"Loud" seems to have become synonymous with "Quality" unfortunately. In the car coming home The Eagles "Desperado" came on and I chuckled thinking "People today would freak over mixing this... it is lacking both loudness and bass!" Terrible wretch of a song....

Sorry to hear that John. I agree with you. I mastered a Jazz album myself recently and I was very careful about this. It is possible to get the overall level reasonably loud but not smashing into flat lining or anythging like that.
 
Jazz needs to breathe and have some decent dynamic range. I listened to a few very nice contemporary Australian albums before I did it and measured the average rms level and it was nothing like -6 or -7 that I have mentioned above. More like -12 to -14. I just did the same.
 
The problem is that the wrong guys are doing Jazz mastering. They do not do their research and listen and compare to really nice mastered Jazz albums before they do it that is the problem. They make the mistake of just applying the same rock heavy approach to everything and they think that is good.
 
Many mastering engineers are pretty stupid actually and do not know a lot about mastering. It is important before you select a mastering engineer to get some of their work and check it out first. A good thing to do is to import their finished tracks into an editor and SEE what is going on. Often it is very revealing.
 

Loudness wars will never end, unfortunately, until a governing body steps in. There's bad in all of that. But, at least we can admit that recordings should be Mastered with less priority on loudness. I miss great sounding releases that you could crank up and would have a dynamic impact second to none.
 
My best advice is to just keep away overs, and mix loud enough that your mix hovers around or just above the middle of the meters. You can push up a little more to taste, but keep out of the red, if possible. If you've got one or two peaks that clip, just drop in a limiter to your Master buss, don't dither, don't change bit-depth or sampling rates, and DON'T adjust the Threshold or Peak! Let the limiter work on ONLY the clipped peaks and nothing else. 
 
A great Mastering Engineer will give it a listen and tell you to make an adjustment. Many of them are enthusiastic about their jobs and want and will work with you. That's what I like about the good ones. They know why they get paid.

Jeff Evans
Sorry to hear that John. I agree with you. I mastered a Jazz album myself recently and I was very careful about this. It is possible to get the overall level reasonably loud but not smashing into flat lining or anythging like that.
 
Jazz needs to breathe and have some decent dynamic range. I listened to a few very nice contemporary Australian albums before I did it and measured the average rms level and it was nothing like -6 or -7 that I have mentioned above. More like -12 to -14. I just did the same.
 
The problem is that the wrong guys are doing Jazz mastering. They do not do their research and listen and compare to really nice mastered Jazz albums before they do it that is the problem. They make the mistake of just applying the same rock heavy approach to everything and they think that is good.
 
Many mastering engineers are pretty stupid actually and do not know a lot about mastering. It is important before you select a mastering engineer to get some of their work and check it out first. A good thing to do is to import their finished tracks into an editor and SEE what is going on. Often it is very revealing.
 

Excellent points Jeff. I for one view jazz with the same respect as classical music. What bugs me is who gave them the right to screw up really great music so it could sound like a pop song? 

I also bought an album from the Airplanes because I liked the song "A Summer Daydreams Part 1". Its a great song that has been so mashed that I had to bring it into Sonar and "fix" it. My version is very listenable. 

 
 

Something that I like to do (I wonder how many mastering engineers do this) is to organise my mixing setup so that I can compare the pre mastered mix to the totally mastered mix at exactly the same volume. ie I lower the mastered version down by the amount I have increased its overall volume.
 
What this does once you remove the level difference is really expose what the mastering is actually doing to the overall sound quality of the mix. It is very revealing. Any distortions due to any of the mastering stages is very exposed now. It is so obvious. Often it can sound worse.
 
Sure the EQ and compression are doing their thing but what I am listening for here is any distortion that may be added especially in the limiting process. It forces you to rethink and continually back things off until you get the two things sounding great and clear.

mike_mccue
I've heard all the explanations about why the mix should be peaking at *insert number here* dBFS and I'm still waiting to encounter one that actually makes sense from either a technical or aesthetic standpoint.

I think some of it is I don't want to have to reduce something in order to allow for processing and then kick it back up again. There's nothing "magic" about -6dB, it's arbitrary. But if someone gives me 6dB or so to play with, that will accommodate most processing I might need to do. If they give me no headroom, then the first thing I have to do is reduce the level anyway to create headroom. Of course I can do this, but if I ask for the person doing the mix to pay attention to levels, it can have other benefits as well. Not all digital metering is calibrated equally, and those raised on analog sometimes thing "hey, if it goes into a red a little bit every now and then, it's not a problem." If they mix to -3 or -6 or whatever, that won't be an issue.
 
As to inter-sample distortion, regardless of whether someone can hear it or not, I just don't think it's good practice to let a master out the door that's known to have distortion. It's also possible that isolated events are not "heard," but are "perceived." There's no way to prove that any more than people who swear that only music recorded at 384kHz sounds any good can prove their point. But, if you can check for and make sure there's no inter-sample distortion, then you don't have to worry about whether it will matter or not.

Jeff Evans
Something that I like to do (I wonder how many mastering engineers do this) is to organise my mixing setup so that I can compare the pre mastered mix to the totally mastered mix at exactly the same volume. ie I lower the mastered version down by the amount I have increased its overall volume.

Agreed, it's a great technique. Wavelab 7 actually introduced a feature that does this - I think it's called "smart bypass."

brconflict
Loudness wars will never end, unfortunately, until a governing body steps in. There's bad in all of that. But, at least we can admit that recordings should be Mastered with less priority on loudness.

You'll be happy to hear that when acts ask for a master that's as loud as possible, I do the version they want to hear but then I also do a version I want to hear. Most of the time when given a choice, they'll take the version with more dynamics. Not all...but most.
 

Yes, by all means let's have the loudness police check our mixes and conform them to an accepted standard.  Just hope that standard embraces your own tastes and doesn't change over time.
 
And one can hope that intermodulation distortion will cause deafness in the millions that prefer it, along w/ the bands and producers demanding it.  That will leave a smaller yet more tasteful market.
 
@