Loudness Wars again

[SY]

Just to show how it SHOULD be done, here's two outputs from Keith Howard's excellent Music Dynamics Analysis software. Two songs with similar instrumentation (male vocal, acoustic guitar). One is a commercial release, one is home-made. Which do you think sounds "better"? Which do you think sounds more "real"?

[kevinkr]

Call me clueless but I have no idea how to interpret those graphs..

Edit: Looking at it again more closely I think that the one on the right is actually the commercial release, and the one on left the amateur recording which I would also expect to sound better.. Am I right?

[SY]

You're clueless.

X axis is the level. Y axis is the amount of time the signal is at that level. Severely compressed stuff gets pushed upward. Here's some documentation on the software and graph interpretation.

[theAnonymous1]

I'm even more clueless than Kevin, but with your explanation it looks like the first one has a significant amount of headroom for peaks compared to the second one. But then I guess that's what this is all about, isn't it?

[SY]

Headroom for sure, but a wide range of levels- that's dynamics. Significant tails toward the low side as well- maybe the "microdynamics" or "downward dynamic range" so beloved of audio reviewers?

It would never sell. Too natural. It just sounds like a guy playing a Yamaha guitar and singing from a normal perspective.

[Pano]

That's why we have mastering engineers. To make recordings like that palatable. Real doesn't sell.

[DF96]

Unreal doesn't sell (well, not to me anyway). Lifting quiet passages above the noise level, and dropping loud passages below the clipping level, is mastering. Heavy compression to just below clipping (or just above, in some cases) is merely vandalism.