Do most TVs now support 4:4:4? Which ones? (original thread not updated). - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 104 Old 12-12-2013, 06:56 AM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644

 

NOTE:

The previous "official 4:4:4 thread" was maintained

by someone who hasn't posted an update to the

list in the top post since 2012, so any new information

is difficult to find.  It was a valiant attempt at the time.

 

What's the current support for 4:4:4 out there?  Which best describes the situation?

  • Few (unlikely I think because of games)
  • Half
  • Most
  • Almost All
  • All

 

I wonder about this. I know of the ways to test for it, but I'm concerned that the information about this isn't always easy to come by.

 

Is it common enough now that we now need a warning list in a "No 4:4:4 support for the following TVs" list?

 


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
Sponsored Links
Advertisement
 
post #2 of 104 Old 12-12-2013, 03:08 PM
 
MonarchX's Avatar
 
Join Date: Jul 2012
Posts: 669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
It became a very common thing for TVs to support 4:4:4 chroma subsampling, using PC modes or similar. In my case, using PC Mode disables colorspace calibration. I prefer using 4:2:2 Game Mode because it is a process-free mode that keeps text and pixels very sharp, while Movie 4:2:2 mode produces red or blue or green sub-pixel lines or whatever they are called.
MonarchX is offline  
post #3 of 104 Old 12-12-2013, 06:01 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
YCbCr 4:4:4 offers zero image quality advantages. All you do with 4:4:4 is send more information to the TV than you can see.

Everything visible to humans is present in 4:2:2

There is no 4:4:4 when you are in RGB mode, 4:4:4 applies only to YCbCr. The important thing is whether the TV produces better looking images in RGB more or YCbCr mode. There SHOULD be no difference. But sometimes stuff happens and maybe 5% of the TVs out there just plain look better when they are in RGB mode than they look if you send either YCbCr format.

By the way, Blu-ray disc is encoded in 4:2:0 right on the disc. So 4:2:2 is an expansion that doesn't really add any new image information. And 4:4:4 is just a waste of bandwidth.

There are no cases I'm aware of where 4:4:4 looks any better than 4:2:2. If it happens, something is wrong as our best video source today is 4:2:0 and going up to 4:2:2 or 4:4:4 accomplishes nothing but adding bits with no added information... once the information is removed in the process of getting from 4:4:4 in the original down to 4:2:0 on the disc, that information is gone forever. So, a display that only worked in 4:2:0 would produce Blu-ray images that look the same as 4:4:4 images from Blu-ray on some other disc.

So, really, it doesn't matter if a display has a 4:4:4 mode or not, it will not do anything for image quality. RGB is like 4:4:4 in that there is far more information encoded in RGB data than we can actually see. That's why it is very rare to see data stored in RGB format... it just wastes space. YCbCr 4:4:4 and RGB have the same amount of data per frame... and should appear identical. If the conversion to YCbCr 4:2:2 is not broken in any given display, 4:2:2 and 4:4:4 will look identical.
MST246 likes this.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #4 of 104 Old 12-12-2013, 07:42 PM
 
MonarchX's Avatar
 
Join Date: Jul 2012
Posts: 669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Quote:
Originally Posted by Doug Blackburn View Post

YCbCr 4:4:4 offers zero image quality advantages. All you do with 4:4:4 is send more information to the TV than you can see.

Everything visible to humans is present in 4:2:2

There is no 4:4:4 when you are in RGB mode, 4:4:4 applies only to YCbCr. The important thing is whether the TV produces better looking images in RGB more or YCbCr mode. There SHOULD be no difference. But sometimes stuff happens and maybe 5% of the TVs out there just plain look better when they are in RGB mode than they look if you send either YCbCr format.

By the way, Blu-ray disc is encoded in 4:2:0 right on the disc. So 4:2:2 is an expansion that doesn't really add any new image information. And 4:4:4 is just a waste of bandwidth.

There are no cases I'm aware of where 4:4:4 looks any better than 4:2:2. If it happens, something is wrong as our best video source today is 4:2:0 and going up to 4:2:2 or 4:4:4 accomplishes nothing but adding bits with no added information... once the information is removed in the process of getting from 4:4:4 in the original down to 4:2:0 on the disc, that information is gone forever. So, a display that only worked in 4:2:0 would produce Blu-ray images that look the same as 4:4:4 images from Blu-ray on some other disc.

