Volume control at the speaker?

[Pooh Bear]

Except that that link may not have been too clever;(. Follow this one.
Of course not aimed at domestic consumers but as to CD-v-FM
quality....

http://www.broadcast.harris.com/radio/transmission/exciters.asp

I'm not sure what your point is.

No 'processor' can get over the basic problems relating to FM transmission /
reception that are inherent to the technique.

Are you aware of how the 'stereo' is extracted from the one transmitted
signal ? It isn't exactly a 'clean' process !

Graham

 

[Keith G]

That's what I thought you'd say. OK, I know where you're coming from
now... I
just hope you realize that, by volume, there's a lot more "shit sold by
Wal*Mart" than pro-sumer stuff! (Tried to buy an SVHS VCR lately?)

I had a debate years ago with a fellow college student about the best way
to
make money selling electronics... he wanted the Wal*Mart approach (low
cost,
high volume, "good enough") whereas I advocated the "boutiquey" approach
(lower volumes, noticeable higher cost, honest-to-God specs that would
meet
with the approval of even the most discerning Pooh Bears). I still prefer
the
later, but I can see now how, from many angles, my friend's approach is a
lot
easier to implement for the average Joe, even if there is a lot more
competition.

Typical snobby view that gets 'audiophiles' a bad name - pound for pound
WalMart 'prosumer' stuff probably gives much more satisfaction/pleasure to a
vastly higher number of people than so-called 'high end' gear does and is
probably a lot more 'honest' too!

As to approach, if you want to make money it is *beyond question* that the
'pile it high, sell it cheap' is by far the most effective method - ask
Walmart (256.3 BILLION USD t/o for 2004 according to:

http://www.walmartfacts.com/docs/738_NewsDesk04AnnualReport_1497483218.pdf

???)..... ;-)

 

[tony sayer]

I'm not sure what your point is.

Just to point out that you aren't that correct in the statement that,

Certainly a *big* reduction in quality compared to CD.

Isn't that accurate as such.

No 'processor' can get over the basic problems relating to FM transmission /
reception that are inherent to the technique.

Its not a processor at all!. Its just an exciter, that's the bit that
makes up the drive (carrier and modulation) that you then amplify up to
the desired power level required for the service cover you wish to
provide in an FM transmission system.

The processor will be in front of that, but that in itself is used for
other reasons.

Are you aware of how the 'stereo' is extracted from the one transmitted
signal ? It isn't exactly a 'clean' process !

It isn't as bad as you seem to think. FWIW I had one of these units
outside the listening room because its quite noisy, cooling fans etc,
and drove that with a Sony CD player with a pro AES/EBU output and
another output direct to a power amp driving a pair of Quad ESL63
speakers. In front of an audience of Six hi-fi types very few could tell
the difference, let alone repeatedly, between direct and off-air via an
Audiolab tuner.


Of course this may not a a very worthwhile experiment/demo as very few
will get to hear an FM TX of that calibre directly driven and without a
Processor in line.....

 

[tony sayer]

Very impressive, but...

(1) A great FM exciter does not guarantee a clean signal as
received. There's many a slip between the cup and the lip.

Yes agreed..

(2) As big and expensive as this Harris puppy is, it can
only do 78 dB dynamic range. That can be matched or beaten
by a $70 PC sound card. I'm talking stereo performance
compared to stereo performance.

Again, add antennas, signal path, and even an exceptionally
good FM receiver, and real-world performance through the
whole loop is not all that wonderful by modern standards.

It's streets ahead of the poxy DAB system that we suffer in the UK
Arny;!...

Admittedly the basic performance specs of 78 dB dynamic
range and 0.5 dB FR aren't shabby, but it doesn't compare to
what you can do with digital coding for a tiny fraction of
the price.

We are talking about "radio" broadcasting ?..

(3) The worst thing about FM broadcasting is what they do to
the signal intentionally, not accidentally. Of course we
can't blame the medium for that.

Agreed;( or perhaps ;(Of course the original poster was stating that FM was nowhere near CD
quality.

For real "CD quality" you need UK DAB !.

Still the BBC are doing very well knocking out the Proms on VHF FM and
with the old Optimod's switched out and very fine a sound it is too)..

Even despite the odd acoustics of the old Albert Hall!.....

 

[Mark]

Very impressive, but...

(1) A great FM exciter does not guarantee a clean signal as
received. There's many a slip between the cup and the lip.

