Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!decwrl!sgi!karsh@trifolium.esd.sgi.com
From: karsh@trifolium.esd.sgi.com (Bruce Karsh)
Newsgroups: comp.dsp
Subject: Re: 180-deg phase shift -- A study in Usenet technical advice
Message-ID: <107564@sgi.sgi.com>
Date: 2 Jun 91 01:10:13 GMT
References: <51184@prls.UUCP>
Sender: guest@sgi.sgi.com
Reply-To: karsh@trifolium.sgi.com (Bruce Karsh)
Organization: Silicon Graphics, Inc., Mountain View, CA
Lines: 65

In article <51184@prls.UUCP> max@prls.UUCP (Max Hauser) writes:

>Preserved in its entirety, this exchange makes an instructive example
>about the Usenet, especially when you see the tones of decisiveness and
>authority that so many of them took.  These exchanges are so neat I will
>try to think up some more innocent questions, have them posted, and
>watch what ensues.

It can get much worse than this.  Here's one from rec.music.misc that
makes the whole 180 degree discussion seem profound!

In article <4571@meaddata.meaddata.com> dedek@meaddata.com (Mike Dedek) writes:

>  Now, I know that CDs have some major advantages over vinyl in terms
>  of wear -n- tear, and background noise due to dust, etc.  My friend
>  and I came up with a theory why digital music will never reproduce
>  sound as well as analog devices.  Since music is made up of sound
>  waves which range over a *continuous* spectrum of frequencies, it
>  seems that digital music will suffer in any frequencies *in between*
>  the ones recognized by the digital equipment. 

In article <4306@csccat.cs.com>, larry@csccat.cs.com (Larry Spence) writes:

> This is quite true.  The analog-to-digital converters are calibrated to
> pick up the frequencies of _correctly tuned_ guitar strings, piano keys,
> etc.  These are reproduced unerringly.  However, it is seldom the case
> that instruments are in perfect tune.  If, for example, a A is 441 Hz
> rather than the standard 440, a digital recorder will not reproduce it
> at full volume, and it may be slightly "muffled."  This is the key to
> the oft-heard complaint that "grungy" music (e.g., Sonic Youth) sounds
> too thin and not very powerful on CD relative to LP.  In the case of
> SY, they often use non-standard guitar tunings, which, since they do not 
> go to the expense of using specially-calibrated A/D converters, are
> poorly converted to digital waveforms.  A group like Floyd would certainly
> be in perfect _relative_ tune in the studio, but possibly not in perfect
> _absolute_ tune.  The effect would be the same -- dull sound from the
> "falling between the binary cracks" effect.

> If, during the remixing from two-track analog masters to 24-track digital
> tape, the mastering engineer adjusts the numeric representation used in
> the digital domain, a fairly accurate conversion can be accomplished.
> In pratice, a straight transfer is done with little care, and the result
> sounds "thin," as essential information was lost.  This is probably the
> effect that you have noticed.  It may seem odd that Mobile Fidelity Labs
> didn't get this right with Dark Side of the Moon, but it is rumored that
> this was due to their use of fixed-point arithmetic in the D/A converters
> due to an engineering oversight at the time (DSOTM was one of MFSL's
> earlier efforts, alas).  The Steely Dan album I don't know about.

>  analogous to the problem of representing certain numbers in binary,
>  such as repeating decimals. 

> The new Sony MD recordable discs avoid this problem entirely by using
> a continued-fraction representation of the signal.  This is possible
> because the MD format is not purely optical (as with CDs), but rather
> magneto-optical, thereby incorporating more of the continuous-domain
> information from the analog signal, much as an analog magnetic tape
> gives a perfectly continuous playback of an analog signal.  Thus, we
> can expect the new format to have much less of a "binary" problem than
> current CDs.

> I hope this cleared things up.

			Bruce Karsh
			karsh@sgi.com