I am not reporting anything new here, but I had an interesting experience.  I sometimes use compression on the Master bus just to make CD copies for myself  to evaluate before the mastering engineer creates the final product.  (Naturally, I take that compression OFF before making the copy for the mastering engineer.)
 
On a recent track, I had EZMix (with a mastering preset) and Boost 11 in the master fx bin so I could see which one sounded best for this song.  Boost 11 worked best in this case, so EZMix was off and Boost 11 was on.
 
Out of curiosity, while the track was running, I clicked EZMix on so both units were on—and the track suddenly sounded incredibly loud even though the master bus meter should no increase in level.
 
Of course, we all know that compression makes things seem louder, but I had no idea how much louder until this little experiment.  Usually, the compressor/limiter makes the mix sound louder but also raises the overall level as metered.  I was shocked to find that the mix sounded twice as loud to my ears without any peak increase. 
 
In case you are wondering, it did NOT sound good—just loud.

konradh
 
In case you are wondering, it did NOT sound good—just loud.

Tip removed from To Do list.

bapu

konradh
 
In case you are wondering, it did NOT sound good—just loud.

Tip removed from To Do list.

I bet the waveform looked *interesting*

Just.  It was almost a square wave, like an Oasis song.  Still, since the peak was already was already almost in the red, I couldn't believe the difference in perceived loudness without any peak increase.  It gave me a lot of insight into the loudness wars.  It also made me realize that while I didn't think my music had a lot of dynamic range, compared to typical pop it must. (That last thing is a subtle point.)

Just because a thing *can* be done, doesn't mean it *shouldn't* be done.
 
Wait, what?

DAW level meters typically shows peak amplitude which doesn't accurately represent perceived loudness.  The ear uses a combination of RMS detection and frequency weighting in determining how loud something sounds.  The K weighting filter was created as a way of measuring perceived loudness. (Measured in LKFS/LUFS)  LUFS metering plug-ins are available:
https://www.google.com/webhp?sourceid=chrome-instant&ion=1&espv=2&es_th=1&ie=UTF-8#q=perceived%20loudness%20meter%20vst&es_th=1
They're mostly found in the broadcast industry and used to monitor levels between dissimilar program content, such as when a commercial is inserted in a TV show.

I think if you examined the WAV file in a hex text editor you'd see this:
 
FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF 01 02 01 03 01 00
 
(the last part is the fade out, with dithering enabled)