Jump to content

Where Do Instruments Sit In The Mix?


Player99

Recommended Posts

  • Members

There are also other things that are crucial to our perception of depth and distance from a sound source. Indoors, the most crucial are probably early reflections. Their level, as well as other aspects of the early reflections (length, spacing, time before the onset of reverberation, etc.) give our ears / brain clues as to the character and size of the acoustical space that the instrument (or other sound source) is located within, as well as its placement within the room relative to the listener's position.

 

And that is indeed where some of the art of mic placement comes in. Also, it means that it's a good idea to have some plan that you're shooting for before you start recording. If you decide that you want to make a record where the drums are up close, you can mic accordingly and it'll be much more convincing than if you mic it up in some random way and try to change the sound later in the mixing stage. ;)

 

 

As far as panning, it is indeed whatever works for you and your aesthetic! What works for me is what I alluded to at the top of this post - Hard Left, Centre, or Hard Right for important mix elements. That's because I'm shooting for big sounding mixes. If I want something more subtle and mashed together I'll soft pan more. But what I find with hard panned/ centre mixes is that you can get powerful sounding mixes, but still have space in the soft panned areas for more subtle elements that might get lost on an equally busy mix where everything was soft panned.

Link to comment
Share on other sites

  • Members

I appreciate the wisdom almost as much as the wise-assing.

 

The mic placement is interesting but for what I am up to now I am not sure if I can use it. My drums are vst Superior Drummer. It does have a mixer and mic options, but I doubt I can do any better than the presets at this time. My bass and all other instruments are recorded direct using a Roland GR-55 guitar synth. My guitar is my guitar-pedals into amp. I am micing the amp with a sm57. I could experiment with a 2nd mic, but for now the 57 is working out pretty good.

 

Anything else to help with mixing?

Link to comment
Share on other sites

  • Members

 

It's completely up to you, there are no rules except ones you impose on yourself.
If you listen to Radiohead's early albums they conform to the 'thom yorke's head' formula; thom's guitar & vocals, and the drums and bass are centered, with whatever johnny's doing panned left and whatever ed's doing panned right. It's a nice and simple way of quickly getting a mix that approximates the live sound.

 

 

This!

Link to comment
Share on other sites

  • Members

 

I appreciate the wisdom almost as much as the wise-assing.


The mic placement is interesting but for what I am up to now I am not sure if I can use it. My drums are vst Superior Drummer. It does have a mixer and mic options, but I doubt I can do any better than the presets at this time. My bass and all other instruments are recorded direct using a Roland GR-55 guitar synth. My guitar is my guitar-pedals into amp. I am micing the amp with a sm57. I could experiment with a 2nd mic, but for now the 57 is working out pretty good.


Anything else to help with mixing?

 

 

Automation can really liven up a mix a bit as far as instrument placement goes, like a lead guitar part might come back a bit while a singer is singing, then take centre stage when they're not, that kind of thing.

Link to comment
Share on other sites

  • Members

Anything else to help with mixing?

 

Get the best monitors you can afford. Improve the room you're mixing in through a combination of sound absorbing materials (foam is only good for high end, DIY yourself some rockwool sheets to control lower stuff, make corner traps to improve bass) and sound reflecting materials (diffuser panels - google is your friend here). Once you've done that, you can trust what you're hearing and from that point it's all about practice, using your ears, and learning how to use various tools to get from what you hear to what you want to hear.

 

The problem with forum advice in mixing is that it's all guesswork if you can't actually sit there in the room and hear the mix. I could say something like "I always compress the drum tracks together as a group" but how does that help you if the drum sounds you choose are different to mine, and you're aiming for a different end result to me? Or "Long reverb sounds better on slower, sparser songs". Well, that might work a lot of the time but sometimes it won't. Hope you don't think this is wise-assing. ;)

 

Some general things that definitely helped me get better at mixing: (i'm still {censored} but you should hear where I started from! :eek: )

 

- Have some idea of the sound you ultimately want right from the start of the project.

 

- Before you start messing with panning, compressors, eq, effects in general, do a mono fader only mix that you think best represents the song. Maybe the guitars want to be prominent, maybe they want to be buried. Maybe the drum sound should be mostly overheads, maybe it should be close mics with overheads only used to fill in the missing pieces. DON'T make the mistake of thinking a good mix is a balanced mix where everything is equally audible. It's not. Emphasise some things, reduce others to a supporting role. The only arbitrator here is your own taste in what you think makes the song stronger.

 

- Once you've got the fader only mix as good as you think it'll be, get cracking. Break out the effects, do whatever you need to do to get the sound to where you want it to be. Don't make the mistake of doing something for the sake of doing it - a very common mistake is automatically EQ'ing + Compressing ever channel. Listen and only do what NEEDS to be done. Take drums for example - often a good balance of the individual drum tracks going into one stereo compressor will be quicker, easier to manage and sound better than separate compression on each individual track.

 

and that links on to my next point -

 

-Keep it simple. Once you've got some internal balances sorted, make group tracks - eg. All guitars going to one stereo track, all drum tracks to another, all backing vocals to another... you get the idea. That way further down the line when you realise the guitars are a little louder than you want, it's a one second job to fix that with one fader. This is important because the longer something takes, the more likely you are to forget why you're actually doing it, lose your perspective of the overall mix, and start to lose faith in yourself.

