..With the single context you're speaking of in the prior quote the relationship you're stating is fine, but to take that voltage to digital value relationship and say that it's universal to any situation as defined by Rec 601, that is my disagreement with what you've posted.I see that unity calibration is beyond your comprehension:

100/0 .7V p-p Y’ (16-235) coded into a Rec 601 8-bit digital payload

100/0 .7V p-p CbCr (16-240) coded into a Rec 601 8-bit digital payload

This represents the industry standard YCbCr 100% color bar “digital” signal per Rec 601. My Leader 5750 SDI Multi Format Waveform monitor has a built in and “calibrated” ITU-R (RGB/YCbCr) matrix. In this mode of operation I don’t care to see a pseudo NTSC waveform with (714 mV) for white and or 7.5% setup.. or Sony Betacam or Panasonic MII presentation because they represent incompatible scaling factors.

I gave you this explanation and definition "reference" standard color bars:
http://broadcastengineering.com/mag/broadcasting_color_bars_puzzle/

Nobody.. and I repeat nobody is encoding Sony Betacam standard, Panasonic MII standard or NTSC (714 mV for Wht) like Ron is suggesting below. Rather I’m looking at what is actually (YCbCr) being encoded into the payload.
The designers have chosen to represent this pseudo waveform:p
or
The product can be configured for the different analog component standards:p.

Look at the specifications at the bottom of the page.

Y Signal Level 700 mV or 714 mV (switchable, +/- 1 dB adjustable)
Pb, Pr Signal Level 934 mV (Betacam) or 700 mV (SMPTE) or 648 mV (MII):D

Nobody.. and I repeat nobody is encoding Sony Betacam standard, Panasonic MII standard or NTSC (714 mV for Wht) like Ron is suggesting below. Rather I’m looking at what is actually (YCbCr) being encoded into the payload.

This forum is titled Display Calibration. Why do you think it's completely clear to make statements that deal with only one single situation (encoding of defined analog signals), which doesn't necessarily always fit with how others might use the forum? I'm fully aware that you continually don't state your context, but why do you think others should implicitly understand a context that goes unstated?


unity calibration is beyond your comprehension

Again, when when was it specifically stated that something called "unity calibration" was the entire realm of discussion? My statements are seemingly in line with the title of the forum, which can apply to many various topics including such things as US NTSC analog levels and Rec 601 quantization. This is why I've harped on the word context repeatedly and used terms like universal and absolute in describing my objection to how you've stated your position. How do you figure others are to understand unstated context? If someone was to look at Rec 601, they would clearly see that the document is not written around one single expectation of 0.7V p-p analog levels for 16-235 as you've again stated.

Her is an email response from John Luff to my question below, which is a word for word quote of a question I emailed to Michael Robin:

My name is Thomas Brunet and I'm a Broadcast Engineer for 25+ years now and I'm currently working with Savannah College of Art & Design:
http://www.scad.edu/
If you will, I would like to ask a question and I understand in advance you probably get many and I may not get a reply from you.

I'm in an on-line debate regarding the concept of a unity ITU-R BT.601 codec model. I guess I'm trying to make a distinction between normalization (0-1) and the "standard" .7V p-p analog YCBCR waveform that I see on my SD-HD SDI waveform monitor graticule. My position is that when I'm encoding Y' excursion for a given frame of video per ITU-R 601, for example an 8-bit Y' excursion (16-235) is defined to be 0V and 700mV respectively and +/-350mV CBCR for codes (16-240)?

I'm being told that ITU-R BT.601 does not define any specific given component amplitude i.e. an 8-bit decimal Y' code value of 235 does not = 700mV? Rather ITU-R 601 only states White is 1V and Black is 0V. So in reality there's no unity .7V p-p analog model. That gain and offset per output is the only time 700mV is relevant?
The simple answer is the if I understand your argument correctly you are right.The issue is the NORMALIZED values before coding vs the input components, assumed to be RGB. The standard defines the encoded color space as well as the transform to 4:2:2 coding.

