Mastering a studio mix?

[Sigma]

I've made a few multitracked mixes in the past that I would call "studio mixes" as it's more like production than mixing. The mixes I've made have either been hip-hop or D&B and I've never had any issues with mastering them.

I'm currently putting together a soul mix. I'm finding it quite difficult to get the levels right. Using the gains on the mixer doesn't tell the full story, cos some of these tracks sound way louder than the others, even though they peak at the same level on my mixer - obviously, this happens with all music, but it's not been an issue with hip-hop/D&B.

So obviously, my goal is to get the mix sounding even throughout in terms of volume, as you would do for any mix, but I don't really know the best way to approach that this time. Let's say I have tracks A and B and A sounds a lot louder than B, despite them both peaking at the same spot on the mixer. I would lower the volume of A so it's down to the level of B based on what I'm actually hearing, but then that's kind of dictating the volume of the overall mix if you know what I mean. I don't want the mix to sound quiet overall.

I don't think I've done a very good job of explaining this, lol, but if you do know what I'm talking about, do you have any tips?

[drop1]

I would go by ear off of the quietest track and then use a limiter/compressor to raise the volume of the entire mix as a whole. Just make sure your tracks sound balanced while mixing, ignore the meters except to use as a reference to stay under 0db.

[Sigma]

Right, that's what I was going to do, but it's the "use a limiter/compressor" part that I need some tips on. Thanks man.

[PONTUS.2]

shit used to happen to me whenever id bust out 36 chambers, the LED would be shining into red but the track would be quiet as fuck.

sorry i didnt offer any help but yeah i feel you.

[Manu]

set your compression, ratio and threshold according to the average loudness (RMS), not peak. Get your tracks all sounding as flat as possible to make to mastering process easier and more efficient. I presume you're using old school tracks, when they would mix with more dynamics and less compression.

[Sigma]

The tracks come from the late 70s to the mid-90s, plus I'm using some live performances too. The actual sound quality is great, cos aside from the live performances it's all from lossless sources, so it's just the way that the tracks were originally mastered that varies.

And I know what you're talking about in your post, but I don't know what I should be using to achieve that. With the mixes I've done in the past, I've never bothered with any stuff like this as I've always felt that the less you mess with the original sound, the better, but here it's gonna be a necessity pretty much.

[Manu]

You shouldn't need much as the old "less is more" still applies here. Regardless of the software: I'd start with a threshold of -6 dB, compression ratio of 4 to 1 and treat it like that. What it does basically is shaving off the top 6 dB of sound and levelling it down.

You can normalise the file again afterwards, assuming it peaks at -0dB in the first place. Also in your case it's better to start it from -0 DB peak as they are old files: you don't want to lower the signal to noise ratio or even try to get a headroom to start with, as it will mess up with the floor noise. Doing it this way will not take too much dynamics away while levelling the tracks towards a common point of perceived volume.

Mac: Waveburner
PC: Soundforge or Bias Peak Pro.

The general rule of thumb is: the better your final mix is, the easier it is to master.

[Sigma]

Thanks man. I've still got a lot to do on it, but I'm sure I can get it sounding OK.

[Interracial Tea]

My process so far has been to get the levels as close as possible in the mix, then open the mix up in Ableton and add a limiter/compressor without it sounding like ass. It's worked so far.

[Austin GoGreen]

Limiter works REALLY well in FL Studio. I'm sure it's just as nice if not better in Ableton. A limiter allows you to raise the overall volume of the mix to a equal level but also allows you to put a ceiling over the mix. You do have to play around with it to find that sweet spot. Track out each record so you can adjust the song accordingly. I put the limiter on the master track.

What I do is get a good solid mix layed out, adjust the levels of each track in my FL mixer manually, then go back and bring the level up on the whole "studio mix" or master with the limiter. Again, right at 0db is where u want to end up.

I spin a lot of electro/house remixes when I'm out DJing and I can always tell which tracks were produced really well with a limiter/compressor just by looking at Serato's built-in db meter.