(2) As big and expensive as this Harris puppy is, it can
only do 78 dB dynamic range. That can be matched or beaten
by a $70 PC sound card. I'm talking stereo performance
compared to stereo performance.

Again, add antennas, signal path, and even an exceptionally
good FM receiver, and real-world performance through the
whole loop is not all that wonderful by modern standards.

Admittedly the basic performance specs of 78 dB dynamic
range and 0.5 dB FR aren't shabby, but it doesn't compare to
what you can do with digital coding for a tiny fraction of
the price.

(3) The worst thing about FM broadcasting is what they do to
the signal intentionally, not accidentally. Of course we
can't blame the medium for that.

To cover 1 house you can use 10 mW and have essentially noise free
reception...

and you can use ordinary RF receivers and even portable radios outside
if you like.

I have an FM link at home and it is very handy.

I agree, you can get better performance on paper using a digital link
with special receivers etc etc etc. But its probably not relevent for
the application.

Mark

 

[Zak]

Zak wrote:

It's *possible* - just not currently commercially available. That's the raw data
rate btw. It'll take a lot more bandwidth to get it wirelessly from A to B.

hmmm... what if you'd use say an Asus WL-HDD as a wireless file
server/access point ($90, need to add laptop drive) and say a wl-500g
(or another wl-hdd) with a USB audio dongle as playback? Will need some
programming but it runs linux. But it does not have a user interface at
all...

These all have wired ethernet as well. Fun toys.


Thomas

 

[Pooh Bear]

hmmm... what if you'd use say an Asus WL-HDD as a wireless file
server/access point ($90, need to add laptop drive) and say a wl-500g
(or another wl-hdd) with a USB audio dongle as playback? Will need some
programming but it runs linux. But it does not have a user interface at
all...

These all have wired ethernet as well. Fun toys.

The original requirement was a *simple* way to get audio from A to B. This chip is
designed to do the job. Only recently released.

http://www.nvlsi.no/index.cfm?obj=document&act=display&doc=242

Graham

 

[TokaMundo]

The original requirement was a *simple* way to get audio from A to B. This chip is
designed to do the job. Only recently released.

http://www.nvlsi.no/index.cfm?obj=document&act=display&doc=242

Graham

Try http://www.avalonrf.com/ They have been doing it for years.
Their audio links even have sub channels with data streams.

Top of the line, true diversity multiple band choices, and the
aircraft carrier boys use it as well as cops and race car teams/tracks

On the carriers, one guy wears a hard drive DR system, and all comms
get recorded on the ship as well.

 

[TokaMundo]

Sorry, that's:

44,1000 samples/sec * 16 bits/sample * 2 channels = 1,411,200 bits/sec

or

44.1 ksamples/sec * 16 bits/sample * 2 channels = 1,411,200 bits/sec

In any case, it's 1.4 Mbits/sec, NOT 6 Mbits/sec.

6 is almost enough to pass DVD quality A/V streams.

 

[TokaMundo]

Which is why you have a receive buffer and there's latency between transmission and reception.

Digital transmission schemas that do not have resend capacity for
broken packets utilize FEC. Pretty basic.

 

[TokaMundo]

I think that's the point.

You could lose *any* given sample. The simplest method of dealing with that is to continually
broadcast the data twice. If the previous error rate was 1 in 10^-6 then that method ( assuming
random errors ) should reduce it to1 in 10^-12. Not *perfect* but pretty damn good !

Sending the data twice also removes the need for negotiation over missed samples between transmitter
and receiver. It sounds like a done deal to me.

That also explains the need for a much higher data rate than 1.4Mbps.

One does not need to send the entire datagram twice. Just package
it up with FEC and let the error correction manage lost packets.

Once the bit error rate gets up over around ten percent, one would
then hear dropouts or other anomalous audio.

 

[Fred Abse]

You don't have to use TCP/IP, and you don't need, or even want, lost
packet redundancy, at least not for streaming audio. If you need a
protocol at all, use something like UUCP.

Don't you mean UDP?

 

[TokaMundo]

Trust me - uncompressed digital audio wireless links are not yet available. I
expect the MP101 compresses wav files on the fly.

Sure they are, just not at the consumer level, unless you gotz lotz
o money.

 

[TokaMundo]

Who wants to walk down the hall to insert a CD when it can be instantly
available on a file server?

Not only that, but who would think that the folks in all 8 rooms
want to hear the same thing? Choice is king!