 

- Act quickly. Listen, decide, do. Don't get bogged down with indecision.

 

- Unbalanced mixes can be fun. Don't be afraid of them. And this can be unbalanced in volume, panning, or depth.

 

Some less general advice;

 

- Don't spend much time listening to tracks solo'd. You'll lose perspective. Solo tracks to identify problems you heard in the mix, then unsolo them while you're actually fixing the problem because you need to know what effect your action is having on the whole mix.

 

- Kick and Bass are the foundation of the mix. Try to pick sounds that work together. If you haven't managed that, EQ them apart. Warm, wooly bass? Try a more defined kick sound and vice versa.

 

- Set up some gentle compression on the mix bus once you've got your fader only mono mix and mix through it. You'll use less individual compression, and the mix may end up more cohesive. Of course, it might not work, you might not find the right settings, but it's all trial and error and ultimately experience.

Link to comment
Share on other sites

  • Members

 

Automation can really liven up a mix a bit as far as instrument placement goes, like a lead guitar part might come back a bit while a singer is singing, then take centre stage when they're not, that kind of thing.

 

 

Yep, very good point. I'd say a lot of the time Automation is pretty much essential to getting a good exciting mix. Even before there was automation they'd have people adjusting things on the mixing desk manually while they recorded a mixdown. Something else they'd do is set up the mix so it sounds good at the start, then record a mixdown until they heard something that needed to change - say the opening mix sounded great until the verse ended, at which point the backing vocals needed to come up and a guitar needed to go down. They'd record the mix up to that point, stop the tape, set the desk up for the next section, record that, and edit the two parts together.

 

That way can work quite well with modern computer recording - if you set up the mix so the opening sounds great, then press play, you can go through the song, automating bits as they need automating, starting again from the top, making sure the automation you've drawn in does work and doesn't throw adjacent sections out of kilter...

 

The key in this is to be excited about the music so that you're on the same page as it. If you don't like it or care about it, you won't be able to work out how to make it better.

Link to comment
Share on other sites

  • Members

With EQ, don't over think it. If something's overbearing, cut it. If it's lacking something, boost it. Sometimes there will be nothing to boost, then you've got to find another way round the problem.

 

Try to EQ without looking at the pretty graphics showing you a pretty EQ curve. It's easy to think of that as the shape of the noise. It's not. If you were mixing different tracks of white noise it might be. But you're mixing complex signals that already have a unique shape which changes constantly. If possible, hide the pretty graphics while EQing and only use your ears to judge whether what you're doing is right or wrong. If your EQ change makes one part of the track sound better but another part sound worse, either automate the EQ section by section, split the track in two and process each part differently, or work out what else is going wrong to make that happen - does your volume balance make sense? So much that we think is a problem of processing is just a volume problem that you've lost perspective on.

 

Try to maintain perspective. Don't EQ everything really bright or really bassy. Current internet wisdom is that you should high pass most things, I personally don't think you need to if the recording is half decent - I'm more likely to use a low shelf to control the bottom end without throwing away that bass information entirely.

Link to comment
Share on other sites

  • Members

Compression.

 

 

Fuuuuuuuck.

 

 

Just turn the knobs until it sounds good. And the difference between helping the mix and totally ruining it can be pretty subtle, so all you can do is really listen to what it's doing. And that means just playing with compressors enough that you know what to listen for.

 

Don't compress something because COMPRESSOR. Compress it because it actually sounds liek it needs compressing - telltale signs might be that a track won't sit right in the mix, maybe you can't get it sounding powerful enough without it being stupidly loud, maybe it just feels wrong.

 

You can use compression for different things. Use it with longer attack to bring out the punchy start of a note or drum hit. Use it with a shorter attack to reduce the volume of that note start and make the sustain seem louder. Adjust the threshold to decide the point at which the compressor starts doing it's thang. Adjust the ratio 'til it sounds good. The Release is the hardest part to get right for me - it's a subtle thing. If the release is right, the track will just feel right - it'll become part of the groove. If it's wrong, it'll fight the overall feel of the playing, it'll push when it should be pulling... sometimes you won't know what's wrong, but it won't feel right.

 

Adjust the makeup gain so the volume is the same in bypass. Louder things always sound better, so if you've set the compressor up terrible but it's boosting the volume, you won't be able to properly judge whether what you've done is good or bad. This is true of most processing, and definitely do the same for EQ as well!

 

DONT use compression to sort out volume fluctuations - EG a vocal that is too quiet in the verses and too loud in the chorus, or that in inconsistent from line to line. Use automation to fix that sort of problem. A compressor's good for changing the envelope of notes - attack, sustain and release. Not for doing the mixing for you. ;)

 

Edit: something worth mentioning; having used some cheap hardware compressors (RNP, RNLA, TLA-50), I think there's something about hardware compression that i can't get from plugins. Hard to describe.

Link to comment
Share on other sites

My drums are vst Superior Drummer. It does have a mixer and mic options, but I doubt I can do any better than the presets at this time.

