Re: arsclist Tubes will not die

[Mike Richter <mrichter@xxxxxxx>]

At 04:50 PM 11/13/2002 -0600, Premise Checker wrote:
> >From the Transhuman news list:
>
> ---------- Forwarded message ----------
> Date: Wed, 13 Nov 2002 17:40:20 +0100 (CET)
> From: Eugen Leitl <eugen@xxxxxxxxx>
> Reply-To: transhumantech@xxxxxxxxxxxxxxx
> To: transhumantech@xxxxxxxxxxxxxxx
> Subject: [>Htech] Re: Tubes will not die (fwd)
>
> ---------- Forwarded message ----------
> Date: 13 Nov 2002 08:56:29 -0800
> From: James Rogers <jamesr@xxxxxxxx>
> To: fork@xxxxxxxx
> Subject: Re: Tubes will not die
>
> On Tue, 2002-11-12 at 20:07, Jon O. wrote:
> >
> > James, this is a JOKE! Perfect CD reproduction?
> >
> > How can you say this -- one is bits and one is vibration. They both have
> > their limitations, but a CD sampling rate is similar to a Mega Pixel
> > rating on a digital camera. It isn't the thing, but a representation
> > of the thing. So is a CD. I suppose you can tell me that a 20 mega pixel
> > digital camera is just like "seeing" photo and 512 bit sample of music on a
> > CD is equivalent to the live production.
>
>
> The best analog recording formats (which are actually digital when you
> get down to it) have less signal fidelity than an ordinary Red Book CD.
> You can record Red Book audio that maintains perfect fidelity WELL below
> the noise floor and beyond the frequency range of vinyl, so a perfect
> reproduction of any signal there is trivial.  I would humbly suggest
> that you study up on sampling theory, signal processing, and other bits
> of relevant trivia.  You are seem to be demonstrating a classic
> misunderstanding of how sampling actually works.

Bah, humbug!

There is a good deal of truth in the preceding paragraph, but in all it is 
not consistent with the tests I ran many years ago. I will address here 
only the frequency argument.

It is certainly true in linear analysis - the sort we engineers find 
comfortable - that frequency response on CD-DA is superior to that on 
vinyl. However, we do not hear pure tones and linear analysis does not tell 
us all that we need to know for sound reproduction.

The issue here is not frequency response per se but response to waveform 
and, with that, to phase error. I have repeatedly run the following simple 
demonstration. Assume that the individual asserts that he can hear nothing 
above 15 KHz.

1. Set up two signal generators, both at 8 KHz, one for sine waves, the 
other for square waves

2. Using wide-band amplifiers and headphones, challenge the subject to 
adjust one of them - amplitude and frequency - to match the other.

It will not happen. Whether the tones are concurrent or sequential, the 
subject will hear a difference. However, the engineer will inform you that 
the only differences are in third and higher harmonics - 24 KHz and above. 
Related tests with triangle waves are comparably interesting but need not 
be explored here.

Similarly, with good analogue equipment and a quality tape deck, record 8 
KHz sine and square waves at low amplitude (say 10 db below saturation). 
Viewed on a scope, the two are distinctly different in waveform with the 
square wave showing a flattening after reproduction. We "know" that the 
recorder cannot reproduce 24 KHz and that there are no terms to cause the 
change of waveform below that frequency, so obviously what we see is due to 
conspiracy of the LP makers.

I first ran these tests with a friend who was a member of AES back when the 
standard for CD-DA was being established. Since then, the practical results 
are before us. The best analogue recordings reproduce the unique timbre of, 
say, a Boesendorfer Imperial Grand faithfully where I have never heard a CD 
played back to do so over the range of the instrument. The difference may 
not be important to the vast majority of listeners, but the simple 
demonstration proved valuable in persuading people of the need for accurate 
playback even in that prehistoric era.


Mike
mrichter@xxxxxxx
http://www.mrichter.com/


-
For subscription instructions, see the ARSC home page
http://www.arsc-audio.org/arsclist.html
Copyright of individual posting is owned by the author of the posting and
permission to re-transmit or publish a post must be secured
from the author of the post.