Petition to stop FM being switched off

In message , Dave Higton
writes
The main criticisam of DAB is that information *is* thrown away in such
a manner that it can never be completely recovered, not even
theoretically, and this is done by the broadcasters themselves before
the signal even gets to the transmitter.

Yes, it's thrown away. But it doesn't necessarily result in a
reduction of quality, which is all down to perception.

You appear to be avoiding my point that FM, being analogue and
being demodulated by a system that is non-linear, inevitably
introduces non-linear distortions: intermodulation and harmonic
distortion, to the audio. The DAB system can be engineered so
that the non-linearities are arbitrarily small, perhaps just a
few parts per million.
A lot of this is rubbish, so what if curtailed sidebands cause
distortion in FM The filters in the receiver filter out everything over
15kHz, leaving a nice clean signal for the listeners.
--
Clive

Clive wrote:
In message , Dave Higton
writes
The main criticisam of DAB is that information *is* thrown away in such
a manner that it can never be completely recovered, not even
theoretically, and this is done by the broadcasters themselves before
the signal even gets to the transmitter.

Yes, it's thrown away. But it doesn't necessarily result in a
reduction of quality, which is all down to perception.

You appear to be avoiding my point that FM, being analogue and
being demodulated by a system that is non-linear, inevitably
introduces non-linear distortions: intermodulation and harmonic
distortion, to the audio. The DAB system can be engineered so
that the non-linearities are arbitrarily small, perhaps just a
few parts per million.
A lot of this is rubbish, so what if curtailed sidebands cause
distortion in FM The filters in the receiver filter out everything over
15kHz, leaving a nice clean signal for the listeners.

I see you simply have not followed the technical details.

ALL of the stereo information is at 38Khz +-, and has sidebqands going
out to god knows where.

Any phase shifts over the IF passband lead to quite sever distortion all
over the place.

FM is better than AM, but real quality it ain't. Wiki reckons its about
96Kbps equivalent data rate, and I reckon that's not far off.

In article , The Natural Philosopher
wrote:
FM is better than AM, but real quality it ain't. Wiki reckons its about
96Kbps equivalent data rate, and I reckon that's not far off.

Does this mean that FM is supposed to have audio quality that is
"equivalent" to that of audio encoded at 96kbps. If this is the case, then
I wonder how they are judging it, because my ears tell me otherwise. It's
easy to make up numbers that "prove" things, but unless it can be shown
how the numbers relate to real life, they are meaningless.

Rod.
--
Virtual Access V6.3 free usenet/email software from
http://sourceforge.net/projects/virtual-access/

In article , Clive
wrote:
In message , Dave Higton
writes
The main criticisam of DAB is that information *is* thrown away in
such a manner that it can never be completely recovered, not even
theoretically, and this is done by the broadcasters themselves before
the signal even gets to the transmitter.

Yes, it's thrown away. But it doesn't necessarily result in a
reduction of quality, which is all down to perception.

You appear to be avoiding my point that FM, being analogue and being
demodulated by a system that is non-linear, inevitably introduces
non-linear distortions: intermodulation and harmonic distortion, to the
audio. The DAB system can be engineered so that the non-linearities
are arbitrarily small, perhaps just a few parts per million.
A lot of this is rubbish, so what if curtailed sidebands cause
distortion in FM The filters in the receiver filter out everything over
15kHz, leaving a nice clean signal for the listeners.

I am afraid that you have missed the point that the distortion products are
folded back as complex intermodulation by the FM and stereo decoding
processes. Hence the distortion can indeed appear below 15kHz in the actual
output.

And in other cases it appears at harmonics of the intended frequency, so a
2kHz tone can give distortion at multiples of this up to 15kHz.

Slainte,

Jim

--
Please use the address on the audiomisc page 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

In article , The Natural Philosopher
wrote:
Clive wrote:


FM is better than AM, but real quality it ain't. Wiki reckons its about
96Kbps equivalent data rate, and I reckon that's not far off.

I would not be as critical as that of FM provided it is broadcast with care
and the link/RX work well. It can deliver excellent sounding results. I've
certainly enjoyed broadcasts for many years. Only real anoyance for me
tends to be the limited dynamic range and the tendency for ignition
interference to be audible in quite passages on R3.

If you ignore the nonlinearity problems then the nominal information
bandwidth is rather more than 96kbps. Assuming good RX SNR, etc.

Slainte,

Jim

--
Please use the address on the audiomisc page 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

In article en.co.uk,
Roderick Stewart wrote:
In article , The Natural Philosopher
wrote:
FM is better than AM, but real quality it ain't. Wiki reckons its
about 96Kbps equivalent data rate, and I reckon that's not far off.

Does this mean that FM is supposed to have audio quality that is
"equivalent" to that of audio encoded at 96kbps. If this is the case,
then I wonder how they are judging it, because my ears tell me
otherwise. It's easy to make up numbers that "prove" things, but unless
it can be shown how the numbers relate to real life, they are
meaningless.