So, really, it doesn't matter if a display has a 4:4:4 mode or not, it will not do anything for image quality. RGB is like 4:4:4 in that there is far more information encoded in RGB data than we can actually see. That's why it is very rare to see data stored in RGB format... it just wastes space. YCbCr 4:4:4 and RGB have the same amount of data per frame... and should appear identical. If the conversion to YCbCr 4:2:2 is not broken in any given display, 4:2:2 and 4:4:4 will look identical.

It has HUGE advantages for text. It may have no advantage in actual DVD/HD content from several feet away, but for those who want to use TVs as monitors, 4:4:4 does have a big advantage in sharpness. I've already posted what one pixel-wide lines look like. You can be a pro calibrator all you want, but people on these forums constantly make posts about how they notice oddities with text and then they find out about PC Mode and 4:4:4, switch to it, and see a difference. You can't just ignore that. These statements are as valid as things you said about video cards and firmware, which was utterly ridiculous.
MonarchX is offline  
post #5 of 104 Old 12-12-2013, 09:35 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
There are so many issues here... one of which is whether you are using an _VGA 15-pin connector from a computer or an HDMI cable from a consumer source component or an HDMI cable from a computer. (more on computers later)

It would be impossible for black and white text to look better in 444 than 422

Cb and Cr carry only color difference information (see YCbCr in Wikipedia for an example of a color image plus the same image as 3 separate images in Y Cb and Cr.

If you were displaying black and white text Cb and Cr would be all zeros whether 422 or 444... no information since all the information for the B&W text would be in the Y channel.

Color text on a color background... MIGHT be problematic if the luminance of the text and background was the same, but that would make for very low contrast text that is difficult to read.

If you were displaying text from a Blu-ray disc and your TV looks better in 444 than 422, text should REALLY look like crap from a Blu-ray disc because the entire disc is encoded as 4:2:0. Since there is plenty of text on many Blu-ray discs and it always looks sharp as a tack, I think it's pretty safe to say that 444 and 422 make NO DIFFERENCE in and of themselves in what text looks like.

If you see a difference in B&W text between 444 and 422, SOMETHING in the chain is messed up. It is NOT the fault of 422 vs 444

It would even be difficult to mess up text if there was a solid color background and text of a different color since al the colors we can see are present in 422 and all the resolution of the image is in the Y channel.

That said, things can be broken... that's why maybe 5% of video displays just happen to look better when you send RGB compared to either YCbCr format... those cases don't mean there's anything inferior about YCbCr, it just means something in the TV is poorly engineered.

The other issue is that computers are notoriously bad with YCbCr, ESPECIALLY over a 15-pin connector... HDMI is less of a problem but can still be problematic. I just checked my laptop with a 4 or 5 year old nVidia board and see ZERO difference between a Microsoft WORD document in RGB, YCbCr 4:2:2 and YCbCr 4:4:4 via an HDMI connection from the computer as displayed on a 4 year old Samsung plasma TV.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #6 of 104 Old 12-13-2013, 06:30 AM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by Doug Blackburn View Post

There are so many issues here... one of which is whether you are using an _VGA 15-pin connector from a computer or an HDMI cable from a consumer source component or an HDMI cable from a computer. (more on computers later)

It would be impossible for black and white text to look better in 444 than 422

 

Of course not, because the luma information isn't subsampled.  But text isn't merely black on white on PCs or games. And you're grossly over-simplifying what chroma subsampling does.  1 pixel wide lines of one color over another are commonplace in PC's.  Yes, Blu-Ray is encoding in 4:2:0, but that's not my question.

 

I was hoping that this thread wouldn't turn into a discussion of what Chroma subsampling is or isn't....I was hoping to confine this to what is available.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #7 of 104 Old 12-13-2013, 06:34 AM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by MonarchX View Post

It has HUGE advantages for text. It may have no advantage in actual DVD/HD content from several feet away, but for those who want to use TVs as monitors, 4:4:4 does have a big advantage in sharpness. I've already posted what one pixel-wide lines look like. You can be a pro calibrator all you want, but people on these forums constantly make posts about how they notice oddities with text and then they find out about PC Mode and 4:4:4, switch to it, and see a difference. You can't just ignore that.

 

Agreed.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #8 of 104 Old 12-13-2013, 06:46 AM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644

 

Quote:
Originally Posted by MonarchX View Post

It became a very common thing for TVs to support 4:4:4 chroma subsampling, using PC modes or similar.
That seems to be the case, but it's hard to find actual documentation on which ones don't.  All the links I find are to old lists.  But I don't believe that this is because the number of 4:2:0 only TVs has shrunk to zero.
 
Quote:
In my case, using PC Mode disables colorspace calibration. I prefer using 4:2:2 Game Mode because it is a process-free mode that keeps text and pixels very sharp, while Movie 4:2:2 mode produces red or blue or green sub-pixel lines or whatever they are called.

