Mixing question: Optimal meter values - once again

[Harry_HH]

Probably I'm not the only one who has red that
you should leave enough headroom before mastering,
typically -6db.
I've always had difficulties to follow this rule (i.e. my mixing levels are
typically much higher) ,
therefore I'd like to ask about your practices:

1. What is your Peak level max typically at (after) the mixing stage?
2. Do you try to maximize the RMS/PEAK -difference - that is often mentioned
to be sign of a good mix?
3. Do you use just Peak level high or also Peak average and RMS when
you optimize the level in you mix?
4. At the mastering stage, do you often/always target the level to 0 db (Peak) or do you leave more headroom also after the mastering?

Harry

[BertKoor]

0) The '-6dBfs rule' is much debated and imho irrelevant. Any mastering engineer can watch the peak level in an instance or apply normalising to -6dBfs without doing any harm at all to the sound quality. He needs that bit of headroom for his EQs etc, but there are lots of points in the chain left to adjust the volume (which is one of the main tasks of mastering)

1) During mixing peak levels should always be below 0dBfs to prevent clipping in the DA converters. A good rule of thumb is to target for peaks no higher than -6dB so you have a bit left for safety. Some people put a limiter on the master bus during mixing, I think you better don't.
Another approach is to forget peak level but target at an RMS level of something inbetween -24 and -20dB. That should leave plenty of headroom, and the RMS level is a better indicator of loudness than peak level.

2) Don't look at the numbers but use you ears to judge what sounds good and what needs improvement. Your mixes don't need to be squashed already.

3) see 1)

4) Mastering is an art, best left to professionals. Fresh ears can spot better what needs to be done to make it sound awesome.
Anyway, nowadays you hardly see anything mastered with a 2-digit RMS level anymore. These 1-digit RMS levels simply cannot be reached with peak levels below 0dBfs.

[mkdr]

No big difference for the mastering engineer if there's empty space up there.. the real thing is that you shouldn't mix that "loud" in the first place. Just turn down your channels and kick up your amp. The extra air you have when mixing like this can translate to more air in the final mastered track in the end. Try it.

[kritikon]

I'll go with the headroom is irrelevant argument. You still hear that MEs want some headroom, but I really can't figure out why. All he's got to do to give headroom is put your track on a pair of channels, pull that fader down and apply his master FX over the master buss for all the headroom he wants. instead of just applying his master FX directly over the mix. So if an ME can't master a track that's presented to him at 0dBFS peaks, it just smacks of bone-idleness on his part IMO. Or as mentioned, just normalise it down, if he absolutely can only work in his own inflexible way. Reducing simple volume should be no problem to anyone these days.

I can also understand the desire for a little headroom...more along the lines of peaking below zero just so that you don't get any un-noticed overs. Most DAWs won't show small clips that are very quick. They will show them on the waveform, but not on the meters, and it's quite normal for most people to use meters as a mixing aid, rather than peering at a waveform readout after it's done. So an ME might stress mixing at lower peaks just so that you're less likely to send him a clipped mix.

Absolutely I can understand that if you're going to use an ME, then you probably don't want to overcompress your mix, and preferably don't limit it at all. It actually makes sense to not use any dynamics tools at all over your mix - let the ME earn his money. And odds are that he'll know how to use a compressor and limiter far better than you do anyway. But headroom is a non-issue.

So...

1. What is your Peak level max typically at (after) the mixing stage?

I generally aim for close-ish to zero, but TBH I don't really care what the peak level ends up at. I can change the gain when I master. And I mostly don't make quiet and subtle music, so I figure my quiet bits aren't that quiet and levels again don't matter.

2. Do you try to maximize the RMS/PEAK -difference - that is often mentioned
to be sign of a good mix?

No. I mix until it sounds good, and the RMS and peaks change drastically according to what style of music I make.

3. Do you use just Peak level high or also Peak average and RMS when
you optimize the level in you mix?

Depends if I'm doing it in separate mixing/mastering stages, or mixing and mastering in one go. TBH if you're mixing, peaks, RMS, whatever doesn't really matter. Again - if you're only going to do the mixing and seriously use a pro ME, then peaks and RMS is his concern, not yours. Your mix needs to be as good as you can get it, but RMS etc isn't that important at this stage. If I'm seriously going to seaparate out my mixing and mastering, then when I'm only mixing I don't ever look at figures on the master meter at all. Except to spot reds.

4. At the mastering stage, do you often/always target the level to 0 db (Peak) or do you leave more headroom also after the mastering?

