Thread: Planning a mix: whats your thought process?

That's a pretty good list. I find that it is best to consider most of these concerns before tracking begins. For example, whether a guitar will comprise the main rhythmic & harmonic function in the song vs. if it will be but one of two or more such functioning elements will perhaps dictate whether you track the guitar in stereo or mono. Same goes for other instruments.

If you're building up a song by overdubbing, try to get the rhythm section bass, rhythm guitar, drums) down first and preferably playing all together. Get a good preliminary balance between these elements, it'll make the rest of your job fall into place more easily.

Lastly, don't overthink this! Trust your ears and your gut. Apply knowledge according to what they tell you, not just because you have it. Become intuitive in the process, just like when one becomes truly proficient playing an instrument.

Hope this helps!

Jeff

I respectfully disagree with Jeff's initial comment... considering that you mention that you're mixing, and make no mention of tracking AND mixing.

If you're tracking AND mixing... the yes, getting those bed tracks captured live is usually a better way than the battle of trying to make those tracks sound live - IF the genre benefits from it. Some do, some don't, and some is doesn't matter.

Which brings up the matter of whether you are the producer, or not. If you are NOT the producer, you have to be cautious of stepping into making production calls that may not meet with the producer's vision.

While your list might work for one song or 200 songs (over the course of a career, that is), it certainly has not been my experience that ANY formula works all the time, if you are truly doing your job as a mix engineer - and doing what's best for the song.

About the ONLY "formula" that I've found that actually works – talk to the band and producer, to get a feel for their personality, then get a list of music from both - that they feel that their music is "closest to" - either in style, genre, feel, etc.... including what they think they want individual songs to sound like.

Once I have that influence firmly in my head, I'll determine the most important aspect of each song as a collection... whether it's going to be vocals or some other instrument. (You can't approach a jazz ensemble-sans vox, the same way you would a 40 voice choir with a 12 piece band behind them, nor a 3 piece power rock band.)

The first thing I generally do with the tracks is to pull everything up broomstick and listen. I don't listen loud, nor do I listen quiet... something midlevel that I can easily talk over.

The song will tell you where to start. I might start with the vocals and start pulling other things down around them, or I might start with something else and build up to the vocals... each song is different and should get individual attention based upon the song itself... not some prefab construct.

I ALWAYS have paper and pencil handy... I may make notes before I start, I may not make a single note on that paper for the entire project - maybe only putting notes in the comments section of the tracks. Every song is different.

I might get an immediate impression for some instrument's importance, placement in the sound field, etc. But I try to make it a rule to NEVER marry to that impression, unless it's overtly obvious.

If the project is supposed to have a "consistent sound", I'm more inclined to strictly make notes on the first pass listen - until I've reviewed all the songs.

Again, I'll maybe listen 2-3 times to the songs before I do anything - or I might start pushing things around immediately.... there isn't a "formula" to go by... except the one that applies to the work as a whole - and that only comes after you've had a chance to listen to what each song is telling you it needs.

One thing I do, is to map out the songs into sections... Verse, chorus, solo, intro, outro, bridge, etc... but if there is a special point of interest... I don't hesitate to drop a marker at those points. That way, I won't miss it when I go back to start tweaking. Even if it's a long note like; "Pull bass and duck kik w/vox", etc... Mapping the song is more efficient usage of your time, IMHO. But it's not anything that you can really do ahead of the game - it's more of a real time thing that your gut and your heart will tell your brain, if you listen to the song.

My humble $.02 worth.

that's like a dollars worth

It sounds like you have a set of tunes already tracked and somewhat mixed.
I think making a layout and record of each song either on paper or Word doc is a great idea!
Especially if your looking to create a flow or consistency throughout your album.

I would create a page for each song, list details of anything that strikes you while your listening to it.
You could certainly add any technical details as far as you want. It a great way to recall settings too!
The amount of detail and the time you want to put into it is up to you. Anal retentive comes to mind though so...

The mixing part is between your ears and in your head. So do what sounds good and flows with you.
You should be focusing on creativity and spontaneity not trying to follow notes or have some preconceived notion as to what you "think" it should sound like. Trying to plan that part will just suck all the creativity out...
Once you've got what you want...then update your song page with your thoughts.
Then review your notes/thoughts afterwards and then make subtle changes.

Your approach is sound, but Max points out a critical element - everything is different, depending on circumstances.

I'll simply reiterate his point that sometimes the best starting point is pulling raw tracks up one by one and hearing what you've got.
Sometimes a song (and it's tracks) will demand attention to the vocal, the kick, the guitar, etc.
Your mapping is a good idea in the sense that you can track what does and doesn't work, to some degree.
At least when listening back to an old project, you'll have notes to help your memory.

For me, anything musical comes down to this:
Having as great a library of skills as possible, knowing when to use certain skills/tricks, and using your ear to tell you when the data is lying to you.

After re-reading your post again, I spotted a couple of things I missed... all good points/ideas you expressed - yet again, in what may, or may not be the correct order...

[/quote]
1. "Ask the producer". Define sound priorities and wanted results.
2. Song Structure analysis: parts and intense moments
3. Density of arrangement: How many instruments and layers are there in the rythmic and melodic sections?
4. Define priority of elements in step 3. Divide them in "background support", "lead" or "interest queues"
5. Amplitude map: Which elements should be louder. Define dynamic priorities
6. Frequency map: Define the frequency response of each element. Not by instrument, but by element. For instance, guitar riff between 100-400Hz. Group elements in "frequency bands".
7. Stereo Image: The X axis. Make a "panning map" to define where will the elements "live" in the stereo image.
8. Depth: The Z axis. Define which elements should be closer and which farther
9. Song Structure implementation: Repeat steps 5-8 for every section of the song and plan transitions between each section.[/quote]


A couple of real life observations;
1. In sparse track counts, it's imperative that everything has a place in the sound field... to fill it in... unless it's not the correct treatment for that song.
2. In large track counts, it's imperative that everything has a place in the sound field... to fill it in... unless it's not the correct treatment for that song.

In GENERAL... I start with basic panning to get stuff where it belongs/needs to be to get out of the way of other things. Then I'll attack the sounds/instruments to either corral the frequency spectrum, or open up the spectrum of that "voice". The whole time, I prefer to keep an eye on the overall 2 buss, and periodically adjust levels up/down to attend to the amplitude of what should be dominant in what order. In essence, 7/8 then 6 and 5.

THEN and only then, do I worry about your #2 through #4... I can't refine what I don't know is there.

Is the kik fighting with the bass? Are guitars covering up vocals? Is the Hammond crashing the party that the horns are having? Is that tuba under powered? Are the cello's masking the acoustic guitar? Does that Strat sound like a cat that has it's tail caught in a door?

So, I generally try to shape the voices first, then worry about bringing the automation of the song structure into play.