France to test DAB+
DAB sounds worse than FM
"January 16th
The French regulator, the Conseil Supérieur de l'Audiovisuel, has authorised
two companies to conduct test transmissions using DAB+ (aka DAB v2) and DRM
and DRM+. VDL is authorised to conduct Band III test transmissions in Paris
using DAB+ (using the AAC+ codec) until February 13th, and using DRM in the
26 MHz band in Lyon and Valence until March 13th. TDF is authorised to
conduct test transmissions in Rennes, first in DRM and then in DRM+ in Band
I until September 12th."
Irony of all ironies: M. Croiset works for VDL, which is carrying out the
DAB+ tests on the system he has argued against using for so long.
Aaaaaaagh, the sweet smell of victory wafts around alt.radio.digital once
again. It is a beautiful smell - like cycling through the smells of rose
pettles, pear drop sweets, or freshly-baked biscuits (you just can't beat
the smell of the McVities factory on the A6).
--
Steve - www.digitalradiotech.co.uk - Digital Radio News & Info
Find the cheapest Freeview & DAB prices:
http://www.digitalradiotech.co.uk/freeview/freeview_receivers.php
http://www.digitalradiotech.co.uk/dab/dab_radios.php
Nicolas Croiset
<dab.is@dead> wrote:
>http://www.wohnort.demon.co.uk/DAB/
>
>"January 16th
>The French regulator, the Conseil Supérieur de l'Audiovisuel, has authorised
>two companies to conduct test transmissions using DAB+ (aka DAB v2) and DRM
>and DRM+. VDL is authorised to conduct Band III test transmissions in Paris
>using DAB+ (using the AAC+ codec) until February 13th, and using DRM in the
>26 MHz band in Lyon and Valence until March 13th. TDF is authorised to
>conduct test transmissions in Rennes, first in DRM and then in DRM+ in Band
>I until September 12th."
>
>Irony of all ironies: M. Croiset works for VDL, which is carrying out the
>DAB+ tests on the system he has argued against using for so long.
>
>Aaaaaaagh, the sweet smell of victory wafts around alt.radio.digital once
>again. It is a beautiful smell - like cycling through the smells of rose
>pettles, pear drop sweets, or freshly-baked biscuits (you just can't beat
>the smell of the McVities factory on the A6).
Don't imagine something !! You don't impact anything about what we do.
We are working on these projects since many years but it was
impossible to say anything before it is official.
Faites attention, vous avez les chevilles qui gonflent !!
davidr...@postmaster.co.uk
> Don't imagine something !! You don't impact anything about what we do.
> We are working on these projects since many years but it was
> impossible to say anything before it is official.
Was it also impossible _not_ to say "mp2 is really good, it's good
enough, you don't need anything else really"? Because from memory you
tried to make that point again and again.
If you knew something better was coming but couldn't say anything, why
didn't you keep your mouth shut? Why did you choose to say silly
things?
Cheers,
David.
DAB sounds worse than FM
> On Wed, 17 Jan 2007 13:53:40 GMT, "DAB sounds worse than FM"
> <dab.is@dead> wrote:
>
>> http://www.wohnort.demon.co.uk/DAB/
>>
>> "January 16th
>> The French regulator, the Conseil Supérieur de l'Audiovisuel, has
>> authorised two companies to conduct test transmissions using DAB+
>> (aka DAB v2) and DRM and DRM+. VDL is authorised to conduct Band III
>> test transmissions in Paris using DAB+ (using the AAC+ codec) until
>> February 13th, and using DRM in the 26 MHz band in Lyon and Valence
>> until March 13th. TDF is authorised to conduct test transmissions in
>> Rennes, first in DRM and then in DRM+ in Band I until September
>> 12th."
>>
>> Irony of all ironies: M. Croiset works for VDL, which is carrying
>> out the DAB+ tests on the system he has argued against using for so
>> long.
>>
>> Aaaaaaagh, the sweet smell of victory wafts around alt.radio.digital
>> once again. It is a beautiful smell - like cycling through the
>> smells of rose pettles, pear drop sweets, or freshly-baked biscuits
>> (you just can't beat the smell of the McVities factory on the A6).
>
> Don't imagine something !! You don't impact anything about what we do.
Yeah, I must have been imagining those hundreds of posts where you stuck up
for using MP2 and were against using AAC/AAC+.
> We are working on these projects since many years but it was
> impossible to say anything before it is official.
And yet right up until WorldDAB actually announced that it was adopting AAC+
and it had become increasingly obvious that France would end up using AAC+
(on whatever system), you *still* continued to argue in favour of using MP2
and being opposed to using AAC+.
> Faites attention, vous avez les chevilles qui gonflent !!
Google translates that to:
"Made attention, you have the ankles which inflate"
What a bizarre thing to say! And I can assure you that my ankles are not
inflated whatsoever - I did once break an ankle, however, and it did balloon
up to a gigantic size, but it returned to normal proportions soon after the
operation, and it has remained that size ever since. I do thank you for your
concern about the size of my ankles, though, but there is nothing to worry
about, and, if I may say so, they are in fact rather shapely ankles.
DAB sounds worse than FM
> Nicolas Croiset wrote:
>
>> Don't imagine something !! You don't impact anything about what we
>> do. We are working on these projects since many years but it was
>> impossible to say anything before it is official.
>
> Was it also impossible _not_ to say "mp2 is really good, it's good
> enough, you don't need anything else really"? Because from memory you
> tried to make that point again and again.
Absolutely.
> If you knew something better was coming but couldn't say anything, why
> didn't you keep your mouth shut?
Couldn't agree more.
>Why did you choose to say silly
> things?
Because he'd dug himself into the mother of all holes by arguing so strongly
in favour of MP2. Now of course he's trying to deny any of it ever even
bloody happened, as is the way of the Croiset...
Boltar
Nicolas Croiset wrote:
> Faites attention, vous avez les chevilles qui gonflent !!
"Pay attention , you have inflatable ankles". Huh?? Must be some french
saying I'm not aware of.
B2003
Nicolas Croiset
<davidr...@postmaster.co.uk> wrote:
>Nicolas Croiset wrote:
>
>> Don't imagine something !! You don't impact anything about what we do.
>> We are working on these projects since many years but it was
>> impossible to say anything before it is official.
>
>Was it also impossible _not_ to say "mp2 is really good, it's good
>enough, you don't need anything else really"? Because from memory you
>tried to make that point again and again.
Yes I maintain what I said many times, if you use MP2 encoder
carrefully you can do something which is equal or superior to FM even
at low bitrate.
Nicolas Croiset
wrote:
It can be interpret as : You have a swollen head.
;-)
Doug McDonald
> Yes I maintain what I said many times, if you use MP2 encoder
> carrefully you can do something which is equal or superior to FM even
> at low bitrate.
>
>
REALLY?
FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
in excess of 70Db with less than 0.3% distortion, stereo. That
0.3% is very low order harmonic distortion.
MP2 is capable of that, of course ... but not at any bitrate
less than 196 kb/sec, which is not "low".
Please do not try to wiggle out of your lie. Equal or
superior means equal or superior. It does not mean
"I think it may be equal perceptually". It means
"or better" .. which says that it has to meet the
same specs, tested in the most robust way. And this means
comparing program material directly, output versus input.
At 16 kHz bandwidth, FM is capable of perfectly flat
frequency and phase response (with digital prefiltering
and multiplex signal generation.)
Doug McDonald
tony sayer
<ncro...@vdldiffusion.com> writes
What's the French for "Bollocks"?...
--
Tony Sayer
DAB sounds worse than FM
Then you are simply confirming that you are either a liar, an incompetent or
both - I really don't care which case applies.
DAB sounds worse than FM
Les bollocks?
DAB sounds worse than FM
No, I've simply been completely vindicated in what I've been saying since
2002, which is that the old DAB system is shit and should never have been
used.
Nicolas Croiset
<mcdonald@SnPoAM_scs.uiuc.edu> wrote:
>Nicolas Croiset wrote:
>
>> Yes I maintain what I said many times, if you use MP2 encoder
>> carrefully you can do something which is equal or superior to FM even
>> at low bitrate.
>>
>>
>
>REALLY?
>
>FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
>in excess of 70Db with less than 0.3% distortion, stereo. That
>0.3% is very low order harmonic distortion.
>
No FM is not able to do that.
The bandwidth is 20 to 15kHz (not 16kHz) and the bandwidth is not
linear due to the pre-emphasis / de-emphasis functions.
The S/N is not 70dB in stereo (and mono also), the best value is
around 40dB, not more. For the distortion I will not announce anything
because you will be horrified.
The stereo separation is less than 40dB when you add the loss due to
the encoder and the loss due to the receiver.
Richard Evans
spelling, and my spell checker doesn't know either).
On the other hand, not many French people would realise that in England
a Cul De Sac, is a road that doesn't go anywhere. In French it
apparently means nothing more than the opening of a sack.
Strange how words are used completely differently in different languages.
Richard E.
Silk
On the other hand, not many French people would realise that in England
> a Cul De Sac, is a road that doesn't go anywhere. In French it
> apparently means nothing more than the opening of a sack.
And "Chicken in a Basket" means exactly that in England, not just fried
reformed battered chicken and chips, as it seems to in France.
Silk
> Because he'd dug himself into the mother of all holes by arguing so strongly
> in favour of MP2. Now of course he's trying to deny any of it ever even
> bloody happened, as is the way of the Croiset...
