How to Mix A Song

[JonPorobil]

So, rather than snarking at the newb, I'll try to be productive by listing here some of the lessons that I've learned from my reading these last two weeks or so, which were absent from HeadShot's original answer. Most of my trouble, by the way, came from Step 3:

3. If you have each track seperately with as little noise as possible, and no mistakes in the recording, all that's left to work on is deciding for yourself what should be louder than anything else. for example: if you can't hear the drums, they're either not loud enough or everything else is too loud. You want to be able to hear everything clearly with nothing overpowering anything else. This can be tricky with vocals, as often your vocals will have to be louder than everything else in order to be able to make out the lyrics clearly.

For instance, I've learned that it helps to layer tracks one at a time. I mixed my "Amsterdam" in the following order: Drums, bass, guitar, tambourine, handclaps, hand drums, cowbell, backup vocals. (Even so, I managed to mix the bass way too low, which proves, if nothing else, that Step 3 isn't nearly as easy as it sounds.)

I've learned a bit about judicious use of panning (completely absent from HeadShot's discussion): bass and lead vocal need not be panned, for instance, but in general, nothing else should be centered. If two instruments play into each other, or seem to be "having a conversation," one might pan them opposite each other for best effect.

I've learned a trick for mixing vocals: lower the volume of your monitors as low as you can, and make sure the vocals are audible and understandable even on that low volume; some people will be cranking the volume, but most will be listening at more reasonable levels, and you need to make sure your vocal sits right in the track no matter what the master volume is.

Things I'm still really fuzzy on, and would like to read up further as I experiment:

EQ - when to use it, when NOT to use it, and how.
Reverb - I still need to find that "sweet spot" with reverb, where it doesn't sound like it was recorded in a tin can, but neither does the track sound too muddy.
Other delay effects - Is it even possible to use effects like the phaser or flanger without sounding completely over-the-top gimmicky?
Compression - How to use it without completely destroying the dynamic quality of the track. Also, what also the terminology on my plugin really means in terms of recording.
Mastering - Gosh, this ought to be a whole other thread.

[irwin]

For instance, I've learned that it helps to layer tracks one at a time. I mixed my "Amsterdam" in the following order: Drums, bass, guitar, tambourine, handclaps, hand drums, cowbell, backup vocals. (Even so, I managed to mix the bass way too low, which proves, if nothing else, that Step 3 isn't nearly as easy as it sounds.)

This is good advice. I always try to mix the rhythm section first (drums/bass), followed by whatever is the
primary instrument .

I've learned a trick for mixing vocals: lower the volume of your monitors as low as you can, and make sure the vocals are audible and understandable even on that low volume; some people will be cranking the volume, but most will be listening at more reasonable levels, and you need to make sure your vocal sits right in the track no matter what the master volume is.

This is also really good advice. Always check your mix at a range of volumes to make sure the important tracks (largely vocals) are still clear enough. Remember also that by the time you get to mixdown, while you know the words to your song (after doing eleventy bajillion takes and mixing passes), the listener does not. So err on the side of having them be more prominent than you think.

Also, check your mix on a number of different speakers/headphones.

Finally, always re-check your mix after a break. Listening with fresh ears will often reveal things you hadn't noticed before.

Things I'm still really fuzzy on, and would like to read up further as I experiment:

EQ - when to use it, when NOT to use it, and how.

That's a whole 'nother thread.

Reverb - I still need to find that "sweet spot" with reverb, where it doesn't sound like it was recorded in a tin can, but neither does the track sound too muddy.

Here's why I try to do (key word: try): Turn up the reverb until you can hear it and then back it off slightly. The idea is that reverb should be felt, not heard. It's there because the world is not an anechoic chamber, so your brain is used to hearing things this way. If the song is too dry, it sounds out of place because of this. But- because your brain is used to hearing it everywhere (and using it for locational cues), it will be filtered out before you consciously "hear" it. So the idea (unless you are aiming for an effect) is to hit that spot where it feels natural, but you don't actually hear it.

Other delay effects - Is it even possible to use effects like the phaser or flanger without sounding completely over-the-top gimmicky?

Subtle stereo delay and flange effects can be used to make tracks stand out in the mix without making them louder.

[Reist]

This is good advice. I always try to mix the rhythm section first (drums/bass), followed by whatever is the
primary instrument .

Word to that - I always start with bass. I don't do drums in one fell swoop though.