Artifacts are common, and you're already almost certainly aware of this, but be careful that your PC mode doesn't leave on some half-assed interpolation that itself leaves artifacts.  The bottom end Sony R55 (on the 70" version at least) is showing blue issues during motion if you're not careful.

 

PC modes really do well with some of the pulse motion-blur mitigation out there, which serves to defeat the sample-and-hold smear against the retina.  Many TVs supply a form of this completely without interpolation that might be on during PC mode.  If you don't have that on, look for it.  Those cause 0 ms addition lag (to what it already has) because there is no computation performed for the pulse.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #9 of 104 Old 12-13-2013, 10:23 AM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
Quote:
Originally Posted by tgm1024 View Post

Of course not, because the luma information isn't subsampled.  But text isn't merely black on white on PCs or games. And you're grossly over-simplifying what chroma subsampling does.  1 pixel wide lines of one color over another are commonplace in PC's.  Yes, Blu-Ray is encoding in 4:2:0, but that's not my question.

I was hoping that this thread wouldn't turn into a discussion of what Chroma subsampling is or isn't....I was hoping to confine this to what is available.


Oh come on... what is the difference between a page of text from a Blu-ray disc that looks perfect on a video display when the page of text was encoded at 4:2:0 and looks perfect on the video display versus the same page of text from a PC that looks better in 444 vs 422... it is NOT possible unless something is broken in the PC signal path. I never say that wasn't possible and that it wasn't happening. I just said if it is happening, it is NOT the fault of 422 sampling navitely looking worse than 444. 4:2:0 decomates color information A LOT, and when that color information is gone, it does not come back even when the disc player converts the signal back to 422 or 444.

I didn't say Blu-ray encoded at 420 was your question... the point is that 420 encoding on Blu-ray will produce a beautiful page of multi-color text, no problem. If 420 looks great, there is nothing to be gained by a page that originates as 422 or 444 and if 420 looks great, 422 might look a LITTLE better but by then, the 422 is carrying all the information human vision can see and 444 won't get any better.

As I said... I can sit here and display 422 and 444 text all day on an old(ish) plasma TV from an old(ish) laptop via HDMI and there's no different.... Black & White or color. Color combinations that look bad in 422 look just as bad in 444. This is using multicolor text pages inclided in DisplayMate with Motion, or Microsoft WORD. I never said you couldn't have a situation where there was a visible difference between 422 and 444, I am saying if there is a difference, something isn't right somewhere. I have see computers that seem to do odd things with resolution and you find out later that you are seeing some lower resolution upconverted to HD with 422 but when you switch to 444 you get HD resolution... assuming the computer even supports 1920x1080.... and the worse looking text was a result of upconverting a lower resolution, not because 422 somehow looks worse. That's the sort of thing I mean about SOMETHING in the chain being broken. Because I see plenty of examples where 422 and 444 look identical, 422 vs 444 is NOT the problem.... if there is a problem it is caused by something else. If 444 looks better than 422 then 420 from Blu-ray should be total crap. Does Blu-ray look like total crap? I think we all agree that it does not. And that text originating in 420 from a Blu-ray looks fine. That's the point. When you have an example that should be even worse than the problem you claim exists and in that case the worse example does NOT look worse, something is wrong with the conclusion that 422 somehow looks worse than 444. I edit MS WORD documents on TVs and projectors often and I always look for differences between RGB and YCbCr 422 and 444. Sometimes there are differences, most of the time there aren't differences (nor should there be differences).

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #10 of 104 Old 12-13-2013, 10:38 AM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Quote:
Originally Posted by Doug Blackburn View Post
 
Quote:
Originally Posted by tgm1024 View Post

Of course not, because the luma information isn't subsampled.  But text isn't merely black on white on PCs or games. And you're grossly over-simplifying what chroma subsampling does.  1 pixel wide lines of one color over another are commonplace in PC's.  Yes, Blu-Ray is encoding in 4:2:0, but that's not my question.

I was hoping that this thread wouldn't turn into a discussion of what Chroma subsampling is or isn't....I was hoping to confine this to what is available.


Oh come on... what is the difference between a page of text from a Blu-ray disc that looks perfect on a video display when the page of text was encoded at 4:2:0 and looks perfect on the video display versus the same page of text from a PC that looks better in 444 vs 422... it is NOT possible unless something is broken in the PC signal path. I never say that wasn't possible and that it wasn't happening. I just said if it is happening, it is NOT the fault of 422 sampling navitely looking worse than 444. 4:2:0 decomates color information A LOT, and when that color information is gone, it does not come back even when the disc player converts the signal back to 422 or 444.

