Limiter on rough mixes

[losthighway]

I'm sure I'm not the first to bring this up but there hasn't been a ton of conversation about it:

When rendering a mix, rough mix, first pass, second pass etc. for a client to listen to, or just to enjoy while driving do you use a limiter or not?

Reasons for:
Obviously, get the overall volume up somewhere reasonably close to a commercial release so a client isn't thrown off. Also to get things above your noise floor when driving an aging vehicle on a poorly maintained piece of American infrastructure. Get a little extra 'oomph' so clients can have a bit more of a visceral experience when celebrating the completion of tracking.

Reasons against?
How much is it changing the balance and presence of the instruments? Is it picking off transients from drums that should be more compressed or mixed lower in the first place? Is it shifting the overall tone of the mix, enhancing lows or mids? *In other words is it informing the clients revision notes, or your own perception of progress in a way that isn't very accurate?*

Obviously there's the how many dbs question, and there's no point in printing a brick for critical listening that will later be mastered with finesse. But then again if you were going to go so light on the limiter as to only get another db of volume for the reference mix, you might as well not worry about it anyways. Please share your thoughts and practices.

[vvv]

I do it, as said, to get closer to a standard average volume. I'll typically take about 4dB off the top and bring it up to -.3 so if I convert to *.mp3 for a mini player or email it doesn't go over 0.

I still use CEP2.1 and it's known for the native transparent limiter - it's really good.

[A.David.MacKinnon]

I do it too. I use the Massey L2007. I use it to get the level up with as little gain reduction as possible. I find it's pretty invisible if it's not working hard (just taking off the stray peaks). Once I'm into the real mix I'll sometimes crank it up just so I have an idea of what the mastering process might bring out. Those versions are just for my own benefit and don't get shared with clients.

I got tired of explaining why rough mixes and unmastered finals were quieter then commercial releases.

[drumsound]

Sometimes I normalize to get the level up when I'm sending something out for evaluation.

[digitaldrummer]

for most projects I do, I'm typically mixing and mastering as I go along - so if the limiter becomes part of the sound, so be it. so yeah, even for rough mixes I start shaping the sound to try to get it close to what the end product will sound like.

[lyman]

Only if there are crazy transients, if not then I just normalize. I am not a commercial operation though so these recordings are just for me/bandmates. Could understand how those with clients have different needs.

[MoreSpaceEcho]

Speaking as a masterer.....I understand why people do this and it's fine, just a couple things:

1. When you're cranking the mix up with the limiter, just make sure you're cranking it up to something reasonable. Sometimes people send me the 'quick master for clients' versions along with the unmastered mixes, and they're crazy loud, like someone just slapped a limiter on and didn't reference it to anything.

I know none of you nice people would do such a thing. Just sayin'!

2. I know people like to put a limiter on to "see how the mix will hold up after mastering" and again, this is fine, BUT just keep in mind that your masterer is probably not getting all the level from limiting. I'm rarely doing more than 1db here, and almost never more than 2. Once in awhile there'll be a record where a bunch of limiting is the best way to go, but most of the time I'm getting most of the level before the limiter.

And just on a related note, if you like to slap a mastering preset in Ozone (or whatever) on the mixes for clients, please, I beg you, take the time to a/b it, level-matched, with the unprocessed mix. I say this because sooooo many times I've heard the "quick master" of something and it sounds CRAZEE, artificially wide, weird hollow phasey bass that makes my ears go crosseyed, and usually too much 4-8k.

I listen to that and think how the hell am I going to do anything with this and then I listen to the dry mix and always, every time, it sounds just fine. The quick masters usually sound like the engineer slapped ozone on at the end of the day when they were super tired and they didn't really listen to what it was doing. Which, I get it, we've all been there. Again, I'm just sayin'.

[Recycled_Brains]

I got tired of explaining why rough mixes and unmastered finals were quieter then commercial releases.

This is why I do it. Barely any reduction though. I don't feel comfortable at all with peak limiters, so I just don't let it do much. Massey L2007 is the one I use.

[losthighway]

The quick masters usually sound like the engineer slapped ozone on at the end of the day when they were super tired and they didn't really listen to what it was doing. Which, I get it, we've all been there. Again, I'm just sayin'.

