Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!linus!philabs!polaris!herbie
From: herbie@polaris.UUCP (Herb Chong)
Newsgroups: net.audio,net.rumor
Subject: Re: CD vs vinyl (long, sorry)
Message-ID: <508@polaris.UUCP>
Date: Wed, 30-Apr-86 20:05:13 EDT
Article-I.D.: polaris.508
Posted: Wed Apr 30 20:05:13 1986
Date-Received: Sat, 3-May-86 04:29:31 EDT
References: <2679@pixar.pixar>
Reply-To: herbie@polaris.UUCP (Herb Chong)
Organization: IBM TJ Watson RC
Lines: 115
Xref: linus net.audio:7740 net.rumor:1854

In article <2679@pixar.pixar> good@pixar (You're only as good as your stereo) writes:
>	Since when have CDs been inherently inferior in areas essential to
>high-fidelity?  Since their inception, I'm afraid.  The people designing and
>selling digital audio are clinging slavishly to the idea that all the bandwidth
>you need is 20 KHz.  This is based on steady-state tone tests done in the 1940's
>for crying out loud.  Recent tests run on modern equipment in Europe have
>indicated that many people under many conditions can hear sine waves into the
>40 KHz range.  But even that is irrelevant to a point.  The human auditory
>system is very sensitive to slew rate, or transient response.  The steepness
>of the leading edge of a pulse is very easy to discern.

slew rate and frequency response are the same thing when it comes to
modifying a signal.  maximum slew rate and rolloff frequency at the
upper end is directly related.  as for the tests, unless you can prove
that subharmonics are not produced by your 40kHz test signals, you
haven't proved anything.  hearing by bone conduction doesn't count as
we never listen that way to our music.

>	When you translate the transient response needed to reproduce sounds
>found in real, acoustic music such as bells, plucked strings and other
>percussive instruments into a sine wave type bandwidth you get numbers way the
>heck up in the many hundreds of KHz.  CD players are falling apart well below
>20 KHz, and they rapidly fall into the mush above that figure.  Finer phono
>cartridges behave quite reasonably up to the 500 KHz region in this respect.

according to whom?  frequencies produced by instruments such as bells
and such may be greater than 20kHz but several hundred kHz is
stretching it.  and even if it did, if the ear is bandlimited to 50 kHz
(in your optimistic case), it matters not at all what happens above
that frequency.  (nit pickers will insist that IM distortion will play
a role.) any CD player will reproduce easily with 0.5dB up until 19kHz
with a small amount of phase shift.  by design, none have higher than
20.5 kHz.  there are very few cartridges that claim response beyond
75kHz.  there is only one i know of that claims beyond 100 Khz and it
stops at 125 kHz (a discontinued Technics MC).  real world measurements
indicate that for normal stereo records, almost anything recorded at
about 50kHz to about 80kHz is totally swamped by vinyl resonance.
(CD-4 records take advantage of this to FM the other two channels with
a reasonable amplitude even though the cartridges were physically not
able to reproduce flat to 50kHz without the compliance of the vinyl
factored in.)  couple that with the cutter head resonance somewhere
between 35kHz and 50kHz and you have total mismash.  add on top of all
that the bandlimited master tape used to produce the record and you see
that it is just not possible and/or relevant to talk about cartridge
behavior above 40kHz except in terms of how all these resonances affect
conventional audible frequencies by means of IM effects.

>	Also, for what ever reason, when you compare a CD to a good phono
>reproduction chain the CD sounds very dynamically compressed: dynamics in
>the music are simply lost.  Please don't flame about "96 db dynamic range".
>For one thing, that is the S/N, not the dynamic range (we aren't talking math,
>we're talking music, remember).  For another, useful information is available
>deep into the noise floor of an analog system, so the dynamic range can be
>greater than 100 db without much trouble.

i have a set of VERY expensive and well kept records.  i also have a
so so CD player (can't afford a good one yet 8-().  with proper
dithering, a usable dynamic range of some 80+ dB can be obtained with
the CD system.  an extremely quiet record can reach 70 dB (at least
some of mine can).  close, but no cigar.  maybe it's the distortion when
a cartridge mistracks that you're missing.  also, have you measured the
dynamic range of some good DD records?  i have.  45 dB.

>	As for un-measurable equating to un-audible, this is pure hokum.  For
>years and years new measurements have been coming along and identifying
>previously audible phenomena.  It is still possible to hear things you can't
>measure.  Unless you take an oscilliscope with you to the symphony you should
>prioritize that which you can hear above that which you can measure.  This
>doesn't mean that measurements aren't an extremely valuable thing.

what people can imagine also far exceeds what people can measure.
careful tests have shown that the majority of audible differences
come down to frequency response and IM distortion.

>	Many of the points made by the above author are quite valid, and I
>never said they weren't.  But most of them indicate that the CD player is
>a better consumer product than the turntable.  No big argument there.  CDs do
>now represent a good value as a convenience medium, and their longetivity is
>truly one of ther greatest assets.  I only object to people touting them as
>state of the art in audio reproduction, because they just aren't.  They are
>simply an alternative which has both advantages and disadvantages compared to a
>competing technology.  It is sad that the Japanese industrial complex has
>succeeded in evoking such a religiously zealous devotion to their marketing
>efforts.

depends on what you mean by state of the art.  they are the highest
technology form of music reproduction available to the consumer today.
as for the sound quality, well, i'm open to persuasion.  most of the
flaws i hear are due to poor master tapes and analog sections.  the
digital sections inherently are better than any currently available
analog means of reproduction.  and, for what it's worth, i can hear
slight differences between CD players.  i'm talking magnitude, not
importance.  (to some, any difference is a significant difference.)  i
happen to favor the 4x oversampled Philips approach but i think they
need to put higher quality analog components in their units.

what i objected to in this posting was the amount of nonsense in the
"technical discussion".  a slight amount of thinking would
have made the stupidities obvious.  note that i'm not saying
that records are better than CD's or vice versa, just that the
arguments used above showed no thought and a lot of hearsay.
for the anti-digital camp, it just ruins the credibility of any
others.

Herb Chong...

I'm still user-friendly -- I don't byte, I nybble....

VNET,BITNET,NETNORTH,EARN: HERBIE AT YKTVMH
UUCP:  {allegra|cbosgd|cmcl2|decvax|ihnp4|seismo}!philabs!polaris!herbie
CSNET: herbie.yktvmh@ibm-sj.csnet
ARPA:  herbie@ibm-sj.arpa, herbie%yktvmh.bitnet@wiscvm.wisc.edu
========================================================================
DISCLAIMER:  what you just read was produced by pouring lukewarm
tea for 42 seconds onto 9 people chained to 6 Ouiji boards.