I didn't say Blu-ray encoded at 420 was your question... the point is that 420 encoding on Blu-ray will produce a beautiful page of multi-color text, no problem.

 

And then you bring it right back to blu-ray.  And regarding stationary text: Only if they were careful creating it in the first place.  If they weren't, the moment the edges of the text start to stray into odd pixel boundaries, you run into a situation where the subsampling has lowered the color component resolution enough to show artifacts.  Plenty of examples of this online.  Enough of this nonsense please.
 

Quote:

If 420 looks great, there is nothing to be gained by a page that originates as 422 or 444 and if 420 looks great, 422 might look a LITTLE better but by then, the 422 is carrying all the information human vision can see and 444 won't get any better.

 

I'm sorry, but you're derailing the discussion.  Please stop trolling.  I am not talking about starting with 4:2:0 and trying to make it better by displaying it at 4:2:2 or 4:4:4.

 

Now back to what TVs do or don't support 4:4:4.  Any surprise 4:2:0-only TVs out there?


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #11 of 104 Old 12-13-2013, 11:00 AM
Advanced Member
 
sawfish's Avatar
 
Join Date: Feb 2013
Posts: 762
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 103
Quote:
Originally Posted by Doug Blackburn View Post

As I said... I can sit here and display 422 and 444 text all day on an old(ish) plasma TV from an old(ish) laptop via HDMI and there's no different.... Black & White or color. Color combinations that look bad in 422 look just as bad in 444. This is using multicolor text pages inclided in DisplayMate with Motion, or Microsoft WORD. I never said you couldn't have a situation where there was a visible difference between 422 and 444, I am saying if there is a difference, something isn't right somewhere.

Try displaying red text on a blue background in a PC program without anti-aliasing or ClearType. To get my 50ST60 to display it properly, I need to enable the TV's 1080p Pixel Direct feature, and when I do, it's as clear as black text on a white background, contrast notwithstanding. If the source were 422 instead of 444, I don't think anything could help it. Is that right?
sawfish is online now  
post #12 of 104 Old 12-13-2013, 11:08 AM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
Once again, the support for 444 is entirely unimportant, and it's just about universal anyway, at least in the "better" models being sold.

The appearance of 422 vs 444 was brought up in this thread as a reason you'd want 444. I am merely responding to that assertion.

Since there are no cons8mer video sources that have 444 chroma subsampling, the only way to get it would be to create video on a computer and view that on a TV or projector because 444 capability will not make 420 sources like Blu-ray or RGB sources like cable/satellite boxes look any better.

Ask a question with meat on the bones and you might get a better response. But asking to generate a list of TVs that do or don't support 444? What is the point? Why not make a list of TVs that have Sharpness control sliders that run from 0-100 versus from -50 fo +50 versus 0-20. Must be some performance difference there too. I mean, really? Who would want a TV with a 0-20 sharpness control? And that's sarcasm in it's finest form. You make it sound like 444 support is going to get you some bit performance advantage and my point is.... it does not do that, at all. And if it DOES do that, something somewhere in the playback chain is messed up because the amount of color decimation won't be visible if everything is working right and the color decimation is no worse than what is done on Blu-ray disc.

Watch who you are calling a troll. This is the Display Calibration thread. TVs having 444 mode or not really isn't relevant to calibration. You are going to get replies from industry professionals you clearly aren't prepared to deal with. I have 34 years experience as an engineer on professional imaging system, including cinema products where calibration of everything was required. Plus another 6 years of training and experience in calibration with ISF and THX certifications, and 18 years experience as a professional reviewer. I don't have time to troll, But I do have time for the truth. If you can't handle the truth... eh, not my problem.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #13 of 104 Old 12-13-2013, 11:34 AM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
Quote:
Originally Posted by sawfish View Post

Try displaying red text on a blue background in a PC program without anti-aliasing or ClearType. To get my 50ST60 to display it properly, I need to enable the TV's 1080p Pixel Direct feature, and when I do, it's as clear as black text on a white background, contrast notwithstanding. If the source were 422 instead of 444, I don't think anything could help it. Is that right?

OK, so you want to put the computer in the worst possible mode with one of the worst possible combinations of text color possible and try to make it look good on a TV? OF COURSE you can diddle with computer settings and make things look crappy on a TV. This is not a big revelation.