I set up the bass, then eq the kick/snare/toms. After that, I let the song play as I mix in those drum tracks. Then I eq and add in the guitar to whatever level I feel is appropriate. If I have keyboard, I add it in after that. At this point, I add in the overhead drum tracks - they're not really required for finding the levels of the other tracks, but it seems to be important to have them in the mix for getting vox levels.

[Reist]

EQ - when to use it, when NOT to use it, and how.
Reverb - I still need to find that "sweet spot" with reverb, where it doesn't sound like it was recorded in a tin can, but neither does the track sound too muddy.
Other delay effects - Is it even possible to use effects like the phaser or flanger without sounding completely over-the-top gimmicky?
Compression - How to use it without completely destroying the dynamic quality of the track. Also, what also the terminology on my plugin really means in terms of recording.
Mastering - Gosh, this ought to be a whole other thread.

Man, those are solid questions. I realize each one should probably have its own thread, but I'll give some quick tips (although I'm sure I'm not qualified to do so).

EQ - I use EQ on every track I mix (largely because I don't have the capabilities to get exactly the sound I want). I use ReaFir (free download) to EQ, since it has a really easy learning curve and a visual theme so it's easy to mess around with it and figure stuff out yourself. If a track feels overly bass heavy, boomy or muddy, I drop the bass section down 10 or 20 dB. Don't make the tracks sound tinny though - drop the bass just enough to retain the "realness" of the track while eliminating unwanted muddiness in the lower frequencies. This way each track can fit into its own frequency section (I don't know the terminology for that, obviously), and the overall mix will feel more spacious at the same time. I up the treble lots on kick, snare and toms since my current recording setup leaves them feeling dead - treble from 5k up to about 11k can really liven up dead sounding drums. With vox, I make a small curve, starting in the midrange of frequencies up to about 5 dB in the 11k range - the more high frequencies vox has, the more it stands out in the mix. I've been noticing more and more that professional mixes have the bgvs mixed with less treble so they don't impose on the lead vox, so I'm keeping that in mind for my next mix. I guess the key with EQ is (as far as I know) - each instrument has its own place and frequency range - it shouldn't be muddying up space where it doesn't belong. Bass shouldn't have a strong presence in the treble range, and vox and guitars shouldn't be too bass-heavy. Don't up the treble so much it sounds fake or burns your eardrums, and don't drop the bass so much that the track sounds totally wimpy.

Reverb - I don't know what to say - I have the same problem as you. I'll say this though - if a reverb effect can't help but make a track sound tinny, then it's probably not a great reverb effect. I've used some pretty good ones, and they sound great at any level. I'm back to using free reverb effects though now because my new computer won't accept the good effects I was using before, so I'm sure my stuff'll sound like crap again now.

Other delay effects - Using just a bit of delay and flange on backing tracks can beef out the song and give the song a more creative/experimental feel, but I'm always worried about using the those effects on anything else. I've heard some pretty good songs use flange on the snare drum during fills to give it a futuristic, almost electronic sound, but I've never tried that myself.

Compression - I use compression sparingly on vox, guitars, bass and keyboards, as they really control the dynamics of the song. I compress the crap out of drums, as it's not the volume of the drums that controls its dynamics, but more the technique and the different cymbals being used. For all my compression needs, I use Classic Compressor (Free Download), which (like ReaFir) is really easy to learn and use. Really, if you don't want to lose the dynamics of the song, don't kill the vox, guitars, bass and keyboards in the mix, and be careful while mastering. Don't worry about compressing drums, they don't need giant jumps in volume anyways (but make sure they keep a little pop and bang - otherwise they might as well not be there).

Mastering - For all my mastering needs, I use ReaFir, Classic Compressor, and GClip (all free downloads), in that order. I up the treble about 5 dB until 11k in Reafir (to brighten up the sound, usually my mixes sound a bit muffled before mastering), then compress it in Classic Compressor down to about -24 at 2:5 (the threshold usually varies from song to song depending on what you want, but I never master using higher than 2:5), then pump up the volume using a combination of Classic Compressor and GClip. That's my formula for pretty much every song I've mixed on a computer, but I still experiment with those plugins and sometimes with adding a touch of reverb on the final mix. That's what I used on my last Nur Ein song, and it felt pretty huge in the end, so I think it might be a good call.

