Setting Peak Limiter Ceiling at -2 Master fader

[Tim]

When I go mix for a client, usually the first thing I (have to) attack is lowering the levels of everything, which can easily mess with the existing mix, as Cont. 7 changes aren't the same with all instruments.
It's so easy to end up with a project with everything louder than everything else, what with MIDI track faders defaulting to 127, and VI sounds being normalized to the max... like I need a solo clarinet at full digital scale (it's like the sample libraries have volume wars like the record companies).

I's nice that Kontakt has a preference to default to -6db.

It'd be nice if DP had a preference for MIDI faders to default to a specified level - something with some headroom, like 90 or lower.

It's not about a Proteus and a 1080 anymore... or a 16 bit dat machine.

Headroom is your friend.

[davedempsey]

mmmm....headroom

[b2]

Hi.

If you're hearing distortion in your mix, then the mix is too loud to begin with. You need to lower the 60 channels. Now, you don't have to lower each one individually. You can buss like-minded elements to a series of auxes and then lower the auxes together. DP has a FANTASTIC buss routing system---very flexible and lots of options. And, this is very useful because it makes it easy to produce stems/splits later if necessary. You should attempt to build a mix template that allows routing of all elements through a set of final auxes, and then changing the overall level of the mix is VERY easy----no brain power required.

The BIGGEST common mistake made in mixing is starting with elements that are too loud that lead to a mix that's too loud, which then ultimately limits your options. I know it can seem weird to mix with such soft elements, but as you're seeing, loud elements add up to being too loud, and then you're trying to put a band-aid on a broken leg. This is why 24 bit audio is so useful. 24 bit audio allows you to use much softer sources without worrying much about bit-quantization artifacts. The answer to the problem of a mix that's too loud is easy----lower the elements. Turn up your monitors to make up for the lower volume. I have a general range where my monitor volume sits when I'm mixing, and the knob is set WAY louder than when I'm mastering. Which leads to my next point----->

Regarding your other question---why not master while you mix?: Mixing and mastering really are two separate processes. When mixing, you should concentrate on one thing-----making a GREAT mix----not a loud mix. Starting softer, with softer elements gives you the option to do WHATEVER you want to the mix to make it sound better. You can really make a snare pop over the mix during the middle of a mix because YOU HAVE ROOM TO DO SO. I'll bet you currently find yourself saying "I wish I could make such-and-such-a-thing louder, but the mix is peaking...dang"....so you try making a couple other things softer and then you can maybe push the element a tad more which isn't really what you wanted, and isn't making a great mix. Or you just throw a limiter on to hold it all back....You're just making compromises based on the fact that it's simply too loud to begin with. If you start lower, you never think about clipping and then can make whatever good mix decisions you want. Again, you're trying to make a great mix. Leave the make-it-loud part for mastering when you can concentrate on it. And again, bussing everything in the mix through various auxes makes adjusting the level of the ENTIRE mix absolutely painless. In fact, you may find at the end of the mix that you still had more room to go louder, so with the auxes you can RAISE the level of the mix easily.

Once you have a mix you like, record it down to a master-mix stereo file, THEN you can go make the slammo-jammo version with the mastering process, but at least you have a nice, open, big sounding mix to start with, not a squashed overly compressed, distorting mix that doesn't leave you any options during the mastering. And you really need to approach mastering with a clear mind-----meaning you're done with the mix and you're not distracted with the 9 billion things you worry about during a mix. Now, you just have a stereo file in front of you and you're using every tool in your arsenal to make it sound as GREAT as it can----mastering is more than just making something loud. This might mean brightening and limiting and compressing and experimenting with pumping basses, scooping away muddy bottoms, adding saturation/distortion, changing stereo width and who knows what else, but now you're applying a bunch of tools simply to make that stereo file sound awesome. You'd be surprised how different a file is when it's done being properly mastered---more so than just being louder. It's actually shocking sometimes. That's because it is it's own process. Now, I'm not trying to freak you out and say you can't master, and that you should just send it to a mastering house......actually quite the opposite. I think mastering is fun and something people should experiment and learn for themselves. It's the final fruition to all the hard work! Mastering is all about experimenting. Every cue of music has it's own challenges---there's no single textbook way to master every cue. But as the content producer, you know what you would like it to sound like and if you learn what the tools can do to a signal, then who better to bring it to the final fruition? People always ask us who masters our music because the cues sound outrageously loud and over the top.....the answer is---we do. But the music didn't start that way. The original files are not loud at all. And sometimes it takes a bit of debate and a few go-arounds to get the sound right.

