Menu
What's new
By:
Filters

Increasing the final mix volume - Limiter/Maximizer?

L

lr11here

New Member
Hello guys, my first post here, hope in the right place. So, I produced my song using samples and my mix was done in a very low volume, mostly because some of the instrument samples were already in a low volume, but the ones that were not in a low volume I brought the faders down in the mix. So whenever I needed to listen the whole track in a high volume, I simply turned up the volume knob on my audio interface. At some point, I was totally happy with the whole thing, the mix was all good and ready to go, but now I'm facing the problem that I am actually not sure how to make the song in a volume that is commercial. I referenced my track against another artist track that I found to be the volume that I wanted, then I tried to simply put the Maximizer plug-in (I am using Cubase) in the post-fader bus in the stereo out and turned up the Optimize knob until I found the volume that I wanted. The problem is that if I do that, trying to match the artist volume by turning up the Optimize knob, the most heavy parts reaches (or goes beyond) the output level that I set (which is 0.1 db), so it clips (not that horrible noise because I set the output level to 0.1db but still the low parts are cut in not a good way, specially percussion).

So guys, sorry for the already long text, but I am not sure how to make my song be in the volume that I want. Maybe the Maximizer plug-in, which I think has the same purpose as a Limiter, wouldn't even be the best way? It's not about loudness war, just matching a volume that the listener will not need to increase their speakers volume heavily. I read that there's compression and even EQ that's used in certain ways to make the mix/instruments louder(?), but as I mentioned, I am completely satisfied with my mix when I turn on the volume knob in my audio interface to be louder.

Thank you.
Click to read more...
 
While I am still learning about the whole process, I continued searching and apparently I was doing/thinking about it in the wrong way. I think what I really wanted is a tool like Ozone which will give me the threshold option where the compression would start working when the frequencies reached a certain point. Took some time to start understanding the concept.
 
lr11here said:
While I am still learning about the whole process, I continued searching and apparently I was doing/thinking about it in the wrong way. I think what I really wanted is a tool like Ozone which will give me the threshold option where the compression would start working when the frequencies reached a certain point. Took some time to start understanding the concept.
Click to expand...

Well, Ozone's threshold is like any other compressor's threshold, it's just that it slams everything that shoots past it down (because it's a limiter). Then it auto-boosts the overall volume by the threshold value.

Limiter/Maximizer is just a name for a compressor with an infinite ratio.

Here I posted about metering and targets for mastering.
 
If you are mixing/mastering, read into LUFs standards for broadcast and streaming. Izotope has lovely lectures on their youtube channel that go into detail about this. Just google, "Izotope What is Loudness?".

Cubase lets you view all of that information from the control room.

If you are slamming everything into a final limiter on your master fader and you are running into issues with specific frequency ranges somehow peaking after the fact, go back and check your mix! Find the problem child and try a few things. Personally, I will first move the fader down a bit to see if that fixes the issue without changing the sound too much. If that does not work, some EQ or Multiband compression may tame the amplitude for those sections. Automation is your friend. Sometimes you can get away with fixing these things on the mix you are mastering, but I often find it easier and cleaner to tweak it at the mix level.

**Disclaimer, I am not an audio engineer.**
 
Follow what Aaron posted.

You want to hit the loudness levels in your mix, not mastering.
Only use your loudness meter in the master track and use a/b comparison for loudness targets.

Kontakt defaults to -6db so you might have to up the levels on each track to balance a proper loudness level while composing or mixing.

Don’t use the maximizer, try to use volume levels in your tracks first. Get it close to the target loudness and then use compression or maximizer to fix a few dB .
 
The concepts you need to read up on and grasp is "gain staging" and "headroom". Lots of vids on that.
 
I don't know if I'm misunderstanding, but you set the output to 0.1db on the limiter/maximizer? I can't figure out why you would do that. If it goes above 0 it clips. So you should be setting the output to 0db or -0.1db. Then you get basically the same loudness (unless you can actually hear a 0.2db difference) but with no clipping.

