Dmitry

You may want to set the peak limiter on your finalizer at - 0.5 dB so that you know you never actually hit 0 dB which will cause distortion.

Makes use of the bits but not using more bits is more accurate..but even then to a point.

Experiment.

Run a 96K/24Bit file of silence and run one of music at any level and compare with the same running time. Bits used! (Figure)

The goal from a mastering engineers perspective (ME) is to get a premas. at a solid -3 to 5 peak and a RMS running around 15dB less than peak.

It depends on track, genre' and a lot of other things..I am working one now that is stomping ass and I am running (premas) of peaks on the 2-bus of -3 and RMS of around -21.

When I master this..the very top peak of the peaks will be at 0.05dB and I expect the RMS to lay close to -18.7

This is how I do it.
This is not how I do it...
I have some finals (sent to me) at 0.0dB Peak and -2.1 RMS

I think compressing at any level where RMS (with music) above -15dB is really heavy compression

I use the 0dB to -16RMS as a reference generally (again depending on the tune) as the most scrunched I will do it.

Depending on tune has a lot to do with how much of the tune is loud chourus work verses intro, balad..etc..

I am a dynamic indivigual

Bill, From a ME's perspective how does the premastered RMS requirements change when mastering something really heavy? ie. when the end result is supposed to be "needles go to 11 and never move" type stuff- I myself hate this kind of "music" and find it fatigueing at best to listen to, but I can't seem to escape the fact that I have clients that want this as an end result. With other, more dynamic material it seems like -18 RMS or so pre mastered works pretty well. It's probably inexperience on my part but often I have a hard time getting my RMS lower than this in a mix and still feel things breathe a bit. I guess the question is does this kind of level make it problematic for you the ME to get the desired end result?
Bob Green
Area 51 Recording Studio

Bob, It really does make things problematic. I actually have to restore dynamics before I can do a really presentable work.

What should happen (maybe many can enforce this)

Make a scrunched mix for the one off for the artist...etc..

Make your dynamic mix with peaks not going past -1.0 (I even like -6dB peak for prosterity)

I just looked at a metal project I did that really works for everyone and I ended up (on a track that has a lot of sustained) with -0.1 peak and RMS was -15.48. I had no one complain of the level of the whole CD and it still has a bunch of punch.

Schrunched mixes render the mastering engineers work rather useless. It is a chore to redynamisize the mixes like that and get them back to some form of organic expression.

Hope this helps!

It sure does help Bill. Like you I'm not only a huge fan of dynamics, but to me they are really the essence of the listening experience! A lot of the hyper-compressed, "louder is better" stuff that I've heard in recent years not only puts my teeth on edge, but after a while I swear it's like listening to an alarm clock that won't shut up. To me loud means nothing if there isn't something else (at the very least not-as-loud ) to reference it to. Thnx very much for your wisdom.
Bob Green
Area 51 Recording Studio

I use the term Signal to signal ratio.

I also use the term Floor to threshold ratio.

And Floor to peak ratio.

Looking at my meters (the main set I use in mastering) are true VU meters because I am so comfortable with their ballistics. They go from minus 30 to plus 3. If they are hovering in the minus 7 to -4 range..that is pretty cool. I have them tuned to my liking. I have the calibration of them set to my system. A peak over the top will almost certainly be an overload of the digital recording.

Now looking at the digital meters -144 to 0. I look at the "fall markers" and I like to see the minus 30 to 34 light up at times. The meters look rather rediculous when only a very small fraction of them is actually moving..but really, dynamics must sound natural and organic rather than sterile and stale.

The worst thing about the super compression is that their are very few mastering engineers (of the ones that do this) that realizes what happens to the EQ when all that schrunching is going on..yes the 3K rises dramatically, the low end is rolled off and the top has no place to go and is masked. This is the problem as well.

How do I decompress a compressed mix?

It is done by using WILD eq settings at a lower gain to find something that smells not of death reek and to make those sounds (give them CPR) live again. The another process of WILD eq. to make another range organic.

Think of your EQ as your dynamic range expander, compressor. IF you use minimum phase shift eq's and have monitors that are flat to below 20HZ, you can work magic. I do essentially all of my dynamic prosessing using the equalizer.

Once I get the track back to some soret of reality (hundreds of edits in fine tuneing and pasting and sorting through passages) then It is time to truly master the project.

It is something I enjoy when I compare the final with the prefinal..but really , Cats *must* lay off the 2-bus compression if they want a real track with real sounds and let your mastering engineer do the blending and put the final brushstrokes on the project to make it the real deal.

Thank you for providing the oppurtunity to state something that has been beat to death....but rarely have I seen anyone nail down the culpret. It is exposed right here.

Satisfy your client in mix by making a compressed 2-bus for them and the A/R guy..and then take the project the right direction. I have had no one complaining doing it this way.

as "anpther groups" has mentioned... lt us blame the MASTERING MAFIA
LOL

I set my mastering output to -0.05dB and the mixes peaking at -0.5db. The music brathes.
:p