Don Pearce wrote:
On Tue, 07 Nov 2006 12:12:27 +0000, Serge Auckland
wrote:

Dave Plowman (News) wrote:
In article ,
Glenn Richards wrote:
Arny Krueger wrote:
Back in the 1980s when people used to buy the LP and the CD of the
same album, play them both and compare the results, they weren't
really comparing the two media. Instead, they were comparing the
(generally quite separate) mastering decisions--EQ, limiting,
etc.--behind the two products, plus the particular characteristics of
their LP and CD playback equipment.
So what that posting is basically saying is that CD is capable of far
better quality sound than vinyl, but due to sloppy mastering (loudness
wars anyone?) vinyl generally sounds better? Because it hasn't been
compressed to within an inch of its life?
You can't generalise. Indeed in my experience this isn't the case - but
then I stopped buying LPs when I got my first CD player. This 'loudness
wars' thingie with CD mastering is relatively recent and mainly applies to
some pop releases.

(Bit of a vinyl fan myself actually...)
But then there are the inherent problems with vinyl which no mastering can
get round. So you're not starting from an even playing field.

Sadly, the loudness stupidity isn't just limited to pop releases. Over
the weekend, I was making some measurements, and decided to leave my
bitstream analyser connected whilst listening to music. I was astounded
at how many of my CDs regularly clip. Amy Winehouse "Frank" lights the
0dBFS light on almost every beat, Diana Krall Love Scenes clips often
per song, as does Norah Jones. My early CDs like Dire Straits keep a
good 3dB headroom, whilst Chesky's Valerie Joyce had 6dB headroom. I
finally disconnected the analyser before even more of my CDs upset me.

I just don't understand why these sorts of CDs need to be mastered into
clipping. I can understand a CD being normalised to 0dBFS, but that
would mean one hit at 0dBFS once per CD, or at worse once per track, if
tracks are mastered individually. There's just no excuse for it.

However, try as I might, I can't hear the clipping in the Diana Krall
and Norah Jones, even sighted, knowing when it takes place. The Amy
Winehouse is , however, very obvious.

S.



Do remember that lighting the top bit light does not necessarily imply
clipping - it is just another value, and if the signal isn't trying to
go beyond that, it hasn't clipped.

Pop recordings use heavy compression, and when this is done in the
digital domain it is quite possible to have sufficient control to peak
to the same value every time. There is no reason not to normalize the
result up to max level.

Still sounds like ****, of course.

d


Don, I agree that hitting 0dBFS doesn't necessarily mean clipping, but
displaying the waveform on a 'scope looks awfully like clipping to me,
on more than one CD. I estimate from extending the slopes of the
waveform before and after clipping that it can go some 2-3dB into
clipping.

My main complaint is not that's it's done on pop recordings, but that
it's done on jazz or other less loudness-concious material where I feel
it's completely unnecessary to compress, digitally or otherwise. It
didn't seem to be done in the eighties before digital signal processors
became available, (analogue compression was obviously used, but the
converted digital signal still had some headroom left) it just seems to
me that as they have the tools, they feel the need to use them even
where it is not needed.

Grumpy old man mode off.

S.