FORUMS: list search recent posts

Broadcast Technical Specifications

COW Forums : Broadcasting

<< PREVIOUS   •   VIEW ALL   •   PRINT   •   NEXT >>
Brian Byrd EL
Broadcast Technical Specifications
on Sep 15, 2006 at 9:35:19 pm

had orginally posted this in the Premiere Pro Forum, but someone there replied: "It looks like these are actually PAL specs. You may have a better chance getting the right answer in the After Effects Forum. There are quite a few more European Mods who should be able to help you with that." Then, when I posted it in the After Effects forum, it was suggested that I post it in the Broadcast Video forum, so here I am.

I am preparing a show to be broadcast on satellite TV. Would anyone care to explain these technical specifications from the broadcaster in plain English, and how to achieve these specs in Premiere Pro 2 and/or After Effect? Thank you very much for any assistance.

Some specific questions I have are:

1.) When I create a new PAL project in Premiere Pro, and then create a new Bars and Tone clip, how do I know if it's 75% or 100% or what? What are these percentages refering to?

2.) The spec below seems to be saying that the tone is at 0dB. Is this what they are refering to as the "line-up signals (or level)"? Then they say that the maximum sound levels should not be more than 8dB's above the line up level. I can't imagine that they want the audio levels at +8dB. Maybe they meant that the audio signal of the show should be no more than 8dB BELOW the line-up signal? I'm confused, and then there are terms I don't know like PPM4, PPM6, and -4VU.

"The tape should have as a minimum, one-minute of EBU 625 75% bars &
0dB tone(s) - i.e. the tone(s) should be recorded at PPM4 or minus
4VU. The sound levels on the recording should be in correct
proportion to the line-up signals and maximum levels should not
exceed PPM6 i.e. - they should be no higher than 8dB


Return to posts index

tony salgado
Re: Broadcast Technical Specifications
on Sep 16, 2006 at 2:09:07 am



First and foremost please hire an experienced on line editor who can comply with all the necessary broadcast specs or else you will get your program rejected.


If you are doing the work yourself you need to consider hiring a qualified video engineer who can guide you through the specifics on how to properly meet broadcast specs.

You really need to be shown how to properly determine proper video levels via a waveform monitor which is fundamental to the entire process.

For some background on how to read a waveform monitor I suggest visiting the Textronix's website and search around for their informational papers on the subject of understanding the use of a waveform/vectorscope as well as video levels etc.


Good luck,


Tony Salgado


Return to posts index

Chuck Reti
Re: Broadcast Technical Specifications
on Sep 16, 2006 at 3:33:39 am

quick links:
Tektronix "PAL Video Measurement" (PDF doc)
Tektronix "NTSC Video Measurements" (excellent guide to video measurement, relevant in many ways to PAL)


Return to posts index


rafalaos
Re: Broadcast Technical Specifications
on Sep 17, 2006 at 8:16:33 am

Hi Brian, I answer to your message to the AE forum, but I wrote last night 3 AM. probably not much easy to understand.
- Talking about BARS. what means 75 or 100 %. What means that?
That can means AMPLITUDE OR SATURATION. The PAL bars are 75% AMPLITUDE and 100 % SATURATED (except the White Bar that is 100 % amplitude, "0" saturation. In the Gray scale we can not talk about saturation. We got pure luminance. The Chroma components are 0).
Thats means that for example the RED BAR will be PURE RED (100 %SATURATED) with an amplitude of 75%. Talking in "RGBl 8-Bits" the higher value of Red, Green and Blue you can get is 255. So this would be Red of 191 (de values differs if we talk in YUV).
So if you use a Color Picker you would see that your bars are
- RED BAR : R= 191 G= 0 B= 0
- GREEN BAR : R= 0 G= 191 B= 0
- BLUE BAR : R= 0 G= 0 B= 191
The Yellow, Magenta an Cyan. similar:.
YELLOW BAR: R= 191 G= 191 B= 0. Etc.
The WHITE BAR would be R= 255, G= 255, B= 255.
NTSC BARS are the same, but the WHITE BAR is 75% amplitude (R=G=B= 191)