Hope that at least helps someone. If not, it only took me like half an hour so at least I haven't wasted my whole weekend.

Now for me to ask a question - does anyone know how to use a de-esser properly? All my songs have burning s's, and I can't figure out how to fix it.

[JonPorobil]

Now for me to ask a question - does anyone know how to use a de-esser properly? All my songs have burning s's, and I can't figure out how to fix it.

Well obviously, stand further from the mic when recording vocals.

But what I normally do, if rerecording the vocals is absolutely out of the question for whatever reason, is spot-check it with a spectral analyzer. If the sibilants are all that hot, they'll be obvious on the graph; just highlight the bright spots and reduce by 5-10db. Maybe not the absolute most precise way of going about it, but I've gotten away with it quite a few times.

[Rabid Garfunkel]

...and, since we know you've got the volume in your lungs, reposition the mike! Lower it so you're singing over the top of it, or move it to the side and angle it so it's pointing at your mouth and sing past it.

Eliminate the straight line from mouth to microphone capsule... making deep throat kissyface with the mike onstage is okay, but in the studio it just ain't a good thing

[JonPorobil]

making deep throat kissyface with the mike onstage is okay, but in the studio it just ain't a good thing

I've always been curious as to why that's "okay" to do live, but not in a recording.

[jast]

I've always been curious as to why that's "okay" to do live, but not in a recording.

One thing to consider is the type of mic you've got. Condensers are much more sensitive than the dynamic mics typically used on stage, so you'll usually have to be more careful with them. Generally, studio mics are more geared towards producing sound with as many nuances as possible, whereas the design of stage mics is focussed on ruggedness, reduction of handling noises, etc. etc. I think you'll often find stage mics to be somewhat less bright than studio mics (ribbon mics being an obvious exception).

[JonPorobil]

I've always been curious as to why that's "okay" to do live, but not in a recording.

One thing to consider is the type of mic you've got. Condensers are much more sensitive than the dynamic mics typically used on stage, so you'll usually have to be more careful with them. Generally, studio mics are more geared towards producing sound with as many nuances as possible, whereas the design of stage mics is focussed on ruggedness, reduction of handling noises, etc. etc. I think you'll often find stage mics to be somewhat less bright than studio mics (ribbon mics being an obvious exception).

Yeah, but I used to record all my vocals with dynamic microphones (and still do every once in a while with the Shure SM58), and even then, you can't get close to the mic like you could performing live. What's up with that?

[jeff robertson]

Yeah, but I used to record all my vocals with dynamic microphones (and still do every once in a while with the Shure SM58), and even then, you can't get close to the mic like you could performing live. What's up with that?

Because when you're singing live, nobody notices things like sibilant esses over all the other loud noises going on?

[jeff robertson]

I use Classic Compressor (Free Download)

Just out of curiosity, what led you to choose Classic Compressor over any of the several compressors that were already bundled with Reaper?

[Reist]

I use Classic Compressor (Free Download)

Just out of curiosity, what led you to choose Classic Compressor over any of the several compressors that were already bundled with Reaper?

Ease of use and simple to learn. The labelled knobs really helped me figure out the basics of compression, while ReaComp's user interface was a lot more confusing to me (as I was pretty much teaching myself). Once I figured out Classic Compressor, I came to realize that it works really well and I just stuck with it.

[Rabid Garfunkel]

Yeah, but I used to record all my vocals with dynamic microphones (and still do every once in a while with the Shure SM58), and even then, you can't get close to the mic like you could performing live. What's up with that?

Two words... Sound Guy

Have you tried running two mikes at the same time while tracking vocals? One condenser, one dynamic? Extra options with a single take, heh...

[HeadShot]

Have you tried running two mikes at the same time while tracking vocals? One condenser, one dynamic? Extra options with a single take, heh...

Anyone who ever did stage crew in high school had to know how to manage up to and more than 15 condensers at one time, for the musicals. Let me tell you, that was a bitch.

[Eric Y.]

Your drama club did not have body mics? You used a combination of "up to and more than 15" overheads and PZMs?

I can see how that would have been difficult. But certainly that would not have been the experience of "anyone who ever did stage crew in high school."

(Sorry for continuing to be off-topic, mixologists.)

[HeadShot]

They were face mics, yeah, but they were condensers, each with their own channel and unique calibration

[Rabid Garfunkel]

Y'all had microphones? Pffft. Kids these days...