Dynamics and level compression - FM vs DAB
 

"Jim Lesurf" wrote in message
...

The problem here, I suspect (again as IIRC Hawker indicates), is that the
broadcasters and set makers were 'promoting' FM for many years and this
was uphill work [pun].
Eh? Sorry I am being thick here! Clue please.

This was fair enough as the competition in those days
was AM, and so FM was pretty likely to be better. But it may mean they
glossed over - and then forgot about - these problems and just how likely
they are. Again fair enough if the choice is FM with some multipath versus
the interference-ridden AM.

I don't remember Pat Hawker's articles, but I have done a bit of work in
the past with a multipath 'scope display. I have to say that I am a bit
surprised that the problem is as severe as you describe. I nearly always
found that multipath could be made acceptable with the right aerial in the
right direction. But I wonder if the change to mixed or circular
polarisation might have made multipath more difficult to get rid of.

As regards the engineering decisions, multipath is not nearly so much of a
problem in mono as in stereo, as it is usually worst for high audio
frequencies with a large S content. So I think the problem mainly arose
with the change to stereo rather than the change from AM to FM. There was
pressure on the BBC to do stereo, and I suppose they thought that anyone who
was interested enough to get a stereo receiver would also get (or already
have) a good outside aerial, which would nearly always be needed anyway for
adequate signal to noise ratio. But I'm sure the engineers never forgot
multipath, which is one reason why they thought DAB was such a good idea.
--
Tony W
My e-mail address has no hyphen
- but please don't use it, reply to the group.

Dynamics and level compression - FM vs DAB
 

In article , Tony
wrote:
"Jim Lesurf" wrote in message
...

The problem here, I suspect (again as IIRC Hawker indicates), is that
the broadcasters and set makers were 'promoting' FM for many years and
this was uphill work [pun].
Eh? Sorry I am being thick here! Clue please.

Pun based on the preference for transmit and receive antennas being high.

The problem which I was meaning is that for many years the general public
tended to stick with AM and ignored FM. But the BBC - and set makers - were
trying to get them to buy FM sets.

This was fair enough as the competition in those days was AM, and so
FM was pretty likely to be better. But it may mean they glossed over -
and then forgot about - these problems and just how likely they are.
Again fair enough if the choice is FM with some multipath versus the
interference-ridden AM.

I don't remember Pat Hawker's articles, but I have done a bit of work
in the past with a multipath 'scope display. I have to say that I am a
bit surprised that the problem is as severe as you describe. I nearly
always found that multipath could be made acceptable with the right
aerial in the right direction. But I wonder if the change to mixed or
circular polarisation might have made multipath more difficult to get
rid of.

This is one of the points he deals with. He reports German work that showed
that moving away from H polarization to V or non-planar *does* tend to make
multipath worse. Quite interesting to read what he says as he makes clear
he is aware that because he worked for the IBA he might be felt to engaging
in BBC bashing to question their decision to move away from H polarisation
for VHF/FM.

As regards the engineering decisions, multipath is not nearly so much of
a problem in mono as in stereo, as it is usually worst for high audio
frequencies with a large S content.

Agreed. This also means it is easier for people to overlook if they aren't
familiar with what FM can sound like when there is no multipath. This was
also something Hawker discussed.

So I think the problem mainly arose with the change to stereo rather
than the change from AM to FM.

The problems probably grew worse when stereo was introduced.


There was pressure on the BBC to do stereo, and I suppose they thought
that anyone who was interested enough to get a stereo receiver would
also get (or already have) a good outside aerial, which would nearly
always be needed anyway for adequate signal to noise ratio. But I'm
sure the engineers never forgot multipath, which is one reason why they
thought DAB was such a good idea.

My recollection is that they were aware of fading and flutter problems for
FM, but I can't recall those involved saying much about multipath
distortion on FM being an audible problem. The multipath resistance of
digital transmissions was, I think, mainly to ensure reliable reception
cover without fades (or ignition interference). Of course, the goalposts
moved here as we have gone from DAB for cars to DAB for general use!