Thanks Scott. This is exactly what's happening sometimes when at the end of a long day someone says, 'Yo we have 10 minutes can we bounce this real quick and get some files' and with my fried ears I think, 'I'll throw a little L2 Maximizer on there so they can really hear it rip' and then later I think 'That might be a totally inaccurate reference mix.' At which point some of the weird artifacts you are talking about could be present, and then the clients list to begin mixing will be based on the last 10 minutes of bad decisions, and not the previous 10 hours of careful work.

[A.David.MacKinnon]

The quick masters usually sound like the engineer slapped ozone on at the end of the day when they were super tired and they didn't really listen to what it was doing. Which, I get it, we've all been there. Again, I'm just sayin'.

Thanks Scott. This is exactly what's happening sometimes when at the end of a long day someone says, 'Yo we have 10 minutes can we bounce this real quick and get some files' and with my fried ears I think, 'I'll throw a little L2 Maximizer on there so they can really hear it rip' and then later I think 'That might be a totally inaccurate reference mix.' At which point some of the weird artifacts you are talking about could be present, and then the clients list to begin mixing will be based on the last 10 minutes of bad decisions, and not the previous 10 hours of careful work.

That's why I like the Massey L2007. It's really easy and quick to dial in and the metering is decent. I can safely set it after a 10 hour day and be confident that it's just grabbing the big peaks not crushing the shit out of everything or adding any unwanted "finalizer" type effects (widening or whatever)

[Snarl 12/8]

I haven't recorded strangers in 30 years, but back when I did I had trouble getting paid if I gave my clients too good of a rough mix.

[MoreSpaceEcho]

'I'll throw a little L2 Maximizer on there so they can really hear it rip' ...

If it's just a limiter and it's only doing a couple few db reduction it's likely fine, that's not going to change the perception/perspective of the mix too much.

It's the 'mastering preset' stuff that makes shit get all bonkers. I only mention it because I feel like I've heard the same sort of artifacts on lots of 'quick masters', that weird phasey bass, etc. It's like it's "mastering preset #1" in whatever DAW, so people are like 'well this must be good' and they put it on after mixing for 12 hours when no one can hear anything anymore and hey alright it's louder and brighter now! We can go home and eat something finally....

[Nick Sevilla]

I use the Waves Ultramaximizer L2 across the mix buss normally.

I set it to threshhold of -1.0 dB, and an output of -0.2 dB.

This accomplishes the goal of not allowing transients to overload the mix print, avoiding distortion. It also avoids coloring or messing up the mix, as it only makes it louder by 1.0 dB, and normally, my mixes peak at about -3 dBFS maximum, -6 dBFS usually.

It also accomplished the goal of not ever allowing the mix file to reach 0 dBFS, or the maximum level.

I researched this years ago, and many of my peers agreed that making CDs and mp3s required that NO maximum volume should
be allowed in the print file, to help avoid problems with the encoding processes of both compact discs and mp3 files.

Further, before the current Fraunhofer mp3 encoding technology, the company which has the best mp3 encoder, currently used by companies such as Avid (maker of Pro Tools) and other companies, including consumer grade level mp3 converters and encoders in consumer computers, all of us professional engineers tested and agreed that making a full scale volume mp3 file would often result in an inferior, distorted mp3 file. Thus, at least until mp3 encoders got good enough to handle full scale input files, we would output a mix file at -3 dBFS maximum for the mp3 encoding process.

Today, I am able to output a mix file all the way to -0.2 dBFS with no fear that current mp3 and other encoders will distort it. This said, it is always helpful and necessary to double check your work before handing it to your client. Take the time to listen to your final output file or product. Or eventually, it will bite you in the ass.

[MoreSpaceEcho]

I researched this years ago, and many of my peers agreed that making CDs and mp3s required that NO maximum volume should
be allowed in the print file, to help avoid problems with the encoding processes of both compact discs and mp3 files.

This is true, but there's overs on loads and loads of CDs and they sound fine. I'm talking big time, famous, legitimate barcoded records. I've analyzed a ton of them and overs are more the rule than the exception.

[Nick Sevilla]

This is true, but there's overs on loads and loads of CDs and they sound fine. I'm talking big time, famous, legitimate barcoded records. I've analyzed a ton of them and overs are more the rule than the exception.

Yes. This is because, especially years ago, the loudness wars and the clients preference basically overruled technical considerations. So it would follow that you would see this a lot.