What's wrong with ringing?
 

On 2008-07-08, Eeyore wrote:
John Phillips wrote:

I often see texts that imply or state outright that ringing in time-domain
audio waveforms, usually associated with time-domain discontinuities,
is to be avoided. I wonder what is the basis for this.

Ringing on square wave inputs may be a sign of maginal stability.

Indeed. For ringing in an "analogue" sense. I was actually thinking
of digital systems and I see now the multiple uses of "ringing" so I
should have clarified.

--
John Phillips

What's wrong with ringing?
 

John Phillips wrote:

Eeyore wrote:
John Phillips wrote:

I often see texts that imply or state outright that ringing in time-domain
audio waveforms, usually associated with time-domain discontinuities,
is to be avoided. I wonder what is the basis for this.

Ringing on square wave inputs may be a sign of maginal stability.

Indeed. For ringing in an "analogue" sense. I was actually thinking
of digital systems and I see now the multiple uses of "ringing" so I
should have clarified.

In digital it could be anything, even down to fundamentals of the theDSP filter
structure.

Graham

What's wrong with ringing?
 

On 2008-07-08, Jim Lesurf wrote:
In article , Eeyore
wrote:
John Phillips wrote:
I often see texts that imply or state outright that ringing in
time-domain audio waveforms, usually associated with time-domain
discontinuities, is to be avoided. I wonder what is the basis for
this.

Ringing on square wave inputs may be a sign of maginal stability.

Key phrase here is "may be". I recall reading in the past more than one
review of a power amp where the writer assumed that 'ringing' on a square
wave was a sign of marginal stability. Whereas in many cases it simply
showed that the amp included a series inductor on its output to ensure safe
and stable operation into naster RF loads. Another example where it was
possible for writers to jump to conclusions based upon limited
understanding.

My impression is that it has been common for digital filtering in domestic
items like CD players to use FIR arrangements. Stability should not be a
problem for these. But may be more of an issue for IIR designs. Indeed, if
I recall correctly, some of the high order delta-sigma modulator (i.e. ADC)
arrangements published by Sony and Philips for DSD/SACD had potential
instability unless used correctly.

So, in summary it seems that there are some causes of time-domain audio
signal ripples ("ringing" as a description of the observed phenomenon)
that may indicate problems in an audio system. One of those problems
may be "ringing" as a cause of the observed phenomenon, associated with
marginal stability in both an analogue and a a digital sense.

However it also seems clear that the Pavlovian response sometimes seen
in audio contexts to the observation of a "ringing" waveform is probably
not necessarily appropriate.

--
John Phillips

What's wrong with ringing?
 

In article , John Phillips
wrote:

So, in summary it seems that there are some causes of time-domain audio
signal ripples ("ringing" as a description of the observed phenomenon)
that may indicate problems in an audio system. One of those problems
may be "ringing" as a cause of the observed phenomenon, associated with
marginal stability in both an analogue and a a digital sense.

I'd say the distinction here is between a symptom - visible ripples or
ringing associated with something like a transient - and assigning a
'cause' of that observed behaviour. The problem is that there are various
possible causes.

However it also seems clear that the Pavlovian response sometimes seen
in audio contexts to the observation of a "ringing" waveform is probably
not necessarily appropriate.

Yes. In practice this is perhaps another example of a common problem in
areas like domestic audio. That a commentator may decide they have spotted
a symptom, and then jumps to a conlcusion about the cause - perhaps being
completely mistaken. Similar to the case of 'Dirty Digital' in another
thread, where someone does a measurement in an inappropriate manner, and/or
misunderstands the results they see. The symptom (FFT analyser display in
that case) is correct - for the given situation - but this symptom then
prompts an incorrect diagnosis of the cause. :-)

Alas, it is one thing to give someone a lot of nice test kit. It is
something else for them to use it correctly, and recognise what the results
do (or do not!) tell them.

Slainte,

Jim

--
Change 'noise' to 'jcgl' if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html

What's wrong with ringing?
 

On 2008-07-07, Jim Lesurf wrote:
In article , John Phillips
wrote:
As long as a filter perserves all human-audible frequencies I cannot see
an objection (to this form of ringing, at least). The only argument I
can think of for objecting is the possibility that the non-linear
behaviour of the ear may result in specific audibility issues which
wouldn't be heard with fully linear hearing. ...

See also the articles on this on audiomisc in the 'hearing' section which
also considers hearing nonlinearity in this context.

Thanks for the pointer. I have looked at your "Ringing in the Ears"
article. I believe it confirms precisely what I suspected. There's
a possibility that some effects which in a linear hearing system would
be innocuous may be audible with real-world non-linear hearing. At
least it seems that the possibility cannot be excluded without
further checks.

I was wondering also about the issue of whether the linear system
equivalence of frequency-domain and time-domain signals might not be
preserved in non-linear human hearing. That is, does the "normal"
20 Hz - 20 kHz human-audible bandwidth imply an equivalent ability to
resolve time-domain differences or does non-linearity break that
equivalence?

--
John Phillips

What's wrong with ringing?
 

In article , John Phillips
wrote:
On 2008-07-07, Jim Lesurf wrote:


See also the articles on this on audiomisc in the 'hearing' section
which also considers hearing nonlinearity in this context.

Thanks for the pointer. I have looked at your "Ringing in the Ears"
article. I believe it confirms precisely what I suspected. There's a
possibility that some effects which in a linear hearing system would be
innocuous may be audible with real-world non-linear hearing. At least
it seems that the possibility cannot be excluded without further checks.

I was wondering also about the issue of whether the linear system
equivalence of frequency-domain and time-domain signals might not be
preserved in non-linear human hearing. That is, does the "normal" 20
Hz - 20 kHz human-audible bandwidth imply an equivalent ability to
resolve time-domain differences or does non-linearity break that
equivalence?

Afraid I can't say. There are various bits of work that support such ideas,
but I've not seen anything that seemed conclusive in showing they actually
matter in practice.

Various things in this area are 'possible' in the sense that you can
propose a hypothesis that isn't clearly falsified by what we know, and
might explain some claims. But this isn't the same as the hypotheses being
correct descriptions of what actually occurs. I fall back on the fact that
I've never heard the 'differences' involved so far as I can tell, and I've
not seen reliable evidence that anyone else has, either.

So I regard it as fair enough to consider such possibilities, at least for
the purpose of seeing if they have any plausibility. But I don't then
assume they must be so in practice. For that, I await evidence which can be
shown not to be due to one or more of the usual errors people make when
making claims.

Slainte,

Jim

--
Change 'noise' to 'jcgl' if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html