Usually yes - or at least very close to zero. Used to be it was advisable to master at slightly under zero as some older CD players couldn't cope with lots of peaks at zero. AFAIK that hasn't applied to players for a number of year now though. No point leaving headroom after mastering at all, unless you are going to do more processing on it, or send it to an ME. If you're going to do more processing, then you've not finished anyway so the point is moot. Don't render it yet. If you're going to use an ME, I'd say don't do your own mastering at all anyway.

[lfm]

I can't see the problem with where own mixes end up in peak levels.

All proper routing would be to have all tracks go through Master track.

Make mix ready to go - but just lower fader on Master track x dB before rendering - and done.

If to leave to ME efter this, maybe good if not all maximizing in loudness is done. If mix is flat loudnesswise here, then is not very much to work with for ME.

Maybe talk to ME and tell that you want an ordinary mix for album, and one for radio compressed to death.

[BertKoor]

Maybe talk to ME and tell that you want an ordinary mix for album, and one for radio compressed to death.

The latter is absolutely obsolete. Radio stations play ordinary album tracks quite fine. The broadcast processor will take care of the mutilation of the dynamics, no need to do that yourself.

[Tarekith]

The only reason MEs want "about" 6dB of headroom is just to ensure the track is not clipping, it's that simple. It has nothing to do with outboard processing or normalizing, it's just a rough guideline to make sure the producer hasn't accidentally clipped somewhere and not realized it. In terms of workflow for the ME, it doesn't matter what the final level is.

[lfm]

Maybe talk to ME and tell that you want an ordinary mix for album, and one for radio compressed to death.

The latter is absolutely obsolete. Radio stations play ordinary album tracks quite fine. The broadcast processor will take care of the mutilation of the dynamics, no need to do that yourself.

There seem to be producers that are not sure about that fact then.

I also read that radio stations have a standard compression going on - since radio in stereo is maybe -35dB dynamics or close to that. But soldiers of the loudness war are not pleased with it - or they would not maximize their productions too.

When CDs were the common source material one could understand that broadcasting had to do something to reduce dynamics or more silent parts would be covered in noise.

I guess that a lot of material are not CD now, maybe even mp3 - I don't know. One would think some lossless like FLAC at least would be standard.

Even my mp3-player have settings for dynamics - how you want material to be post-processed sort of.

So no productions at all would need to boost loudness in the way they seem to nowadays.

[Harry_HH]

Thank you, good point all - and especally nice to read
KVR user practicies/opinions, these are more interesting than theories or
used slogans.
At least two "open" things I'd like to hear more: 1) The needs/rules of
radio & tv, I've also got comments demanding more headroom, 2) Mastering -
I also prefere to let this for "outside" professionals,if not for anything else
but to get "an other opinions and point of view" to the track/soundlandscape, listening fatigue is very common towards own work. But you don't have
always money/time/patiente and quite ofthen you end mastering your tracks, too.
How much do you master your work and which tools you use (single plugins, ready-made packages type Ozone, outboard gears etc...?

Harry

[BertKoor]

1) The needs/rules of radio & tv, I've also got comments demanding more headroom

You should ask them what levels they expect

Anyway, this might be a good read sidely related to this topic:
http://www.digido.com/level-practices-p ... ystem.html

[Tarekith]

In general really compressed music is going to sound worse once processed at the radio station. They have rather complex processing chains (Orban, etc) designed to level out program material in place, and hyper-compressed songs don't fair as well as songs that are left with a bit more dynamic range than what you expect from your average CD version (for instance). Usually submitting a really hot master for radio play to make it nice and loud can have the opposite effect.

Harry, try this:

http://tarekith.com/assets/pdfs/Mastering.pdf

MIght be a good primer on self-mastering for you.

[rocksure]

Personally I think it is best to work with sensible gain staging throughout the process of recording and mixing
http://rocksuresoundz.com/2012/03/07/au ... n-staging/

Mix things to busses, and compress them if necessary, but don't use a peak limiter on your stereo track. Just make sure that your transient peaks stay below zero. Most times the level of those peaks is well above the average RMS level of the track, and it's the average level that you want to try and get right, rather than obsessing whether you get close to 0 dBFS or not with your momentary peaks. As long as they are below clipping they can be compressed and limited during mastering.
There is no point in trying to make your mixes sound like masters, so leave yourself a bit of room and some dynamics. It is during mastering that the final loudness and limiting are best done.

[james_mcfadyen]

Just make sure you never 'red line'. Never