 

Some of the virtual drum plugins (such as Strike and BFD) have the ability to adjust the balance from the close mikes, overheads, and room mikes. This does give you some control over the amount of "distance" you hear. It's worth looking into - if your drum VSTi has a separate control for the room mikes, try raising that up and listen to the difference it gives in terms of a sense of distance - and remember, depth is distance. :)

 

My bass and all other instruments are recorded direct using a Roland GR-55 guitar synth.

 

The world's your oyster. You can modify and process these (especially the synth stuff) to your heart's content with plugins in the DAW to taste. Bass generally doesn't need reverb, delay is not often used either, but just a touch of something - a tiny bit of a ambience or room setting on a verb can make it sound more like it was in the room with you, and I sometimes prefer that rather than the audible giveaway of a dead-dry-DI.

 

My guitar is my guitar-pedals into amp. I am micing the amp with a sm57. I could experiment with a 2nd mic, but for now the 57 is working out pretty good.

 

Nothing wrong with the SM57 on a guitar amp. Zillions of tracks have been waxed that way. One reason why this is probably working out well for you is it's one of the few (if not the only) element where you're utilizing an actual mic and moving some air and diaphragms. :) Nothing wrong with trying another mic later too if you'd like. If you'd like to occasionally record an acoustic, or track some vocals too, get a condenser. If you're primarily working with guitar amps, I'd suggest getting a ribbon mic.

 

Anything else to help with mixing?

 

What can you tell me about the track?

 

Stuff that's generally relevant / important:

 

Genre?

 

Instrumentation? Heavily overdubbed / multitracked or not?

 

Did you play with or without a click track or tempo map?

 

Do you have multiple takes? If so, have you "comped" an optimized take for each player? Note I said optimized, not perfect.

 

Vocalist - is it layered?

 

How many overall tracks did you use - 8? 24? 64? More?

 

What DAW are you using?

 

Got clips / MP3?

 

Any other particular issues (besides panning) that you're concerned or wondering about?

Link to comment
Share on other sites

The Release is the hardest part to get right for me - it's a subtle thing. If the release is right, the track will just feel right - it'll become part of the groove. If it's wrong, it'll fight the overall feel of the playing,

 

Ever try timing the release to musically appropriate time values? There are ways to calculate the time between hits based on the tempo, and adjust your compressor's release time to correlate... :idea::)

Link to comment
Share on other sites

  • Members


Ever try timing the release to musically appropriate time values? There are ways to calculate the time between hits based on the tempo, and adjust your compressor's release time to correlate...
:idea::)

 

I haven't... what a good idea! :idea:

 

When you do that, do you then subtract the attack time you've got dialed in? I guess if it's a slow, warm bass and the attacks at 100ms or something then would that have enough of an effect to matter?

 

I'm gonna try this next time I'm mixing. :)

Link to comment
Share on other sites

Here's another thought for you... a lot of multitrack mixing is illusion. You're creating the illusion of a live performance that never actually happened. That "performance" can be as wild and fanciful or as true to life as you want to make it. IMO, the music largely dictates that... but that's ultimately an artistic decision. But when you create the illusion, you set the stage. If you want that DI bass to sound like it was in the room with you, or sound like it was played in a bathroom on Mars, then you'll have to create the illusion of that acoustical environment. It can be hyper-realistic, or it can be hyped to hell and gone. A lot of that is in how you apply the tools at your disposal - mainly reverbs, delays, EQ and compressors. You can add just a tiny bit of something and it's perfect, or a lot and it's too much - and sometimes the reverse is true - nothing less than a lot will give you the overblown effect or heavy impact you're looking for.

 

Oh, and a big +1 on the comments about the importance of good monitors and a good monitoring environment. It makes a huge difference!

Link to comment
Share on other sites

I haven't... what a good idea!
:idea:

When you do that, do you then subtract the attack time you've got dialed in? I guess if it's a slow, warm bass and the attacks at 100ms or something then would that have enough of an effect to matter?


I'm gonna try this next time I'm mixing.
:)

 

I think you'll like it. :)

 

Use the numbers as a starting point. I like to fine-tune it by ear... often I'll have the release let go completely a few to several milliseconds before the next hit occurs... although sometimes an overlap "feels" better. Try both and see what you think in each specific case. "Doing the math" can really come in handy. You can really control the compression quite a bit, and it really does make a difference in terms of whether the compression is working with, or fighting against the groove and feel of the music.

Link to comment
Share on other sites

  • Members

Here is a good little article on EQ:

http://www.recordingeq.com/EQ/req0400/OctaveEQ.htm

 

Bob Dennis has other articles with Motown techniques and stories that are definitely worth a read:

http://www.recordingeq.com/motown/

 

In terms of EQ, I like to start by thinking about giving each instrument its own (frequency) space in the mix. For example, a bit of HP filtering on the guitar so that it stays out of the bass' way. Boost EQ is great as an effect, and I often use more radical EQ while recording to commit the altered sound to "tape". A habit from 4/8 track tape days when the EQ would change how the tape saturates, but also helps me separate sound shaping from blending the sounds together.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...