Is Mixing what I need to learn?

TTony

I've been playing around with some multi-track recording.  Nothing too fancy, just drums, bass, keys, guitars & vocals.  No more than 10 tracks, and probably only 4-5 at any one time.

I'm using Presonus Studio One (Artist version), on a MacBook Pro, and generally using a pair of over ear Bose headphones (NOT noise cancelling).

When I listen to properly recorded stuff, the different instruments/tracks are usually clearly defined or separated and each can be heard and followed individually.

With my recordings, I'm struggling to achieve the same level of clarity.

I've experimented with EQ, and separating tracks to different degrees across the stereo L/R range, but not had any great "wow, that's it" moments.

Any suggestions? 

poopot

You need to be listening on something decent…

ideally monitors in a treated room, however that’s not always possible…

a decent set of headphones will do…

when you mix your stuff have a reference track in the project… doesn’t matter what the track is as long as it’s the “sound” you are after… use that track, listen to it, really listen to it… where are the guitars sitting, what’s the bass doing… 

make your mix like the reference track… 

poopot

I’m going to get another mix challenge going… try mixing those tracks… you won’t be as attached to the song so you won’t be as critical of what you are doing…

WiresDreamDisasters

Big big big topic. There are no short-cuts for this kind of thing, and it's not always the most straight forward thing to discuss.

So for me.... your first two things are level and pan. You can do a lot with those - don't be afraid to move them around and experiment. A lot of people fall into the trap of locking them into place, and then trying to fix a ton of stuff with EQ and compression, when all they really needed to do was be a bit more limber.

But EQ and compression are the next things. EQ controls the frequency content of your audio, and compression controls the dynamic range of your audio.

Take a kick drum and a bass guitar. Each of them in isolation sound good, and you think.... why aren't these working together??? Could be lots of reasons, but the most common is frequency masking - each element shares the same frequencies, so one element masks the other and you can't hear it. One solution might be selective EQ'ing to remove the frequencies from the element causing the masking, so you can hear the thing that is being hidden.

And it gets really FUN (read: annoying as fuck) when these relationships change from second to second!

Take a clean guitar for example... you might notice that some notes jump out a lot louder than others. Or some are too quiet and can't be heard. That could be a good place to use compression.


A kind of "standard" approach that a lot of people start with, is by having a high-pass filter on every element except the kick drum. This can result in frequencies below 60hz say, being filtered out on all instruments except the kick. This can clear space for the kick. 

Finally, comparisons are actually a good idea. I spent years learning how to EQ a kick-drum to sound like some of my favourite records. 

It's hard to give any specific advice. But yes... you could learn mixing. It's a deep topic. And you're always learning something. It's akin to learning a language in my opinion.

WiresDreamDisasters

Oh, and to add - the power of mixing really lies in being able to identify problems. Being able to understand exactly what the issue is, and then figuring out how you can fix it.

TTony

WiresDreamDisasters said:

It's hard to give any specific advice. But yes... you could learn mixing. It's a deep topic. And you're always learning something. It's akin to learning a language in my opinion.

Cheers @WiresDreamDisasters.
I feared that it might be something of a rabbit hole.


I've heard that mixing skills are well regarded, so whilst I didn't expect to get good quickly, I'd hoped that it'd be a bit longer before I got frustrated!  But I've got some time to play, it'd just help if I understood some of the rules that I need to play by.  Any recommended sources / sites / channels to look at for a start?  

I've been playing with compression/limiting too, to try to get a less variable volume level on the base tracks.  When using EQ, do you generally boost/cut on fairly narrow or wide range of frequencies?  And, I'm assuming, what you boost on one track, you aim to cut on another?

spark240

Post the tracks let us lot mix them....if any of them turn out how you think they should sound...we tell you what we did ;-)

WiresDreamDisasters

TTony said:

WiresDreamDisasters said:

It's hard to give any specific advice. But yes... you could learn mixing. It's a deep topic. And you're always learning something. It's akin to learning a language in my opinion.

Cheers @WiresDreamDisasters.
I feared that it might be something of a rabbit hole.


I've heard that mixing skills are well regarded, so whilst I didn't expect to get good quickly, I'd hoped that it'd be a bit longer before I got frustrated!  But I've got some time to play, it'd just help if I understood some of the rules that I need to play by.  Any recommended sources / sites / channels to look at for a start?  

I've been playing with compression/limiting too, to try to get a less variable volume level on the base tracks.  When using EQ, do you generally boost/cut on fairly narrow or wide range of frequencies?  And, I'm assuming, what you boost on one track, you aim to cut on another?