Now getting the loudest parts of your track to hit that -0.1db ceiling with out overcompressing/limiting is a matter of mixing well and compressing the master carefully. I would consult some of the things people linked above for that.

But I can't think of any practical reason to ever set your master limiter ceiling/output above 0db.
 
StevenMcDonald said:
I don't know if I'm misunderstanding, but you set the output to 0.1db on the limiter/maximizer? I can't figure out why you would do that. If it goes above 0 it clips. So you should be setting the output to 0db or -0.1db. Then you get basically the same loudness (unless you can actually hear a 0.2db difference) but with no clipping.

Now getting the loudest parts of your track to hit that -0.1db ceiling with out overcompressing/limiting is a matter of mixing well and compressing the master carefully. I would consult some of the things people linked above for that.

But I can't think of any practical reason to ever set your master limiter ceiling/output above 0db.
Click to expand...


random conversation i had w a friend that made me remeber 10+ years ago but does anyone remember when DAWS didn't let you go above 0 dbfs? i remember anytime in pro tools going over 0 will give a huge noise burst. now it seems all daws have some sort of soft limting.
 
The first thing you must do is stop relying on mix bus plugins to do the heavy lifting when mixing. As was stated before, get your mix where you like it before you start futzing with your mix bus. An old audio production mentor of mine always considered the mix bus as sacred ground. Be gentle!

Spend some time and learn some basic fundamentals of audio mixing. It's a lifelong field of study; I'm always trying to make better mixes. It is a very different hat to wear compared to composing/writing for sure.

Good luck! :)
 
StevenMcDonald said:
So you should be setting the output to 0db or -0.1db
Click to expand...

Just to chime in here, set your tracks' ceiling not above -0.2dB due to intersample clipping (it's actually fine if you're mixing in the project, but if you're exporting stems either to mix in a different session or to send to an engineer, make sure they don't exceed -0,2dB and your engineer will love you).

If you're rendering to WAV and plan to have it that way (it's going to CD), your master can have a -0.3dB ceiling. If the output is going to get converted and compressed at some point (uploading to Soundcloud, storing as MP3, etc.), set your ceiling to -0.6 dB. Artifacts and noise get added in processing and you really, really don't want digital clipping.
 
gsilbers said:
random conversation i had w a friend that made me remeber 10+ years ago but does anyone remember when DAWS didn't let you go above 0 dbfs? i remember anytime in pro tools going over 0 will give a huge noise burst. now it seems all daws have some sort of soft limting.
Click to expand...

It was like that because it was a fixed point system.

Today you'll still be clipping if your master goes above 0, but tracks won't because of the floating point system.
 
I'm sorry, my mistake, in fact I really meant -0.1db on the output level, so that I would avoid any bad distortion, but apparently a -0.5db would be safer.

Anyway, now it makes much more sense to me that everything should be handled in the mix as much as possible before going to plug-ins or even master. I think what confused me was the fact that I always heard, or though that I heard, that we should mix in very low levels so we would have headroom to improve things anytime, so once I finished a mix in a very low volume I could simply use tools to increase the volume without clipping, and I thought that limiter would be the tool that would do that.

Well, what I did now: I cleared my mind, spent some time listening over and over again to the aggressive part of my track and even comparing it to the soft part, understanding how much difference they had between them, so I did some adjusts, bringing down faders. I was really impressed how loud the aggressive part was compared to the rest of the mix and I was not noticing before, maybe because I spent hours and hours listening to that way while recording? Anyway, then after re-balancing the mix, I pushed all the faders up. It was not still the final volume I was looking for, but maybe because the reason @gsilbers mentioned - Kontakt being in -6db as default. So then I used the maximizer with the default value from cubase (25%) and everything sounded in the volume that I wanted without any clipping/distortion in the loud parts. Maybe I could have pushed things up a little bit more in the mix, so the maximizer would even not be needed? I have to say thank you for the help until now, although I still have so much to learn and even maybe receive a confirmation from you guys if I am in the right path now.
 
lr11here said:
I think what confused me was the fact that I always heard (or though that I heard) that we should mix in very low levels so we would have headroom to improve things anytime, so once I finished a mix in a very low volume I could simply use tools to increase the volume without clipping, and I thought that limiter would be the tool that would do that.
Click to expand...