TBH I am not sure about the engineers. May depend on the era you have in
mind. Only aware that the broadcasters essentially fell silent on the
matter. The articles by Hawker are the only ones I have found thus far.
I've been reading mags like the audio ones and WW on and off for decades
and I can't off-hand recall any other articles that examine multipath in
anything more than general terms - and assume it is a minor problem.

Interestingly, Hawker does mention some research the BBC engineers did, but
this was never officially published! That, I think, also says something
about the attitude at the time, but it is hard to know who made the
decision to not publish the results as a normal BBC paper.

Also my experience is that when I have in the past read about or asked
about multipath the 'standard response' has been along the lines of your
own initial comments. i.e. That it isn't much of a problem, and can
generally be cured by a good RX and carefully aligned antenna.

However, I now seriously doubt that is so for many people. Hence my
suspicion that the early engineers did know about this, but felt it didn't
matter, or wasn't something to bother people about. Then - later on - as
stereo and FM grew the mindset may have been established that multipath
wasn't a problem. So the then current engineers tended to assume it wasn't
a problem as FM had been going for years and no-one had told them it *was*
a problem.

As with my 'uphill' pun, I think the mindset was to promote FM and then
Stereo, not to make a meal of any snags. ( ahem You might like to draw a
parallel with DAB here. ;- )

Part of the problem is that a general analysis of this would have been
quite difficult before computers were commonly available. Easy now to
forget this and that the world before about 1970 was different to today in
this respect. General modelling of FM can be quite difficult.

Another part is that a statistically useful survey of the problem takes a
significant amount of time and effort. So not something engineers would do
unless they already thought there was a good reason.

Perhaps the history here is a parallel with 'how to cook a crab' stories...
:-)

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

Dynamics and level compression - FM vs DAB
 

In article , Jim Lesurf
scribeth thus
In article , tony sayer

wrote:
In article , Jim Lesurf
scribeth thus

For example, have you ever read Pat Hawker's 1980/81 WW articles on
multipath?

Well nothing that a directional aerial won't help much towards for
serious listening;)..


Your reply indicates you haven't read his articles!

No as it 'appens .. I haven't..

Got an online reference for them?..

No, afraid not. After the references were given in uk.tech.digital-tv I
found them in my uni library and took xerox copies. However if you send me
your postal address by private email I can perhaps post you a printed copy
if you wish. Snag here is copyright as it is a WW article so I assume it
would be wrong to simply put a copy online - but maybe someone has done
this. If so, I don't know about it. It seems OK to make one or two copies
for research purposes. But not to make it openly available, I fear.

Thats very kind of you to offer .. I'll drop U a mail..

Similarly, I'd like to have copies of some of the refs he quotes, but fear
these may be difficult to track down. I will be giving it a try, though.

FWIW The copyright situation for such things does vex me at times. There is
a lot of interesting technical data in old WW, or HFN issues. But their
status isn't quite the same as academic journals as it would be easy tread
on the toes of those who own copyright.

Personally, I'd love it if copyright law allowed all technical journal
articles to be freely republished after, say, 10 years. Would make finding
reference material much easier and avoid wheel reinventions. Alas, those
who have a cash interest and own the copyright for magazine articles can be
- quite understandably - against this. I would wish to respect their wishes
as I accept the material is theirs to dispose of.

I must admit that the more I have looked at this topic, the more
curious I have become that it has largely been ignored by
broaddcasters, etc, over the years.

Well its not really a broadcaster problem Jim after all what can they do
about it?..

IIRC one of the comments Hawker makes is along the same lines. The classic,
"Well, it was alright when it left us." :-) However it strikes me as
somewhat naughty if they are saying this *knowing* that the results may
well be much poorer for a large section of the audience for reasons outwith
the listener's control. ...unless they move house!

Yes but even so what can they do?..