To be perfectly honest with you, it's been over a decade since I had to learn anything from scratch like this. I'm probably not the best person to ask for guides as such. 

However, the best thing that I've ever read on mixing is a book called 'Mixing With Your Mind' by Mike Stavrou. However it's ludicrously expensive now!!!

I just found this after a bit of Reddit digging: https://archive.org/details/GuideToMixing/mode/2up
This looks to be pretty good on face value, but I haven't read it thoroughly.

Diagrams like this can be useful but they're not a bible, so don't take them too much to heart:


Whether I'm boosting or cutting narrowly or widely sort of depends on the job at hand - if I want to control the general tonality of an instrument, then I'm using wider cuts and boosts. Like if I want some high-end air on cymbals, I'll do a fairly large shelf boost on my overheads.

But if I'm being surgical with tom drums for instance, then it's narrow cuts and boosts all the way.

One common approach is what is known as subtractive EQ - so basically... never boosting, only cutting. This can help in the beginning, because you're less likely to push frequencies higher which may excite signal processing further down the chain. Less likely to clip your channels too.

When recording material, I always aim for my peaks to be between -18dbFS and -12dBFS on the meters in whatever recording software I'm using. Because this means that I can pretty much record as much material as I want, and I'm never going to end up with a super loud and super clipped mix as my default setting. This is kind of standard practice too.

Listen to each element in your mix, try and judge for yourself what you think is wrong with it. Maybe there is a note on the bass guitar that is just absolutely piss-boil-irritating to listen to?? Maybe every time it happens you need to nudge your fader down by 8dB, or maybe you just need to notch out the offending frequency? There's two solutions that will result in quite different outcomes. Because notching out a frequency also affects all the other notes that might have content for that frequency too.

Maybe you've got a violin. Why I don't know, because they sound like shit. But maybe you've got a violin... and the low notes don't have enough body to them - bass boost.... 

A lot of what mixing can be is finding the fundamental frequency of instruments that are annoying, and bringing them down. And finding the fundamental frequency of instruments that aren't present enough, and selectively boosting them across the track.

Gating..... gating is quite important too. Let's say you've recorded a cranked Fender Deluxe.... and let's say the guitarist plays 1 note the entire song.... and the rest of the song the amp is just making noise, and that noise is getting all over your track. Chop it out!!! It sounds simple... but a lot of people don't even do this, and it's one of the hallmarks of crappy amateur recordings.

@octatonic might have some good resources too, as he's more regularly in the game than I am, and has worked on a wider variety of stuff. @nerine and @Danny1969 would have some good input too hopefully.

WiresDreamDisasters

And @Cirrus and @guitarfishbay ;

They got skillzzzz too.

TTony

Thanks for the detailed comments.

Part - probably a large part - of the problem is not knowing what I don't know, and therefore what I need to learn and understand.

I could spend some more time experimenting - that's easy enough - but it sometimes feels that I'm experimenting on the "infinite monkeys, infinite time" principle.

TTony

spark240 said:

Post the tracks let us lot mix them....if any of them turn out how you think they should sound...we tell you what we did ;-)

I might just try that @spark240 if people have got the time to play around with them.

WiresDreamDisasters

Maybe as an exercise try this - loop a chunk of the song.... bring down all the faders... then one at a time bring them up for 5 seconds and then back down.... and think about what you heard and what you didn't like tonality-wise. Try and ignore the music, and focus just purely on the sound.

It's an ear training thing as much as a tech thing.

blobb

Start off by creating a 'level and pan' mix using a pink noise reference:

https://www.soundonsound.com/techniques/mixing-pink-noise-reference

Alternatively, download Mixbus :-)

TTony

Thanks @Blobb.

I'm going to skip Mixbus though, if I'm right in thinking it's "just another" DAW?  I'm using Studio One, which I'm getting to understand reasonably well now.    I  probably need to learn/understand some more basic concepts rather than another piece of software.

Unless there's a magic plugin out there ...

Cirrus

The biggest thing is being able to hear what you're working with, bar nothing.

When I was at uni I was totally shit at mixing, but I was able to work in good studio rooms with acoustic treatment and excellent speakers - so I could hear what I had and I could hear what I was doing. Those mixes were amateur as hell, but the basic balances and eq decisions were sound and translated out into the real world.

5 years later, working from home with no acoustic treatment and budget monitors, I was making more advanced decisions and had more of an idea how to make a coherent emotive production, but... I couldn't hear what I was doing properly, and the resulting mixes were worse out in the real world. They were turds rolled in glitter.