Well, complementing, maybe I can still work in low levels if I want to do so in order to be more flexible (e.g., when changing things during the recording sessions), but not necessary anyway, and the most important thing is really the balance in the mix and train the ears. Is that correct?

Although I still don't get why Kontakt is -6db, do you keep the Kontakt volume in this default volume? I checked some videos on the internet and I noticed that some people leave the way it is, and some people increase it. Wouldn't a higher volume than 0db there in kontakt make the instrument distort at some point?

Another thing, I was referencing my track against some tracks in soundcloud and then I went to spotify and listen the same track there, I noticed the huge difference in the loudness, at least for me. What would be the best to reference to?

And one more thing that crossed my mind about my track. I was using samples from completely different dynamics, even recorded in different places, some softer, some more agressive, and i was trying to mix them. It seems to me that if I wanted to have things as natural as possible (like a classical piece, for example), I would respect much more the difference between the soft part and the aggressive parts, not tweaking everything to sound more similar. My track is not a classical piece, so I really felt the need to place things in a similar range, but still having the dynamics between the soft and aggressive parts. That makes me reflect on things that I heard, specially regarding compression, where there are mixes that make everything sound the same, which is not good. Now I am reflecting here, sorry.
 
lr11here said:
Well, complementing, maybe I can still work in low levels if I want to do so in order to be more flexible (e.g., when changing things during the recording sessions), but not necessary anyway, and the most important thing is really the balance in the mix and train the ears. Is that correct?
Click to expand...

Yeah, the balance in the various tracks is the most important, but it don't hurt to think ahead. I have no set dB level for any particular track, but I do make sure the level is high enough for each track so that I don't run out of room with the fader levels. I use Reaper and there are many ways to do this, however, I do avoid normalizing, with Reaper there's no need for it.

Also I don't touch Reapers main MASTER, neither the fader level, which is always at 0.0dB, nor do I put FX on it. Instead, I create my own bus track for the Master, and I don't touch that either. I also have 2 Sub Master tracks, SubMast1 & SubMast2. SubMast1 is where I will have all my mastering FX, usually just an EQ and a limiter. SubMast2 is also where all my tracks and aux tracks are routed to. SubMast2 is routed to SubMast1.

Besides what I've mentioned above, I will also have a bus set up for my Reference material. Not only that, but I will aslo have a SubReference bus I call "SubRef" that goes to the Reference bus.

Before I go any further, I should explain that I've been an audio engineer for nearly 50 years and the high end in my hearing is gone. So some of what I'm explaining here is due to that. In the midst of this I will have a "Span" track where both the SubMast1 and SubRef busses will be routed to so that I can continually see how their spectrum analysis compares.

In summary, all the audio tracks go to SubMast2, which goes to SubMast1, which goes to my Master bus. All my mastering FX goes on SubMast1.

Why do I leave Reaper's main MASTER alone? Because Reaper has a Monitor FX, and I want to be able to hear both my mix and the reference mix through that Monitor FX for comparison. My Reference mix is also routed to the main MASTER.

So now the SubMast2 and SubMast1 relationship. SubMast1 has the limiter on it, so consequently where the level of SubMast2 is at will determine how hard I hit the Limiter on SubMast1. This allows you to adjust the actual mix throughout the time you are mixing. I also have a VCA track that will adjust the level of SubMast2, so in the end, I can automate not only the level of the mix, but also how hard I hit the limiter.

Now my Master bus and the Reference bus have their "mutes" grouped so I can easily compare them. Also since the Submast2 bus and the SubRef bus are routed to the Span analizer, I can see the comparison of their frequncy analysis at all times, which is very important to me.

As has already been mentioned, most DAWs use floating point math, so if you go in the red on a few of your project tracks it won't matter too much, but I do try to avoid that and will go out of my way to have that not happen.

Although I still don't get why Kontakt is -6db, do you keep the Kontakt volume in this default volume? I checked some videos on the internet and I noticed that some people leave the way it is, and some people increase it. Wouldn't a higher volume than 0db there in kontakt make the instrument distort at some point?
Click to expand...

