LUFS

[[email protected]]

Everybody's "signed-up" for this, seemingly, but is anybody *actually* broadcasting with this system?

Even BBC R3 (the classical music station) doesn't appear to be using it.

I've started experimenting but I'm still sending off tracks made in old style max RMS (or whatever it is/was).

--
gilorms.com

[Scott Dorsey]

> wrote:
>Everybody's "signed-up" for this, seemingly, but is anybody *actually* broadcasting with this system?
>
>Even BBC R3 (the classical music station) doesn't appear to be using it.
>
>I've started experimenting but I'm still sending off tracks made in old style max RMS (or whatever it is/was).

Well, the idea is that the actual levels you're getting on the air, shouldn't
be any different. You're still setting modulation for peak level and then
managing the loudness once you have the peaks set, like you always did with
VU meters. You just have a different way of measuring loudness.

The question that needs to be asked is whether anyone is following a
particular LUFS level for broadcast... and that I can't answer.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

Scott Dorsey:

If you peak normalize two sources, one
with a high crest factor and one with a
low crest factor, the one with low crest
factor will most likely sound louder to
your ears.

Normalizing to, say -16 or -20LUFs,
means that the source with the lower
CF might not peak anywhere near 0
DBFS.

[[email protected]]

On Sunday, February 19, 2017 at 10:50:05 AM UTC-5, wrote:
> Everybody's "signed-up" for this, seemingly, but is anybody *actually* broadcasting with this system?
>
> Even BBC R3 (the classical music station) doesn't appear to be using it.
>
> I've started experimenting but I'm still sending off tracks made in old style max RMS (or whatever it is/was).
>
> --
> gilorms.com

You want to get to RMS level, no "normalizing" (whatever that means) will accomplish that. Trim those peaks, they do nothing for audio quality.

See, many don't listen. People listen to "radio" at work, where they keep the volume low. YOU need to compensate for that, otherwise, they'll think you went off the "air".

Jack

[Scott Dorsey]

In article >,
> wrote:
>
>If you peak normalize two sources, one
>with a high crest factor and one with a
>low crest factor, the one with low crest
>factor will most likely sound louder to
>your ears.

Right, and the VU meter will indicate this. The VU meter was intended to
measure perceived loudness, and although it doesn't do all that well doing
so with non-voice signals, it's still pretty good for making rough judgements
about that.

>Normalizing to, say -16 or -20LUFs,
>means that the source with the lower
>CF might not peak anywhere near 0
>DBFS.

Right, but you've got two meters, one measuring LUFS and one measuring peak
level. You set the peak level to match the broadcast requirements given you
by the ITU. But now, in the New European Era, you'd be setting the average
LUFS level to match the broadcast requirements given you by the EBU. So all
the stations on the dial should have more or less the same perceived loudness
as well as the same peak levels.

Yes, if you did no processing, just moved a gain control at the transmitter,
you wouldn't be meeting your peak envelope if you were getting the correct
loudness. But we live in the age of digital processing, where we can do some
processing to bring the loudness up. BUT, because of the standard, it will
not be to anyone's advantage to overprocess because you can only bring the
loudness up so far before you'll have to bring it back down to meet the EBU
requirements.

The whole idea here, as detailed in the EBU documents, is to prevent the sort
of FM loudness wars that we got in the US.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

"Yes, if you did no processing, just moved a gain control at the transmitter,
you wouldn't be meeting your peak envelope if you were getting the correct
loudness. "

Does that imply that all radio stations
must constantly fill their peak envelope,
regardless of average/RMS level?

[Scott Dorsey]

> wrote:
>"Yes, if you did no processing, just moved a gain control at the transmitter,
>you wouldn't be meeting your peak envelope if you were getting the correct
>loudness. "
>
>Does that imply that all radio stations
>must constantly fill their peak envelope,
>regardless of average/RMS level?

Yes. Let's say you're a classical station that is playing a recording
with very wide dynamics and minimal processing. You set the levels so
the crescendo hits 0DBFS, then your ppp parts may be 30dBFS down from
that.

If you -didn't- hit the peaks at 0dBFS, then your ppp parts would be
even _farther_ down and that would be bad.

In reality, the noise floor on FM being what it is, you need a little
bit of compression to bring those ppp parts up another 10 or 15 dB.
That might be served by manual gainriding or by an AGC compressor.

Now... this is actually kind of oversimplified but it's exactly what
happens with mono FM broadcast. The channel is limited by deviation
which limits your peak baseband audio level, and so you want to have
your loudest parts hitting those peaks so your soft parts don't disappear.

With stereo FM it gets a little bit more complicated because you are
limited by deviation which limits the total _stereo composite_ audio
level, and so you can juggle maximum levels vs. maximum stereo width
back and forth as well. I gather the EBU doesn't care so much about
that because they aren't living in a world where people are trying to
eke out every possible dB like stations in the US are.

But the _peak_ levels are set by the channel characteristics, the _average_
levels are set by program directors.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

>
> But the _peak_ levels are set by the channel characteristics, the _average_
> levels are set by program directors.
> --scott
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

and the channel characteristics are set by the FCC (in the US)

yep, any modern FM radio station is hitting 100% deviation regularly... like every syllable.

They can't legally go over and nobody goes under for any significant length of time.


the variation between loud stations and not so loud stations is simply how regularly and how often they hit that 100%.

