Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!elroy.jpl.nasa.gov!sdd.hp.com!caen!umich!sharkey!news.iastate.edu!IASTATE.EDU!mpurtell
From: mpurtell@IASTATE.EDU (Purtell Michael J)
Newsgroups: comp.dsp
Subject: Re: DSP for Audio Marketing
Message-ID: <1991Apr8.015749@IASTATE.EDU>
Date: 8 Apr 91 06:57:49 GMT
References: <1991Apr5.172259.11257@vaxa.strath.ac.uk>
Sender: news@news.iastate.edu (USENET News System)
Reply-To: mpurtell@IASTATE.EDU (Purtell Michael J)
Organization: Iowa State University
Lines: 83

In article <1991Apr5.172259.11257@vaxa.strath.ac.uk>,
cnbs30@vaxa.strath.ac.uk writes:
> 
> I am currently collating information on the use (or abuse!) of DSP for
the
> marketing of digital audio.  One look in recent Audio and Hifi
> magazines will confirm that digital signal processing (DSP) jargon
> is used for technical marketing purposes.  CDs and DATs are
> described as having D/A converters, digital filters, 8 x's
> oversampling, references to quantisation, sampling,
> frequency responses and so on.  And the latest references to
bitstream;
> presumably the average buyer does not understand bitstream, but is
> impressed by this latest description, which makes his system better?
> 
> How much of this is just marketing gimmicks?  4 x's versus 8 x's
> oversampling, can we hear the difference?  Oversampling is described
> in some Hifi magazines as simply inserting extra samples between
> adjacent samples in order to increase the resolution?  I always
> thought that oversampling was to avoid having to use expensive
> analogue filters at the input (anti aliasing) and outputs??

Inserting interpolated samples increases the sampling frequency which
eases
the constraints for the lowpass filter at the output.  Sometimes the
circuit
that does the interpolation also increase the resolution (see below).

> Is DSP terminology being used to hype up 
> 
> How can companies realise 20 bit outputs from D/As, when the data on a
CD
> is only 16 bits?  Just where is this extra information supposed
> to be materialising from?

This is done by the interpolation circuit (usually a digital filter,
ones
I've read about are FIR filters).  Imagine you have two samples with the
values 1 and 2.  Now you want to insert 7 values inbetween these for the
8 times oversampling.  The values you'd probably use if you had
unlimited
resolution and were doing linear interpolation (for simplicity) would
be:

(real) (binary)
1.125  1.001
1.25   1.010
1.375  1.011
1.5    1.100
1.625  1.101
1.75   1.110
1.875  1.111

This is where the extra bits of resolution come from.

Whether or not you can hear the difference is a debate I won't enter.
:-)

> --
> 
>
========================================================================
=========
> Bob Stewart           |JANET:   cnbs30@uk.ac.strath.vaxa
> EEE Dept.             |E-MAIL: 
cnbs30%vaxa.strath.ac.uk@nsfnet-relay.ac.uk
> Strathclyde University|BITNET:  cnbs30%vaxa.strath.ac.uk@UKACRL
> Glasgow G1 1XW        |UUCP:    cnbs30%vaxa.strath.ac.uk@ukc.uucp
> Scotland.             |
>
========================================================================
=========

--
-- Michael Purtell --  | "In a hundred years, | There's an Old Irish
Recipe for
mpurtell@iastate.edu   |  we'll all be dead." |   Longevity: Leave the
Table
Iowa State University  |  -- The January Man  |  Hungry.  Leave the Bed
Sleepy.
                "slow is real"                |    Leave the Tavern
Thirsty.