Sometimes you have to make compromises when you have a job.
DAB sounds worse than FM
These 2 posts perfectly summarise your very long-standing and incompetent
support of MP2, because you first claimed that MP2 at 128 kbps can sound
better than FM, and now you're lying that the maximum SNR is 40 dB, and you
then criticise the stereo separation *on FM* when MP2 at 128 kbps uses
intensity stereo for all frequencies above, IIRC, 2.25 kHz, which means that
all frequencies above this are pan-potted mono.
You've pretty much always been on your own in suggesting that MP2 at 128
kbps is going to sound better than FM, so I'll ignore that as being
ridiculous nonsense. But the bits in this last post just show that you're
either a liar or don't actually have any idea what you're going on about.
For example, you're suggesting that the SNR is a maximum of 40 dB on FM,
whereas I can remember reading in an EBU Tech Review article written by one
of your pro-MP2 incompetent mates that said you need 40 dB SNR as a MINIMUM
for stereo!
To sum up: Nicolas, your opinions/suggestion have ALWAYS been INCOMPETENT,
but you've now TOTALLY LOST THIS ARGUMENT, so please be on your way, because
you've wasted far too much of my time.
Nicolas Croiset
Hi,
a "cul de sac" means in french exactly what you said :-)
Bye.
Nicolas Croiset VDL
http://www.vdl.fr/
DAB sounds worse than FM
I think you're mistaking me for someone that gives a shit about your
opinion.
Silk
> Silk wrote:
>> DAB sounds worse than FM wrote:
>>
>>> Because he'd dug himself into the mother of all holes by arguing so
>>> strongly in favour of MP2. Now of course he's trying to deny any of
>>> it ever even bloody happened, as is the way of the Croiset...
>> Sometimes you have to make compromises when you have a job.
>
>
> I think you're mistaking me for someone that gives a shit about your
> opinion.
>
>
Richard Evans
Was it part of Nicolas' job to try to sell the mp2 codec to us here on
this NG?
Nicolas Croiset
>Nicolas Croiset wrote:
>> On Wed, 17 Jan 2007 11:21:56 -0600, Doug McDonald
>>> FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
>>> in excess of 70Db with less than 0.3% distortion, stereo. That
>>> 0.3% is very low order harmonic distortion.
>>>
>>
>> No FM is not able to do that.
>>
>> The bandwidth is 20 to 15kHz (not 16kHz) and the bandwidth is not
>> linear due to the pre-emphasis / de-emphasis functions.
>> The S/N is not 70dB in stereo (and mono also), the best value is
>> around 40dB, not more. For the distortion I will not announce anything
>> because you will be horrified.
>> The stereo separation is less than 40dB when you add the loss due to
>> the encoder and the loss due to the receiver.
>
>
>These 2 posts perfectly summarise your very long-standing and incompetent
>support of MP2, because you first claimed that MP2 at 128 kbps can sound
>better than FM, and now you're lying that the maximum SNR is 40 dB, and you
>then criticise the stereo separation *on FM* when MP2 at 128 kbps uses
>intensity stereo for all frequencies above, IIRC, 2.25 kHz, which means that
>all frequencies above this are pan-potted mono.
This is your assertion without any good materials with professional
encoders and professional audio sources. Once again, what you listen on
BBC is not a pure audio source but an audio source already compressed by
different audio processors and the result is very bad in my opinion.
And you forget the mandatory audio processing in FM if you want to not
exceed the maximum deviation of +-75kHz which imply a low audio dynamic.
>
>You've pretty much always been on your own in suggesting that MP2 at 128
>kbps is going to sound better than FM, so I'll ignore that as being
>ridiculous nonsense. But the bits in this last post just show that you're
>either a liar or don't actually have any idea what you're going on about.
>For example, you're suggesting that the SNR is a maximum of 40 dB on FM,
>whereas I can remember reading in an EBU Tech Review article written by one
>of your pro-MP2 incompetent mates that said you need 40 dB SNR as a MINIMUM
>for stereo!
Yes the SNR is around 40dB when you receive FM very well without any
multipath. There is not so many people which can receive FM in this
condition. The SNR could be better if you use a pre-emphasis of 75盜
instead of 50盜. In Europe the spec is 50盜. Remember that all theorical
values give on a FM receiver are made in laboratory with perfect
conditions which does not exist in real life and it is 400 Hz !
For example, in our company we have a professional FM receiver which is
able to detect the intensity of multipaths and it is very important ot
have less than 5%, if it is not the case the audio/stereo/RDS
measurements are wrong.
>
>To sum up: Nicolas, your opinions/suggestion have ALWAYS been INCOMPETENT,
>but you've now TOTALLY LOST THIS ARGUMENT, so please be on your way, because
>you've wasted far too much of my time.
Don't reply if it waste your time.
DAB sounds worse than FM
I really don't care what you think, and the only reason you've not been
killfiled is because you use an email address that blocks quite a few other
people's posts as well.
Silk
> Silk wrote:
>> DAB sounds worse than FM wrote:
>>
>>> Because he'd dug himself into the mother of all holes by arguing so
>>> strongly in favour of MP2. Now of course he's trying to deny any of
>>> it ever even bloody happened, as is the way of the Croiset...
>>
>>
>> Sometimes you have to make compromises when you have a job.
>
> Was it part of Nicolas' job to try to sell the mp2 codec to us here on
> this NG?
Do you think insulting someone in a position of influence is going to
win them around?
Silk
> Aaaaaaagh, the sweet smell of victory wafts around alt.radio.digital once
> again. It is a beautiful smell - like cycling through the smells of rose
> pettles, pear drop sweets, or freshly-baked biscuits (you just can't beat
> the smell of the McVities factory on the A6).
You should try soap, that smells even better.
tony sayer
<ncro...@vdldiffusion.com> writes
What was the French for "Bollocks" again?....
--
Tony Sayer
DAB sounds worse than FM
The audio you're referring to obviously goes out on FM as well, but the
difference is that the audio quality is higher on FM - everybody apart from
you seems capable of realising that that in itself proves that FM sounds
better than DAB at 128 kbps.
> And you forget the mandatory audio processing in FM if you want to not
> exceed the maximum deviation of +-75kHz which imply a low audio
> dynamic.
It is NOT mandatory to do anything - broadcasters CHOOSE to process their
audio on FM, and there's very, very, very few stations that don't process
their audio on DAB.
>> You've pretty much always been on your own in suggesting that MP2 at
>> 128 kbps is going to sound better than FM, so I'll ignore that as
>> being ridiculous nonsense. But the bits in this last post just show
>> that you're either a liar or don't actually have any idea what
>> you're going on about. For example, you're suggesting that the SNR
>> is a maximum of 40 dB on FM, whereas I can remember reading in an
>> EBU Tech Review article written by one of your pro-MP2 incompetent
>> mates that said you need 40 dB SNR as a MINIMUM for stereo!
>
> Yes the SNR is around 40dB when you receive FM very well without any
> multipath. There is not so many people which can receive FM in this
> condition.
You said that 40 dB SNR is the MAXIMUM, which is yet another Nicolas Croiset
lie / made up fact to suit your argument. The reality is that thermal noise
power is a constaant that depends on things like banwidth:
Noise power = 4 k T B
k = Boltzmann's constant (constant)
T = absolute temperature (constant for a given environment)
B = bandwidth (constant for FM reception)
You've then got to account for the noise figure of the receiver, which is a
few dB. So we basically have a constant noise level, so if the signal level
increases then the bloody SNR will obviously increase, or don't you even
understand that SNR is given by the following equation:
SNR = signal power / noise power
and it will continue to increase as the signal level increases (e.g. if
there's 2 receivers and one is closer to the transmitter than the other then
the closer one will have a higher SNR), so please give a
scientific/engineering explanation that accounts for this "MAXIMUM" 40 dB
SNR on FM. (I'm ignoring overloading here)
This has always been my main problem with you, because you just make things
up because they sound good for your argument, but in reality you just never
back up your claims with any explanation for why something is as you
suggest - it's because you cannot back up your claims, because the reality
is that you don't actually understand the engineering issues involved.
So come on, explain why there is a maximum SNR of 40 dB, or admit that you
just made that up - one or the other.
> The SNR could be better if you use a pre-emphasis of 75盜
> instead of 50盜. In Europe the spec is 50盜. Remember that all
> theorical values give on a FM receiver are made in laboratory with
> perfect conditions which does not exist in real life and it is 400 Hz
> !
Is this another of your big claims? Can you provide a reference to this
claim, or is it just another claim that we have to just take your word for
(remember that I don't believe a single thing you say, because I know that
you lie all the time).
>> To sum up: Nicolas, your opinions/suggestion have ALWAYS been
>> INCOMPETENT, but you've now TOTALLY LOST THIS ARGUMENT, so please be
>> on your way, because you've wasted far too much of my time.
>
> Don't reply if it waste your time.
No, YOU waste my time by making incorrect claims that you can't back up and
which I feel obliged to shoot down.
DAB sounds worse than FM
Tony, you've got a lot of experience with using FM kit, so what's the
highest SNR you've ever seen on FM? Croiset claims that 40 dB is the
*maximum* SNR possible, so I think it's time to demolish this claim.
DAB sounds worse than FM
Hahahahahahahahhahhahahahaha, Croiset influential? Don't make me laugh ...
hard!
DAB sounds worse than FM
What??
