Petition to stop FM being switched off

"Bill Wright" wrote in message
...
"The Natural Philosopher" wrote in message
...
Mark wrote:
There is no perfection in nature.

You obviously haven't seen my body.
Bill

Apparently nether has Maddy Prior

Steve Terry

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

The only time you dont need something like this is where the multicast
is local to a subnet and doesnt need to cross a router - IGMP itself
and various routing protocols such as OSPF and the main ones i see
like this in practice.

Which is the case for a localised radio network. The back channel stops
at the transmitter..

What if there are no listeners to that station? Is it not efficient to
give the bandwidth to the ones that do have listeners?

Surely, that's what you were advocating?

Er, no.

You cant actually have a backchannel with true broadcast radio, so you
must perforce transmit all channels.

The trade-off between that system, and the sort of cellular TCP/IP
strategy, is that under adverse conditions the spectrum can get blocked
with retransmit requests etc. Degrading things more.

No, the spectral efficiency was simply on the basis of NOT allocating a
fixed bandwidth pipe to each radio station, but allowing them to use
what they needed in the context of the complexity of what they were
sending audio wise.

E.g.a more or less respectful silence with the odd nose blow before the
conductor picks up his baton, is very low in information. Rapturous
applause once he has finished, is very high.That is the basis of all
packet switched variable length packet network operation. To get a
better utilisation in transfer of information from disparate sources
with a high peak to mean information ratio. That is the bandwidth needed
tends towards the AVERAGE content times the number of channels, rather
than the PEAK content times the number of channels.
As an aside there was an interesting article I read some years back
comparing fixed timeslot multiplexing with variable packet switched
multiplexing on optical fibres..sadly the actual increases that packet
switching would have given were nullified by the fact that they simply
couldn't switch fast enough! TDM being simpler, could.

Anyway, looking at DAB in toto rather than as its now implemented,
really all I was saying is that the current system is crap because it
allocates fixed bandwidth channels, that are too small for quality, and
are not optimal in compression.

Which leads me to some basic assertions

1/. That doesn't mean DAB has to be crap forever.

2/. FM is limited by the S/N ratio of the channels its on to a quality
it cant ever exceed. My back of envelope calculations suggest that CD
quality could be transmitted in an FM channel EASILY at a S/N ratio of
25-30dB or so.

3/. Better bandwidth could be achieved overall by using statistical
multiplexing - packet switching - than fixed multiplexing over a
broadcast medium.

The most highly developed stat mux technology we have for these sorts of
data rates is IP, and it contains already the basic hooks to be used as
a one way broadcast medium, so why not use it? and all that goes with it
like the codecs which are all well understoiod and in the public domain?

And the Beeb is also pretty heavily into IP technology with I-player and
internet radio..

Now that is currently TCP/IP based, but there will come a point where if
it is popular enough, the overhead of carrying multiple TCP/IP streams
will exceed the overhead (on the internet) of carrying broadcast or
multicast streams. All you then need to do is terminate the stream at
local transmitter, and build a receiver that can decode it. That
receiver front end would look exactly like a multicast or broadcast
enabled IP router, limited to broadcast traffic..after that the rest is
pure IP and any internet capable subsystem could utilise it.

In short your DAB radio might also have an ethernet port and a wifi
aerial, and be capable of picking up teh same information through those
via teh Internet, through on-air transmitters, wherever it was.


In this context, the actual transmitters become simple last-ten mile
links of a one way UDP over IP broadcast stream.

Its the same paradigm shift that BT are now contemplating, instead of IP
being a special case of voice utilizing one or more fixed voice channels
to propagate, make the whole network IP and see voice, as VOIP, as the
particular case rather than the general. I.e. instead of delivering an
audio citrcuit that happens to also do broadband, deliver a brodaband
circuit that happesn to do voice.

So the beeb SHOULD concentrate on delivering broadcast or multicast
internet radio, some or all of which MAY use RF as the last ten miles of
hop.

It really has no business supporting and developing a proprietary DAB
standard that is obsolescent, doesnt satisfy its discerning customers,
is bandwidth inefficient, used dated compression standards and is likely
to never be a great sucess..

"Dave Plowman (News)" wrote in message
...
In article ,
jasee wrote:
I can't remember being particularly impressed with the original quad
electrostatics particularly with full orchestras and organs, as in
certain organ concertos, really almost disappeared. I really couldn't
see what the fuss was about. And I remember the quad 405 amplifier I
bought becoming so noisy that I sent it back.

You're a few years apart. The original '57 would have been driven with
Quad II valve amps. Or even just one. The first Quad transistor amp was
the 303. Early 405s used a poor op amp which was changed for a better one
shortly after introduction.

