stuff according to the convention that's best for your workflow
- Look at which elements might need to have the same processing done to them and give them their own bus, which will both serve to get a static mix faster and save on processing. That will usually give one master bus for drums, one for bass, guitars, keys, vocals and so on. These can be composed of a mixture of subgroups or all tracks of the same "type" routed to said bus. This helps in situations where you need to lower just the toms without changing individual volumes for them and similar things.
- Once all that's set, you put all faders except for the master on zero and start bringing them up one by one and balancing them until they sit well. Panning is optional.
- The previous step can be done element by element of the whole mix at once. If I decide to do the latter, I don't touch vocals until I have the instrumental portion right.
- Once the volumes are "correct," start adding your processing to each track as necessary without soloing any tracks unless absolutely necessary. You want to listen to how that processing sounds in the context of the whole mix. Adjust volumes as necessary.
- Once all that is done, then you'll have a good idea as to which tracks need compression and which ones don't, as well as necessary saturation and other elements to enhance stuff.
- That should give you a pretty dry mix that's hopefully gained correct to the point you don't have to add compression or limiting to the master bus to catch overloaded peaks or transients. At this point, you should bring the vocals and start strategizing how to fit them in the mix using whatever processing is necessary. Don't touch any reverb or delay at this point.
- After the vocals look like they're seating well, start adding your reverbs and delays.
This approach is neither bottom up or top down, but shares elements of both and hopefully uses the advantages of each, on top of making easy to make it more top down or more bottom up if necessary. Why use one tool or the other when you can use both?

[@channelmames]

Why are you saying it’s not top down if you clean it up and get the project ready first? It’s not against the rules and it’s still top down if you get the project set first😂You can do an initial balance/ remove mud and obvious issues first in the editing process before thinking about top down. There are tons of great reasons to work topdown, especially saving cpu. Might use automation differently, but ultimately should have to use less of it.

[@martinmixing]

You're right, it is definitely not against the rules :) That's how I did it in the past as well. But then you're still left with the send/reverb/delay issues, plus the possible complications with making future adjustments, stems, and revisions of more complex nature, especially once you include compression on multiple busses. CPU should not be an issue these days with any mixing method though.180 tracks bottom up mixes with a 4 year old mac with no glitches at all.