Re: [ARSCLIST] CD writing programs and DC Offset

["Copeland, Peter" <Peter.Copeland@xxxxx>]

Dear All,
    I have two comments about the discussion on "CD Writing Programs", one
technical, and one non-technical.
    As an analogue engineer who often has to convert sound to digital, my
"archival hat" makes me think that the anti-aliassing filter often does the
worst damage to the sound. Therefore my former employer acquired a pulse
generator, specifically for documenting the performance of anti-aliasing
filters. The idea was to send a number of widely-separated 1-microsecond
pulses (from a Thurlby Thandar Instruments type TGP110 analogue Pulse
Generator) to the analogue-to-digital converter considered as a whole, and
record the results in the digital domain. (This operation can be done
*after* a dodgy recording has been made, so long as the machine is still
available!). Although we do not yet have the technology to reverse such
distortions, it is clear that if the artistic performance, the waveform
resulting from the pulses, and the information about use of a particular
machine can all be stored together, future generations may have the
wherewithal to reverse the side-effects.
    The "non-technical" comment is to *listen*!  If you have the analogue
sound and its digitised equivalent available on a changeover switch, and you
switch from one to the other (whist moving your head to catch any nuances
such as stereophonic corruption or particular faint distortions), you can at
least check that you are doing the least damage to the analogue sound
yourself - with little or no need for sophisticated test procedures! Does
no-one on this listserv do this?
    I would also like to add a comment to Mike Richter's posting about DC
offsets. In my experience, the clicks at edits are the principal symptom
(rather than even-harmonic distortion); but many noise-reduction processes
(especially cheap ones) become less effective if they don't "know" where
zero (silence) is - because "direct current" has become added!  I have had
to write some software myself for neutralising DC offsets on alien
recordings, which operates below the frequencies for the lowest-pitched
audible sound, while allowing for the inevitable "drift" in the offset.
Peter Copeland

-----Original Message-----
From: G. W. Ulrich Sieveking [mailto:sieveking@xxxxxxx]
Sent: 21 March 2003 23:10
To: ARSCLIST@xxxxxxxxxxxx
Subject: Re: [ARSCLIST] CD writing programs

stevenc@xxxxxxxxxxxxxx schrieb:
>
> ----- Original Message -----
>The information available to any
> third-party observer is a summary of what the listening party
> THINKS he/she/it hears; as such, it is of limited value in a
> scientific sense.
> Steven C. Barr

You are completely right with this statement as seen from a scientific
point of view.

On the other hand this scientific view at least partially misses our
objective. We are, or should be, dealing with - hopefully recorded -
sound and not with its technical representation. At least that should
ideally be, what we are cumulating in our libraries. We only use the
representation, because this is the best we can achieve. One might even
argue, that the original sound only exists in the listener's ears and so
does a replay of its technical reproduction.

Thus there are indeed two separate lines of investigation and two
separate ways of gaining additional knowledge:
There is the scientific quest for the best possible representation of
sound by technical means and there is sound itself, which defies
scientific treatment, because it is impossible to treat sound
scientifically without using a electronic - or any other human-free -
representation of it.

In other words: We are dealing with something, which we want to preserve
for posterity, but which only really existed during its creation.
Anything we do depends on our ears and minds as well as it depended on
the ears and minds of the players and those, who recorded it.

It is of course our first obligation to do anything we can to make a
representation of a performance as accurate as possible and this can
only be done with scientific means applied to all parts of the
recording, preservation and replay processes. Even if one can only
measure, what one wants - or expects - to measure.

Anything else has to be left to the device for which any sound was
meant, the ear. No serious judgment made with this instrument should be
dismissed. Even if the results are not quantifiable, we must consider
them real, because there is no way of falsification.

I do not want to start any 'flame wars' or any needless self-repeating
discussion about basic facts or decisions, but I want to make a clear
distinction between the technical processes involved in recording and
playback and the decidedly non-technical and non-scientific processes
involved in listening to/using/understanding any sound, which must
define how we think of it and how we measure its technical
representation.

Greetings,

U. Sieveking


**************************************************************************

Free exhibition at the British Library Galleries :

Magic Pencil : Children's Book Illustration Today (to 31 March) original
graphic work of 13 contemporary artists

*************************************************************************

The information contained in this e-mail is confidential and may be legally
privileged. It is intended for the addressee(s) only. If you are not the
intended recipient, please delete this e-mail and notify the
postmaster@xxxxx : The contents of this e-mail must not be disclosed or
copied without the sender's consent.

The statements and opinions expressed in this message are those of the
author and do not necessarily reflect those of the British Library. The
British Library does not take any responsibility for the views of the
author.

*************************************************************************