Cheers.
-gabe

Excellent advice! Thank you Gabe! I'm going to do it your way. Headroom yes! Options yes! Hurray!

[stephen1212b]

Great advice not to worry about maximizing levels until mastering. This would be an example where saving the mix as 32 bit Floating Point could really give you the most options by preserving the full 24 bit dynamic resolution all the way into the mastering where the dynamic range issues and final balance can be addressed.

[Kubi]

Unless you are overloading plug-ins somewhere before in the signal flow, you should be able to simply lower DP's master fader until the distortion stops, thanks to DP's internal 32bit float processing. Or you can insert a Trim plug at the top slot of the master fader.

If you do have distortion earlier in the chain, you could also insert a Trim-plug at, say, -6dB in the last slot on each channel. (If a single channel is overdriving a plug-in all by itself, insert the trim plug in the first slot instead, but be careful if on that channel you are using compressors or similar plugs that react to input level - you'll have to re-adjust the threshold accordingly.)

Either way, this could be done fairly quickly - simply insert one Trim plug, set it to -6dB and keep option dragging to the other channels. However, if you are using any buss compressors, saturators or anything else that reacts to incoming level, you will indeed change the sound of the mix a little by doing this.

All in all I wholeheartedly agree with the advice of mixing with plenty of headroom, you can always boost at the end.

[cloudsplitter]

I like to start my mixes with the hotest component reaching about -5 -7 on the vu's..then I work from there...I will also make to stereo mixes to master from...I use Ozone..so I will take one of the two master mix tracks and get it to sounding as good as I possibly can...and then listen to it over the next few days periodically writing down everything that I think may make it sound a tad better....then I set up ozone on the second master mix and compare my first mix with the second...I am an amateur though so I still do everything the hard way...LOL but, having said that, I have never had any complaints about my mixes/masters

[Gabe S.]

Excellent advice! Thank you Gabe! I'm going to do it your way. Headroom yes! Options yes! Hurray!

hehe yay! I'm glad you found something useful in my endless ramblings... Check back in and let us know how it goes.

btw, another small piece of advice------invest in great plugins. Your "sound" can only get as good as your plugins. In my humble opinion, the older DP plugins are not very useful for high performance use. The newer ones like the MasterWorks EQ are VERY nice. But older ones like the MW limiter or eVerb-----oooh, my advice is stay away from those and find some nicer 3rd party ones. In the year 2009, there are plenty of options, and if the budget is tight, you still have quite a few nice choices.

Cheers.
-gabe

[slicerecords]

a great limiter with a -0.2 ceiling won't give you inter-sample peaks. I have mastered numerous records over that past 8 years using various limiters and have never encountered that as an issue. Whoever is experiencing that issue needs to check out the limiter they're using. Also, there's a nice inter-sample meter made by SSL for free. Called the SSL X-ISM, it's an AU plug. Comes in handy.

Also, you should be checking what RMS level you're limiting to, sometimes people just push it too hard. It sounds like you're just clipping your master buss though.

[Splinter]

Also, you should be checking what RMS level you're limiting to, sometimes people just push it too hard. It sounds like you're just clipping your master buss though.

Release times can also make a difference. If they are to short you will certainly distort the transients as the limiter snaps back too quickly. You can see this in the waveform. If you lengthen the release time a bit, the limiter will breathe more and not "breakup" the transients as much. It's always a matter of finding the sweet spot because if the limiter release times get too long, it will pump like crazy and sound mushy. This, my friends, is what we call SKILL. Something a preset will never have

[davedempsey]

This is my go to mastering limiter ... not cheap, is from the big W, but nothing (in my experience) comes close to it.