To understand the standard fully you have to think of it in different terms. First, the input to BT.601 is analog component video (RGB), the same as specified in SMPTE 125M, which also defines 4:2:2 sampling, and is somewhat more clear in presentation. The analog input values are specified in the standard as 1.0 V for luminance and each of the chroma difference signals. Part of the standard requires that prior to quantization the signals (1.0 V for luminance and the matrixed values for Cr and Cb “…(E’R – E’Y ) + 0.701 to – 0.701 and for (E’B – E’Y) a range of + 0.886 to – 0.886 ”) are normalized. Effectively it is those values that are normalized to 1 volt for peak excursions and then encoded using the parameters in the chart below (10 bit samples).

Part of the confusion comes from the fact that it is possible to encode signals using ITU-R BT.601 which would exceed the legal color gamut. That is the reason that white is defined less than full excursion, in addition to reducing ringing on sharp transitions.

John Luff
Television Technology Consultant
SMPTE Fellow
SMPTE Conference VP

I'm being told that ITU-R BT.601 does not define any specific given component amplitude i.e. an 8-bit decimal Y' code value of 235 does not = 700mV

This is not exactly the statement being presented here. Component video has defined analog levels. Within a defined context where certain analog levels are used, Rec 601 can relate between analog levels and digital values. The thing is that there are also other defined analog levels relevant to Rec 601, such as US NTSC levels. Much of that sort of context is generally legacy, but my players still contain composite and you can bet some people still use it. I've read that some DVD players have also allowed for component to use setup in order to use a similar range as US NTSC, which I guess was maybe for displays that only allowed one brightness setting for all inputs. All these issues are not necessarily relevant for what you're doing, but in general they apply to a forum titled Display Calibration.

ITU-R 601 only states White is 1V and Black is 0V.

The standard uses 1 and 0. It generally avoids the voltage term. V does appear once I know of, but it's listed as an example. The standard also states that it can apply to quantization of digital levels, so that's also likely related to why Rec 601 is based around normalized levels of 0-1 for E'Y in going to 16-235.

Again, when when was it specifically stated that something called "unity calibration" was the entire realm of discussion?
http://www.avsforum.com/avs-vb/showthread.php?t=1009493&page=2

Posted 03-27-08, 07:10 PM
Originally Posted by tbrunet
I said at least three times already for a hypothetical ideal A/D to D/A unity.

Originally Posted by tbrunet
If I don’t measure 0mV (aka blanking level) for the quantization level Y’=16 , Cb=128, Cr=128, then either the D/A is not (unity) calibrated or the test pattern is non standard. Greg I also understand the concept of AC or DC coupling.

Originally Posted by tbrunet
Nominal excursion for ITU-R BT.601 has blanking level = 0mV and reference white 0-700mV respectively.


http://www.avsforum.com/avs-vb/showthread.php?t=1038332&page=5

Posted 07-01-08, 11:05 AM
Originally Posted by tbrunet
Here are several examples of Greg Rogers attempting to prove that I can’t measure (analog voltage per a unity calibrated chain) the original (digital) encoded IRE level that’s embedded on a given DVD test disc.
Alluringreality just give it up already…John Luff has validated my position 100%

Pun intended:)

The thing is that there are also other defined analog levels relevant to Rec 601, such as US NTSC levels.Not for the encoded standard YCbCr signal that I've been defining for months now.

http://www.tek.com/Measurement/App_Notes/DigitalTV/25W-7203-3.pdf
:rolleyes:
Rec. 601 described the sampling of the signal. Electrical interfaces for the data produced by this sampling were standardized separately by SMPTE and the EBU. The parallel interface for 525/59.94 was defined by SMPTE as SMPTE Standard 125M .

Digital-to-analog conversion range is chosen to provide headroom above peak white and footroom below black as shown in:

Figure 2-5 Luminance quantizing
Figure 2-6 Color difference quantizing

alluringreality, in the link above I dare you to report the analog volts and depicted digital code words in Figure 2-5 and 2-6, I’ll convert hex to decimal if needed.

