Re: Broadcast sound question

[Michael Drainer]

What exactly do you mean by maximum defination? Anytime you are dealing
with the sattelite/broadcast world, it is critical that you keep your output
peak levels below 0 dBm or -20 dBFS. If you are running much hotter, you
WILL get calls from those who receive the downlink. If you are looking to
maximize the percieved "Loudness" of the program material, you can compress
the snot out of the signal then boost the gain but depending on the program
material, you will be doing more harm than good.

MD


"Gian Luca Specchia" > wrote in message
...
>
> I'm working for a satellite television that use a mixer Yamaha DM 2000 .
> I would like to work with the maximum level possible before clip.....I
mean
> near 0 dbfs
> Do You think it's wrong or Should I follow another way to achive the
maximum
> definition ?
>
> --
>
>
>
>

[Michael Drainer]

What exactly do you mean by maximum defination? Anytime you are dealing
with the sattelite/broadcast world, it is critical that you keep your output
peak levels below 0 dBm or -20 dBFS. If you are running much hotter, you
WILL get calls from those who receive the downlink. If you are looking to
maximize the percieved "Loudness" of the program material, you can compress
the snot out of the signal then boost the gain but depending on the program
material, you will be doing more harm than good.

MD

"Gian Luca Specchia" > wrote in message
...
>
> I'm working for a satellite television that use a mixer Yamaha DM 2000 .
> I would like to work with the maximum level possible before clip.....I
mean
> near 0 dbfs
> Do You think it's wrong or Should I follow another way to achive the
maximum
> definition ?
>
> --
>
>
>
>

[Brothermark]

What exactly do you mean by maximum defination? Anytime you are dealing
with the sattelite/broadcast world, it is critical that you keep your output
peak levels below 0 dBm or -20 dBFS. If you are running much hotter, you
WILL get calls from those who receive the downlink. If you are looking to
maximize the percieved "Loudness" of the program material, you can compress
the snot out of the signal then boost the gain but depending on the program
material, you will be doing more harm than good.

[Scott Dorsey]

Gian Luca Specchia > wrote:
>
>I'm working for a satellite television that use a mixer Yamaha DM 2000 .
>I would like to work with the maximum level possible before clip.....I mean
>near 0 dbfs

Why? If you are working that close to the line, you have to be very careful
not to go over the limit. Leave yourself some headroom.

>Do You think it's wrong or Should I follow another way to achive the maximum
>definition ?

Most of the video guys keep peaks around -20dBFS. This gives you some room
to play if things go wrong. And with a 16 bit system keeping peaks at -20,
you still have 76 dB of dynamic range to play with, which is a lot more than
any TV program will ever use.
--scott

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

[Marc Wielage]

On Thu, 4 Sep 2003 12:47:06 -0700, Gian Luca Specchia wrote
(in message >):

> I'm working for a satellite television that use a mixer Yamaha DM 2000 .
> I would like to work with the maximum level possible before clip.....I mean
> near 0 dbfs...
>-----------------------------<snip>----------------------------<

Some of the other messages in this thread are incorrect, or at least far
beyond my experience here in LA. The specs typically given for most of the
American cable/sat services to which I've delivered jobs (particularly HBO
and Showtime) usually request an operating level of -20 dB on a digital
scale, with maximum peaks no louder than -10. I assume we're talking about a
standard Lt/Rt stereo mix, as opposed to 5.1 surround.

I have seen shows rejected for air when explosive peaks exceeded -8, but
they'll usually allow you to go a little bit beyond -10, especially if it's
under rare occasions. Home video mixes will usually give you a little more
leeway, but I would make sure the maximum peaks are at least a couple of dB's
under 0. And foreign deliverables seem to come under more scrutiny than
domestic, for some reason.

Dialog intelligibility is actually a much bigger concern. Some networks have
average levels at which they'd like to see their dialog (typically -30 to
-22, or thereabouts), but a lot of that is a question of taste and intent.

For more, check over on the "Post" section of the Digidesign User Conference;
I've seen several good discussions of "acceptable broadcast levels" over
there. Also, you should ask the specific satellite TV network you're dealing
with for a written copy of their specs, just so that you know precisely what
they expect from you as far as channel assignments, set-up tones, identifying
slates, average levels, peak levels, and so on.