Doug McDonald
> On Wed, 17 Jan 2007 11:21:56 -0600, Doug McDonald
> <mcdonald@SnPoAM_scs.uiuc.edu> wrote:
>
>> Nicolas Croiset wrote:
>>
>>> Yes I maintain what I said many times, if you use MP2 encoder
>>> carrefully you can do something which is equal or superior to FM even
>>> at low bitrate.
>>>
>>>
>> REALLY?
>>
>> FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
>> in excess of 70Db with less than 0.3% distortion, stereo. That
>> 0.3% is very low order harmonic distortion.
>>
>
> No FM is not able to do that.
>
> The bandwidth is 20 to 15kHz (not 16kHz)
Nothing, these days, limits it to 15 kHz. With a 19 kHz
pilot, 18 kHz is easily feasible.
> and the bandwidth is not
> linear due to the pre-emphasis / de-emphasis functions.
Uh ... preemphasis is a linear function. Distortion is
non-linear, changes in frequency response are linear, and undoable.
> The S/N is not 70dB in stereo (and mono also), the best value is
> around 40dB, not more.
Utter and complete poppycock, assuming standard US
preemphasis. I have actually MEASURED 80 dB in stereo,
RMS noise of a blank channel (input to multiplex encoder
not connected to a source) versus RMS of peak signal
of same station playing test signals. This is a real over the air test,
using a high quality late 1970's tuner. Note, however, that it
is in the US ... Europe would of course be lower S/N due to less
preemphasis.
> The stereo separation is less than 40dB when you add the loss due to
> the encoder and the loss due to the receiver.
There is no loss due to encoder, at least, not significant, in the range
70-80 dB, if you use good oversampling and high quality D/A convertors
at the output. Receiver is similar, if using high quality oversampled
(say 120 kHz bandwidth) A/D convertors.
While, conversely, at 128 kb/sec MPEG has simply terrible distortion.
Otherwise, of course, it is quite nice. At adequate bitrate it can be
as transparent as desired. It's not a bad idea or implementation ...
if you have enough bits.
Doug McDonald
Rayzor
"DAB sounds worse than FM" <dab.is@dead> wrote in message
news:KUwrh.62344$UC.3...@newsfe5-win.ntli.net...
> Silk wrote:
>> DAB sounds worse than FM wrote:
>>
>>> Aaaaaaagh, the sweet smell of victory wafts around alt.radio.digital
>>> once again. It is a beautiful smell - like cycling through the
>>> smells of rose pettles, pear drop sweets, or freshly-baked biscuits
>>> (you just can't beat the smell of the McVities factory on the A6).
>>
>> You should try soap, that smells even better.
>
>
> nic alone your jeolous of him
Stephen
news:RNwrh.62341$UC.1...@newsfe5-win.ntli.net...
low budget equipment are important and probably make up the majority. For
them, discussions about the technical details of perfect FM reception are
irrelevant. Their FM reception is on a budget hi-fi with no stereo defeat
switch and a dangly wire aerial, and is always truly appalling with gross
distortion, mushy hf, and a great deal of hiss. When they upgrade to a mini
hi-fi with DAB (still using a dangly wire aerial) they will be pleased with
the improvement in sound quality.
While I quite agree that we should be using higher bitrates and/or better
codecs for digital radio, I think it's important not to underestimate the
severity of the shortcomings of the standard FM stereo system in everyday
use, and how these shortcomings can easily make the existing DAB system
sound better than FM, particularly for those people using budget receivers.
Nicolas Croiset
>Nicolas Croiset wrote:
>> On Wed, 17 Jan 2007 11:21:56 -0600, Doug McDonald
>> <mcdonald@SnPoAM_scs.uiuc.edu> wrote:
>>
>>> Nicolas Croiset wrote:
>>>
>>>> Yes I maintain what I said many times, if you use MP2 encoder
>>>> carrefully you can do something which is equal or superior to FM even
>>>> at low bitrate.
>>>>
>>>>
>>> REALLY?
>>>
>>> FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
>>> in excess of 70Db with less than 0.3% distortion, stereo. That
>>> 0.3% is very low order harmonic distortion.
>>>
>>
>> No FM is not able to do that.
>>
>> The bandwidth is 20 to 15kHz (not 16kHz)
>
>Nothing, these days, limits it to 15 kHz. With a 19 kHz
>pilot, 18 kHz is easily feasible.
>
Except the specs of stereo FM encoders and stereo FM decoders.
when you build a stereo encoder it is possible to a filter which start a
little bit higher its filtering, but you also have the same filter on
the receiver and its quality is not the same (due to the price
essentially), so the limit of 15kHz is the real limitation and it is
also the mandatory limits in the specifications.
>
>> and the bandwidth is not
>> linear due to the pre-emphasis / de-emphasis functions.
>
>Uh ... preemphasis is a linear function. Distortion is
>non-linear, changes in frequency response are linear, and undoable.
>
>
Except that you have dispertion in components so the result is not the
same on each receiver, for the SAME transmission.
>
>> The S/N is not 70dB in stereo (and mono also), the best value is
>> around 40dB, not more.
>
>
>Utter and complete poppycock, assuming standard US
>preemphasis. I have actually MEASURED 80 dB in stereo,
>RMS noise of a blank channel (input to multiplex encoder
>not connected to a source) versus RMS of peak signal
>of same station playing test signals. This is a real over the air test,
>using a high quality late 1970's tuner. Note, however, that it
>is in the US ... Europe would of course be lower S/N due to less
>preemphasis.
>
And the receiver was directly connected on the transmitter right ?
You are giving values of receivers at a very high price which is not the
common receivers buy by a lot of people.
>
>> The stereo separation is less than 40dB when you add the loss due to
>> the encoder and the loss due to the receiver.
>
>
>There is no loss due to encoder, at least, not significant, in the range
>70-80 dB, if you use good oversampling and high quality D/A convertors
>at the output. Receiver is similar, if using high quality oversampled
>(say 120 kHz bandwidth) A/D convertors.
There is loss from the encoder AND the receiver. The function L+R and
L-R is not always perfect on both, even if it is better on the encoder
side.
Nicolas Croiset
>Nicolas Croiset wrote:
>
>> And you forget the mandatory audio processing in FM if you want to not
>> exceed the maximum deviation of +-75kHz which imply a low audio
>> dynamic.
>
>
>It is NOT mandatory to do anything - broadcasters CHOOSE to process their
>audio on FM, and there's very, very, very few stations that don't process
>their audio on DAB.
It is mandatary not to exceed +-75kHz (ETSI). In France for example it
is written in the license and if you don't respect this value you can
pay a fee.
Radio stations for masking the FM background noise use audio processors
and put the system to its maximum of capabilities and when you measure
the audio dynamic level it is less than 4dB !! So yes DAB at low bitrate
without audio processing is better than FM.
Richard Evans
What I was getting at was, what compromise would he have been making by
simply not coming trying to tell us how good mp2 is. (Unless it was his
job to do so).
Exactly what that has to do with insulting him, I don't know.
Richard E.
than unless it was his job to come here telling us how good mp2 is, why
Richard Evans
> Silk wrote:
>
>>DAB sounds worse than FM wrote:
>>
>>
>>>Aaaaaaagh, the sweet smell of victory wafts around alt.radio.digital
>>>once again. It is a beautiful smell - like cycling through the
>>>smells of rose pettles, pear drop sweets, or freshly-baked biscuits
>>>(you just can't beat the smell of the McVities factory on the A6).
>>
>>You should try soap, that smells even better.
>
>
>
> What??
>
>
out with comments like that.
(PS. Sorry I can't get the spell check to recognise hippocrit)
Don Pearce
<R.P.Evan...@NTLWorld.com> wrote:
>(PS. Sorry I can't get the spell check to recognise hippocrit)
Has it occurred to you to wonder why? Of course you could always just
press the button that adds this neologism to your personal dictionary;
then you'd have no problem.
d
--
Pearce Consulting
http://www.pearce.uk.com
Richard Evans
> On Thu, 18 Jan 2007 08:32:09 GMT, Richard Evans
> <R.P.Evan...@NTLWorld.com> wrote:
>
>
>>(PS. Sorry I can't get the spell check to recognise hippocrit)
>
>
> Has it occurred to you to wonder why?
Spell check to recognise it.
> Of course you could always just
> press the button that adds this neologism to your personal dictionary;
> then you'd have no problem.
If you mean that it is not a proper word, then that is something I
didn't know.
Richard E.
Don Pearce
<R.P.Evan...@NTLWorld.com> wrote:
The word is hypocrite. It comes from hypo - under, which is common
enough and krinein, which means to decide between options. Your first
part - hippo - means horse as in hippopotamus.
Matthelas
Stephen a écrit :
> While I quite agree that we should be using higher bitrates and/or better
> codecs for digital radio, I think it's important not to underestimate the
> severity of the shortcomings of the standard FM stereo system in everyday
> use, and how these shortcomings can easily make the existing DAB system
> sound better than FM, particularly for those people using budget receivers.
Considering the difference between what systems can do in theory and
what is their actual use is a very interesting point. I don't know for
radio but for TV one of the big asset for digitalization is sometimes
coming from the actual state of the old analog reception. When antennas
are self set, it's usually ok in digital (or you don't get anything)
whereas it is usually not that good in analog, at least not as good as
it could be.