If you can hear what you're doing, you can hear when you're screwing things up. If you can't hear what you're doing, you overthink and stop actually listening without realising - you mix what you *imagine* it sounds like, but you can't hear through bad acoustics and bad speakers/ headphones. They simply don't give you enough information to work with. It's like painting with green tinted glasses that are the wrong prescription and have masking tape covering most of one eye. But your ears don't alert you to missing information in the same way your eyes do.

octatonic

Agree with everything Drew has said.
I learned to mix by trial and error, a lot of the latter, frankly.

I would advocate goal oriented free experimentation sessions.
Figure out how to make snares sound good in the context of other instruments, then guitars, then bass, vocals, etc.

It takes years.
Good monitoring helps, as does a good room but you can work around it to a degree.

Rowby1

I’m going to swim against the tide a bit here. 

I think that familiarity with your monitoring and environment is actually more valuable than great monitors and room treatment. 

Listen to a lot of well recorded reference material in the room you’ll be mixing in on the monitors you’ll be using and become very familiar with how it sounds. Once you know what you can and can’t hear it’ll inform your mixing decisions.

Otherwise I’d echo a lot of what’s been said above. Just don’t let lack of expensive monitoring or room treatment put you off trying.

CoffeeAndTV

blobb said:

Start off by creating a 'level and pan' mix using a pink noise reference:

Panning is good for separating the instruments.  I would look at that too. 

TTony

@Cirrus ; - thanks for the reply  & suggestions.

I get the message, but most of my work is done via headphones rather than monitors in an acoustically treated room.  I realise that's going to be sub-optimal, but it's a constraint that I'm working within. 

Happy to spend some more ££s on more suitable headphones (which is why I mentioned I'm using standard Bose OEs currently), if that's going to be a big improvement in being able to hear more detail.  I'm not going to be getting into  setting up a proper studio environment though.


Cheers @octatonic - understand the "it takes years".  I'm hoping that the first 50% of the journey (to better mixing!) can be completed relatively quickly  though.  I'd probably stop at that point - I've no ambition to attain your level of skills, just enough to turn something out that I think sounds OK.

WiresDreamDisasters

Rowby1 said:

I’m going to swim against the tide a bit here. 

I think that familiarity with your monitoring and environment is actually more valuable than great monitors and room treatment. 

Listen to a lot of well recorded reference material in the room you’ll be mixing in on the monitors you’ll be using and become very familiar with how it sounds. Once you know what you can and can’t hear it’ll inform your mixing decisions.

Otherwise I’d echo a lot of what’s been said above. Just don’t let lack of expensive monitoring or room treatment put you off trying.

I agree with this. I don't think you need the perfect room and the best (read: most expensive monitors) - I've done great work on cheap Tannoy Reveals and cheap Focal Alphas.

I'd also say, don't worry about not being perfect out of the gate. It's a learning process. Make something... get feedback... action on it... rinse and repeat.

Cirrus

TTony said:

@Cirrus ; - thanks for the reply  & suggestions.

I get the message, but most of my work is done via headphones rather than monitors in an acoustically treated room.  I realise that's going to be sub-optimal, but it's a constraint that I'm working within. 

Happy to spend some more ££s on more suitable headphones (which is why I mentioned I'm using standard Bose OEs currently), if that's going to be a big improvement in being able to hear more detail.  I'm not going to be getting into  setting up a proper studio environment though.

I guess I should make clear - I'm not saying you might as well not bother unless you've got >10k to spend on professional acoustic design and studio monitors. I'm just saying, you might as well know where the biggest source of uncertainty and confusion is going to come from. @Rowby1 is right - you can mitigate it a bit by knowing your system, and by referencing in other places. It's just that it adds a layer of second guessing and uncertainty at every stage.

Specifically, you talked about getting clarity and seperation in a mix. If you're listening on headphones that naturally cloud up certain frequency ranges, pro mixes might sound fine but they won't tell you what you need to know in your tracks to identify where space is needed and where conflicts need to be resolved.

These days I'm mixing in a box room that's only just made bearable with some unsightly rockwool panels, and I've just given that up for a nursery so I'm going to be on headphones a lot more, too!

Bose gear I've heard in the past has had a sound that's anything but transparent - usually it's hyped in various places to give the impression of bigness at the expense of clarity. I bet if you spent £150 on some headphones by Sennheiser, Shure, Beyerdynamic etc it'd be a big step up in terms of being able to critically listen to what you've got.