It is interesting to hook an oscilloscope directly to the discriminator of a receiver (before the demphasis) and see how it is smashed against the legal limit. Every station hits the limit, the variation in perceived loudness corresponds more to the DENSITY of the modulation rather than the actual upper limit. They all have the same upper limit set by the FCC.

m

[Mike Rivers[_2_]]

On 4/13/2017 5:14 PM, wrote:
> the channel characteristics are set by the FCC (in the US)
>
> yep, any modern FM radio station is hitting 100% deviation regularly... like every syllable.
>
> They can't legally go over and nobody goes under for any significant length of time.

And they start their records in the middle so there's never a moment of
silence.

--

For a good time, call http://mikeriversaudio.wordpress.com

[[email protected]]

Scott Dorsey:

So does any of that explain the following
difference: On a CD of my favorite
classic rock group, the snare and kick
sound like, well, a snare and kick -
albeit slightly compressed - but still
crisp and impactful. When I hear
the same song on the local classic
rock station, the drums sound
really mushy, distorted, and the
snare has a'crunch' to it instead
of a clear crack.

[Scott Dorsey]

> wrote:
>Scott Dorsey:
>
>So does any of that explain the following
>difference: On a CD of my favorite
>classic rock group, the snare and kick
>sound like, well, a snare and kick -
>albeit slightly compressed - but still
>crisp and impactful. When I hear
>the same song on the local classic
>rock station, the drums sound
>really mushy, distorted, and the
>snare has a'crunch' to it instead
>of a clear crack.

No, it really has nothing to do with it.

We're talking here about metering standards, not processing standards.
LUFS is a different standard for _measuring_ average levels.

The fact that you're listening to some station where average levels are
held to only a few dB lower than peak levels has little to do with how
you measure the levels.

And the overcompression problem you describe basically doesn't exist in
countries where the EBU standards are being promoted anyway.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[[email protected]]

Mike Rivers wrote: "And they start their records in the middle so there's
never a moment of silence. "

LMAO! If you're talking about the
5-15 seconds they edit off for the
'radio' cut, I get your point. Not
to mention songs like 'Hang On,
Sloopy', where the entire second
verse is cut.

[geoff]

On 14/04/2017 10:04 AM, wrote:
> Scott Dorsey:
>
> So does any of that explain the following
> difference: On a CD of my favorite
> classic rock group, the snare and kick
> sound like, well, a snare and kick -
> albeit slightly compressed - but still
> crisp and impactful. When I hear
> the same song on the local classic
> rock station, the drums sound
> really mushy, distorted, and the
> snare has a'crunch' to it instead
> of a clear crack.
>


What are you listening on ?

geoff

[[email protected]]

geoff wrote: "- hide quoted text -
On 14/04/2017 10:04 AM, wrote:
> Scott Dorsey:
>
> So does any of that explain the following
> difference: On a CD of my favorite
> classic rock group, the snare and kick
> sound like, well, a snare and kick -
> albeit slightly compressed - but still
> crisp and impactful. When I hear
> the same song on the local classic
> rock station, the drums sound
> really mushy, distorted, and the
> snare has a'crunch' to it instead
> of a clear crack.
>


What are you listening on ?

geoff "

What does it matter?

[None]

"geoff" > wrote in message
...
> On 14/04/2017 10:04 AM, wrote:
> What are you listening on ?
>
> geoff

Stop feeding the tard. And his dead hobbyhorse.

[Michael Beacom[_4_]]

On 2017-02-19 15:50:02 +0000, said:

> Everybody's "signed-up" for this, seemingly, but is anybody *actually*
> broadcasting with this system?
>
> Even BBC R3 (the classical music station) doesn't appear to be using it.
>
> I've started experimenting but I'm still sending off tracks made in old
> style max RMS (or whatever it is/was).

NPR's sattellite system requires -24 db LUFS.
Member stations were complaining about variable apparent levels between feeds.

[geoff]

On 16/04/2017 12:31 AM, wrote:
> geoff wrote: "- hide quoted text -
> On 14/04/2017 10:04 AM, wrote:
>> Scott Dorsey:
>>
>> So does any of that explain the following
>> difference: On a CD of my favorite
>> classic rock group, the snare and kick
>> sound like, well, a snare and kick -
>> albeit slightly compressed - but still
>> crisp and impactful. When I hear
>> the same song on the local classic
>> rock station, the drums sound
>> really mushy, distorted, and the
>> snare has a'crunch' to it instead
>> of a clear crack.
>>
>
>
> What are you listening on ?
>
> geoff "
>
> What does it matter?
>

Would you be happy to judge the qualities of a fine painting through
dirty smudged scratched sunglasses ?

geoff

[Trevor]

On 16/04/2017 5:44 PM, geoff wrote:
> On 16/04/2017 12:31 AM, wrote:
>> geoff wrote: "- hide quoted text -
>> On 14/04/2017 10:04 AM, wrote:
>>> Scott Dorsey:
>>> So does any of that explain the following
>>> difference: On a CD of my favorite
>>> classic rock group, the snare and kick
>>> sound like, well, a snare and kick -
>>> albeit slightly compressed - but still
>>> crisp and impactful. When I hear
>>> the same song on the local classic
>>> rock station, the drums sound
>>> really mushy, distorted, and the
>>> snare has a'crunch' to it instead
>>> of a clear crack.
>>>
>>
>> What are you listening on ?
>>
>> What does it matter?
>>
>
> Would you be happy to judge the qualities of a fine painting through
> dirty smudged scratched sunglasses ?
>

Would it matter to a blind man?

Trevor.

[[email protected]]

geoff wrote: "
Would you be happy to judge the qualities of a fine painting through
dirty smudged scratched sunglasses ?

geoff "


What do you THINK I'm listening on
when I made that comparison? A
boombox?? No - a decent component
stereo system of course. Common
sense.