Ok this is the issue i'm facing when i'm mixing

I set the first track...to a random volume i think its ok to listen to

and slowly slowly i'm bringing other tracks in and starting to realize i don't really have a method of setting volume and sooner or later i'm gonna clip and all that jazz

its always a failing game for me

can someone advice me a way so this doesn't happen all the time

 

What genre(s) do you work in? Even better, do you have a clip of a track that you're working on that you could post - that'll help us give you much more you-specific advice.

If you want some general pointers: If your music has vocals, then for most genres, the most important elements tend to be the vocal, the beat, and one main melodic element.

The ear can only really focus on two, and at a stretch, three things at once, so you have to decide what the most important elements of a track are, and then work the mix to feature those elements, and have all the other stuff complement and support them. (To further complicate things, the thing that's most important may change across the course of the track, and even if it doesn't, you should be trying to find ways to subtly change things up a bit for the listener, because the brain very quuickly starts to tune things out that it perceives as being unchanging).

As I said, for most genres, including rock, pop, hip hop, most edm, etc, these things will be the vocal, the beat, and then typically a melodic element - often a bassline, but could be a guitar, piano, synth, loop, whatever. This will vary slightly between genres, and from track to track - eg in rock, you usually want the snare to be the most prominent element of the drumkit, whereas in most techno (and a lot of pop) genres it's usually the kick.

A lot of mixing is essentiially creating an illusion in the listeners brain - what you're hearing in commercial mixes as everythign being huge, and full-range, and everything bright and loud, etc is usually not entirely what you're actually hearing, because as you've discovered, you only have a finite amount of headroom to work with when mixing. So you have to figure out how to ration out that headroom in a way that creates the impression of 'everything bigger than everything else' without the mix sounding like a muddled, shouty mess!

A lot of this happens in the production / programming / tracking stages, ie choosing and recording the right sounds. But you can do a lot in the mix as well:

One of the most important pieces of advice I'd give anyone is learn to eq things in the context of the rest of the mix - the solo button is NOT your friend here! What I mean is, it's not that hard to get a single, solo'ed element to sound really great, but often what sounds good in isolation sounds crap iin the context of the full mix.

Say you've got a distorted rrock guitar. Now, if you play that in isolation and try to eq it, what you'll probably do is dial in a bunch of low end, make it really thick and creamy in the mids, maybe cut some of that "harsh sounding" 3k, etc. Now, take that guitar track out of solo, and listen to it in the track - it almost completely disappeared, right? That's because when you're listening to a mix, you think you're hearing a guitar that has big, beefy low end when it chugs, and it sounds smooth up top. What you're actually hearing is the guitar working with the bass and the kick to give you the impression that it's fuller than it actually is.

If you then eq that same guitar while listening to it in context with the rest of the track, when you get it to the point that it actually sounds good, try soloing it now. What you'll most likely hear is a fairly thin sound, with almost piercingly-present upper midrange, basically something that probably doesn't sound all that amazing on its own.

But here's the most important point - the only person who's ever gonna hear it like this is you. Put it back in the track, and it sounds killer and you sound like a genius lol.

That advice obvs isn't just applicable to guitar - the same concept apples to every part of a mix, whether it's a vocal, a synth bass, and whether you're eq'ing, tweaking compression, or addiing delay/verb.

You have to split your brain into about 6 different pieces when mixing - you have to be able to focus in to small tweaks and corrections on individual sounds, while at the same time always remembering to stay focussed on the mix as a whole. Tweaking one affects the other, and it's basically a massive juggling act to get everything working together at the same time.



There's no real 'one size fits all' approach to mixing a track - if you can post a sample of the track you're working on, we can cerrtainly try to give you some pointers.

 

Last edited: Jul 27, 2022

Two mistakes at once firstly, don't start with the first track but with the one which should be loudest secondly, to listen to??? You set the level you're listening to with the monitors/interface, not with the tracks' levels.

Here's how I'm doing this:

1. Turn down the volume (monitors/interface),
2. play all tracks with faders at 0dB. This will clip the mainout like hell (that's what step 1 is for),
3. turn down all faders until the mainout isn't clipping anymore,
4. set pan and level roughly for all tracks only by pulling the faders down. This will lead to something like -10 or 12dB in the mainout,
5. pull all faders up until the mainout is at ~ -3dB (you can leave this out if you want),
6. start mixing without raising the peak in the mainout.