Have a nice day:)

Alluringreality just give it up already…John Luff has validated my position 100%

Pun intended:)

I don't know why you need him to validate your position... after all, it's only a matter of time before he disagrees with you and you turn against him like you did Greg. Funny how you have a new "authority" every week.

May I make a suggestion. Instead of typing this:

Sorry but Ron, Greg, hwjohn are wrong

why don't you just forego your ever growing list and use this instead:

Sorry, but I'm the only one who is right, even if you agree with me.

John Luff has validated my position 100%

I must have missed where you discussed how "the standard requires that prior to quantization the signals are normalized."


I dare you to report the analog volts and depicted digital code words in Figure 2-5 and 2-6

The supposed challenge is unnecessary, because the context clearly gives that component voltages are being discussed. I didn't notice any poorly-written comments about Rec 601 that I have issue with in the PDF.

I must have missed where you discussed how "the standard requires that prior to quantization the signals are normalized."
I see.. you did not comprehend that the sampled analog electrical signal or YCbCr components represent a prior normalized 100/0/100/0 signal condition so that Y values take on the values between 0 and 1? Maybe you should have read the links I gave you from the many on-line Broadcast Engineering Mag articles I’ve provided for you? NO?
http://broadcastengineering.com/mag/broadcasting_color_bars_puzzle/
The fifth row of Figure 1 shows the formation of the 100/0/100/0 E´Y signal from the primary E´G, E´B, E´R signals. The amplitudes of the eight steps are expressed in percentages of the full-amplitude signal (700 mV).

These signal amplitudes are reduced by scaling factors to meet specific signal amplitude range requirements. When the allowed amplitude range of these signals is 700 mV (±350 mV), as specified in the EBU N10 Standard and the ITU-R BT 601 recommendation, the color-difference signals are given by the following expressions:

E´CB = 0.564 (E´B - E´Y), also known as PB in North America

E´CR = 0.713 (E´R - E´Y), also known as PR in North AmericaSearch back in this thread and see how many times I defined a analog 100/0/100/0 color signal to you and please note or count how many times I defined exactly that p-p analog signal voltage.
You're not actually measuring mV with your scope from a digital input.This of course is wrong!
Digital levels are not in volts. That's pretty damn elementary. A waveform monitor that accepts digital signals (i.e. SDI) can have a graticule that shows equivalent analog volts in any standard that relates digital levels to analog voltsThis to is wrong! A industry standard SDI WF monitor "calibrated" per ITU-R standards..the (RGB/YCbCr matrix) .7V graticule is not a pseudo representation:)
The designers have chosen to represent this pseudo waveform

..Thers no "real" blanking in SDIand that would be wrong and wrong as well.

http://broadcastengineering.com/mag/broadcasting_color_bars_puzzle/
In http://www.avsforum.com/avs-vb/showthread.php?p=14422760#post14422760 I didn't look at the link because I considered your comments not relevant to the point of what you quoted like usual. When you linked that a second time I forgot how all your comments have to be filtered through an unstated implied context of only speaking about encoding component video, and so the prior paragraph seemed at best a bit convoluted because something like US NTSC could be said to be an "industry standard" from my perspective. Anyway, certainly that second article does define E'y of white relative to 100% or 1 and answers the question I asked prior.

This of course is wrong!

I figure I get what you've generally attempted to say, but I'm still not so sure you get the point of others have stated. My point of what you quoted was that if you actually sampled a digital signal in terms of mV you would basically get alternation between two levels. Sure your scope can represent the digital information in other terms and use a variable mV scale, but you're not actually sampling mV. That's my whole point, so I have completely no idea how that could be said to be wrong because it's just digital basics. Anyway I figure you'll have some issue with these statements, and my ignore list has grown to two.

Anyway I figure you'll have some issue with these statements, and my ignore list has grown to two.Better make that at least three:p

1) tbrunet
2) Michael Robin
3) John Luff