I always say "Spice is the spice of life" and that variety helps flavor our experiences, especially in audio production.
I receive many recordings into Eclectica Studios for mixing that have good intentions sprinkled all over them, but usually these good intentions do not work out as the original recording engineer had intended or hoped for.
An example would be an acoustic guitar that was tracked once, panned Left, then the audio was copied to a second track and panned Right. This second track is then delayed or moved slightly later in time to make a "Stereo sounding" part... ...Or another would be when 10 tracks of electric guitars are recorded to make the "Wall of sound" and are done by just tracking the same thing over and over again...
In the first example this is simply the same guitar part with one side coming out of the speaker a few millisecond later than the other speaker. This does not cause the illusion of 2 guitar players playing, but simply creates panning to one side when listened to through headphones. When listening with speakers, if you are anywhere but exactly in the center of the two speakers, it sounds mono and tunnel like.
The second example is even worst, if you track the same guitar player playing the same guitar and parts 10 times, you actually get a smaller sound then just two properly done takes. This is because the sound starts cancelling itself out after the first few tracks are layered.
That bring us to the topic of variations.
It's always good to layer parts that warrant it, but you do not want to simply record a part the same way twice. Lets take the electric guitar example. I find the best way to track 2 guitar parts and get them to sound big and gorgeous together is to first dial in the guitar and amp to the sound you want, mic it and track it. Then play the part again as close as possible to the way you performed it the first time and simply hard pan these take, Left/Right.
For this second track though, you want to vary the elements as much as possible; Different guitar, different amp, or maybe different speaker cabinet, different microphone, even a different preamp for the mic, any change will help.
When you swap out an element in the recording chain, even just one of the above listed elements, it changes the spectrum and tones, even the timbre of the part and therefore sits together with the first part beautifully.
The frequency spectrum of the two parts now compliment each other rather than cancel out each other due to similarities in frequencies. The more variation, the bigger the combine sound is.
When tracking acoustic guitars, I always use 2 different guitars, but often use the same ADK A51TC mic into my MANLEY preamp. Since the guitar is the main element, for me its enough to swap that out. I also usually use my solid maple Guild Dreadnought for the first take and then my Gibson J100 Extra jumbo body or even my Oscar Schmidt wood bodied Dobro for the second take. I also will capo one of the guitars at the 4th, 5th or 7th fret to make sure I play different chordal registries on the one of the takes.
When a vocalist wants to double up a part, depending on the style, I will sometimes change the mic that we are using. I may have them stand back from the mic or move closer, just to vary the tone by way of the proximity affect of the mic. Sometime turning their head to face off angle with the mic to get more 'room' sound into the take. Again, any and all variations will help because the same voice tracking the same part twice, when they come together in a mix you get a thinning out of the sound from some phase cancelling, and it's usually the lower tones. You end up with a thinner sound, while all along you were trying to make it thicker.
These same kind of principles are for mixing too. Last night I mixed a track that had 2 electric guitar rhythm parts, panned Left and Right, but they had the EXACT same sound from using the same patch on a Line6 POD.
To counter this I sent one through a UAD NEVE 1081 EQ and the other into a UAD API Channel. First I solo'd and then centered the Left guitar to hear it independently. I open the 1081 EQ and emphasis the Hi-mids and lows around 100hz. Then moved it back to Left and unsolo'd it. Next I solo'd and panned center the Right guitar and with the API Channel I High Passed at 100Hz, and dialed in more Mids around 1KHz. I then panned it Right again and unsolo'd. Now when played back together, the two guitar tracks sound fuller and compliment each other.
A key here is to use not only two different EQ settings but also 2 different EQs. These two modeled after their hardware counterparts that have their own personality and ways of affecting the sound.
I have seen many people using the same Channel strip across their entire mix, example would be the UAD API channel strip on every channel in in a mix in order to get "That API sound". It looks good on your screen and gives you that "I am working on a console" vibe, but even though each track can be affected in a manor that is like an API console, a very big ingredient is missing; Variations in the sound from channel to channel that the hardware console exhibits.
For instance...24 channels of a real API Vision console will have 24 channels that sound basically the same, but not exactly the same due to component within the channel stip. These little variation make the overall tone and sound blend together and fill in. I have a friend who works on a SSL 4000 console and he knows which channels sound best for certain things; "Channel 6 sounds best for snare, 17 and 21 for electric guitar..." This is because as the components age and change, the sound changes too.
This is one place where Plugins are not yet emulating the hardware: Unit to unit Variations.
Great River EQ-2NV to my master buss for mixing. Its fed by a Mytek D/A and capture with an Apogee A/D. Beyond the wonderful sounding EQ curve I get from the EQ-2NV, the sound stage or stereo image and depth are all enhanced and affected greatly. The 2 channels are discrete and therefore slightly different I am guessing, but I am sure both channels are dialed in slightly different from each other too, do to the setting of the EQ knobs on each channel by hand with no automatic recall. Just dialing them in to the best of my ability.
Regardless of what contributes to it, the variations between the Left and Right create something wonderful! It makes me wonder if any of my plugins, while running in stereo, are not simply the same mono channel that was modeled being doubled to make the two channels for a stereo instance?
My UAD MANLEY Massive Passive plugin sounds great as an EQ but does not affect the depth and stereo stage even 20% of what the Hardware Great River does, could this be why? I may research how UA models stereo units if I get a chance.
Regardless, there is a wonderful sound pallet that comes from the imperfections of analog gear.
So there you have it. Switch it up, use different elements in your productions, and then send me your tracks to Mix and master :-)