Klarinet Archive - Posting 000186.txt from 2002/06

From: "Benjamin Maas" <benmaas@-----.com>
Subj: RE: [kl] Clarinet sampling
Date: Thu, 6 Jun 2002 01:20:00 -0400

> -----Original Message-----
> From: William Wright [mailto:w7wright@-----.net]
>
> Mattias, a few years ago I very much wanted a realistic clarinet sample
> for the sake of an experiment with spectrographs. I was hoping to have
> a long tone on clarinet that I could reproduce exactly whenever I wanted
> it. I was also hoping to be able to change the vibration recipe in a
> reproducible manner --- as part of the experiment.
>
> So I went to the local electronic music store, which *did* have dozens
> of different keyboards, but none of the pre-programmed 'clarinets'
> sounded even remotely like a real clarinet.

We need to be careful mixing our terminology. Samples versus Synthesis.
These are two ENTIRELY different things. Synthesis is the act of building
sounds that may (or may not as the case may be) sound like something else.
This can be done a number of different ways. Sampling is the act of
recording something and reproducing it by changing pitch using math (DSP).

> And as someone (Benjamin Maas?) has already posted, the price tag for
> equipment on which a person could build his own sample (allegedly!
> according to the very enthusiastic salesman who was unable to actually
> make the equipment perform as advertised) was earth-shattering --- for
> my budget at least. And it was obvious that I would need a minor
> education in order to use the expensive equipment.

What I am describing is not even building your own sample. Gigastudio (see
info at http://www.tascam.com/products/gigastudio/giga_160.php) can be used
to record samples but also use sample libraries that can have been made by
third parties.

> Rather than give up, I went to a professional sound studio and asked how
> much they would charge to manufacture a pair of clarinet samples with
> the same fundamental pitch but different partials. I was thinking: "I
> can afford to pay this fellow for an hour or two of work." The fellow
> had electronic gizmos covering all four walls and several large
> countertop consoles with hundreds of switches and buttons on them.

This actually would not be the best way to go about doing your samples. The
very best samplers libraries out there use full chromatic sampling of the
instrument. The most intensive sample library for this is a string one: the
Garritan Orchestral Strings (http://www.garritan.com/home.html) They are
incredible samples... Sampled different instruments in a good space (Allie
Tully Hall, I believe) all chromatically with different articulations and
different ensemble configurations (string orchestra, sections, solo, with
and without vibrato, etc...)

> He just smiled and said that a clarinet would be 'fairly difficult' and
> it wasn't his line of work anyway.

The engineer was probably a rock guy and didn't know much about acoustic
music. Even then, building sample libraries is a pretty involved task.

--Ben

Benjamin Maas
Freelance Clarinetist and Recording Engineer
Los Angeles, CA
benmaas@-----.com
http://www.fifthcircle.com

---------------------------------------------------------------------

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