 

Well, to grossly simplify things: if the concern is clipping the master out, you can always just take a step back and turn everything down. You can do it by manipulating the gain within a plugin in the master channel, even, and the result should be the same. Is that a great strategy? Not really – as you said, you're lacking control over your audio, leaving things up to chance. It's good to always be aware of your peak and RMS levels, for one – the tool I use for that is the Klanghelm VUMT – and I'd also advise to always keep a limiter on, with a ceiling lower than 0. This way, you can check what's hitting it and instantly gauge if that reduction works for the source and is good for the song overall, and you can then go on to inserting a limiter (or compressor, saturator) in the individual channels (and busses) that could use it and prevent the master limiter of working too hard when everything is summing together.

These videos should be good resources. This is definitely something you have to study and experience, and not keep looking for 'tips and tricks'. It takes time.

 

did u see someone using this method? any tutorial or something? would be interesting to see


also this volume thing is done before compression right?

so all the peaks/rms whatever they are called will change with compression afterwards...so won't all that change the volume of the tracks again, so its back to square 1?

 

You don't really need a tutorial for this mate, it takes all of 5-10 mins to just try it yourself. This isn't something you learn by watching others, you learn it by doing it many, many times, with as many tracks as you can get your hands on.

And most mixing engineers will pretty much use some variation of this method - it's usually referred to as getting a 'rough balance'.

What you're doing is very roughly gain staging your tracks first, so you're not clipping your 2-buss.

Then you're making quick balancing moves to get a sorta half-decent, listenable version of your track. It's best to try and do this quickly, without overthinking it - just bounce around from element to element until things start to sound good to you. Once you've done this with the levels, you can sketch in some rough panning if you want.

At that point, you'll have a better feel of the mix, and can then start thinking about eq, compression moves, etc.

Some people at this point like to then do more of a 'rough mix' - ie applying the same "move quickly and try not to overthink it" approach to basic eq, compression, group and buss treatments. This helps to get a more polished, vibey mix going in maybe the first hour or two. Once you're at that point, you can then start focussing in more surgically on issues you're hearing.

You really need to develop your own workflow by doing it, and seeing what works for you - this is just one approach, other people prefer getting the really anal, surgical stuff outta the way first, and then "zooming out" to build the full mix. There's no right or wrong way, just that way that gets you the results you want.

 

https://www.hornetplugins.com/plugins/hornet-thenormalizer/

 

If ever wind up in that situation I just attenuate every channel by, say, 4 or 5 dB. Gets headroom back and mix remains exactly the same.

 

Simple rule :
3dB = double volume = double track number

So basically, you add 3dB every time you double track number
1 track -0dB on track fader
2 tracks -3dB on each track
4 tracks -6dB
8 tracks -9dB
16 tracks -12dB
32 tracks -15dB
64 tracks -18dB
...
You see the "gain build up" now

And now you understand WHY a lot of ppl advise -18dB as a starting point on each track.
It is safe for "usual" mixing situations with 64 tracks or lower.
Of course, it works with buses.
You can "lower" a bunch of tracks by just lowering bus level.

Second trick :
LOWER levels as a starting point, don't "push" : you will saturate master bus.

That's MY way of "seeing" things as starter advices

Another trick : use project templates as starter
OR select all tracks (or all buses) and lower them all at the same time on already made projects

 

Getting a "static balance" is more challenging than most people realise. There's a course on the sister site called "Mixing With Your Eyes" which uses meter levels to set track levels. Bobby Owsinski also has some videos describing a similar method. If you do a search for Static Balance you'll come across all sorts of tips and tricks - for example, set the kick to -3VU then bring up the bass until the meter hits 0VU when both play at the same time. Listening at very quiet levels can also reveal if anything is sticking out. Also, listen to references in the genre you're working in. If you can get stems then you can even assess relative balances within songs. Streaky (Mastering Engineer) has a good tip for setting vocal levels (move to the side of your monitors). See below for a few other tips that come from Mike Senior:

SOME TRICKS FOR JUDGING FADER LEVELS

If you can’t make your mind up about a particular fader, then here are a few helpful tricks. The first comes courtesy of Gus Dudgeon: “I find the quickest way to decide whether something is actually loud enough or not is to run the mix and turn the signal on and off. If you turn it on and you can hear it, but it’s not smack bang in your eye, it’s probably where it should be. . . .The best thing to do [if an instrument is too loud] is to take the bloody fader out and start again. Creep it in until you think that’s probably where it should be and then try switching it on and off.” Mike Stavrou offers another good one: “Instead of focusing all your concentration on the fader being moved, listen to the neighboring instruments instead. While trying to hone the perfect level of the snare drum, for example, do so while listening to the bass drum. . . . You will quite often find a more decisive point of balance this way. That’s not surprising, because when you think about it that’s exactly how the listener listens — his (or her) mind is elsewhere until you attract his attention.”