This has nothing to do with Display Calibration. If the subsampling in the computer from RGB to YCbCr 422 was done WELL, the quality of the text you described would be no different on the TV compared to the 444 version of the text. There are only 2 colors present. Red and Blue. WIth 422 there are 2 colors. With 444 there are 2 colors. There's no more color information in 444 vs 422 when you are talking about 2 colors. If there is difference in image quality, it is caused by the computer, not by 422 or 444. That said, with your compiter and your TV you might have to use 444 to get the best looking red on blue text (who want's to read that anyway). Find some red text on a blur background on a Blu-ray disc and have a look at THAT in 422 vs 444, won't be a difference... and that video is encoded as 420, even LESS chroma information than 422.

Once again, this is the Display Calibration thread and this entire discussion really isn't relevant to display calibration. It might be an image quality related topic - but it is pointing out peoples' tendency to think if something isolated happens in their system, it must lead to something like 444 ALWAYS having some advantage over 422. Using 420 examples of how great something can look with HUGE color decimation as an example of why it is HIGHLY unlikely that differences you see in 422 vs 444 images from a computer (especially when you go out of your way to find the worst possible scenario) is really the fault of 422 vs 444, when the actual problem is most likely in the computer and how well or poorly it is handling video. Switch to RGB mode... most computers do better in that domain anyway... and in RGB mode there is no color decimation (aka chroma subsampling).

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #14 of 104 Old 12-13-2013, 12:01 PM
Senior Member
 
praz's Avatar
 
Join Date: Sep 2001
Posts: 375
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 26
Any of these discussion involving a computer as the source is really a waste of time. Most do not understand the interaction and limitation of computer components to output a proper video level signal. And if the signal is properly sent there is always the risk that some future firmware, driver or software update will break it.
praz is online now  
post #15 of 104 Old 12-13-2013, 12:01 PM
AVS Special Member
 
PlasmaPZ80U's Avatar
 
Join Date: Feb 2009
Posts: 7,140
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 63 Post(s)
Liked: 195
what category do PC video games fall under? I'm guessing it's not consumer video sources (as it's designed for PC monitors first and TVs second)
PlasmaPZ80U is offline  
post #16 of 104 Old 12-13-2013, 12:23 PM
Advanced Member
 
sawfish's Avatar
 
Join Date: Feb 2013
Posts: 762
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 103
Quote:
Originally Posted by Doug Blackburn View Post

Quote:
Originally Posted by sawfish View Post

Try displaying red text on a blue background in a PC program without anti-aliasing or ClearType. To get my 50ST60 to display it properly, I need to enable the TV's 1080p Pixel Direct feature, and when I do, it's as clear as black text on a white background, contrast notwithstanding. If the source were 422 instead of 444, I don't think anything could help it. Is that right?

OK, so you want to put the computer in the worst possible mode with one of the worst possible combinations of text color possible and try to make it look good on a TV? OF COURSE you can diddle with computer settings and make things look crappy on a TV. This is not a big revelation.

This has nothing to do with Display Calibration. If the subsampling in the computer from RGB to YCbCr 422 was done WELL, the quality of the text you described would be no different on the TV compared to the 444 version of the text. There are only 2 colors present. Red and Blue. WIth 422 there are 2 colors. With 444 there are 2 colors. There's no more color information in 444 vs 422 when you are talking about 2 colors. If there is difference in image quality, it is caused by the computer, not by 422 or 444. That said, with your compiter and your TV you might have to use 444 to get the best looking red on blue text (who want's to read that anyway). Find some red text on a blur background on a Blu-ray disc and have a look at THAT in 422 vs 444, won't be a difference... and that video is encoded as 420, even LESS chroma information than 422.

Once again, this is the Display Calibration thread and this entire discussion really isn't relevant to display calibration. It might be an image quality related topic - but it is pointing out peoples' tendency to think if something isolated happens in their system, it must lead to something like 444 ALWAYS having some advantage over 422. Using 420 examples of how great something can look with HUGE color decimation as an example of why it is HIGHLY unlikely that differences you see in 422 vs 444 images from a computer (especially when you go out of your way to find the worst possible scenario) is really the fault of 422 vs 444, when the actual problem is most likely in the computer and how well or poorly it is handling video. Switch to RGB mode... most computers do better in that domain anyway... and in RGB mode there is no color decimation (aka chroma subsampling).

I was talking about RGB mode.
sawfish is online now  
post #17 of 104 Old 12-13-2013, 12:51 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
Arrrrrghhhh....