 

[TokaMundo]

[snip]
Who wants to walk down the hall to insert a CD when it can be instantly
available on a file server?
[snip]

Some of us have a 300 CD jukebox ;-)

...Jim Thompson

Who wants to walk down the hall to select a CD in the jukebox when it
can be instantly available on a file server?

It doesn't hurt to get some exercise you know !

Besides who wants to so nerdy as to run a file server just to listen to music
? You won't get laid that way !

A file server is far and away better than a CD jukebox. One can
convert ALL of one's collection, store it and more on the server, and
access it from any client in the house, including the media server
that runs through the TV.

 

[Pooh Bear]

Try http://www.avalonrf.com/ They have been doing it for years.
Their audio links even have sub channels with data streams.

Top of the line, true diversity multiple band choices, and the
aircraft carrier boys use it as well as cops and race car teams/tracks

On the carriers, one guy wears a hard drive DR system, and all comms
get recorded on the ship as well.

Thanks for the link but their stuff seems to be targeted at generalised data comms.

I was looking for ( and have now found ) an application specific solution to my
reqirement that doesn't involve me in doing all the protocol stuff and whatever. It's
virtually 'ready to go'.

Graham

 

[Pooh Bear]

6 is almost enough to pass DVD quality A/V streams.

Which is *compressed* !

I need *uncompressed* with guaranteed QoS as Nordic Semi call it. Hence the overhead on
the data rate.

Graham

 

[Pooh Bear]

One does not need to send the entire datagram twice. Just package
it up with FEC and let the error correction manage lost packets.

Once the bit error rate gets up over around ten percent, one would
then hear dropouts or other anomalous audio.

Yup. There are various solutions to the problem though. Some are more oriented to fixing corrupted data (
the occasional bit or so ) - other techniques can deal with lost packets.

Here's an extract from the data sheet of the part I'm now rather interested in.


4.3.3 Quality of Service engine
The primary function of the quality of service engine is to deliver a robust communication channel
between the audio transmitter and audio receiver in an audio streaming application. Several data streams
with different properties are handled. The available bandwidth is shared between audio data, service data
and remote data.

Data integrity is ensured through a number of RF protocol features:

1. Packets of data are sent in frames and integrity of each packet is ensured as every packet has a
complete build of RF address, payload and CRC.

2. Packets that are lost or received with errors are handled by the error correction level of the quality
of service engine; a two way, acknowledge protocol:
When a packet is received by ARX, it’s registered and CRC is checked.
After ARX has received a frame it sends a packet back to ATX
acknowledging the packets that where successfully transferred. Packets lost
or received with errors will be re-sent from ATX in the next frame.

3. Finally the information (audio data) is spread over the 2.4 GHz band by use of an adaptive frequency
hopping algorithm. Through this a nRF24Z1 link can handle RF propagation challenges like reflections and
multi-path fading and not least avoid heavily trafficked areas of the 2.4 GHz band. The 2.4 GHz band is a
world wide open RF band and co-existence with RF systems such as Bluetooth, ZigBee,
WLAN/WiFi as well as other nRF applications, is increasingly important.

nRF24Z1 constantly monitors the quality of the RF link and numbers indicating total link quality are
available for external control devices in registers. nRF24Z1 can also be set to interrupt external
controller devices upon poor link quality before RF link is lost. An external controller device can hence
take further actions to improve link quality or warn end user if RF link margins are poor.

The secondary function of the QoS module is to run a link initialization algorithm which manages initial
connect and re-connect if link is lost (ex: out of range) between paired nRF24Z1’s. Several schemes are
available to enable nRF24Z1 connection without end user involvement.


Graham

 

[Pooh Bear]

Sure they are, just not at the consumer level, unless you gotz lotz
o money.

That was indeed what I meant ( see the context of the thread ). ;-)

Graham

 

[Pooh Bear]

Jim Thompson wrote:

[snip]
Who wants to walk down the hall to insert a CD when it can be instantly
available on a file server?
[snip]

Some of us have a 300 CD jukebox ;-)

...Jim Thompson

Who wants to walk down the hall to select a CD in the jukebox when it
can be instantly available on a file server?

It doesn't hurt to get some exercise you know !

Besides who wants to so nerdy as to run a file server just to listen to music
? You won't get laid that way !

A file server is far and away better than a CD jukebox. One can
convert ALL of one's collection, store it and more on the server, and
access it from any client in the house, including the media server
that runs through the TV.

I generally like the idea - it just means you need a PC to listen though. That may
not be as attractive so some ppl ( often called wives or girlfriends ) as you
find it yourself.

Graham