“It is always easier to tell when something is wrong than when something is right,” says Roger Nichols. “While listening to the mix, one at a time change the levels of each instrument in varying combinations. Turn the bass up 1dB. Does the mix sound worse? If the answer is yes, then turn it back down. Turn the bass down 1dB. Does this sound worse? If the answer is yes, then the original level was right for the bass. Now try [the other instruments]. . . . If you can change the level of an instrument in the mix by a tenth or two-tenths of a decibel and you can hear the change that you made, the mix is getting pretty good.” (If that “pretty good” feels like a bit of an understatement, just bear in mind it’s coming from an eight-time Grammy winner!)

Another nifty hint comes from Roey Izhaki. “Take the fader all the way down. Bring it up gradually until the level seems reasonable. Mark the fader position. Take the fader all the way up (or to a point where the instrument is clearly too loud). Bring it down gradually until the level seems reasonable. Mark the fader position. You should now have two marks that set the limits of a level window. Now set the instrument level within this window based on the importance of the instrument.”33 What’s particularly useful about this tip is that the size of the “level window” can also provide advanced warning of tracks where further mix processing is required—as we’ll see later.

The level of the lead vocal is probably the most common bone of contention, because there’s often a fine line between “loud enough” and “too loud.” As Dave Pensado notes: “If it’s too loud, the track sounds wimpy. If it’s too soft, the vocalist sounds wimpy.” One rule of thumb from Bob Power is that the vocal should feel a little too quiet when you turn your monitors up really loud, but it should also feel a little too loud when your monitors are turned down really quiet. Given that the lead vocal usually carries the totality of a song’s publishing copyright, i.e. the melody and the lyrics, there’s clearly commercial sense in making sure it’s clearly audible, but outside chart-centric releases a lower vocal level can actually encourage listeners to turn things up and listen more attentively. “You try to draw the listener in,” explains Dave Bottrill. “I learnt that on those early Tool records. You’ll notice on Aenima sometimes the vocal’s quite quiet . . . The idea was that if the listener’s thinking “What did he say? What did he say?” then suddenly he’s facing into the mix.”

HTH!

 

No, why should I need a tutorial for this? If it works it works.

Yep.

Sure, that's why you do gainstaging.

 

As a ME I have the vocal as a constant reference. If the tracks are instrumental I go by percieved loudness and dynamic progression.

Same goes for when I mix. The vocals (or a lead instrument) is the reference. I set that as focus. Then it's just healthy gainstaging and putting everything else underneath to support it.

 

Last edited: Aug 3, 2022

You should consider the following two points:
1. never make random decisions about level. Make sure you have enough headroom.
2. it's a good practice to set the levels BEFORE the faders.

Every DAW has a way to set levels before the faders.
Be it the clip gain, the pre-FX level, a level plugin in the FX insert, or the VSTi output level.... - use it!
This gives you a lot of flexibility and saves you a lot of trouble on the way to a finished mix.

1. Set all faders of your tracks to Unity.
2. Select the instrument you want to be loudest in your mix.
3. Set this instrument to a level of -12 dB peak using one of the above methods.
4. Now use the same method and bring the other instruments to the desired level in relation to your loudest instrument.
5. That's it. Now you can start mixing!

You can now enjoy the following benefits:

• Sufficient headroom when mixing.
• Good resolution of plug-in controls (less fiddling).
• Plenty of fader resolution. Especially important for fine adjustments and level automations!

Have fun!

 

Last edited: Jul 28, 2022

I simply start with a VU meter and tracks that are supposed to be the loudest ones like drums, bass and vocals. Then I level them to sound good together and hit the 0VU which is -14db VU-RMS. Everything else in the track gets the level according to these tracks. I generally end up with a track peaking between -6dBFS to -3dbFS and that's perfect for true peak limiting to -1dB afterwards.

LUFS? Depends on the track/genre and possibly "pop-iness"? . I like to get everything sound right in the mix. Final limiting is not mastering. It's just making things loud and possibly f-ing up the good sounding mix. Since my mixes sound good to me already, all I want in the "mastering stage" is cut the too loud TPs and make the track -13dB LUFS compliant which is the best for all platforms.