Did you READ my other posts? 422 and 444 have NOTHING to do with RGB, there's no such thing in RGB mode. RGB and YCbCr 444 carry the same amount of image data. 422 and 444 exist ONLY for YCbCr.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #18 of 104 Old 12-13-2013, 12:58 PM
 
MonarchX's Avatar
 
Join Date: Jul 2012
Posts: 669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Quote:
Originally Posted by praz View Post

Any of these discussion involving a computer as the source is really a waste of time. Most do not understand the interaction and limitation of computer components to output a proper video level signal. And if the signal is properly sent there is always the risk that some future firmware, driver or software update will break it.

Are you serious? Videocards on PCs are more powerful and capable than any video cards in BD players, Roku and similar boxes, or TVs themselves. Are you seriously saying that even gaming consoles cannot output a proper video level signal? Consoles were designed for HDTVs - they use the same HDMI cables as PCs and BD players! Consoles use slightly different versions but of the same videocards from nVidia and AMD. They do an incredible job at outputting a proper limited (16-235) and full range (0-255) video level signals, but they are nowhere near as powerful as most high-end PC videocards. If anything, PC video cards are designed to output at full RGB/ 4:4:4 0-255 range over DVI and either 0-255 or 16-235 over HDMI. Some cards, like nVidia cards, need a registry entry to properly use 0-255 range for HDMI devices.

This is ridiculous, seriously ridiculous - TV manufacturers create PC or similar modes because they are fully aware that 4:2:2 does not look that great for PC users, otherwise they would have not made such modes! It has nothing to do with PCs being PCs, but the fact that text and interface are designed for 4:4:4, which all monitors use. There was even a big thread in this exact section dedicated to chroma subsampling and it provided specific examples with images that showed the difference between what is being seen with 4:2:0, 4:2:2, and 4:4:4. People consistently report issues with 4:2:2 and see them resolved with 4:4:4. This whole "There should be no difference" is pointless because there is. Again, its so damn stupid to rely on these theories when in practice things are different.

Again, unless a user calibrated his/her TV using either ICC profiles or videocard image settings, no new driver will change a calibration performed on the TV itself, using user or service menu. No desktop PC cards from AMD and nVidia, which own 90% of the market share, require any kind of firmware or BIOS updates. I do not know about Intel and other integrated graphics cards (in laptops), which should not be used for HTPCs as its really slow and are a minority in the world graphics adapters/cards. Only a handful of experienced users seeking to increase voltage for overclocking or unlock features on a card will mess with updating their cards' BIOS. Unless such an update is targeting changing or fixing any output problems (chances are slim to none), nothing will change for TV calibration. This has been the case for many years now! I have installed 3 sets of drivers since I calibrated my TV and I even formatted my hard drive with Windows 7 and installed Windows 8.1, which uses drivers differently. Guess what? My TV calibration is still fine! My ArgyllCMS 3DLUTs still produce identical results. Now, if I switch to my integrated Intel HD4000 card, then my RGB WB will be off, but that is fine because it is a different source! If I were to connect a PS3 to my TV, then it would also need adjustments, maybe in different areas. That's is completely normal and that is why there are WB settings saved for each and every source on most TVs. CMS tends to stay the same for all sources, at least on my TV.
MonarchX is offline  
post #19 of 104 Old 12-13-2013, 01:51 PM
Advanced Member
 
xvfx's Avatar
 
Join Date: Feb 2013
Location: UK
Posts: 610
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 64 Post(s)
Liked: 64
I'm curious, why is there no 4:2:0 setting on Blu-Ray players. I did set mine from default 4:4:4 to 4:2:2 but I honestly couldn't tell any difference but left it on 4:2:2 since so many say it's preferred and the closest to 4:2:0.

Especially from years old discussions on internet searches that Doug has discussed.
Quote:
Originally Posted by PlasmaPZ80U View Post

what category do PC video games fall under? I'm guessing it's not consumer video sources (as it's designed for PC monitors first and TVs second)

Yep. Serious gamers use high end displays. Usually. Not TV's and those that usually use TV's on PC are usually racing simulator fans who only care about latency and game physics for the simulators. Who usually don't care about image fidelity. I'm not saying there isn't some that wont care but it's not something they usually talk about if at all. Car cockpit simulators isn't practical on desk displays.

FPS shooters will never use a TV, except console owners. The hardcore will always use computer displays. Most usually again don't care about image accuracy. Only fast responses for their kills.

Plus NVIDIA and ATi/AMD are usually good at breaking things. What was once fixed is now broken or other bugs rectified with new ones introduced. They used to get lots of complaints years ago with people and their HDTV's hooked up to their computer via HDMI for their HTPC setups.

Gamers usually only care about more FPS. Even if it means some sacrificed bugs.
xvfx is offline  
post #20 of 104 Old 12-13-2013, 02:28 PM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by Doug Blackburn View Post

Watch who you are calling a troll.