Your specific questions;
1) The bars generated by PP must be 75%. The 100% Bars, are ILEGAL. They fall out of the broadcast specs. They can only be used for testing but not broadcasted. There is not point to make them available in PP, FC or so. Any way. How to know if the bars are 75 or 100%? You can use the Color Picker method I told you, but the easier way is to use the Vector-scope (FC has, I don't know PP). The Vector-scope of FC is set for 75% Bars. When you set the bars on the time-line, the lines that represents each of the bars, must reach the small squared boxes.
If you use the one of SyntethicAperture, be careful because you can adjust it for 75 and 100% bars (just set it at 75 and check that the lines reach the boxes).
2) Your second question: This people had made a big mess when they ask the specs. they are mixing audio terms (db's and VU) with video (0.3 V + 0,7 V is the amplitude of the full video signal. 1V from peak to peak).
SO, DON'T WORRY, DON'T WORRY.DON'T WORRY..
They just have made a big mess trying to explain what they want. These seems instruction for printing video in an anagogic desk, Beta-cam or so (they talk about Vertical interval time-code, for example)
They are just asking for a standard PAL signal with legal levels. You just have to see that the levels of the Luminance an Chroma are legal (FC warns you when they are not. And if not legal, just to make a simple levels correction).
So just pic your time-line, insert the PP PAL bars and (if you got correct levels) when you print to video anagogic, your video card will take care that all the parameters (Syncs,horizontal length, horizontal and vertical blanking, etc) will be OK.
Talk with some expert about the audio levels, because they are talking about analog levels and they are completely different from the digital.
They talk as well about the 80% Safe Area. When you play any thing in a TV set, a god piece of the image is cropped. So think about that will be cropped a 10% each of the sides. They want your film to fill the screen but don't miss any action (or titles or so0 that could be cropped.
Well, I hope i didn't mixed you up a bit more.
salud,
rafael
PD: I'm finding than at great deal of the problems that people bring to the forums is because the lack of a basic knowledge of the Video signal (PAL and NTSC). People think digital, works digital but they shake when anagogic. But Video is anagogic. We see and hear (and feel) anagogic. We film anagogic (although we record digital) and we project in any screen anagogic.
Digital people are scare of a technology that didn't change in 50 years and that can be understood reading five pages of paper. Some times I think to post a letter in defense of the anagogic video. Is one of the most brilliant and "human" technologies ever developed by the human kind. And even been quite old, still alive. in fact all the new formats, standards etc are based in that all system.




Return to posts index

rafalaos
Re: Broadcast Technical Specifications
on Sep 17, 2006 at 5:03:34 pm

Chuck,
Thanks for the link of tecktronick. The document (at leaast the PAL) is great. Few tims I've tried to find something like that on the web.
salud,
Rafael


Return to posts index

glenn chan
Re: Broadcast Technical Specifications
on Sep 22, 2006 at 4:52:06 pm

1.) When I create a new PAL project in Premiere Pro, and then create a new Bars and Tone clip, how do I know if it's 75% or 100% or what? What are these percentages refering to?
In 75% color bars, the chroma amplitude is 75% of the maximum chroma amplitude allowed. The bars however are always a pure color (i.e. the red bar never has blue or green in it).

It's all a little confusing, since saturation is an ambiguous term. It refers to chroma or to chromaticity, which are NOT the same thing. We tend to think of saturation in terms of chromaticity, which is related to the purity of a color. Chroma on the other hand is "proportional" to luma... higher luma values need higher chroma values if you are to maintain constant hue and saturation/chromaticity.

Premiere makes 75% bars... 100% bars are rare. You can verify the correctness of color bars on an external waveform monitor and vectorscope. Software color bars and scopes may be incorrect. In a NLE, do not import color bars as they may be wrong (sometimes this occurs from conversions that occur). The internally generated bars tend to be mostly correct; some of the NLEs differ in the color of the -I and Q bars, although those bars are not useful anymore since no one broadcasts Y'IQ anymore.

