or Connect
AVS › AVS Forum › Display Devices › Plasma Flat Panel Displays › First VT50 review
New Posts  All Forums:Forum Nav:

First VT50 review - Page 4

post #91 of 284
Quote:
Originally Posted by D-Nice View Post

Another estimate? Didn't someone from HDGuru do an estimate on the 2011 CES floor (eyeballed by putting their hands on the display to block out the exhibit floor's lighting) for the VT30 and stated the blacks were blacker than the VT25 and, again, close/equal to the Kuro (2nd gen)??? How did that pan out per real measurements?

I think it is best people wait for actual measurements done with actual equipment that can properly measure the displays instead of eyeballed measurements. I bet we will see HDTVtest.com's review real soon.

+1, i'm waiting for your accurate measurements and David Mackenzie review of the VT50
post #92 of 284
Quote:
Originally Posted by chikoo View Post

Isn't black simply absence of light? Everything else would be a shade of grey.

The meter must have some transmitted light to read.
post #93 of 284
Quote:
Originally Posted by rogo; View Post

I think we've ruled that out actually. I kind of nudged Kevin into checking a Sharp 735 to see if the cyan bug was a Quattron problem that pre-dates the Elite. He actually did the work to figure out it was. It seems, therefore, that the error was always there and that because it's relatively hard to detect, he simply missed it the first time around.

I was wondering about that. Is there confirmation from other experts that the cyan bug is a Quattron problem?
post #94 of 284
Quote:
Originally Posted by chikoo View Post


Isn't black simply absence of light? Everything else would be a shade of grey.

Technically yes. Perhaps we should stick to the correct terminology which is "measuring the minimum luminance level (MLL)".
post #95 of 284
Quote:
Originally Posted by rogo View Post

I think we've ruled that out actually. I kind of nudged Kevin into checking a Sharp 735 to see if the cyan bug was a Quattron problem that pre-dates the Elite. He actually did the work to figure out it was. It seems, therefore, that the error was always there and that because it's relatively hard to detect, he simply missed it the first time around.

I thought it was a result of that yellow sub-pixel
post #96 of 284
Quote:
Originally Posted by 8mile13 View Post

I was wondering about that. Is there confirmation from other experts that the cyan bug is a Quattron problem?

No one else took up my challenge, except for the estimable Mr. Miller. The point is the "bug" is in the Quattron panel, prior to the Elites. Kevin confirmed that. He reported that back to Sharp. I'm not sure why anyone else needs to confirm it, but if someone with the appropriate equipment and access to a 735 is bored, they can do so at their leisure.

Quote:
Originally Posted by TitusTroy View Post

I thought it was a result of that yellow sub-pixel

It's easy to blame the four-pixel design, but that's sort of selecting on the dependent variable. There's a guy at HDJ who insists you can't possibly get the colors right on a 4-sub-pixel design, but I'm not persuaded. Sharp gets nearly every color right and misses cyan across a range of luminance values only. It sounds like they botched something alogrithmically a while back and it's apparently harder to fix than a bunch of laypeople think it is.

Maybe it's a design flaw. Maybe it can never be fixed without a new Quattron panel. Maybe it can't be fixed with a Quattron panel at all. That's a bunch of speculation that I'd rather not get into. Let's see where Sharp goes from here.
post #97 of 284
Quote:
Originally Posted by rogo View Post
It's easy to blame the four-pixel design, but that's sort of selecting on the dependent variable. There's a guy at HDJ who insists you can't possibly get the colors right on a 4-sub-pixel design, but I'm not persuaded. Sharp gets nearly every color right and misses cyan across a range of luminance values only. It sounds like they botched something alogrithmically a while back and it's apparently harder to fix than a bunch of laypeople think it is.
It's theoretically possible to achieve proper color accuracy with any kind of filter sets which matches some particular criteria.
But remember that you want to meet with some standards at the end and all standard color spaces are established on RGB primaries (exactly as it should be for emissive displays).
Using non-RGB or RGB+any filter sets makes it a bigger challenge to provide an additive RGB color space. The extra channels require more raw processing power and the color mixture is much more difficult (if you want to actively use that sub-pixel).
And note that almost every consumer displays suffer from color issues for some degree, so why make it even more difficult when there is absolutely no real benefit of adding anything else over RGB filters?

Don't forget that this particular sub-pixel filters this yellow from the same backlight and at the end, it should fall exactly to the place where a yellow should be when mixed from red and green.
And it's only yellow, a full CMY set (while still redundant) would look a little bit more reasonable (and serve better in over-saturated "shop mode"). But it's just this yellow...


My best guess is that, for some reason, it was somehow convenient for the engineers to divide the pixels into four sub-pixels instead of three (I guess the sub-pixel structure was easier to create if the sub-pixel is not a ~1/3 but ~1/4 of a square) and once they had the spare space they used it to create a new sub-pixel with an extra color instead of keeping the spare space blocked.
I guess they thought it's a good marketing feature and it will eventually show up as some revolutionary thing instead of a simple engineering decision (which could be absolutely irrelevant and transparent for the end users) about the sub-pixel structure. (And I guess it actually does for many people, so...)