 

Sure, ok.  You might not be a troll, but you're certainly acting like one here.

 
Quote:
This is the Display Calibration thread. TVs having 444 mode or not really isn't relevant to calibration. You are going to get replies from industry professionals you clearly aren't prepared to deal with.

 

Bull@#$%.  The thread is here because the people who have been exposed to a myriad of TVs that can and cannot handle 4:4:4 would be here.  The thread is here because the people who have best seen the results of chroma subsampling among various TVs are here.  And I've been clear: I just wanted a thread focused on which TVs do and do not support it, including a rough assessment of (half, most, etc.).  I'm not interested in some ill-conceived rationale you have for why 4:4:4 (YCC or otherwise) doesn't matter.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #21 of 104 Old 12-13-2013, 02:29 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,605
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 167
Quote:
Originally Posted by xvfx View Post

I'm curious, why is there no 4:2:0 setting on Blu-Ray players. I did set mine from default 4:4:4 to 4:2:2 but I honestly couldn't tell any difference but left it on 4:2:2 since so many say it's preferred and the closest to 4:2:0.

HDMI spec only allows for 4:4:4 and 4:2:2 even though DVD and ATSC (HD broadcast) was 4:2:0.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #22 of 104 Old 12-13-2013, 02:35 PM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by Doug Blackburn View Post

There's no more color information in 444 vs 422 when you are talking about 2 colors.

 

Chroma subsampling is a spatial subsampling (effectively a resolution reduction) for the chroma information.  If something starts at 4:4:4, depending upon where those two colors lay, you might or might not be able to respresent it in 4:2:0.  You convert a 4:4:4 image of a single red pixel next to a single blue pixel, you can't always get a red and blue.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #23 of 104 Old 12-13-2013, 03:19 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
WHAT IS YOUR POINT?

You babble about 444 and 422 and it simply does not exist in RGB color that you are talking about. RGB data is NOT subject to chroma subsampling because there IS NO CHROMA DATA IN RGB data, you have red, green and blue represented directly. Period.

Chroma data exits only in Cb and Cr or other color formats that represent color data separate from luminance data.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #24 of 104 Old 12-13-2013, 03:44 PM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by Doug Blackburn View Post

WHAT IS YOUR POINT?

You babble about 444 and 422 and it simply does not exist in RGB color that you are talking about. RGB data is NOT subject to chroma subsampling because there IS NO CHROMA DATA IN RGB data, you have red, green and blue represented directly. Period.

Chroma data exits only in Cb and Cr or other color formats that represent color data separate from luminance data.

 

Please take a look at this .png of PC output onto an LG with and without 4:4:4 enabled.  This is output from a PC.

 

 

It is taken from this overview here in the original 4:4:4 thread.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #25 of 104 Old 12-13-2013, 03:53 PM
 
MonarchX's Avatar
 
Join Date: Jul 2012
Posts: 669
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Quote:
Originally Posted by tgm1024 View Post

Please take a look at this .png of PC output onto an LG with and without 4:4:4 enabled.  This is output from a PC.




It is taken from this overview here in the original 4:4:4 thread.

I don't think he cares about what is actually going on, only what is supposed to be going on. If its not like he says it is, then the problem is somewhere else... apparently for just about every TV manufacturer and model... Maybe manufacturers did get it wrong, all of them did, but the fact remains - 4:4:4 text and other things is much cleaner, crisper than the same text with 4:2:2. Period.
MonarchX is offline  
post #26 of 104 Old 12-13-2013, 05:10 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
OMG - this is getting ridiculous beyond all rediculossness.

Is that image in RGB mode or YCbCr?

Answer that question and only that question.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
post #27 of 104 Old 12-13-2013, 05:26 PM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by Doug Blackburn View Post

OMG - this is getting ridiculous beyond all rediculossness.

Is that image in RGB mode or YCbCr?

Answer that question and only that question.

 

When reading the article, it's connected to a PC video card using a DVI->HDMI cable.  I doubt there's any extra processing going on, nor am I quite sure if there is a way to bypass YCC and go to direct RGB through HDMI to non 4:4:4 TVs.  If there were, then there would be no complaints nor images like the one I just showed you.  And there are.

 

Unless of course you're trying to state that such images are always avoidable on all TVs.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #28 of 104 Old 12-13-2013, 05:32 PM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,605
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 167
Quote:
Originally Posted by tgm1024 View Post

When reading the article, it's connected to a PC video card using a DVI->HDMI cable.  I doubt there's any extra processing going on, nor am I quite sure if there is a way to bypass YCC and go to direct RGB through HDMI to non 4:4:4 TVs.  If there were, then there would be no complaints nor images like the one I just showed you.  And there are.