2.) The spec below seems to be saying that the tone is at 0dB. Is this what they are refering to as the "line-up signals (or level)"? Then they say that the maximum sound levels should not be more than 8dB's above the line up level. I can't imagine that they want the audio levels at +8dB. Maybe they meant that the audio signal of the show should be no more than 8dB BELOW the line-up signal? I'm confused, and then there are terms I don't know like PPM4, PPM6, and -4VU.
You have to pay attention to the units. VU in this case would refer to the VU meters on a betaSP deck. They are a measure for analog audio levels.

Don't confuse that with dBFS, used for digital audio (i.e. Premiere). They aren't the same as VU.

Line up level is a reference level, so that a particular dBFS level will correspond to a particular VU level.

You can actually go over the line up level, since audio systems are capable of handling that / they have a little headroom for sudden peaks and transients. Unfortunately, the standards are a little fuzzy here as there are different ways of measuring peaks.

Peak, PPM, and VU meters have different responses to peaks/transients in your audio. With digital audio, digital meters tend to show you the highest numerical value that a peak hits. This produces higher numbers than the other methods.
VU meters are balistically weighted, so they take a short amount of time to respond to a peak and they don't go as high. Some meters aren't weighted like this (to save cost) so that's something to watch out for.
A Peak Program Meter is another method... I forget its details. I think it's the kind of meter with LEDs lights that you see on analog audio mixers.



Return to posts index


rafalaos
Re: Broadcast Technical Specifications
on Sep 23, 2006 at 11:17:56 am

[glenn chan] "Chroma on the other hand is "proportional" to luma... higher luma values need higher chroma values if you are to maintain constant hue and saturation/chromaticity."
Yes, very confusing.
And I think we add more confusion, mistaking concepts
Chroma and Luma are completely different magnitudes and independent.
To say that higher luma needs higher Chroma values is wrong. Needs higher COLOR values. We can have the higher Luma values allowed (1) with Chroma=0. That is a WHITE.
When you put in your screen a white (R=G=B=1) you got Y=1. So the two "color difference" parameters R-Y and B-Y will be =0 because R=B=Y= 1, So R-Y=1-1=0. And B-Y=1-1=0 So I=0 Q=0 in NTSC and U=V=0 in PAL. The same happens if we have a 25% gray (R=G=B=0,25) etc.
This was one of the premises of the old analog TV. In the gray areas of the image there are not Chrominance at all. Just have a look to your any B&W picture in the Oscilloscope. The wave-form monitor will shows you the Luminance level, but when you have a look to your Vector-scope, you find there NOTHING because there is not any chroma to show. Grayes are pure luminance.
So in any picture (Colored or B&W) when we rise the value of the Luminance, we are rising the values of RGB in the same proportion. But Chroma it doesn't change.
All this talking about the old analog concepts of Luma and Chroma. In the digital world we are working with Luma, but the Color Components (Cr and Cb) are magnitudes completely different that the "Color Difference" vectors used in the analog world.
Salud,
Rafael




Return to posts index

glenn chan
Re: Broadcast Technical Specifications
on Sep 23, 2006 at 9:09:20 pm

What I'm saying is illustrated in the following example.

Suppose you have the following values:
1.00R' 0.00G' 0.00B'
0.75R' 0.00G' 0.00B'
0.50R' 0.00G' 0.00B'
0.25R' 0.00G' 0.00B'

I would say that all four values have the same chromaticity. They are all "red".

However, they do not all share the same color difference components.

1.00R 0.00G 0.00B
Y' (Rec. 601) = 0.299
B' - Y' = -0.299
R' - Y' = 0.701

0.75R 0.00G 0.00B
Y' (Rec. 601) = 0.22425
B' - Y' = -0.22425
R' - Y' = 0.52575

0.50R 0.00G 0.00B
Y' (Rec. 601) = 0.1495
B' - Y' = -0.1495
R' - Y' = 0.3505

0.25R 0.00G 0.00B
Y' (Rec. 601) = 0.07475
B' - Y' = -0.07475
R' - Y' = 0.17525

*Rec. 601: Luma (Y


Return to posts index

rafalaos
Re: Broadcast Technical Specifications
on Sep 24, 2006 at 4:32:46 am

Hi Glenn,
The only problem we have is the terminology. Chromaticity (tone?) is not a magnitude in terms of video. In video we manage two magnitudes perfectly mesurable: Luminance and Chrominance.
The Luminance carry the brigthnnes information, and is modulated in amplitude. On top of that we add a signal that contais the color information: The Chrominance. We get the Chorminance by modulating in Quadrature the Color Difference component. The amplitude of the Chrominance determine the saturation, and the phase of the Chrominance determine the hue.