One of the clear advantage of digitalization is to keep the border line
sharp. Either you get it clean, either you don't get it. Very few will
get it degraded, while it is a continuous way between perfect to snowy
reception in analog.
Cheers,
M.
davidr...@postmaster.co.uk
> Sorry if this is throwing a spanner in the works, but I think listeners with
> low budget equipment are important and probably make up the majority. For
> them, discussions about the technical details of perfect FM reception are
> irrelevant. Their FM reception is on a budget hi-fi with no stereo defeat
> switch and a dangly wire aerial, and is always truly appalling with gross
> distortion, mushy hf, and a great deal of hiss.
It depends where you live, and what you listen to.
In all the places I've lived (4, urban and rural), it's been very
difficult _not_ to get near perfect reception of the local stations,
even on that dangly bit of wire. It gets a bit hissy if you have a PC
on nearby, but that's all.
In those same locations, national stations have varied between very
good, and exactly as you describe, depending on how far away the
nearest powerful national tx was from where I was listening.
> When they upgrade to a mini
> hi-fi with DAB (still using a dangly wire aerial) they will be pleased with
> the improvement in sound quality.
If it works at all. And if it carries the local stations!
> While I quite agree that we should be using higher bitrates and/or better
> codecs for digital radio, I think it's important not to underestimate the
> severity of the shortcomings of the standard FM stereo system in everyday
> use, and how these shortcomings can easily make the existing DAB system
> sound better than FM, particularly for those people using budget receivers.
FM can be a pain, but I would also observe that, in most highly
populated areas in other countries, the FM dial is full of stations
which blast in at the strength of UK local stations. I wouldn't want to
swap Radio 4 for these stations, but I'm just suggesting that for lots
of people, DAB isn't going to improve reception in any particularly
noticeable way.
Cheers,
David.
davidr...@postmaster.co.uk
> On Wed, 17 Jan 2007 11:21:56 -0600, Doug McDonald
> <mcdonald@SnPoAM_scs.uiuc.edu> wrote:
> >FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
> >in excess of 70Db with less than 0.3% distortion, stereo. That
> >0.3% is very low order harmonic distortion.
>
> No FM is not able to do that.
>
> The bandwidth is 20 to 15kHz (not 16kHz) and the bandwidth is not
> linear due to the pre-emphasis / de-emphasis functions.
> The S/N is not 70dB in stereo (and mono also), the best value is
> around 40dB, not more.
Oh don't be so silly! I can connect my FM tuner to my PC and record it
into Cool Edit Pro. SNR is clearly ~60dB, and that's without a proper
aerial!
We don't all live in Paris with its abomination of an FM system!
Cheers,
David.
DAB sounds worse than FM
No, this doesn't throw any spanner in any works, because, like I've noticed
you do many times before, you seem to be discussing something different to
the topic being discussed.
*If* you had bothered to read the last 2 or 3 posts of this sub-thread, you
will have found that it has turned into a theoretical claim made by Nicolas
Croiset that the maximum SNR possible on FM is 40 dB, and whether that is
true or not.
So unless you can provide some information about this, just do not bother to
respond. 'kay?
DAB sounds worse than FM
> "DAB sounds worse than FM" <dab.is@dead> wrote :
>
>> Nicolas Croiset wrote:
>>
>>> And you forget the mandatory audio processing in FM if you want to
>>> not exceed the maximum deviation of +-75kHz which imply a low audio
>>> dynamic.
>>
>>
>> It is NOT mandatory to do anything - broadcasters CHOOSE to process
>> their audio on FM, and there's very, very, very few stations that
>> don't process their audio on DAB.
Whoooa whoooa whoooa there boy! You have completely snipped the important
issue, which is whether the *maximum* SNR on FM is 40 dB or not. Only
keeping in deviation is extremely dishonest, because that's almost an
irrelevance - then again, we are dealing with you, Nicolas Croiset, the
well-known liar, so I shouldn't be surprised at this behaviour.
> It is mandatary not to exceed +-75kHz (ETSI). In France for example it
> is written in the license and if you don't respect this value you can
> pay a fee.
Erm, I'll reply to this point after you've returnd to the issue of maximum
FM SNR.
> Radio stations for masking the FM background noise use audio
> processors and put the system to its maximum of capabilities and when
> you measure the audio dynamic level it is less than 4dB !! So yes DAB
> at low bitrate without audio processing is better than FM.
I'm sorry to have to tell you this, but the audio processing on DAB stations
in the UK seems to be just as high as the same stations on FM. Furthermore,
I know of radio stations that use higher levels of audio processing on DAB
to the level they use on FM - one of my local stations, in fact.
DAB sounds worse than FM
> Doug McDonald <mcdonald@SnPoAM_scs.uiuc.edu> wrote :
>>> The S/N is not 70dB in stereo (and mono also), the best value is
>>> around 40dB, not more.
>>
>>
>> Utter and complete poppycock, assuming standard US
>> preemphasis. I have actually MEASURED 80 dB in stereo,
>> RMS noise of a blank channel (input to multiplex encoder
>> not connected to a source) versus RMS of peak signal
>> of same station playing test signals. This is a real over the air
>> test, using a high quality late 1970's tuner. Note, however, that it
>> is in the US ... Europe would of course be lower S/N due to less
>> preemphasis.
>>
>
> And the receiver was directly connected on the transmitter right ?
> You are giving values of receivers at a very high price which is not
> the common receivers buy by a lot of people.
What you originally said was this:
"The S/N is not 70dB in stereo (and mono also), the best value is around
40dB, not more."
and
"Yes the SNR is around 40dB when you receive FM very well without any
multipath."
So, I'm afraid you DO NOT get to add conditions to your claim afterwards,
because your claim applied to EVERYBODY.
I'm still waiting for this explanation about why the SNR is maximal at 40
dB - hurry up, please.
>>> The stereo separation is less than 40dB when you add the loss due to
>>> the encoder and the loss due to the receiver.
>>
>>
>> There is no loss due to encoder, at least, not significant, in the
>> range 70-80 dB, if you use good oversampling and high quality D/A
>> convertors at the output. Receiver is similar, if using high quality
>> oversampled (say 120 kHz bandwidth) A/D convertors.
>
> There is loss from the encoder AND the receiver. The function L+R and
> L-R is not always perfect on both, even if it is better on the encoder
> side.
So you have the audacity to criticise mid/side coding, which is
theoretically lossless, when you are supporting the MP2 audio codec, which
is the only audio codec I'm aware of that is limited to only using intensity
stereo for joint stereo coding, which is a technique that destroys the L and
R signal envelopes, which mid/side coding would keep?
Nicolas, YET AGAIN you have come out with information that you have just
made up (or you simply don't understand the technology), and for once your
lies have been found out by a number of people rather than the usual case
where it's just me or maybe me and 1 or 2 other people that realise that
your whole argument is based on information you've made up.
The reality is that you COULD NOT have known that FM is limited to an SNR of
40 dB, because it isn't limited to an SNR of 40 dB - you just made it up
because that is what you do in these arguments. Also, you clearly don't
understand the issue of stereo separation or you would not be criticising
mid/side coding when your preferred MP2 coding uses intensity stereo coding.
Basically, you are a fraud - and that's as polite as I'm going to get with
you...
DAB sounds worse than FM
> DAB sounds worse than FM wrote:
>> Silk wrote:
>>
>>> DAB sounds worse than FM wrote:
>>>
>>>
>>>> Aaaaaaagh, the sweet smell of victory wafts around
>>>> alt.radio.digital once again. It is a beautiful smell - like
>>>> cycling through the smells of rose pettles, pear drop sweets, or
>>>> freshly-baked biscuits (you just can't beat the smell of the
>>>> McVities factory on the A6).
>>>
>>> You should try soap, that smells even better.
>>
>>
>>
>> What??
>>
>>
> Mr hippocrit, tells you off for insulting other people, and then comes
> out with comments like that.
Absolutely right.
> (PS. Sorry I can't get the spell check to recognise hippocrit)
Put an online dictionary in your Favorites, such as something like this:
tony sayer
Matthelas <matt...@gmail.com> writes
Well there are advantages with digital transmission but they've been
thrown away by the system and its implementation;(...
--
Tony Sayer
tony sayer
<ncro...@vdldiffusion.com> writes
And what's the French for "more bollocks" again?.
In Angleterre we have a saying that when you are in a hole the best
thing to do is to stop digging!...
--
Tony Sayer
Doug McDonald
>
>>
>> Utter and complete poppycock, assuming standard US
>> preemphasis. I have actually MEASURED 80 dB in stereo,
>> RMS noise of a blank channel (input to multiplex encoder
>> not connected to a source) versus RMS of peak signal
>> of same station playing test signals. This is a real over the air test,
>> using a high quality late 1970's tuner. Note, however, that it
>> is in the US ... Europe would of course be lower S/N due to less
>> preemphasis.
>>
>
> And the receiver was directly connected on the transmitter right ?
Well, no, not exactly. This was a 300 watt station about a half-mile
away ... received through a metal-coated anti-RF glass window
at a bad angle. Still, 300 watts at 1/2 mile line of sight is plenty
of power.
> You are giving values of receivers at a very high price which is not the
> common receivers buy by a lot of people.
We are talking what is capable, and what people who listen
on hi-fis actually used in the 70's and 80's when the stations
tried to transmit good signals. Yes, this was an expensive stereo
in those days. My late-90s NAD is nearly as good. It has
much better stereo separation, for example (the 80 dB one
does indeed have only 40 dB separation, due to phase
problems ... but of course, the error is NOT DISTORTION in
the non-linear sense, nor is it pan-potted mono.)
