Klarinet Archive - Posting 000785.txt from 1997/11

From: Jonathan Cohler <cohler@-----.net>
Subj: Re: Recording the Clarinet
Date: Fri, 21 Nov 1997 13:42:42 -0500

Jerry Korten wrote:

> And it was in the vein of trying to explain why digitally recorded sound
>doesn't capture the full analog waveform that that discussion got started.
>Somehow we got out of the world of A/D converters (Nyquist) and I missed
>that turn. Which is why I was so insistent on showing why A/D converters do
>things differently.
>
>You agree that in fact there are errors in how an A/D captures the signal,
>but these errors are very small. Recording companies are trying to improve
>this, but it is not necessary and they are wasting their time and money.
>(By the way we consumers are using 16 bit, 44 KHz technology, what are
>professional recording studios using?)

The vast majority of professional recording studios are still using 16-bit
44KHz technology. Some recording companies and studios make 20-bit
recordings (and a few are now doing 24-bit), and do all the editing in 20
(or 24-bits) and then reduce it down to 16-bits through a dithering process
to create the final masters, which are all still 16-bits, because that's
what CDs are.

>
>So let's get back to the issue of recording clarinet.
>
>Here is my problem...
>
>I have recordings of clarinet on CD, and clarinet on record. My recordings
>on record sound more like clarinets.
>
>I have spent quite a lot of dough on a D/A converter, (but perhaps not
>enough). What type do you recommend?

My guess is that the problem is not in your D/A converter (if you spent
more than a few hundred dollars on it, or it is built into a reasonably
expensive component). What kind of D/A (or component) do you have?

Also, the A/D used in the recording process would be more of a culprit than
the D/A.

>
>You allude to the fact that it is most likely in the recording setup, but
>how can this be for all my recordings on CD?

I find it to be true that most recordings of the clarinet (LP or CD) are
poorly recorded and don't sound much like a clarinet. But I believe that
99% (or more) of this problem can be attributed to the following factors:

1. Type of microphones used
2. Placement of microphones
3. Space in which the recording was made

Clarinet is a particularly difficult instrument to record, because it's
spectrum changes so dramatically depending on what angle (both the angle
in the horizontal plane and the elevation angle) you are listening from.
This problem is actentuated when one records in a relatively dead room (in
which I find many, if not most, clarinet recordings are made).

Having a concert hall that disperses sound evenly around the room and has a
natural reverberation time of 2 to 3 seconds allows the spectrum of the
clarinet to be "room averaged" as Benade calls it. This way one gets a
much truer picture of the clarinet sound.

Also, because the clarinet has such a wide dynamic range and so much energy
in the high notes, one must be very careful about level control when
recording the clarinet (20-bit digital eliminates this problem by giving
one an extra 24dB of headroom above what the final 16-bit master can take).

> Is there an inherent
>distortion in records that make things sound more life like (or is this an
>unconscious bias to analog so that I can complicate my life by having to
>deal with cleaning records and otherwise make my life miserable)?
>

There are several inherent distortions in records, but I believe this is
simply an unconscious bias on your part. It could also be that the
particular LPs that you own and listen to were recorded better than the
particular CDs that you own and listen to.

A way to test this on yourself would be to set up a double-blind,
level-matched study where you play an LP and a CD that were made from the
same original master. And try to determine which you like better. Of
course, the problem with this, is that it will be virtually impossible to
make it double blind, because as soon as you here the first
snap-crackle-pop, you'll know it's the LP.

I certainly have LP recordings of the clarinet that are better than some of
my CD recordings, but the differences I hear are all attributable to the
factors I mentioned above, and none of them to the digital or analogue
technology.

The quality of the CD recordings, i.e. the faithfulness with which the
digital medium reproduces the original input signal, is superior in all
cases where reasonably high-end digital equipment is used, however.

--------------------
Jonathan Cohler
cohler@-----.net

   
     Copyright © Woodwind.Org, Inc. All Rights Reserved    Privacy Policy    Contact charette@woodwind.org