PAL EX XDCAM is 4:1:1 or 4:2:0??
 

Hi,
Pal dvcam is 4:2:0.
What about Pal xdcam?

4:2:0 sadly.

Very sad indeed :)

Thanks for the reply.

all XDCAM 35 Mbps (EX and disc-based) is 4:2:0.

4:1:1 was never used before or since NTSC DV as far as I know.

The camera can output 4.2.2 through the SDI port.

Sony has a video online where video compression is explained and they show examples of each and 4.2.0 looks much better than 4.1.1 so imo 4.2.0 is a good thing! That's why 4.1.1 is not used for anything but SD DV.

Actually 4.4.4, 4.2.0, etc. technically only applies to standard definition video, this number scheme is just one of the artifacts left over from the early days of digital SD but everyone uses it.

It describes the different types of chroma subsampling regardless of definition.

It is actually referring to the sampling rate of the circuit for each of the components.

A number such as "4.4.4" was based on an image of 720x480 pixels using a sampling circuit of 13.5Mhz, a "2" would refer to a 6.75Mhz sampling circuit, etc.

Since full HD is 1920x1080 pixels you can't sample that at 13.5Mhz you need a much higher sampling frequency - 74.25Mhz.

YES, everyone uses the "4.4.4" numbers for HD but very few could explain the math that generates those numbers, because if they did they would understand how TECHNICALLY it is wrong when talking about anything but a SD sampling rate.

I don't think 4:1:1 is any better than 4:2:0.

True enough.

I've talked with industry guys who have proposed scaling the nomenclature from 4:2:2 to 22:11:11 to maintain the original math.

I guess that must've seemed too complicated

We never guess, we always look it up.
 

The subsampling scheme is commonly expressed as a three part ratio J:a:b (e.g. 4:2:2), that describe the number of luminance and chrominance samples in a conceptual region that is J pixels wide, and 2 pixels high. The parts are (in their respective order):

J horizontal sampling reference (width of the conceptual region). Usually, 4.
a number of chrominance samples (Cr, Cb) in the first row of J pixels.
b number of (additional) chrominance samples (Cr, Cb) in the second row of J pixels.

Note that this is abstract geometry and has NOTHING to do with the actual pixel dimensions ("resolution") of the video frame.

You could look it up instead of arguing about it here.
Chroma subsampling - Wikipedia, the free encyclopedia

No one is arguing, Kevin was simply pointing out the origin of 4:4:4, in which he is totally correct. You guys have simply misunderstood him.

I'd have to agree. In fact, I'd rather key 4:2:0.

You have to look back to the origins of digital video to see how the terminology came about. Initially the composite signal was digitised, and it made sense to lock the sampling frequency to subcarrier - frequencies of 3x sub-carr freq for PAL (13.29MHz) and 4x sub-carr freq (14.32Mhz) for NTSC became the norm.

Move to component, and the desire was to use similar frequencies for digitising the component signals. It was further desired to use the same frequency for the 525 and 625 line systems (which it now becomes wrong to call "PAL" and "NTSC") and 13.5 MHz was settled on as a common compromise.

The "4" and "2" in 4:2:2 etc therefore comes from "no of samples in a cycle of NTSC subcarrier", even though by the time it was first used the move from 14.32MHz to 13.5MHz sampling meant it was referring to longer than a cycle anyway.

That's the origins - the wikipedia definition is how it's now come to used, and it must be one of the worst and most confusing bits of nomenclature in the TV world. All too often it's heard that "X is 4:2:2, so it must be better than Y which is only 4:2:0". That's only true if the luminance aspects of both systems are the same, compare a 4:2:0 full raster system with a 4:2:2 subsampled one and it's completely untrue.