>
> There is loss from the encoder AND the receiver. The function L+R and
> L-R is not always perfect on both, even if it is better on the encoder
> side.
>
Yes, of course. I merely point out that 70 dB is quite easy these days,
with digital processing. It is, as I will readily agree, unnecessary
for real stereo, where even 30 dB is fine if it is merely a case
of intermixing left and right due to phase errors.
Doug McDonald
Nicolas Croiset
<dab.is@dead> wrote:
>Nicolas Croiset wrote:
>> "DAB sounds worse than FM" <dab.is@dead> wrote :
>>
>>> Nicolas Croiset wrote:
>>>
>>>> And you forget the mandatory audio processing in FM if you want to
>>>> not exceed the maximum deviation of +-75kHz which imply a low audio
>>>> dynamic.
>>>
>>>
>>> It is NOT mandatory to do anything - broadcasters CHOOSE to process
>>> their audio on FM, and there's very, very, very few stations that
>>> don't process their audio on DAB.
>
>
>Whoooa whoooa whoooa there boy! You have completely snipped the important
>issue, which is whether the *maximum* SNR on FM is 40 dB or not. Only
>keeping in deviation is extremely dishonest, because that's almost an
>irrelevance - then again, we are dealing with you, Nicolas Croiset, the
>well-known liar, so I shouldn't be surprised at this behaviour.
>
I didn't say it is a maximum, I said it is around.
>
>> Radio stations for masking the FM background noise use audio
>> processors and put the system to its maximum of capabilities and when
>> you measure the audio dynamic level it is less than 4dB !! So yes DAB
>> at low bitrate without audio processing is better than FM.
>
>
>I'm sorry to have to tell you this, but the audio processing on DAB stations
>in the UK seems to be just as high as the same stations on FM. Furthermore,
>I know of radio stations that use higher levels of audio processing on DAB
>to the level they use on FM - one of my local stations, in fact.
The Audio processing on DAB is not a DAB problem but a radio station
problem you make confusion between problems.
Doug McDonald
Processing has nothing to do with FM. IT IS NOT NECESSARY! There is no
requirement anywhere for it, except setting the level right.
If you transmit recorded programming, the only processing necessary
is to look through the program before sending it out, find the
max level, and set it at the max allowable signal modulation.
I.e. set the level. No Optimud necessary.
Doug McDonald
Nicolas Croiset
<davidr...@postmaster.co.uk> wrote:
Effectively Cool edit pro is a well known measurement system which is
better than all Agilent and R&S equipments ;-)
DAB sounds worse than FM
> On Thu, 18 Jan 2007 10:39:40 GMT, "DAB sounds worse than FM"
> <dab.is@dead> wrote:
>
>> Nicolas Croiset wrote:
>>> "DAB sounds worse than FM" <dab.is@dead> wrote :
>>>
>>>> Nicolas Croiset wrote:
>>>>
>>>>> And you forget the mandatory audio processing in FM if you want to
>>>>> not exceed the maximum deviation of +-75kHz which imply a low
>>>>> audio dynamic.
>>>>
>>>>
>>>> It is NOT mandatory to do anything - broadcasters CHOOSE to process
>>>> their audio on FM, and there's very, very, very few stations that
>>>> don't process their audio on DAB.
>>
>>
>> Whoooa whoooa whoooa there boy! You have completely snipped the
>> important issue, which is whether the *maximum* SNR on FM is 40 dB
>> or not. Only keeping in deviation is extremely dishonest, because
>> that's almost an irrelevance - then again, we are dealing with you,
>> Nicolas Croiset, the well-known liar, so I shouldn't be surprised at
>> this behaviour.
>>
>
> I didn't say it is a maximum, I said it is around.
That is a lie. This is what you said:
"The bandwidth is 20 to 15kHz (not 16kHz) and the bandwidth is not
linear due to the pre-emphasis / de-emphasis functions.
The S/N is not 70dB in stereo (and mono also), the best value is
around 40dB, not more."
Note the use of "not more", i.e. maximum.
>>> Radio stations for masking the FM background noise use audio
>>> processors and put the system to its maximum of capabilities and
>>> when you measure the audio dynamic level it is less than 4dB !! So
>>> yes DAB at low bitrate without audio processing is better than FM.
>>
>>
>> I'm sorry to have to tell you this, but the audio processing on DAB
>> stations in the UK seems to be just as high as the same stations on
>> FM. Furthermore, I know of radio stations that use higher levels of
>> audio processing on DAB to the level they use on FM - one of my
>> local stations, in fact.
>
> The Audio processing on DAB is not a DAB problem but a radio station
> problem you make confusion between problems.
I'm not confusing anything, because I'm not the person trying to claim that
audio processing is an FM problem, and thus implying that it isn't a problem
on DAB - you are suggesting this.
Nicolas Croiset
You don't need any explanation. You create a full FM chain with a
measurement system on the receiver side and you will get the result.
Theory is theory, practice is an other world.
>
>>>> The stereo separation is less than 40dB when you add the loss due to
>>>> the encoder and the loss due to the receiver.
>>>
>>>
>>> There is no loss due to encoder, at least, not significant, in the
>>> range 70-80 dB, if you use good oversampling and high quality D/A
>>> convertors at the output. Receiver is similar, if using high quality
>>> oversampled (say 120 kHz bandwidth) A/D convertors.
>>
>> There is loss from the encoder AND the receiver. The function L+R and
>> L-R is not always perfect on both, even if it is better on the encoder
>> side.
>
>
>So you have the audacity to criticise mid/side coding, which is
>theoretically lossless, when you are supporting the MP2 audio codec, which
>is the only audio codec I'm aware of that is limited to only using intensity
>stereo for joint stereo coding, which is a technique that destroys the L and
>R signal envelopes, which mid/side coding would keep?
I am sorry but the FM stereo system destroy also the stereo of the
audio. The stereo separation is largely better in DAB than FM. If you
do a test with a stereo FM encoder and broadcast on both systems (FM
and DAB) a 1kHz sine wave tone on left input only, and you measure the
output value on the right, you will see the result of the real
separation. Even if sometimes in some frames (24ms/frame) the stereo
creates some artefact, the global result is largely better than FM.
Then you can do a similar test with 2 different sound on left and
right input and you will see also a result which is not as good as you
like in FM.
davidr...@postmaster.co.uk
Nicolas Croiset wrote:
> On 18 Jan 2007 02:06:39 -0800, "davidr...@postmaster.co.uk"
> <davidr...@postmaster.co.uk> wrote:
>
> >Nicolas Croiset wrote:
> >> On Wed, 17 Jan 2007 11:21:56 -0600, Doug McDonald
> >> <mcdonald@SnPoAM_scs.uiuc.edu> wrote:
> >
> >> >FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
> >> >in excess of 70Db with less than 0.3% distortion, stereo. That
> >> >0.3% is very low order harmonic distortion.
> >>
> >> No FM is not able to do that.
> >>
> >> The bandwidth is 20 to 15kHz (not 16kHz) and the bandwidth is not
> >> linear due to the pre-emphasis / de-emphasis functions.
> >> The S/N is not 70dB in stereo (and mono also), the best value is
> >> around 40dB, not more.
> >
> >Oh don't be so silly! I can connect my FM tuner to my PC and record it
> >into Cool Edit Pro. SNR is clearly ~60dB, and that's without a proper
> >aerial!
>
> better than all Agilent and R&S equipments ;-)
You're trolling Nicolas. SNR is easy to measure properly.
Perhaps you fed the 19kHz pilot tone and even the 38kHz RDS into your
analyser, and forget to limit the bandwidth?! Or maybe you forgot to
tune your radio in!
For the record, I used a 15kHz LPF to remove the low level junk around
19kHz which my tuner allows through.
Of course, if you were any kind of audio engineer at all, you'd know
what 40dB SNR _sounds_ like, and know that FM is several times better
than this! This knowledge would save you from posting such obviously
incorrect results for everyone to see and laugh at.
Cheers,
David.
Malcolm Knight
news:1169128273.1...@a75g2000cwd.googlegroups.com...
About Nicolas Croiset...
> Of course, if you were any kind of audio engineer at all, you'd know
> what 40dB SNR _sounds_ like, and know that FM is several times better
> than this! This knowledge would save you from posting such obviously
> incorrect results for everyone to see and laugh at.
40db is four or five db worse than what you'd commonly get off the
linear track of a VHS tape and we all know how hissy that can be. An
audio professional suggesting that FM is usually worse than VHS linear
can, as you suggest, only be the subject of mirth.
--
Malcolm
DAB sounds worse than FM
I once rang VDL to have a word with M. Croiset, but he wasn't there. I did
ask what job it was he does, and the girl said that he's a manager in their
software department, so he's basically a software engineer who tries to make
out he understands engineering issues merely by the fact that the company he
works for is associated with engineering. I've asked him on many, many
occasions whether he has a degree or not, and he has never to this day
answered that question, which leaves me with only one conclusion - I'm not
saying it's bad not to have an engineering degree per se, but if he's going
to say definitive things such as A is better than B then he needs to be able
to show that he's more than just some random C programmer, because having
the job he's got won't actually require him to understand any DSP or digital
comms theory.