[drumpusher]

I'd go with limiting over compressing too.

[Manu]

You still want to keep the dynamics of the original track though. That's quite tricky with "older" music. The mixing/mastering techniques were different before when it went almost all digital. When the artists were working in analogue, it was about maximising the input without hitting the point where it starts distorting. It was also about to deal with minimising the analogue signal to noise ration, -70 dB back then compared to -100 onwards nowadays.


Thanks man. I've still got a lot to do on it, but I'm sure I can get it sounding OK.

No worries (I'll still redeem the clown shoes at some point). What you need to do is get your mix about level in terms of perceived loudness and the mastering takes care of the rest.


:edit: you cnut

[mostapha]

Right, that's what I was going to do, but it's the "use a limiter/compressor" part that I need some tips on.

Biggest tip: less is more.


The tracks come from the late 70s to the mid-90s, plus I'm using some live performances too.

That's the issues…several generations of the loudness war in there, along with varying technology and recording methodologies that Manu mentioned.

If you're multi-tracking the mix and you want a lot of practice, you could re-master each track individually. That could be fun…if you're crazy and think that kind of thing is fun (like I do).

Apart from that, one thing I've liked is staged compression. It takes a lot of experimentation, but you can use more than 1 compressor on a channel.

So…see what it sounds like with 3 compressors and a limiter:
-20dB Threshold; 1.5 to 1 ratio
-10dB Threshold; 3 to 1 ratio
-6dB Threshold; 5 to 1 ratio
Brickwall at -0.05dB

You'll have to tune that based on how you recorded and how it sounded…and tweak the heck out of your attack and release times and both input and output gains of each compressor, but it can make things sound more even without killing dynamics.

It's kind of like the knee value in a lot of modern compressors, except I've gotten better results out of doing it this way. And some of the compressor plugins that I really like (it's not like I've used that many) don't have knee settings.

The actual values were chosen out of thin air, so don't take them as anything more…just kind of an idea of where to start and the progression you're looking for. I'd probably tune to have as little gain reduction as possible…just barely to the point where you can hear it. And make sure to bypass each one and the whole chain occasionally so you check to see if you're actually improving on what you have or just tweaking endlessly for no reason.

My last recorded mix, I used a compressor and a limiter set up in a similar way. I don't remember the values.

Also, if you have access, using different vintage-modeled compressors can yield a cool sound. I've done that in the past, usually using an 1176 clone as the first stage and an SSL4000G Master Bus Compressor clone as the last (before the limiter). I've always been happy(ish) with bundled limiters (in Ableton, Logic, and PT).

If you happen to use PT10, the channel strip is actually pretty cool just because it can set a maximum compression depth, which is something I haven't seen too often. Just turn off everything else (or us its EQ if you're going to do that…which sounds okay to my ears).


You still want to keep the dynamics of the original track though. That's quite tricky with "older" music. The mixing/mastering techniques were different before when it went almost all digital.

That's certainly true. And it's something that I think was kind of lost. The technique I described above is based on the stop-gap all-analog pseudo-mastering that you can do if you have a lot of compressors lying around but not an actual analog mastering setup. It sounds kind of amazing. The noise floor on all-digital stuff is really quiet. It all kind of sounds amazingly clean and beautiful and…unforgiving.

The weird thing about this thread to me is that digital is kind of amazing…the noise floor is basically nonexistent. And we're basically talking about going to a lot of effort to squish the whole mix into maybe 20dB of used dynamic range that's barely more than 3-bit audio. I'm not saying that's the wrong decision or anything. It's just weird.

[Hamza21]

I found this back on the old DJforums several months back and that's what I do to my mixes. It works for me. It starts @ 3:01


http://www.youtube.com/watch?v=likk4fbEblg

One thing you don't have use amplify after limiting you can set the amplify within in limiter itself, well at least in my audacity on linux. I'm not sure about a windows version.

[Scrap McNapps]

Have you thought about using automation in your DAW to control the volume?