We should be these days moving to digital systems that offer better
things than the analogue ones that went before, but the reverse is
happening ever since the spectrum was "valued" by Ofcom and the
broadcasters discovered bit reduction;!..


The problem here, I suspect (again as IIRC Hawker indicates), is that the
broadcasters and set makers were 'promoting' FM for many years and this was
uphill work [pun]. This was fair enough as the competition in those days
was AM, and so FM was pretty likely to be better. But it may mean they
glossed over - and then forgot about - these problems and just how likely
they are. Again fair enough if the choice is FM with some multipath versus
the interference-ridden AM.

Indeed, I assume most RF engineers haven't ever really been aware of this
issue in more than general terms, and respond as you have done with the
assumption that a good antenna, etc, will be a fix. This is 'conventional
wisdom'. I accepted it for many decades and only started to feel it was
doubtful when I wanted to write an article about multipath and began to
study the topic for myself. This followed my increasing puzzlement that so
little previous work seemed findable in the literature.

Now, of course, there are many other transmission/distribution systems and
the choice isn't as simple as it was a few decades ago. So perhaps time for
the skeletons in the FM cupboard to be revealed. :-)

Well just what sort of percentage of the population find it a problem in
practice?..


I now suspect the problems are more common than is generally realised,
and that fixing it isn't always a simple matter of having a decent
tuner and good antenna (alignment). That may well help, but isn't a
panacea.

No under serious cases it won't but it does got a long way over and
above those simple Halo jobbies;!..

I agree that in many cases using a good directional antenna - correctly
aligned - plus a good tuner will reduce the effects of multipath. But in
practice I fear it isn't that simple a lot of the time. Hawker has some
comments on this that stuck me as quite perceptive.

Well isn't this one of the reasons they devised DAB for;!...

Slainte,

Jim


--
Tony Sayer

Dynamics and level compression - FM vs DAB
 

In article , Tony
scribeth thus
"Jim Lesurf" wrote in message
...

The problem here, I suspect (again as IIRC Hawker indicates), is that the
broadcasters and set makers were 'promoting' FM for many years and this
was uphill work [pun].
Eh? Sorry I am being thick here! Clue please.

This was fair enough as the competition in those days
was AM, and so FM was pretty likely to be better. But it may mean they
glossed over - and then forgot about - these problems and just how likely
they are. Again fair enough if the choice is FM with some multipath versus
the interference-ridden AM.

I don't remember Pat Hawker's articles, but I have done a bit of work in
the past with a multipath 'scope display. I have to say that I am a bit
surprised that the problem is as severe as you describe. I nearly always
found that multipath could be made acceptable with the right aerial in the
right direction. But I wonder if the change to mixed or circular
polarisation might have made multipath more difficult to get rid of.

As regards the engineering decisions, multipath is not nearly so much of a
problem in mono as in stereo, as it is usually worst for high audio
frequencies with a large S content. So I think the problem mainly arose
with the change to stereo rather than the change from AM to FM. There was
pressure on the BBC to do stereo, and I suppose they thought that anyone who
was interested enough to get a stereo receiver would also get (or already
have) a good outside aerial, which would nearly always be needed anyway for
adequate signal to noise ratio. But I'm sure the engineers never forgot
multipath, which is one reason why they thought DAB was such a good idea.

I expect that mixed polarisation might just make the problem worse.

One of the main reasons for its introduction was for the vertical
component to be used by cars the horiz to be used by fixed aerials.

But one thing it does do especially in urban areas is that when there is
multipath there is sometimes polarisation skew i.e. what was Vertical is
now Horiz and vice versa. Course take a Horiz TX and some reflections
coming as Vertical will now be discriminated against if you see what I
mean..

The good side is that it helps to reduce flutter for mobile systems in
that if the vertical component gets skewed then the horiz one will be -
bent- to Vertical and thus fill in the -missing- as it were....