Another thing to remember is that his company is a *transmission* company,
so the software his company writes just has to implement the standard, which
is simple and requires virtually no understanding of theory whatsoever - the
real engineering happens on the receiver side, where you do have to have a
very good understanding of theory to produce a good receiver.
Basically, putting my view of him as being a liar to one side, he hasn't got
the first clue what he's on about in the vast majority of cases, which is
why he gets everything so wrong. This thread has been good, because it's
exposed his usual practice of making things up as he goes along for all to
see, and thankfully many have commented on it - the vast majority of the
time he makes things up it's about digital technologies, which far fewer
people understand, or at least understan well, so he gets away with the
exact same thing as he's done on this thread other than me telling him he's
wrong, and then usually me getting frustrated with him and calling him a
liar (which he is, or he's incompetent), and *I* then get chastised for
being rude!
Nicolas Croiset
<dab.is@dead> wrote:
Sorry I would say not largely more like 70dB.
So it is effectively around 40dB.
Nicolas Croiset
<mcdonald@SnPoAM_scs.uiuc.edu> wrote:
>Nicolas Croiset wrote:
>> Radio stations for masking the FM background noise use audio processors
>> and put the system to its maximum of capabilities and when you measure
>> the audio dynamic level it is less than 4dB !! So yes DAB at low bitrate
>> without audio processing is better than FM.
>>
>
>Processing has nothing to do with FM. IT IS NOT NECESSARY! There is no
>requirement anywhere for it, except setting the level right.
>If you transmit recorded programming, the only processing necessary
>is to look through the program before sending it out, find the
>max level, and set it at the max allowable signal modulation.
>I.e. set the level. No Optimud necessary.
Totally wrong !!
If you don't have a sound processing dedicated to FM (like a standard
audio limiter) you will go far more away than 75kHz for the deviation.
You MUST NOT GO more than +-75kHz, in the other case you will makes
interferences on the programs which are broadcast near yours.
Due to the pre-emphasis the audio processor MUST use this curve before
its processing.
Imagine first a 100 Hz sound at 0dB on the input of the stereo
encoder, which will generate a deviation of 75kHz on the transmitter
before and after pre-emphasis. The pre-emphasis start around 1kHz, so
that's why there is no problem for the deviation.
Then you apply a 10kHz sound at 0dB, before the pre-emphasis,
everything is fine. But after the pre-emphasis of 50盜 you will not
obtain a deviation of 75kHz but largely more perhaps 160kHz or more (I
don't have time to calculate the result).
A little draw for a FM transmission limitation :
Audio input >----+---------------- VCA ---------> Output
| ^
| |
--- Pre-emphasis --+
and conversion
for the limitation.
So you need to use an audio processor with multi band functionnalities
and a specific limitation which pre-emphase the audio signal before
Then you convert this value which and generate a voltage or current to
pilot by VCA the audio output.
Conclusion : audio processor specific to FM is MANDATORY even if you
check before not to go higher than 0dB on your audio source before
pre-emphasis.
On DAB you need also a limiter but it is only to protect the audio
encoder (or all AES/EBU interfaces) to have no sound higher or equal
to 0dBFs. Normally the standard limitation is -18dBfs, on DAB we use
generally -6dBfs.
Nicolas Croiset
<dab.is@dead> wrote:
>Malcolm Knight wrote:
>> <davidr...@postmaster.co.uk> wrote in message
>> news:1169128273.1...@a75g2000cwd.googlegroups.com...
>>
>> About Nicolas Croiset...
>>
>>> Of course, if you were any kind of audio engineer at all, you'd know
>>> what 40dB SNR _sounds_ like, and know that FM is several times better
>>> than this! This knowledge would save you from posting such obviously
>>> incorrect results for everyone to see and laugh at.
>>
>> 40db is four or five db worse than what you'd commonly get off the
>> linear track of a VHS tape and we all know how hissy that can be. An
>> audio professional suggesting that FM is usually worse than VHS linear
>> can, as you suggest, only be the subject of mirth.
>
>
>I once rang VDL to have a word with M. Croiset, but he wasn't there. I did
>ask what job it was he does, and the girl said that he's a manager in their
>software department, so he's basically a software engineer who tries to make
>out he understands engineering issues merely by the fact that the company he
Sorry not software department, it is digital department, in this
department there is a sub-department for software. I am not in the
sub-department software.
DAB sounds worse than FM
Could somebody please reply to this post in case I'm dreaming that Nicolas
Croiset has just posted that 70 dB is not largely more than 40 dB, please?
In Croiset's tiny mind presumably a factor of 1000 difference is "not
largely more".
Ridiculous - how on earth did he get a job?
Nicolas Croiset
Don't make the confusion between audio and video for VHS.
DAB sounds worse than FM
I have to say that I find it amusing that you've just written a post in
which you attempt to make out that you have a clue what you're talking about
by quoting numbers (of all things) and even providing a pretty little
diagram (although the main point of your post was that FM should use a
limiter, which Doug did mention in his post, and you're wrong that FM needs
to actually have audio processing applied - so, all-in-all, not a very good
post, again, eh), and yet yesterday you didn't even seem to understand that
SNR simply equals signal power divided by noise power, and in your last post
you tried to suggest that 70 dB was "not largely more" than 40 dB, despite
there being a factor of 1000 difference between them.
I suppose I do overuse these words, so they lose some of their meaning, but
I'm afraid you're a buffoon, old boy. So Nicolas, if you wouldn't mind,
could you stop posting on this newsgroup, please, because you distract
people away from logical discussion?
Thanks, and good luck for the future.
DAB sounds worse than FM
But you've just admitted that your original figure was incorrect by 30 dB -
i.e. a factor of 1000 incorrect.
Remind me, why should anybody take any notice of what someone like you says?
>>>>> The stereo separation is less than 40dB when you add the loss due
>>>>> to the encoder and the loss due to the receiver.
>>>>
>>>>
>>>> There is no loss due to encoder, at least, not significant, in the
>>>> range 70-80 dB, if you use good oversampling and high quality D/A
>>>> convertors at the output. Receiver is similar, if using high
>>>> quality oversampled (say 120 kHz bandwidth) A/D convertors.
>>>
>>> There is loss from the encoder AND the receiver. The function L+R
>>> and L-R is not always perfect on both, even if it is better on the
>>> encoder side.
>>
>>
>> So you have the audacity to criticise mid/side coding, which is
>> theoretically lossless, when you are supporting the MP2 audio codec,
>> which is the only audio codec I'm aware of that is limited to only
>> using intensity stereo for joint stereo coding, which is a technique
>> that destroys the L and R signal envelopes, which mid/side coding
>> would keep?
>
> I am sorry but the FM stereo system destroy also the stereo of the
> audio. The stereo separation is largely better in DAB than FM.
So intensity stereo joint stereo coding, which is known to destroy the left
and right channel signal envelopes is better than FM, which uses mid/side
joint stereo coding, which is theoretically lossless (albeit that there's
some loss in practice)?
Yeah, good one.
> If you
> do a test with a stereo FM encoder and broadcast on both systems (FM
> and DAB) a 1kHz sine wave tone on left input only, and you measure the
> output value on the right, you will see the result of the real
> separation.
I don't think you could pick a worse experiment if you tried, because if
there's only a 1 kHz tone present then there's only one subband in use, so
even at very low bit rates the encoder will choose to use discrete stereo,
which means you're cheating on the test, which invalidates your test
results, and really just shows that you haven't got the first clue what
you're talking about - no surprise there then.
> Even if sometimes in some frames (24ms/frame) the stereo
> creates some artefact, the global result is largely better than FM.
Buffoon.
DAB sounds worse than FM
So what do you do? You OBVIOUSLY have no idea about DSP nor about digital
communications nor digital audio coding, so what the bloody hell do you do?
Nicolas Croiset
This is wrong due to the pre-emphasis.
If you watch for each equipment on the chain (receiver only for
example) you can obtain on it a SNR of 70dB (some receiver
manufacturers wrote that), the main problem is when you calculate the
SNR of the full chain including FM, the result is around 40dB.
tony sayer
<ncro...@vdldiffusion.com> writes
>obtain a deviation of 75kHz but largely more perhaps 160kHz or more (I
>don't have time to calculate the result).
>
>A little draw for a FM transmission limitation :
>
>Audio input >----+---------------- VCA ---------> Output
> | ^
> | |
> --- Pre-emphasis --+
> and conversion
> for the limitation.
>
>
>So you need to use an audio processor with multi band functionnalities
>and a specific limitation which pre-emphase the audio signal before
>Then you convert this value which and generate a voltage or current to
>pilot by VCA the audio output.
>
>Conclusion : audio processor specific to FM is MANDATORY even if you
>check before not to go higher than 0dB on your audio source before
>pre-emphasis.
>
>
>On DAB you need also a limiter but it is only to protect the audio
>encoder (or all AES/EBU interfaces) to have no sound higher or equal
>to 0dBFs. Normally the standard limitation is -18dBfs, on DAB we use
>generally -6dBfs.
>
>
Christ he's really lost the plot now!.
Or the translation?....
--
Tony Sayer
tony sayer
<ncro...@vdldiffusion.com> writes
How do you say I think you ought to see le medser of the mind in
French?...