--MFW

[Gian Luca Specchia]

And with a 16 bit system keeping peaks at -20,
> you still have 76 dB of dynamic range to play with, which is a lot more
than
> any TV program will ever use.

Consider guys that the system has 24 bits...........and onair (we do use
mpeg 2 compression) the sound is low in level.
This kind of broadcast has a poor level....if I reach the maximum peak
of -20dbfs nobody will hear the program unless the television has a volume
setted at the maximum.......

Thanks To everyone.........

[Scott Dorsey]

Gian Luca Specchia > wrote:
> And with a 16 bit system keeping peaks at -20,
>> you still have 76 dB of dynamic range to play with, which is a lot more
>than
>> any TV program will ever use.
>
>Consider guys that the system has 24 bits...........and onair (we do use
>mpeg 2 compression) the sound is low in level.

That sounds like you need some limiting somewhere, then. Cranking the
level on the distributed program up won't buy you anything since the
AVC box on the transmitter airchain will just turn it back down again.

>This kind of broadcast has a poor level....if I reach the maximum peak
> of -20dbfs nobody will hear the program unless the television has a volume
>setted at the maximum.......

Then use some limiting to bring average levels up while keeping peak levels
low.

Turning up the volume won't do you any good since the AVC will just turn it
back down. Processing is your only real answer to bring average levels up.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[WillStG]

<< "Gian Luca Specchia" >>
<< I'm working for a satellite television that use a mixer Yamaha DM 2000 .
I would like to work with the maximum level possible before clip.....I mean
near 0 dbfs
Do You think it's wrong or Should I follow another way to achive the maximum
definition ?
>>

Absolutely wrong in a studio facility But you should forget about your
"maximum definition" concept here, that's crap in this situation. You mix to
0VU, which is typically -20dbfs in broadcasting. On a peak reading meter you
will go much higher than -20, but you should never even _approach_ 0dbfs in a
broadcast situation coming straight out of your console.

Mix to 0 on your VU meters, and don't argue with your broadcast engineers.


Will Miho
NY Music & TV Audio Guy
Fox And Friends/Fox News
"The large print giveth and the small print taketh away..." Tom Waits

[WillStG]

<< "Gian Luca Specchia" >>
<< Consider guys that the system has 24 bits...........and onair (we do use
mpeg 2 compression) the sound is low in level.
This kind of broadcast has a poor level....if I reach the maximum peak
of -20dbfs nobody will hear the program unless the television has a volume
setted at the maximum..... >>

My friend, you are screwing up the audio because you don't know the
difference between a peak reading meter and a VU meter. This has nothing to do
with bit rate or broadcast compression. You DO have VU meters, don't you?
When you put tone on the line, the VU meter should read "0", and the peak meter
should read "-20". The peak reading meter should go _way above_ -20 while
you mix, that's normal, but the VU meter should hover around 0. With a good
peak limiter properly set inline the peak reading meter will probably not pass
a level of -8 when you hit 0VU. Some places in the US want the levels not to
pass -12 or -10.

In most broadcast facilities the usused headroom before 0dbfs is
compensated for with broadcast limiters, upstream of any MPEG compression.

You have been trying to keep your levels from passing -20 on a peak
meter? That's the problem right there Dude.

Will Miho
NY Music & TV Audio Guy
Fox And Friends/Fox News
"The large print giveth and the small print taketh away..." Tom Waits

[Gian Luca Specchia]

I can use either the peak meter of the consolle and a vu
meter..........consider this:
a tipical vu meter has 0vu at -20 dbfs if I reach the 0vu means that I have
12 db's more and this means -8 or -7 depending on the peak.....I don't want
to be fixed on 0dbfs my question regarded only occasionally peaks......
thanks for your answers you are very precious..............

[Gian Luca Specchia]

My friend, you are screwing up the audio because you don't know the
difference between a peak reading meter and a VU meter.

I know Will.....maybe I'm little confused my head is about to exploding.
Peak meter read the peaks of the program level....Vu read the avarege level
without reading the peaks, strictly speaking about sound material the
difference between the instruments is about -8db ( that's why usually 0vu
correspond to-9 of Ppm or -8 if you are working with nagra).
Some souncraft's consolle mounted a led integrated in the vu shell to achive
the operator to see the peaks with the vu indication.......

thanks man....... don't you think that the word dude is ancient.....??? :-)