You can see this on a spec analyser whilst driving around an area that
has mixed and vertical only stations. The mixed is quite stable in level
whereas the single Vert is UTP and down like the proverbial...
--
Tony Sayer

Dynamics and level compression - FM vs DAB
 

In article , tony sayer

wrote:
In article , Jim Lesurf
scribeth thus

Well its not really a broadcaster problem Jim after all what can they
do about it?..

IIRC one of the comments Hawker makes is along the same lines. The
classic, "Well, it was alright when it left us." :-) However it
strikes me as somewhat naughty if they are saying this *knowing* that
the results may well be much poorer for a large section of the audience
for reasons outwith the listener's control. ...unless they move house!

Yes but even so what can they do?..

Well, what they *could* have done is:

1) Carried out some decent research on the how much this crops up, and how
that varies with circumstances.

2) Investigated measures to help reduce problems.

3) Publish the results and ensure people (makers, dealers, and end users)
are aware of the outcomes of (1) and (2)

However it seems that the BBC did a bit of (1), then didn't publish the
results. Instead we have had decades of "It isn't really a problem, and if
you get it then buy a better tuner/antenna and waggle the antenna". When
the reports I have seen thus far indicate it often simply isn't that easy.



Now, of course, there are many other transmission/distribution systems
and the choice isn't as simple as it was a few decades ago. So perhaps
time for the skeletons in the FM cupboard to be revealed. :-)

Well just what sort of percentage of the population find it a problem in
practice?..

Hard to tell for various reasons.

One is that I can find almost no research that even addresses questions
like, "How many people have their FM listening affected by a level of
multipath that increases the distortion?" The only report I have seen is
via Hawker. This showed it was the *norm* for reception to be degraded by
multipath. But this BBC work was apparently never published.

Another is that most people aren't aware of the problem. So aren't
listening for it. Non audio-enthusiasts will just be using radios,
portables, etc, and will simply assume what they hear is what is what they
can expect. With no clue or interest in reasons. Audio enthusiasts have
been led over the years to expect FM to deliver good results if they have a
decent tuner and antenna. So, again, tend to assume that what they hear is
OK, and if not, have had no reason to suspect multipath. Nor, indeed, are
likely to be able to test for it or measure it.

As you will know, most people have no real expectation of 'hifi' or have
ever heard what a really good audio system can do when fed excellent source
material. How would they realise they had a 'multipath problem' given they
are have probably never heard the term?

Plus, of course, radio has been largely ignored in audio mags during recent
decades... Perhaps because - apart from R3 - the sound on most music
stations is awful.

So many people may not think they have a problem caused by multipath. But
some of them might be surprised if they heard the same broadcasts *without*
their local multipath and say, "That sounds quite different to my FM
radio!" This is a curious point, given that Vinyl LP also tends to produce
higher levels of distortion for high frequency and amplitude combinations,
and higher for L-R. People may prefer this as a result of habituation and
being led to think it is what the sounds should be like! :-)

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

Dynamics and level compression - FM vs DAB
 

In article , tony sayer


I expect that mixed polarisation might just make the problem worse.

Yes. Your comments are in line with what Hawker wrote. That mix poln is
helpful for mobile reception using V rods, etc. But that it makes multipath
more problematic for fixed reception using H plane antennas.

The statistical trade off seems to be: less likely to get flutter/fades but
with the cost of higher amounts of multipath.

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

Dynamics and level compression - FM vs DAB
 

In article ,
Jim Lesurf wrote:
I expect that mixed polarisation might just make the problem worse.

Yes. Your comments are in line with what Hawker wrote. That mix poln is
helpful for mobile reception using V rods, etc. But that it makes
multipath more problematic for fixed reception using H plane antennas.

The statistical trade off seems to be: less likely to get flutter/fades
but with the cost of higher amounts of multipath.

This was mentioned recently on uk.tech.broadcast as regards using a
vertical FM aerial to get a better DAB signal.