--
Tony Sayer
tony sayer
than FM <dab.is@dead.?> writes
>tony sayer wrote:
>> In article <pfosq2pj3be109f0e...@4ax.com>, Nicolas
>> Croiset <ncro...@vdldiffusion.com> writes
>>> On Wed, 17 Jan 2007 11:21:56 -0600, Doug McDonald>>>>> Yes I maintain what I said many times, if you use MP2 encoder
>>>>> carrefully you can do something which is equal or superior to FM
>>>>> even at low bitrate.
>>>>>
>>>>>
>>>>
>>>> REALLY?
>>>>
>>>> FM is capable of a bandwidth of 0-16 kHz with a S/N ratio
>>>> in excess of 70Db with less than 0.3% distortion, stereo. That
>>>> 0.3% is very low order harmonic distortion.
>>>>
>>>
>>> No FM is not able to do that.
>>>
>>> linear due to the pre-emphasis / de-emphasis functions.
>>> The S/N is not 70dB in stereo (and mono also), the best value is
>>> anything because you will be horrified.
>>> The stereo separation is less than 40dB when you add the loss due to
>>> the encoder and the loss due to the receiver.
>>>
>>>
>>
>
>
>Tony, you've got a lot of experience with using FM kit, so what's the
>highest SNR you've ever seen on FM? Croiset claims that 40 dB is the
>*maximum* SNR possible, so I think it's time to demolish this claim.
>
>
It can do more than 70 under ideal conditions, but I haven't bothered
measuring for some time now but I will if I get a moment over le weekend
for the benefit of our European neighbours;).
Not bothered as its normally good enough for purpose...
--
Tony Sayer
Malcolm Knight
news:8ccvq2dtta5hj4bnf...@4ax.com...
Where do I mention video on VHS? If you were truly an expert on audio
you would know what I meant by the term 'linear track'.
Clue. It's a bog standard audio tape system. Usually mono but not
always. Usually without Dolby processing, but not always. And in Europe
it runs at the amazing speed of 23.39mm/second. Half that of a Compact
Cassette.
Quoting directly from the spec sheet of my JVC VHS machine. Audio SNR
"Better than 43dB" Video SNR "Better than 46dB". Those with any
experience of listening will know that VHS Linear Audio hisses
constantly and never gets better than that. FM Radio generally doesn't
hiss and if it does can be made better.
Don't tell me I don't know the difference between audio and video.
That's just the desperate bleating of someone who is in Tony's big hole
but can't stop digging.
--
Malcolm
DAB sounds worse than FM
Nicolas, don't make the confusion between 40 dB and 70 dB - there's a factor
of 1000 difference between them.
For example: If I bought a car for £10,000, a factor of 1000 increase would
mean that that car would cost £10,000,000, or in other words, you haven't
got the first clue what you're going on about.
tony sayer
>You are giving values of receivers at a very high price which is not the
>common receivers buy by a lot of people.
Ever tried a Denon TU260 MK2 surprisingly good for the money?..
--
Tony Sayer
tony sayer
<ncro...@vdldiffusion.com> writes
Bollocks..
>
>
>>
>>>>> The stereo separation is less than 40dB when you add the loss due to
>>>>> the encoder and the loss due to the receiver.
>>>>
>>>>
>>>> There is no loss due to encoder, at least, not significant, in the
>>>> range 70-80 dB, if you use good oversampling and high quality D/A
>>>> convertors at the output. Receiver is similar, if using high quality
>>>> oversampled (say 120 kHz bandwidth) A/D convertors.
>>>
>>> There is loss from the encoder AND the receiver. The function L+R and
>>> L-R is not always perfect on both, even if it is better on the encoder
>>> side.
>>
>>
>>So you have the audacity to criticise mid/side coding, which is
>>theoretically lossless, when you are supporting the MP2 audio codec, which
>>is the only audio codec I'm aware of that is limited to only using intensity
>>stereo for joint stereo coding, which is a technique that destroys the L and
>>R signal envelopes, which mid/side coding would keep?
>
>I am sorry but the FM stereo system destroy also the stereo of the
>audio. The stereo separation is largely better in DAB than FM.
More of le Bollocks...
>If you
>do a test with a stereo FM encoder and broadcast on both systems (FM
>and DAB) a 1kHz sine wave tone on left input only, and you measure the
>output value on the right, you will see the result of the real
>separation. Even if sometimes in some frames (24ms/frame) the stereo
>creates some artefact, the global result is largely better than FM.
>
>Then you can do a similar test with 2 different sound on left and
>right input and you will see also a result which is not as good as you
>like in FM.
>
Cobblers then?..
--
Tony Sayer
Doug McDonald
>>
>> "Yes the SNR is around 40dB when you receive FM very well without any
>> multipath."
>
> You don't need any explanation. You create a full FM chain with a
> measurement system on the receiver side and you will get the result.
> Theory is theory, practice is an other world.
>
I actually tested what amounts to a full chain. True, the background
was not a programmatic one ... BUT THAT IS IMMATERIAL,
since the S/N of CD is far greater than 80 dB.
>
> I am sorry but the FM stereo system destroy also the stereo of the
> audio. The stereo separation is largely better in DAB than FM.
Well, it CAN be. Or it may not me. FM is plenty good at the
separation business.
> If you
> do a test with a stereo FM encoder and broadcast on both systems (FM
> and DAB) a 1kHz sine wave tone on left input only, and you measure the
> output value on the right, you will see the result of the real
> separation. Even if sometimes in some frames (24ms/frame) the stereo
> creates some artefact, the global result is largely better than FM.
FM can do very well if allowed to. With modern equipment design,
there will be no phase errors of significance in either the
exciter or receiver. Now that ass not true, of course,
of analog stereo generation/reception.
Doug
Doug McDonald
>
> For the record, I used a 15kHz LPF to remove the low level junk around
> 19kHz which my tuner allows through.
>
> Of course, if you were any kind of audio engineer at all, you'd know
> what 40dB SNR _sounds_ like, and know that FM is several times better
> than this! This knowledge would save you from posting such obviously
> incorrect results for everyone to see and laugh at.
>
>
When I did my tests, I used a very high grade 100 kHz bandwidth
16 bit digitizer with a 30 kHz 8-pole Bessel anti-alias filter, and
the results were computer processed to remove amplitude and phase errors
in that filter.
Doug McDonald
DAB sounds worse than FM
Admittedly, as you've just confused 40 dB with 70 dB, then you're unlikely
to be able to add up simple numbers, so I think it's unlikely that you'd
understand what's going through my head about what you could do in order to
set the limiter's level even ... shock horror ... with pre-emphasis and
still avoid using audio processing, so I won't bother, because I'm bored of
morons - there's a few morons in Big Brother at the moment, so I want to see
as little of you as possible.
> If you watch for each equipment on the chain (receiver only for
> example) you can obtain on it a SNR of 70dB (some receiver
> manufacturers wrote that), the main problem is when you calculate the
> SNR of the full chain including FM, the result is around 40dB.
No, Doug's already said that he's got 70 or 80 dB, so your 40 dB is a lie -
try and remember that lying is a bad thing, and that you should not do it.
Doug McDonald
>> I.e. set the level. No Optimud necessary.
>
> Totally wrong !!
>
> If you don't have a sound processing dedicated to FM (like a standard
> audio limiter) you will go far more away than 75kHz for the deviation.
> You MUST NOT GO more than +-75kHz, in the other case you will makes
> interferences on the programs which are broadcast near yours.
>
> Due to the pre-emphasis the audio processor MUST use this curve before
> its processing.
Yes, when you set the level, you have to do that looking at the
preemphasized signal. That's a given, you are correct there.
>
> Conclusion : audio processor specific to FM is MANDATORY even if you
> check before not to go higher than 0dB on your audio source before
> pre-emphasis.
>
In a sense, yes ... but ALL that it has to do is set the level.
For pre-recorded program, that can be done ONCE FOR THE WHOLE
PROGRAM.
Doug McDonald
DAB sounds worse than FM
Okay, nice one. What do you think a "typical" SNR would be?
> Not bothered as its normally good enough for purpose...
Absolutely.
DAB sounds worse than FM
In other words, Nicolas should fermez his bouche.
Nicolas Croiset
>> If you watch for each equipment on the chain (receiver only for
>> example) you can obtain on it a SNR of 70dB (some receiver
>> manufacturers wrote that), the main problem is when you calculate the
>> SNR of the full chain including FM, the result is around 40dB.
>
>
>No, Doug's already said that he's got 70 or 80 dB, so your 40 dB is a lie -
>try and remember that lying is a bad thing, and that you should not do it.
He does not understand what are the consequences of pre-empahsis, so for
the rest...
Nicolas Croiset VDL
http://www.vdl.fr/
Nicolas Croiset
>Nicolas Croiset wrote:
>
>>> I.e. set the level. No Optimud necessary.
>>
>> Totally wrong !!
>>
>> If you don't have a sound processing dedicated to FM (like a standard
>> audio limiter) you will go far more away than 75kHz for the deviation.
>> You MUST NOT GO more than +-75kHz, in the other case you will makes
>> interferences on the programs which are broadcast near yours.
>>
>> Due to the pre-emphasis the audio processor MUST use this curve before
>> its processing.
>
>Yes, when you set the level, you have to do that looking at the
>preemphasized signal. That's a given, you are correct there.
>
If you set this on the linear level and take into account the
pre-emphasis and respect the peak at a maximum level for the deviation
at 75kHz. In fact your result will be around 15dB lower than the same
audio which pass through the audio processing. so the SNR will be lower
as expected with the audio processing.