[djc]

+1 for Gabe's suggestions. Seems to me many folks mix loud because the usual final produce is loud. Mastering-very important intermediate step!

[Armageddon]

I think a lot of distortion (and not just in home-made .mp3 masters, but in commercial ones, as well) comes simply from the fact that it's just too easy to make something too loud now, and both home studios and commercial ones have the mentality of "louder is better". You can load up a lot of waveforms from stuff ripped from CD and see absolutely black bars with maybe a -0.1 dB notch. You can be well below digital zero and still distort the hell out of your stuff.

If you're worried about overs, either do what most people have suggested here and mix with a lot of extra headroom (I prefer to go about -3.0 dB, since I'd rather not have all my volume wholly dependent on my mastering processors, which can introduce a lot of hyped frequencies from your mastering plugs that you may not want), or, failing that, stick a limiter on the end of the chain of your master output, set the threshold down to about -0.3 dB and your output to the same. Turning your fader down may not prevent overs -- let's say, for example, that you set your master fader down to -0.3, but there's a spot in your piece of music where the mix peaks at +1 dB, you're still gonna go over, just not as much. Having a brickwall limiter at the end that does nothing but squash spikes over your intended output level will allow you to output at your intended level without altering the overall sound and dynamics beyond the stuff you don't want. If you have it, the best limiter for the job is Waves' L-series, very transparent.

[stephen1212b]

My favorite was an album by a female bass player where the bass was flat line for over 100 samples in a row, granted at the - 0.01 db level so that a digital meter wouldn't show an over. I guarantee that that album sounded very different on every system that played it!

[Rhetro]

Hello.

I'd figure I'd jump in here since I couldn't find another place to post this. I figure it's close enough. Bare with the history, please.

I've been working on a mix for a punk band now for quite some time. Being punk musicians, of course they wanted everything louder - now, what would one expect!!?? I've been giving them mixes over the course of time, and they're finally satisfied with the product (mixes).
Everything being loud has been just to placate them in the auditioning of the mixes (they are hard of hearing -can't imagine why). To make it louder all I did was place the MW Limiter on the master bus and tweak from there. I don't get any clipping when I play back the mixes in DP7, but when I bounce to disk, once the bounce progress bar is done doing it's wipe, there are some guilty tracks that went over. Is this what people are referring to as inter-sample clipping? Or is this a result of some of the plugins having a reaction to the bounce process? I called MOTU and they said not to worry about it. But that doesn't sound right to me; I'm of the mindset that red lights should never be ignored.
However, I'm now ready to start the mastering phase, and agree that it's time to shut off the limiter, pull back the levels, and bounce. I figure that this 'more headroom' aproach will aleviate the spikes, and I can then move on to the mastering phase.

So to recap, I got the mixes sounding good. I'm just getting red lights on the bounces. I want to make sure I don't blow it at the mastering phase.

So pull back the master fader until I got nothing going over -6? Sounds like headroom city to me! This is what I want. However these punk dudes are always going to want more. Especially when they say how quiet things have become!!! I'll worry about this later.

I'll squash and maximize at the mastering phase (using Ozone btw -can't afford the other stuff).

Unless my logic is off (woudn't be the first time)
Do I kind of got it?

Thanks in advance!

[stephen1212b]

First inter-sample overs refers to the analog signal constructed by the D to A conversion filters. The MOTU metering is digital and does not take this into account. SSL offers a free plug called X-ISM that will calculate likely inter-sample overs. In reality every DAC will vary. You are correct that these overs can be up to 6 db.

If you are seeing red on the MOTU meters you have digital overs. If this is only when checking the bounce to disc try mixing with the exact same settings in real time to a stereo track. If there aren't overs during recording then there shouldn't be overs on playback.

I personally feel however that to have compressed the tracks in order to make them appear to sound as loud or louder in comparison to other CDs and then to simply lower the volume to -6 is the wrong approach. The CD will now loose the loudness wars and you really aren't taking advantage of the headroom in a useful way.