--
*OK, so what's the speed of dark? *

Dave Plowman London SW
To e-mail, change noise into sound.

Dynamics and level compression - FM vs DAB
 

In article , Jim Lesurf
scribeth thus
In article , tony sayer

wrote:
In article , Jim Lesurf
scribeth thus

Well its not really a broadcaster problem Jim after all what can they
do about it?..

IIRC one of the comments Hawker makes is along the same lines. The
classic, "Well, it was alright when it left us." :-) However it
strikes me as somewhat naughty if they are saying this *knowing* that
the results may well be much poorer for a large section of the audience
for reasons outwith the listener's control. ...unless they move house!

Yes but even so what can they do?..

Well, what they *could* have done is:

1) Carried out some decent research on the how much this crops up, and how
that varies with circumstances.

They don't seem to do that or much of it anymore;(..

2) Investigated measures to help reduce problems.

Well -what- in practice can they or anyone reasonably do other then to
use the best RX directional aerial they can muster and hope they have
someone who can deal with it. Our local rigger has been rigging for
years but I rather doubt he'd have anything for assessing it.

His answer to an FM aerial requirement is a Horizontal Halo;!..

For really severe cases theres always satellite:)...


3) Publish the results and ensure people (makers, dealers, and end users)
are aware of the outcomes of (1) and (2)

As number 1 !..

However it seems that the BBC did a bit of (1), then didn't publish the
results. Instead we have had decades of "It isn't really a problem, and if
you get it then buy a better tuner/antenna and waggle the antenna". When
the reports I have seen thus far indicate it often simply isn't that easy.



Now, of course, there are many other transmission/distribution systems
and the choice isn't as simple as it was a few decades ago. So perhaps
time for the skeletons in the FM cupboard to be revealed. :-)

Well just what sort of percentage of the population find it a problem in
practice?..

Hard to tell for various reasons.

Indeed, but you very rarely hear complaints about it perhaps because it
isn't as obvious as say ghosting on analogue TV...


One is that I can find almost no research that even addresses questions
like, "How many people have their FM listening affected by a level of
multipath that increases the distortion?" The only report I have seen is
via Hawker. This showed it was the *norm* for reception to be degraded by
multipath. But this BBC work was apparently never published.

Another is that most people aren't aware of the problem. So aren't
listening for it. Non audio-enthusiasts will just be using radios,
portables, etc, and will simply assume what they hear is what is what they
can expect. With no clue or interest in reasons. Audio enthusiasts have
been led over the years to expect FM to deliver good results if they have a
decent tuner and antenna. So, again, tend to assume that what they hear is
OK, and if not, have had no reason to suspect multipath. Nor, indeed, are
likely to be able to test for it or measure it.

And how many tuners around now have the required outputs on them
anyway?..

Although my Audiolab T8000 has the REVOX B261 or Denon's haven't!..

As you will know, most people have no real expectation of 'hifi' or have
ever heard what a really good audio system can do when fed excellent source
material. How would they realise they had a 'multipath problem' given they
are have probably never heard the term?

Plus, of course, radio has been largely ignored in audio mags during recent
decades... Perhaps because - apart from R3 - the sound on most music
stations is awful.

Indeed though some of the smaller ones like for instance Radio Jackie on
South London go out of their way with regards to audio quality..

And as I've muttered before the BBC services ought to all be available
on high bitrate satellite which is an excellent medium for high quality
Audio..

You really ought to try Bayern Klassik 4 for what digital radio can
do:)) Puts the BBC to shame for detail..can be had from Maplins for
around a 100 quid:)..


So many people may not think they have a problem caused by multipath. But
some of them might be surprised if they heard the same broadcasts *without*
their local multipath and say, "That sounds quite different to my FM
radio!" This is a curious point, given that Vinyl LP also tends to produce
higher levels of distortion for high frequency and amplitude combinations,
and higher for L-R. People may prefer this as a result of habituation and
being led to think it is what the sounds should be like! :-)

Well how many young people are being bought up on a diet of compressed
radio and MP3 players have any idea what it should or could be like?..


