Y/C convergence -- is there a name for this?

Due to my lack of knowledge, for now I'm giving my own name to a phenomena I've noticed that seems to affect image quality on all CRT TVs (whether HD or not). I suspect this should affect all kinds of TVs including LCD, but since I have little experience with them, I'm posting this topic in the CRT board.

I'll call this phenomena Y/C convergence. It has a similar effect to the well-known RGB color convergence issue (although RGB color convergence only affects CRTs). What seems to happen is that the color (C) information in the input signal ends up leading (or trailing) the luminance (Y) information in the input signal, when it comes time to display on the screen. This results in a deconvergence in the displayed image, similar to when the Red/Green/Blue guns are misaligned on a CRT. It's not quite the same effect, but somewhat similar. Pretty much all TVs I've seen seem to be inflicted with this Y/C deconvergence disease to some extent (good TVs perfect or near-perfect, bad TVs terrible). Usually the amount of Y/C deconvergence is greater than the same TV's amount of RGB deconvergence on the sets I've seen, so I think it is an important thing to try to get right.

Y/C convergence is totally unrelated to RGB convergence though, it's just a way for me to explain what it looks like. I think Y/C convergence is related to the inputs on the television, because I've noticed that the same television can have completely different amounts of Y/C deconvergence depending on the particular input used.

What it looks like is most easily explained or seen in a cartoon. Imagine a cartoon image where someone drew a picture on a white screen with a black pencil. Then someone comes and colors in the colors in the appropiate areas. OK, now you've got a perfect image. But, say you then have Y/C deconvergence of 2 pixels. What happens is that the black pencil lines will be shifted to the right (or left, depending which way is your deconvergence) by 2 pixels, but the colored-in image will stay in the same spot. So now your cartoon drawing will look a little bit weird.

Y/C deconvergence will similarly have an extremely noticeable effect on colored subtitles with black outlines, such as those a DVD player might generate on a foreign film. Say your subtitles are supposed to be a yellow color. If your Y/C deconvergence is bad, some parts of the letters might turn white instead (because the color information is shifted onto the black part, and vice versa).

But Y/C deconvergence will have no effect on white subtitles with black outlines, because it only relates to color (unless the black outlines are actually invisibly a dark shade of a color).

Does anyone know what is the proper name for this phenomena?

It’s a pretty simple term it’s called an Y/C delay. Some machines like a BetaSP have an adjustment for it and others like a Digi-beta don’t. This could be happening when the monitor is changing the signal from color deference to RGB. It is very rare that we get a tape in with an Y/C delay.

Do you see this from specific input connections and not others, or is it always there? If it's specific to certain inputs, like Composite or S-Video, it could have to do with the specific processing on that input.

A lot of TV's also have "velocity scan modulation", which is designed to speed up the B+W transitions. It could be that is causing the luminance to lead the chrominance. If there's a menu item for it, you may want to try turning it off.

It's not specific to a certain type of input. It happens on all of Component, Composite, S-Video. The amount of it will vary with the particular TV, probably even the particular unit (quality control), and the particular input you use on that TV. A good TV should have 0 Y/C delay but practically all TVs I have seen have a noticeable amount.

My old JVC analog tube has 4 pixels of Y/C delay to the left (C leads, Y trails) through S-Video inputs. 1 of my recent Toshibas had 1 pixel of Y/C delay to the right through S-Video (Y leads, C trails). All my last 3 Toshibas had 5 pixels of Y/C delay to the left through component, but my 1st Toshiba had 0 pixels of Y/C delay through component, it was spot on.

I always turn off velocity scan modulation. Check back a page on this forum and you'll see I made a topic called "Scan velocity modulation sucks" or something like that. :) I think that was just before you arrived.

It's not specific to a certain type of input. It happens on all of Component, Composite, S-Video. ...My old JVC analog tube has 4 pixels of Y/C delay to the left (C leads, Y trails) through S-Video inputs. 1 of my recent Toshibas had 1 pixel of Y/C delay to the right through S-Video (Y leads, C trails). All my last 3 Toshibas had 5 pixels of Y/C delay to the left through component, but my 1st Toshiba had 0 pixels of Y/C delay through component, it was spot on.There is an impairment in composite color video called Differential Phase and Differential Gain a type of Intermodulation Distortion (IM) - that would cause that problem and others - It is monitored by the broadcaster with a vector scope using colorbar a signal added into VIB of the the composite video -
Unless yourMpeg decoder in your TV or STB, PVR etc. is defective - digitally encoded video that was never analog - shouldn't have it - Some animé (most from what I read) is mastered in analog. When you are talking about the use of S-Video input -then - from this explanation - the cause is simple delay of the 2 (completely) separate signals --using S-Video input for a signal that was ever composite will not restore the Y/C separation or bandwidth lost when it was composite
http://en.wikipedia.org/wiki/Y/C (Redirected from Y/C)
Separate video, abbreviated S-Video and also known as Y/C (or erroneously, S-VHS and "super video") is an analog video signal that carries the video data as two separate signals (brightness and colour), unlike composite video which carries the entire set of signals in one package.
Overview
The luminance (Y; brightness) signal and modulated chrominance (C; colour) subcarrier information are carried on separate synchronized signal/ground pairs.
In composite video, the luminance signal is low-pass filtered to prevent crosstalk between high-frequency luminance information and the color subcarrier. S-Video separates the two, and detrimental low-pass filtering is unnecessary. This increases bandwidth for the luminance information, and also subdues the color crosstalk problem. The infamous dot crawl is eliminated. This means that S-Video leaves more information from the original video intact, thus having a much-improved image reproduction compared to composite video.
Due to the separation of the video into brightness and colour components, S-Video is sometimes considered a type of component video signal, although it is also the most inferior of them, quality-wise, being far surpassed by the more complex component video schemes (like RGB). What differentiates S-Video from these higher component video schemes is that S-Video carries the colour information as one signal. This means that the colours have to be encoded in some way, and as such NTSC, PAL and SECAM signals are all decidedly different through S-Video. Thus, for full compatibility the used devices not only have to be S-Video compatible but also compatible in terms of colour encoding.for gaming - all bets are off - anything can happen in the rendering by the console.