That's been a headache for a long time. I will go into the "Instrument Options" and set them to the level that I think is best, but sometimes that don't work. If it's a real simple instrument with only 1, 2, 3, or 4 groups, I might adjust the Amplifier volumes, but only if I know the instrument well enough that it wont mess anything up.

Another thing, I was referencing my track against some tracks in soundcloud and then I went to spotify and listen the same track there, I noticed the huge difference in the loudness, at least for me. What would be the best to reference to?
Click to expand...

Personally I say download the tracks that sound good to you and set them up like I explained above. You really have to match the levels of the reference and your mix to really know what you are hearing. Pay close attention to not only the "peak" levels, but also the "RMS" levels, that's very important.
 
lr11here said:
Although I still don't get why Kontakt is -6db, do you keep the Kontakt volume in this default volume? I checked some videos on the internet and I noticed that some people leave the way it is, and some people increase it. Wouldn't a higher volume than 0db there in kontakt make the instrument distort at some point?

Another thing, I was referencing my track against some tracks in soundcloud and then I went to spotify and listen the same track there, I noticed the huge difference in the loudness, at least for me. What would be the best to reference to?
Click to expand...

-6 is so there's headroom, which you do still want when working digitally, despite 32 bit float having tons of headroom.

Case in point: If you use anything analog modeled (UAD, Slate, Soundtoys) you want headroom. They are generally coded to behave like they are receiving tracked/analog levels, not "in the box" levels.

(This is where the misunderstanding comes from about the belief that because 32 bit float has hundreds of dB of headroom it's fine to hammer on it and not pay a price...) Totally incorrect, you want to hit modeled plugins with the same respect you would if you were tracking into a piece of equipment...

the easiest way to see the need for gain staging in the box is by using a distortion or amp plugin... The louder the signal you feed either plugin, the more the distortion you get. If you're trying to reamp something and you want it clean it's pretty common you'll wind up bringing the amp plugins input gain down... the higher the input signal the more the distortion on anything that distorts or saturates...
Anything in your chain designed to mimic hardware should be treated with the same respect...

What this translates to is you want to aim for hitting modeled plugins at 0VU. (roughly -18 dBFS, depends on the calibration of metering used...) Anyway.. although you want dynamic range, things sum up beyond the level of any given signal... So a signal averaging -18 RMS and -6 peak would be a healthy tracked signal... That said, a mix where multiple tracks peak at -6 will likely sum up to over 0 dBFS... This is what some are referring to by using their DAW to trim the level using a gain plugin or trim feature if your DAW has one...
So although you might start with -6, typically you'll trim it back as the mix gets fuller and build the mix around levels where you have at least a few dB of headroom on the mix bus....

As far as getting your track to sound the same level you should get in the habit of doing two things:

A: reference mixing to something similar, as you go.
B: Comparing your level to the track's level using a proper level meter.

If your track sounds the same level but has a lower crest factor and sounds kind of distorted you can pretty much be sure you have a ton of rumble, or low end build up somewhere you want to cut... (Assuming your instrumentation and ranges are similar.) If your track sounds as loud and shows a higher crest factor it's possible you cut too much rumble. (Again range and instrumentation dependent...)

If you see big variances in either direction compare your arrangement first and foremost... If you have a very similar instrumentation and the reference mix is what you're going for then you might want to break out an analyzer and see what's going on way down at bottom of the spectrum, or at least compre where your mix differs from the reference on an analyzer...

Cheers and hope this helps...
 
Last edited:
synthpunk said:
Take a look at the Sonnox Limiter and Inflator demo versions. Still the top choice of engineers at Abbey Rd.
Click to expand...

These are both amazing plugins and I have these available for resale if anyone is interested (taking advantage of the topic being discussed.. sorry I don't intend to start commercial discussion). The only reason I'm selling these is I send all my compositions to a mixing engineer who uses outboard gear and I don't need to have my investment locked into these plugins
 
You must log in or register to reply here.
Top Bottom