Slainte,

Jim

Ah!, Your getting old like a lot of us ... who do know better;-))


--
Tony Sayer

Dynamics and level compression - FM vs DAB
 

On 2008-09-12, Jim Lesurf wrote:
In article , John Phillips
wrote:
On 2008-09-10, Jim Lesurf wrote:
I have now put up a page that compares the dynamics and level
compression on FM with that on DAB. The page can be found at

http://www.audiomisc.co.uk/HFN/DABvs...ewithlike.html

[ 20,000 blank lines snipped! :-) ]

Err, sorry to everyone about that. I'd like to blame the technology
but probably it was more to do with "finger trouble" (but I'd still like
to know how I didn't see that many extra blank lines and understand how
they got there in the first place.)

...
These days I tend to prefer the BBC4 TV Proms to R3. I have gained two
impressions wrt ambience. One is that there often seems to be some LF
noise, perhaps due to air conditioning or passing traffic. However it may
be the audience swaying or breathing! :-)

This year I have not really noticed the same degree of difference between
BBC4 and R3/DAB that I noticed a few years ago. Possibly I have not
been listening to the sound so much and listening to the music instead.
Possibly the BBC's audio processing has got to be more consistent?

...
For perhaps obvious reasons such ambient noise seems louder when there is
something like an extended violin solo. ...

At a Prom a few years ago - Mahler 9th Symphony I think - quiet ending on
the strings: Someone close to me in the stalls swivelled on their seat
and it emitted an unlubricated squeak vastly louder than the orchestra. If
that contribution to the ambience had been from me I think I would have
died from embarrassment. I don't know how it came over on the broadcast.

I do sometimes notice level adjustments on the BBC4 prom broadcasts, but
they give me the feeling they are being done by a human who is following
the score and tweaking with intelligence to make the result. ...

On CD I have a few recordings of live opera from the 1960s where there
is some all-too-obvious level adjustment at times. Keeping the peaks
below the tape's saturation level, I assume. Probably there's more
that goes on which I don't notice because it's done with some musical
sensitivity by someone who "knows the score."

Actually I can recall at least one other example (a 1970 recording IIRC)
that gets very "hot" in places, where they *should* have done this.

--
John Phillips

Dynamics and level compression - FM vs DAB
 

In article , tony sayer

wrote:
In article , Jim Lesurf
scribeth thus


Well, what they *could* have done is:

1) Carried out some decent research on the how much this crops up, and
how that varies with circumstances.

They don't seem to do that or much of it anymore;(..

Alas, I agree. Indeed, it now seems close to impossible to even talk to
anyone involved with the engineering side of what the BBC transmit!



2) Investigated measures to help reduce problems.

Well -what- in practice can they or anyone reasonably do other then to
use the best RX directional aerial they can muster and hope they have
someone who can deal with it. Our local rigger has been rigging for
years but I rather doubt he'd have anything for assessing it.

That is one of the problems.


Well just what sort of percentage of the population find it a problem
in practice?..

Hard to tell for various reasons.

Indeed, but you very rarely hear complaints about it perhaps because it
isn't as obvious as say ghosting on analogue TV...

....and my experience in the past is that many people seemed either not to
notice ghosting, or assumed it was 'normal'. Given that it is much more
obvious than the effect on FM sound radio it is hardly surprising that most
people have no idea there might be a problem.

What is less understandable is that the broadcasts have remained shtum
about this, as have the consumer mags.


And how many tuners around now have the required outputs on them
anyway?..

Although my Audiolab T8000 has the REVOX B261 or Denon's haven't!..

Very few do. My CT7000 does, and even allows you to indicate multipath on
its meters, but is unusual in many ways.


Ah!, Your getting old like a lot of us ... who do know better;-))

I must confess to getting old. But I am less sure it means I know much more
that I did. :-)

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