If you ignore nonlinearity problems then you can estimate the inherent
information bandwidth from Shannon. The result depends on what value you
think is appropriate for the output dynamic range. But for the sake of
example if we assume a range equivalent to 13 bits per LPCM sample, and a
minimim sampling rate (per channel) of 30ksamples/sec you end up with
13 x 30000 x 2 = 780 kbps.

The real value will certainly be different to that. But the channel
capacity of a decent TX/RX link should be rather more than 96kbps.

If the above is based on assuming some specific lossy coding system and
type of audible material, then it may tell us more about the choice of
coding system and test material than it does about FM. :-)

FWIW I do now tend to prefer both DTTV and the aac iPlayer to FM for my
listening to BBC R3, R4, etc. But I would not use that to argue that FM has
an information bandwidth below, say, 192kbps.

Slainte,

Jim

--
Please use the address on the audiomisc page 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

Roderick Stewart wrote:
In article , The Natural Philosopher
wrote:
FM is better than AM, but real quality it ain't. Wiki reckons its about
96Kbps equivalent data rate, and I reckon that's not far off.

Does this mean that FM is supposed to have audio quality that is
"equivalent" to that of audio encoded at 96kbps. If this is the case, then
I wonder how they are judging it, because my ears tell me otherwise. It's
easy to make up numbers that "prove" things, but unless it can be shown
how the numbers relate to real life, they are meaningless.

Well I don't know the answer to that.
But certainly even on a good antenna system here, FM is nothing to write
home about.

Rod.

Paul Martin wrote:
In article ,
The Natural Philosopher wrote:
Paul Martin wrote:

DAB and DVB-T both "waste" bandwidth to achieve robustness in the face
of multipath. (This is the guard interval between symbols.)

Ah but multipath can be catered for with 'echo cancellation' with modern
DSPS.

Ah, but can that echo cancellation handle multipath which varies from
millisecond to millisecond, as in reception on the move in a city
canyon?

Pretty much, yes.

In message , Jim Lesurf
writes
I am afraid that you have missed the point that the distortion products are
folded back as complex intermodulation by the FM and stereo decoding
processes. Hence the distortion can indeed appear below 15kHz in the actual
output.

And in other cases it appears at harmonics of the intended frequency, so a
2kHz tone can give distortion at multiples of this up to 15kHz.
Where I live FM rules. Digital may have pretences above their station
(pun not intended) but in straight listening tests are rubbish compared
to the FM in this part of the woods.
--
Clive

In article , Clive
wrote:
In message , Jim Lesurf
writes
I am afraid that you have missed the point that the distortion products
are folded back as complex intermodulation by the FM and stereo
decoding processes. Hence the distortion can indeed appear below 15kHz
in the actual output.

And in other cases it appears at harmonics of the intended frequency,
so a 2kHz tone can give distortion at multiples of this up to 15kHz.


Where I live FM rules. Digital may have pretences above their station
(pun not intended) but in straight listening tests are rubbish compared
to the FM in this part of the woods.

I note your opinion and point out it has nothing to do with what I wrote
that you quoted above it. :-)

My point was to inform you and others that your earlier assertion that

A lot of this is rubbish, so what if curtailed sidebands cause
distortion in FM The filters in the receiver filter out everything over
15kHz, leaving a nice clean signal for the listeners.

Is simply incorrect or misleading. I've re-quoted you here as you snipped
away the context which may have mislead people again.

The simple engineering reality is that the use of a finite bandwidth for
real-world TX and RX reasons leads to distortion for FM links. This arises
from the defined nature of FM. The details of the distortion depend on the
details of the modulation and link.

If you doubt this you can look at the maths, and/or get yourself some IF
filters of various widths and flatness/phase specs and see what they do to
a stereo FM signal. My comments are based on having done both of these
things on a number of occasions in the past. Plus knowing the results of
other engineers who have done so.

Provided signal modulation levels and frequencies are kept modest this
distortion can be kept to a very low level. But higher modulation levels,
etc, particularly for stereo, can lead to higher distortion. Some of this
will appear in the region below 15kHz when the signals are demodulated and
presented as stereo.

Alas, there is a tendency in recent years for broadcasters to level
compress the modulation to make it 'louder'. This will tend to increase the
distortion generated by the FM link. Although I can't say if this is more
of a pest than the way the level compression itself fudges up the sound.
:-)

That said, my own opinion is that - having enjoyed FM for many decades - I
find that nowdays I have come to prefer either DTTV or the aac iPlayer [1]
for listening to BBC radio. Both of these are 'digital'. However this
personal preference isn't because of the distortion that can arise on FM,
so also has nothing to do with the above. It is for other reasons like
interference, background noise level, etc, for R3. Also because in recent
years there has been a tendency for R3 on FM to have more level compression
applied than on DTTV.

Slainte,

Jim

[1] Tennis permitting. :-)

--
Please use the address on the audiomisc page 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