When you says:
I would say that all four values have the same chromaticity. They are all "red". However, they do not all share the same color difference components....

Yes they can never share the same "Color Difference components". and the y can never have the same Luma. THE ONLY THING THE SHARE IS THE PHASE OF THE CHROMINANCE VECTOR. That is what determine the tone in NTSC and PAL. That was the old problem in the NTSC. When demodulating, with any little shift in the phase we will recover the wrong Color Difference components.
In all the gray scale we got Luma but Chrominance=0. Each gray value correspond to a R=B=B value. Yes we have colors, but no Chrominance.:


[glenn chan] "- In the case of 0.75R 0.00G 0.00B, let's assume that the legal range is from 0-1. We could say that
0.75R 0.00G 0.00B is either 100% saturated, which is true if you mean chromaticity"


0.75R 0.00G 0.00B are 75% saturated. !00% saturated would be 1R 0.00G 0.00B. The only think is that this color is ilegal, because when you get your video composite signal, you pass the limits established by the NTSC and PAL standard.
Any way, Glenn, as I sayd before, I think we have a terminology problem. I've learnt the NTSC and PAL system back in 1985. I've got beside me the same old manual I used. Now we all the new digital think, I have to look at the book now and them becouse I get really confused. The think is that in digital it doesn't exist the Chrominance. Is a concept and a magnitude that only appear when play the video out.
Salud,
Rafael
PS: I think that, even with our mistakes and doubts, we must be happy of our knowledge about video. I think that a great deal of the people working in digital video, are really lost.






Return to posts index


glenn chan
Re: Broadcast Technical Specifications
on Sep 25, 2006 at 1:21:53 am

0.75R 0.00G 0.00B are 75% saturated.
Any way, Glenn, as I sayd before, I think we have a terminology problem.

This is the point. To me, 0.75R 0.00G 0.00B is 100% saturated. And I would be correct based on *my* definition of saturation (saturation = chromaticity). Chromaticity is not a "magnitude in video" but it is a useful concept especially in artistic work.

To you, 0.75R 0.00G 0.00B is 75% saturated. And you would be correct if you're talking about the strength/magnitude/amplitude of the resulting signals.

2- In much of digital video, Y'CbCr encoding is used. The Y' component (luma) contains the luma channel, while Cb and Cr are color difference components (chroma). So there is such thing as luma and chroma in digital video. The scale factors and the offsets however are different. Charles Poynton's book has good information on this.
http://www.poynton.com/DVAI/index.html

3- If you're interested, there are differences between the luma of video engineering and the luminance of color science. So you may want to avoid the term luminance, since it generally does not apply.
http://www.poynton.com/papers/YUV_and_luminance_harmful.html

The main difference is in the order of operations. Gamma correction is applied before forming luma. With luminance, gamma correction would be applied after forming luminance.


Return to posts index

rafalaos
Re: Broadcast Technical Specifications
on Sep 25, 2006 at 6:00:41 am

Hi Glenn,
Thanks for the documment on YUV and Luminance. Its clear many things that I couldn't understand.
I learnt the NTSC and PAL more than 20 years ago, but now with the "all digital" I was getting really confused.
Any way I would like to discuss with you about few points, but I'm going for a travel today and I'll be unable to use internet for the next few days (there are still places in the world without it). Could you please send me your e-mail and I'll write you when I"m back?
Thanks.
Rafael


Return to posts index

glenn chan
Re: Broadcast Technical Specifications
on Sep 25, 2006 at 5:57:59 pm

My *gmail* account is glennchan@___.com


Return to posts index

<< PREVIOUS   •   VIEW ALL   •   PRINT   •   NEXT >>
© 2017 CreativeCOW.net All Rights Reserved
[TOP]