Unless of course you're trying to state that such images are always avoidable on all monitors.

So is the PC outputting YCC or RGB?

A requirement for HDMI is the ability to accept RGB.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #29 of 104 Old 12-13-2013, 05:35 PM - Thread Starter
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 6,083
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 238 Post(s)
Liked: 644
Quote:
Originally Posted by sotti View Post
 
Quote:
Originally Posted by tgm1024 View Post

When reading the article, it's connected to a PC video card using a DVI->HDMI cable.  I doubt there's any extra processing going on, nor am I quite sure if there is a way to bypass YCC and go to direct RGB through HDMI to non 4:4:4 TVs.  If there were, then there would be no complaints nor images like the one I just showed you.  And there are.

Unless of course you're trying to state that such images are always avoidable on all monitors.

So is the PC outputting YCC or RGB?

A requirement for HDMI is the ability to accept RGB.

 

I honestly don't know.  But if the PC were outputing RGB, does this mean that a 4:2:0 only TV is required to output it cleanly (after all it has only RGB, so why convert internally to YCC)?  It sounds as if this is what he is implying, but I was guessing that the moment you connect to a 4:2:0 only TV, then YCC is all you got as a transfer option.


WARNING: You have now entered a no @#$%tard zone. Please gather your anti-vaccine propaganda nonsense and slowly back out the way you came in.
tgm1024 is offline  
post #30 of 104 Old 12-13-2013, 06:02 PM
AVS Special Member
 
Doug Blackburn's Avatar
 
Join Date: May 2008
Location: San Francisco - East Bay area
Posts: 3,453
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 226
If you or the original poster do not know if the image was generated in RGB mode from the PC or YCbCr moder, neither you nor the original poster of the image can blame the bad image quality on 422 vs 444 because in RGB mode there is no 422 or 444... IT DOES NOT EXIST.

Computers will operate in RGB mode all the time unless the computer user happens to have a video system that will output YCbCr... many computer video systems DO NOT have a selection for YCbCr mode so they can ONLY output RGB.

DVI computer interfaces are usually RGB but the video subsystem in the computer MIGHT be able to change the RGB to YCbCr... but there is ZERO guarantee that this is done well.

In the TV, there should not even BE a 422 or 444 selection when the TV is receiving RGB because they are not applicable to RGB video. Some TVs (probably most TVs) do not gray-out the choice while receiving RGB but even if you can MAKE the choice it should not be active for an RGB input because it just doesn't exist. If the TV is receiving RGB from the computer and the image changes when you seleet 422 or 444 the TV is really messed up... PERIOD. And what you are seeing has nothing to do with 422 or 444.. once again, because 422 and 444 do not exist for RGB.

If the computer is in RGB mode it could use digital steps from 0-255 or from 16-235... that's a valid setting for RGB mode... but YCbCr is (our SHOULD BE) 16-235 all the time regardless of settings anywhere else along the line.

The topic of this thread is a waste of time for Blu-ray... you send your TV YCbCr and RGB and you pick the one that looks the best. 5% of the time or so, it will be RGB another 5% of the time. 90% of the time, there's no difference between RGB and YCbCr and that is NORMAL/CORRECT/PROPER/EXPECTED for video displays driven by disc players and other consumer sources. If you are talking about computers, ANYTHING is possible because there are so many ways for things to go wrong at some point, and for things to NOT BE WHAT THE USER THINKS THEY ARE, that you can' make any general statements.... just because 444 looks better doesn't mean the problem is caused by 422 vs 444 - ESPECIALLY if you are sending RGB from the TV for the FACT that they do not exist for RGB. A setting may exist for 422/444 in the computer or TV or both, But if you are in RGB mode, there is no chroma subsampling, period. You'd have to convert the RGB to YCbCr or some other color format where luminance and color were separate from each other, do the chroma subsampling, then convert BACK to RGB and all that processing alone can degrade the image... but it is STILL not the fault of 422 vs 444.

Please tell you understand this because I feel AWFULLY repetitive here, but I feel like the facts keep getting ignored by people who aren't interested in understanding the underlying technology... they just want to label something simple the culprit and be done with it,. even if they are wrong.

"Movies is magic..." Van Dyke Parks
THX Certified Professional Video Calibration
ISF -- HAA -- www.dBtheatrical.com
Widescreen Review -- Home Theater & Sound
Doug Blackburn is offline  
Reply Display Calibration

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off