I think you might read this document :
http://www.omniaaudio.com/tech/bs412.htm
Nicolas Croiset
>>
>> There is loss from the encoder AND the receiver. The function L+R and
>> L-R is not always perfect on both, even if it is better on the encoder
>> side.
>>
>
>Yes, of course. I merely point out that 70 dB is quite easy these days,
>with digital processing. It is, as I will readily agree, unnecessary
>for real stereo, where even 30 dB is fine if it is merely a case
>of intermixing left and right due to phase errors.
The best FM stereo encoder (full digital) I know have a separation of
60dB at 1kHz. This value is true and measured.
tony sayer
<ncro...@vdldiffusion.com> writes
Nicholas.. I think anyone on this group who knows anything about FM
systems are perfectly aware of pre-emphasis!.
Do stop digging the hole bigger!....
--
Tony Sayer
tony sayer
Around 65 to 70 odd dependent on various factors. As said earlier I'll
make some measurements this weekend if I get time!....
>
>
>> Not bothered as its normally good enough for purpose...
>
>
>Absolutely.
>
>
--
Tony Sayer
Doug McDonald
I most certainly DO understand preemphasis. In your distorted mind
how do you think I don't understand preemphasis?
My 70 or 80 dB measurement included the preemphasis/deemphasis
in the chain. The S/N level of a max level test sine wave above the frequency
at which preemphasis starts in will indeed be less than 70 or 80 dB,
and if it is 15 kHz, it will be a lot less. But at least in the US
it won't ever quite get as bad as 40 dB.
Doug McDonald
Doug McDonald
>> Yes, when you set the level, you have to do that looking at the
>> preemphasized signal. That's a given, you are correct there.
>>
>
> If you set this on the linear level and take into account the
> pre-emphasis and respect the peak at a maximum level for the deviation
> at 75kHz. In fact your result will be around 15dB lower than the same
> audio which pass through the audio processing. so the SNR will be lower
> as expected with the audio processing.
>
Could you translate that into English?
Doug McDonald
tony sayer
<ncro...@vdldiffusion.com> writes
Is this the Harris unit they claim 65 on sinewave and 60 on more complex
materiel.....
Still not bad compared to that abortion called joint stereo or the new stereo
on UK DAB i.e. Mono!..
http://www.broadcast.harris.com/product_portfolio/prod_media/HAR177338.pdf
--
Tony Sayer
Nicolas Croiset
Effectively I wrote it too fast ;-)
I will explain differently. You have a record with 2 tones at 0dB each
not at the same time. The first one is 500Hz, the second one at 14500Hz.
If you respect the pre-emphasis and the deviation of 75 kHz with this
record. In fact you will play it at around 15dB lower than it is
possible with a FM audio processing.
Doug McDonald
>>
>
> Effectively I wrote it too fast ;-)
>
> I will explain differently. You have a record with 2 tones at 0dB each
> not at the same time. The first one is 500Hz, the second one at 14500Hz.
>
> If you respect the pre-emphasis and the deviation of 75 kHz with this
> record. In fact you will play it at around 15dB lower than it is
> possible with a FM audio processing.
>
Not with linear processing that does not destroy signal
integrity. By destroy integrity I mean, the output of the
receiver is the same as input to the whole transmitter chain.
Of course, if you do alternate 1kHz and 14.5 kHz tones at the same level,
fed into the encoder, and do it linearly, no "processing"
of the usual nonlinear sort, you sure DO have to crank the gain
to where the max S/N of the 1 kHz is well below 70 or 80 dB.
HOWEVER, for most serious music program material, the falloff
above 1 kHz in the actual material us such that the peak level
is determined by frequencies well below 1 kHz. For much orchestral
material, it is below 100 Hz. For piccolo or triangle or cymbal
solo, it's not below 1 kHz, and yes, the overall S/N will be less than
70 or 80 dB.
Doug McDonald
Ken
<dab.is@dead> wrote:
>> Sorry I would say not largely more like 70dB.
>> So it is effectively around 40dB.
>
>
> Could somebody please reply to this post in case I'm dreaming
> that Nicolas Croiset has just posted that 70 dB is not
> largely more than 40 dB, please?
>
> In Croiset's tiny mind presumably a factor of 1000 difference
> is "not largely more".
>
> Ridiculous - how on earth did he get a job?
Good liar?
DAB sounds worse than FM
Yes, very likely.
Richard Evans
> Nicolas Croiset wrote:
>
<SNIP>
>>
>>Sorry I would say not largely more like 70dB.
>>
>>So it is effectively around 40dB.
>
>
>
> Could somebody please reply to this post in case I'm dreaming that Nicolas
> Croiset has just posted that 70 dB is not largely more than 40 dB, please?
Actually I don't think that is what he meant.
I read the same sentence as something like "not a great deal more than
40, and nothing like as much as 70". I think perhaps with English not
being his first Language, he didn't explain it very well.
Whether or not his figure of 40 was accurate however is another matter,
and I'm not really in any position to comment.
Richard E.
DAB sounds worse than FM
> DAB sounds worse than FM wrote:
>> Nicolas Croiset wrote:
>>
>
> <SNIP>
>
>>>
>>> Sorry I would say not largely more like 70dB.
>>>
>>> So it is effectively around 40dB.
>>
>>
>>
>> Could somebody please reply to this post in case I'm dreaming that
>> Nicolas Croiset has just posted that 70 dB is not largely more than
>> 40 dB, please?
>
> Actually I don't think that is what he meant.
>
> I read the same sentence as something like "not a great deal more than
> 40, and nothing like as much as 70". I think perhaps with English not
> being his first Language, he didn't explain it very well.
He said this elsewhere:
"The S/N is not 70dB in stereo (and mono also), the best value is
around 40dB, not more."
where he explicitly states that the SNR on FM cannot exceed 40 dB.
Chris Malcolm
I suspect he's talking about what your average kitchen FM radio might
be capable of. The best quality FM tuners are capable of a lot better,
and sometimes some FM transmissions exploit most of their capacity.
The really interesting question for me is on what technological
platform do the BBC intend to put their highest quality broadcasts,
and is that quality level going to be better than what the current FM
system is capable of, if you have a high quality tuner, good aerial,
and a transmission capable of exploiting those capacities.
I got the impression from BBC adevertising that DAB was actually among
other things going to host their best quality stuff. Since they are
still sometimes putting out better FM Radio 3 transmissions than DAB
at its best is capable of, DAB as it is clearly isn't the vehicle for
their best quality. DAB+ migh be capable of that, but only if the BBC
actually decide to exploit it in that way. Since what wrecked existing
DAB quality wasn't the theoretical limits of the system, but marketing
decisions about how to use itB, it's not clear that DAB+ will be their
quality flagship either. It's also clear that FM is obsolescent.
It's not at all clear to me where the BBC is going with respect to
high audio quality. It's the sort of thing they ought to have a
publicly stated position on.
--
Chris Malcolm c...@infirmatics.ed.ac.uk DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]
davidr...@postmaster.co.uk
> The really interesting question for me is on what technological
> platform do the BBC intend to put their highest quality broadcasts,
> and is that quality level going to be better than what the current FM
> system is capable of, if you have a high quality tuner, good aerial,
> and a transmission capable of exploiting those capacities.
>
> I got the impression from BBC adevertising that DAB was actually among
> other things going to host their best quality stuff. Since they are
> still sometimes putting out better FM Radio 3 transmissions than DAB
> at its best is capable of, DAB as it is clearly isn't the vehicle for
> their best quality. DAB+ migh be capable of that, but only if the BBC
> actually decide to exploit it in that way. Since what wrecked existing
> DAB quality wasn't the theoretical limits of the system, but marketing
> decisions about how to use itB, it's not clear that DAB+ will be their
> quality flagship either. It's also clear that FM is obsolescent.
>
> It's not at all clear to me where the BBC is going with respect to
> high audio quality. It's the sort of thing they ought to have a
> publicly stated position on.
100% agree, but sadly it's exactly the sort of thing which the people
responsible for policy at the BBC know nothing about.
Cheers,
David.
Robert Orban
ncro...@vdldiffusion.com says...
The Belar FMSA-1 can measure 80 dB of left/right crosstalk (separation)
from 10 Hz to 15 kHz. It represents the state of the art in an FM stereo
demodulator.
Specs are here:
We use this monitor as a test instrument when designing FM stereo
encoders.
Bob ORban
Robert Orban
says...
I should point out that the BBC, back in its most "purist" of days, still
used pre-emphasis and deviation limiting on its classical music service.
Indeed, it used a "delay line limiter" of its own design.
In theory, one could ride gain manually while watching the peak carrier
deviation. But this is not practical (the unexpected cymbal crash could
easily overmodulate the transmitter by 300% due to pre-emphasis), even when
the broadcaster is seeking extremely high quality.
Bob Orban
Alan White
<donot...@spamblock.com> wrote:
>In theory, one could ride gain manually while watching the peak carrier
>deviation. But this is not practical (the unexpected cymbal crash could
>easily overmodulate the transmitter by 300% due to pre-emphasis), even when
>the broadcaster is seeking extremely high quality.
If the SM was following a score, reading ahead and 'riding the gain'
which they used to, it's unlikely that a cymbal crash would be
unexpected.
--
Alan White
Twenty-eight miles NW of Glasgow, overlooking Lochs Long and Goil in Argyll, Scotland.
Webcam and weather:- http://windycroft.gt-britain.co.uk/weather