In article ,
Glenn Zelniker wrote:

Robert Morein wrote:

The last time I spoke to a NYC music "producer", which in that market means
a guy who actually does everything, he had very little job satisfaction. The
execs incessantly demanded that the result play loud.

Bingo. The labels often lean on the mastering engineer to
make it L-O-U-D, almost always at the expense of dynamics.
Peak levels contribute little to overall "loudness."
Loudness is conveyed by something more like average or RMS
program level. Peak level on a CD is fixed and the way you
make something louder is by boosting the average/RMS.
Because there's nowhere for the peak to go (due to the fixed
ceiling imposed by the format) we often get to the absurd
situation where the peak is only a wee bit higher than the
average/RMS. And why is this a problem? Because the ratio of
the peak to the average/RMS is what gives a sense of
dynamics. Ergo, "making it louder" with a constrained peak
by necessity decreases dynamics.

(I build digital mastering compressors and limiters and I
want to emphasize that digital dynamics processors don't
kill music; A&R guys kill music).

Or, as a guy on a pro sound forum put it:

"(T)he 'A' and the 'R' don't stand for 'production', 'mastering' or
'telling me how to mix my ****ing album'."

While "no compression" and "no EQ" can be audiophile ideals, in the real
world tasteful use of each can make a big difference.

Hope this helps!

Stephen