The original '57 was very room sensitive. Was your room approaching a
cube? The bass output seemed to disappear in those. Worked best in a long
room with the speakers across the narrow wall. Of course it only went down
to about 40 Hz in the best of circumstances. And the lack of resonances
made it appear bass light compared to a honky cabinet speaker.

I borrowed the quads for about a couple of weeks in London and they were
with valve amps and yes the room was more of less square and the particular
organ had a low 32cycle (IIRC) note which you simply couldn't hear at all.
The 405 I bought myself later but changed for a Raford transistor amp which
was simply worlds better. Never did like valves anyway.

"Steve Terry" wrote in message
...
"Bill Wright" wrote in message
...
"The Natural Philosopher" wrote in message
...
Mark wrote:
There is no perfection in nature.

You obviously haven't seen my body.
Bill

Apparently nether has Maddy Prior

Steve Terry

Last night I was working late in the office. It's a small room with two PCs
and some CCTV stuff. It was very hot indeed so I took l my top off. After a
while I was forced to take my trousers off, so I sat working in my pants.
Hil came in and passed a rude comment, and opened the window. After a while
the room cooled down so I put my top on. Some time later I knocked off, so I
put my shoes on and took the dogs round the field. When I got back Hil said,
"What are you doing, going out without your trousers?" It didn't matter
because it was dark, but I'm glad I didn't meet the lampers or anyone.

Bill

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

E.g.a more or less respectful silence with the odd nose blow before the
conductor picks up his baton, is very low in information. Rapturous
applause once he has finished, is very high.

Experience shows anything other than digital silence (ie. all zeros)
requires almost as much bandwidth as applause. It's the same sort of
signal, but of a different amplitude. As perceptual encoders work in
the frequency domain, not the amplitude domain, they need approximately
the same amount of information to describe both.

Real "silence" will be dominated by things like the rush of
airconditioning systems, flies buzzing, people shuffling their feet,
coughing and the noise from the microphones and other signal processing
before it becomes digitised. Although it's of low amplitude, it's just
as complex to describe (as Fourier components) as any other sound.


I am not sure we use Fourier analysis when digitising sound tho.

The tendency is to use a full bandwidth sample, and then post treat the
resultant bitstream.

Modified delta modulation works pretty well as a basic way to do that.
It only reduces the slew rate of the final signal..so it cant cope with
full power high frequencies, but then typically nothing else in the
normal audio chain can, either.
,

It's easier to encode something strident but simple in nature, which
will mask other sounds.

Again I am not sure that is really the case..but as you realise, it
depends on the actual encoding used.

There is certainly always less information in a same noise, but quieter,
than in the same noise, louder, when in the presence of a fixed noise floor.

Dave Plowman (News) wrote:
In article ,
jasee wrote:
I can't remember being particularly impressed with the original quad
electrostatics particularly with full orchestras and organs, as in
certain organ concertos, really almost disappeared. I really couldn't
see what the fuss was about. And I remember the quad 405 amplifier I
bought becoming so noisy that I sent it back.

You're a few years apart. The original '57 would have been driven with
Quad II valve amps. Or even just one. The first Quad transistor amp was
the 303. Early 405s used a poor op amp which was changed for a better one
shortly after introduction.

The original '57 was very room sensitive. Was your room approaching a
cube? The bass output seemed to disappear in those. Worked best in a long
room with the speakers across the narrow wall. Of course it only went down
to about 40 Hz in the best of circumstances. And the lack of resonances
made it appear bass light compared to a honky cabinet speaker.


IIRC they didn't go down that far at any power at all. ISTR about 100Hz
plus..always needed a subwoofer IMHO, and sensitivity was pretty shocking.

I always felt the quad valves were good, but neither the 303 nor the 405
were in any way special: Certainly I was able to beat them with my own
designs.

No idea what they are like now.

jasee wrote:
"Dave Plowman (News)" wrote in message
...
In article ,
jasee wrote:
I can't remember being particularly impressed with the original quad
electrostatics particularly with full orchestras and organs, as in
certain organ concertos, really almost disappeared. I really couldn't
see what the fuss was about. And I remember the quad 405 amplifier I
bought becoming so noisy that I sent it back.
You're a few years apart. The original '57 would have been driven with
Quad II valve amps. Or even just one. The first Quad transistor amp was
the 303. Early 405s used a poor op amp which was changed for a better one
shortly after introduction.

The original '57 was very room sensitive. Was your room approaching a
cube? The bass output seemed to disappear in those. Worked best in a long
room with the speakers across the narrow wall. Of course it only went down
to about 40 Hz in the best of circumstances. And the lack of resonances
made it appear bass light compared to a honky cabinet speaker.