I claim that the decoders in almost ALL TVs are "defective" to some extent regarding their ability to synchronize the Y/C information. This appears to be a non-trivial task, since so many televisions display some small amount of Y/C delay, including the much-lauded Sony Wegas.

It has nothing to do with the source. And it has nothing to do with the process of color separation, as this phenomena occurs with all types of connectors, including S-Video, composite, and component.

Do you see it on DVD's using the Component inputs? That would be strange as DVD's are encoded as Component data and there should be a near perfect translation between the Component data and the Component signals.

Since the Toshiba TV's upsample everything, including the Component inputs, there are extra conversions and processing going on that could skew the response, but gee whiz, if it's in the digital domain, it should be a constant offset that could be compensated for...

If it varies from set to set of the same model, then it must be analog, but I don't see why it would happen on Component video from a DVD...

Due to my lack of knowledge, for now I'm giving my own name to a phenomena I've noticed that seems to affect image quality on all CRT TVs (whether HD or not). I suspect this should affect all kinds of TVs including LCD, but since I have little experience with them, I'm posting this topic in the CRT board....
What it looks like is most easily explained or seen in a cartoon. Imagine a cartoon image where someone drew a picture on a white screen with a black pencil. Then someone comes and colors in the colors in the appropiate areas. OK, now you've got a perfect image. But, say you then have Y/C deconvergence of 2 pixels. What happens is that the black pencil lines will be shifted to the right (or left, depending which way is your deconvergence) by 2 pixels, but the colored-in image will stay in the same spot. So now your cartoon drawing will look a little bit weird.
Y/C deconvergence will similarly have an extremely noticeable effect on colored subtitles with black outlines, such as those a DVD player might generate on a foreign film. Say your subtitles are supposed to be a yellow color. If your Y/C deconvergence is bad, some parts of the letters might turn white instead (because the color information is shifted onto the black part, and vice versa).Does anyone know what is the proper name for this phenomena?Since DVDs, NTSC, HDV and PAL have only one color pixel datum for every 4 luminance pixels data ( 4:2:0 -the four luminance pixels have the identical hue and saturation but differ in brightness) and OTA HDTV has at least two luminance pixels for each color pixel datum (4:2:2) the inherent ambiguity of the color space would be as much some of the "Y/C convergence error" you are seeing on the various displays digital inputs. In the case of DVDs or SD eg. four adjacent pixels will all be the same color -- In the case of HD content at least 2 adjacent pixels will always be the same color (although different intensity) -
Digital TV Color Resolution - http://members.aol.com/ajaynejr/vidcolor.htm
The necessity of reducing color resolution to fit the allowable transmission or storage space has continued into the digital age. The most common standard for U.S. video, both for HDTV and digital SDTV broadcasting and DVD, is half the luminance resolution, both horizontally and vertically. In other words, there is one color pixel for every 2x2 block of luminance pixels. There is one other common standard where every four luminance pixels in a row share the same color.If you burn a DVD from your DV camcorder - the numbers get even worse 1 color for 8 intensity pixels
and then you may want to look here too Transitioning To A Different Color, Oversimplified (http://members.aol.com/ajaynejr/vidcol2.htm#Transition)

Boby, I did see it on DVDs through my component inputs on my 3 Toshiba 30HF85s that were manufactured in April 2005. But I did not see it through component on my September 2005 unit.

Yes, that confuses me also since component video should already have the color information separated. It could be that only the Toshibas were defective and any other TV is fine through component (the Toshibas are the only TVs I have ever tested through component).

maicaw, that last post you made was REALLY helpful!! Thanks a lot!! So then I should not expect perfect "Y/C convergence" through any inputs. I should generally expect at least 1 or 2 pixels of varying "deconvergence" from any input, even a "perfect" one.

That still doesn't fully explain why on some of my Toshibas the deconvergence was even worse though (5 pixels, and worst through component, of all things). My guess is that there is still some stage at which the TV can make a further mess of things regarding Y/C synchronization, and good TVs won't make a mess. But it's good to know that perfect Y/C synchronization can't occur even in theory on TVs (but I'm guessing on computer monitors it can be perfect, since I've never seen this problem on them).