12dB of headroom is pretty perfect for all pop genres. For classic and jazz it should be more like 18dB LUFS.

Some really great advices for mixing in this thread. Read them, watch them, use them.

 

Last edited: Jul 29, 2022

Voice is a reference in audio domain, because we hear voices everyday.
Like skin tones in photography.

Skin tones and voices are "brain" solid references

 

@SineWave makes great points about loudness just above - people can get very hung up on levels (both micro and macro) with their tracks.

The truth nowadays is really that level, with respect to master level, and LUFS, just doesn’t matter in the same ways it did to a lot of people in the 80s, 90s and 00s.

Mastering nowadays should really never be approached with max loudness as the goal, because it just isn’t necessary.

And much moreso than ever now, it’s way more likely to be detrimental to your track, because it’s not just the side effects of extreme loudness processing you’ve gotta worry about, it’s also the fact that submitting a ridiculously-slammed master to streaming platforms will only result in one thing - a track that ends up being turned way down by their normalisation processes, and ends up sounding small, flat, and weak because the loudness-focused mixing / mastering processes have sucked all the dynamics/life out of it.



With that said, @SineWave’s general outline of starting a mix by taking your two or three most important elements, and getting a balance of them, where they’re hitting around -14 on the 2-buss is a very good rule of thumb to get you in the ballpark.

I’ve not really consciously thought about it in a long time, but I tend to work this way at the beginning of a mix too - I’ve just been doing it for so long that I don’t particularly realise I’m doing it! I don’t use a vu meter on the output, but I’ve been doing this for long enough that I just kinda know what -12>-14 looks like on the dbfs meter on my Cubase 2-buss.

I don’t really pay that much attention to my mix level when I’m working, but again, I’ve don’t this for long enough that my mixes usually turn out where they’re peaking anywhere from about -1 to -3 by the time I’m finished anyway, which is plenty loud enough, and also enough headroom for any good mastering engineer to do their job, without having to fight my mixing decisions.

From time to time, I’ll realise that I’ve pushed a mix louder, even to the point where I’m slightly clipping the output, but I’ve usually done this because I feel it sounds appropriate for the track, and if that’s the case, I’d just tweak the exported file to give me mastering engineer a dab or two of headroom to work with, while retaining the sound I felt worked best for that mix.

I always find it very satisfying when a master comes back, and the engineer has processed my mix more or less flat - where there’s only very minimal apparent changes, and what tweaks they have made are just little bits of polishing here and there, as opposed to any type of major level/balance tweaks - because if they’re feeling like they’re having to make significant level/balance/dynamics decisions, then your mix probably wasn’t working quite as well as you thought it was!

But yeah, the only way to get to that sorta point is to mix a LOT of tracks, and learn from the process and the feedback you’re getting. Tutorials etc can ofc be very useful and inspiring, but at the end of the day, this whole mixing thing is def something that you gotta learn by doing a lot of it!

 

Here's another method from over on the sister site - look for "MyMixLab - Drums and Bass Levels".

Cheers!

 

I think this is the best single bit of advice in the thread so far for a beginner. Especially if you are already using Samples (from multiple sources).

If you manually Gain each sample to the same rough peak, your channels basically start at a 1:1 relationship with any other channel; and at your faders this is reflected because you can start them all out the same -dB from Unity once you have this accomplished.. You can usually set a Fade in/out per sample at this point too, so adding a few milliseconds fade on the front and back ends can reduce clicks before copying out multiple copies of a loop. Being able to mute entire subgroups is very handy to toggle on/off to check volume again.

Using SubGroups (summing stack in Logic), is a very handy way to set things up while you are at this point also. The "volume" relationship is much easier to determine what your target is, if you are comparing it to the other instruments in that subgroup. Your drums are submixed together into a kit, or your Bass comp tracks etc. Then you are matching the relationships between these groups. When something is out of place and needs more fine tuning, you go back into the group and adjust that one channel. Instead of spending 2 minutes even figuring out where that individual channel is, because your project is not organized for you to work efficiently.

Call this old school, but I really try to minimize the number of times I will bounce something. So often when I am doing SinusWell suggestion, I am already going to add my eq/saturation/compression directly to the sample I am bouncing for some other simple correction. This also reduces the number of activated processors you have running.

 

Last edited: Aug 3, 2022