I borrowed the quads for about a couple of weeks in London and they were
with valve amps and yes the room was more of less square and the particular
organ had a low 32cycle (IIRC) note which you simply couldn't hear at all.
The 405 I bought myself later but changed for a Raford transistor amp which
was simply worlds better. Never did like valves anyway.


The Radford WAS good, from distant hazy memory.

These days, with FET outputs and as long as you are prepared fr an amp
that runs fairly hot, you can beat bipolars hands down really, and knock
valves into the middle of the last century, where they belong ;-)

But there is no incentive to do that. People Want Valves, and a tranny
amp handbuilt and tuned costing £700 is simply not there marketing wise
when it come to the pure red glow of a few EL34's..

I liked the ELS for classical at modest volume, but they were hopeless
for jazz or rock. My favorites were horns..good horns, with either bass
reflex for the bottom, or infinite baffle. Never did build any concrete
bass horns..

The Natural Philosopher wrote:
jasee wrote:
"Dave Plowman (News)" wrote in message
...
In article ,
jasee wrote:
I can't remember being particularly impressed with the original
quad electrostatics particularly with full orchestras and organs,
as in certain organ concertos, really almost disappeared. I really
couldn't see what the fuss was about. And I remember the quad 405
amplifier I bought becoming so noisy that I sent it back.
You're a few years apart. The original '57 would have been driven
with Quad II valve amps. Or even just one. The first Quad
transistor amp was the 303. Early 405s used a poor op amp which was
changed for a better one shortly after introduction.

The original '57 was very room sensitive. Was your room approaching
a cube? The bass output seemed to disappear in those. Worked best
in a long room with the speakers across the narrow wall. Of course
it only went down to about 40 Hz in the best of circumstances. And
the lack of resonances made it appear bass light compared to a
honky cabinet speaker.

I borrowed the quads for about a couple of weeks in London and they
were with valve amps and yes the room was more of less square and
the particular organ had a low 32cycle (IIRC) note which you simply
couldn't hear at all. The 405 I bought myself later but changed for
a Raford transistor amp which was simply worlds better. Never did
like valves anyway.
The Radford WAS good, from distant hazy memory.

These days, with FET outputs and as long as you are prepared fr an amp
that runs fairly hot, you can beat bipolars hands down really, and
knock valves into the middle of the last century, where they belong
;-)

Quite right!


But there is no incentive to do that. People Want Valves, and a tranny
amp handbuilt and tuned costing £700 is simply not there marketing
wise when it come to the pure red glow of a few EL34's..

I liked the ELS for classical at modest volume, but they were hopeless
for jazz or rock. My favorites were horns..good horns, with either
bass reflex for the bottom, or infinite baffle. Never did build any
concrete bass horns..

Transmission lines were the answer, Radford built a good one as did IMF
(IIRC), but there were lots of homemade designs.

"Bill Wright" wrote in message
...

"Steve Terry" wrote in message
...
"Bill Wright" wrote in message
...
"The Natural Philosopher" wrote in message
...
Mark wrote:
There is no perfection in nature.

You obviously haven't seen my body.
Bill

Apparently nether has Maddy Prior
Steve Terry

Last night I was working late in the office. It's a small room with two
PCs and some CCTV stuff. It was very hot indeed so I took l my top off.
After a while I was forced to take my trousers off, so I sat working in my
pants. Hil came in and passed a rude comment, and opened the window. After
a while the room cooled down so I put my top on. Some time later I knocked
off, so I put my shoes on and took the dogs round the field. When I got
back Hil said, "What are you doing, going out without your trousers?" It
didn't matter because it was dark, but I'm glad I didn't meet the lampers
or anyone.
Bill

I was down to my Y fronts last night wandering around the house, my Heather
begged me not to go into the kitchen as there aren't any curtains or nets up
on the window at the moment, she was afraid the neighbours would suffer.

Steve Terry

On Fri, 3 Jul 2009 19:55:14 +0100, "jasee" wrote:

Transmission lines were the answer, Radford built a good one as did IMF
(IIRC), but there were lots of homemade designs.

I had a pair of TDL Monitors for some time driven by a QUAD 606 but changed them
for ATC SCM 20SLs driven by AVI monoblocks for reasons which I can't remember.
At that time we were listening to a lot of live classical music and the ATC/AVI
combination was the only set up I've owned which didn't disappoint after
returning home from a concert. We still have them, fourteen years on.
--
Alan White
Mozilla Firefox and Forte Agent.
Twenty-eight miles NW of Glasgow, overlooking Lochs Long and Goil in Argyll, Scotland.
Webcam and weather:- http://windycroft.gt-britain.co.uk/weather