That yellow sub-pixel does not mean cyan can't be correct. That's a different problem. (May be happened because it was more difficult to write color calculation algorithms for the RGBY layout but not because the presence of yellow made it impossible to mix the cyan properly. More difficult - yes, impossible - NO.)



And I can't say this enough times but you can't just take a look at a 2D CIE gamut + a 10p grayscale chart and say "colors are right".
You need to check at least a few points of the color gradients and some mixed colors (not pure primaries or even secondaries)!
Something like this (but this is only an example and a "simple" cal report, not an in-depth display benchmark/analysis...):

 

0k . file
post #98 of 284
Quote:
Originally Posted by janos666 View Post

And I can't say this enough times but you can't just take a look at a 2D CIE gamut + a 10p grayscale chart and say "colors are right".
You need to check at least a few points of the color gradients and some mixed colors (not pure primaries or even secondaries)!
Something like this (but this is only an example and a "simple" cal report, not an in-depth display benchmark/analysis...):

What software produced that report? Seems much more detailed (but less pretty) than a typical calman/chromapure report.
post #99 of 284
Janos, thanks for that excellent explanation.
post #100 of 284
Quote:
Originally Posted by Wizziwig View Post

What software produced that report? Seems much more detailed (but less pretty) than a typical calman/chromapure report.

He's using the open source Argyll CMS package coupled with a display profiling GUI front-end call DispcalGUI. It's a very powerful method for identifying color errors over the entire gamut. As the sharp elite issue has demonstrated, getting the extreme corners of the color space right is no guarantee the rest of it is ok.

I agree with Janos analysis of the yellow sub-pixel. In a 3-d additive system you should be able to use 4 colors as long as they are sufficiently independent. The problem is that the vectors are coupled so finding a solution that minimizes error now depends on how much one color couples with another. I don't think such a system can achieve "reference" (zero error everywhere) theoretically even in the case of just 3 colors so now you add a fourth and it will make the situation worse. I wouldn't be surprised if the global average dE increased with each additional vector you add to such a system. If that is the case then no amount of algorithm magic will fix it.
post #101 of 284
Quote:
Originally Posted by Wizziwig View Post

What software produced that report? Seems much more detailed (but less pretty) than a typical calman/chromapure report.

DispcalGUI with ArgyllCMS (+ my custom HDTV test palette with a slightly modified report style).

The only drawback of this software (if somebody may consider to use it for semi-professional tasks) is the lack of the support for reference grade spectrometers (Konika-Minolta, Jeti, PhotoResearch etc).
But it does let you create and load custom correction matrices for the supported colorimeters, so you can profile a cheap (but still decent) colorimeter (like the i1d3) to a reference instrument.

About the aesthetics, you can easily create some nice 2D CIE gamut and linear or logarithmic tonal response charts from the raw measurement data in let's say MS Excel and attach them to the PDF on a new page...

It's not that difficult to calculate the CIE dE on color gradients or even on random colors. I really can't understand why Spectracal does not offer a "professional reviewer / developer add-on" (for a few hundred(thousand?)$, of course ).
That would make some things easier (like find and understand the so called "color decoding issues" - I still did not find a real decoding error on my Pana G30 - other color problems, sure... not color decoding but color processing... and the
post #102 of 284
Quote:
Originally Posted by janos666 View Post

That would make some things easier (like find and understand the so called "color decoding issues" - I still did not find a real decoding error on my Pana G30 - other color problems, sure... not color decoding but color processing... and the

+1
I don't believe there are any decoding issues with present day sets, it's simple math. Reviewers/calibrators fall back on that statement when they see elevated luminance tracking because they don't understand the limitations of CMS systems that use primaries with correlated spectra. It's a problem because everyone calibrates with 100% saturation patterns. When you desaturate a color you add the other two colors, but the other two colors contain bits of the color you are desaturating so you end up with a too high luminance at lower saturations. 1 + 1 != 2
post #103 of 284
Quote:
Originally Posted by zoyd View Post

I don't believe there are any decoding issues with present day sets, it's simple math.

Then, you really have not measured a lot of "current" displays out there. A really simple pattern to use for checking color decoding problems with be the 709/601 color ramp chart on DVE. If color decoding is correct and the display has a viable RGB only mode (or you have a very good filter like what is provided with the Disney WOW disc) you should not see any problems with that pattern.

Quote:


It's a problem because everyone calibrates with 100% saturation patterns. When you desaturate a color you add the other two colors, but the other two colors contain bits of the color you are desaturating so you end up with a too high luminance at lower saturations. 1 + 1 != 2

Uh, no. If you have ever measured multiple displays at difference stimuli levels, you would see where this is false. All displays do not perform the same way. Have a gander at the Sharp Elite as a reference
post #104 of 284
By current I meant digital displays, it's amazing if manufacturers are still getting the 709/601 issue wrong.

The issue we were discussing with luminance and/or xy tracking I was postulating is a color processing issue based on correlated primaries, with the elite being a prime example of said problem.
post #105 of 284
Quote:
Originally Posted by rogo View Post

It's easy to blame the four-pixel design, but that's sort of selecting on the dependent variable. There's a guy at HDJ who insists you can't possibly get the colors right on a 4-sub-pixel design, but I'm not persuaded. Sharp gets nearly every color right and misses cyan across a range of luminance values only. It sounds like they botched something alogrithmically a while back and it's apparently harder to fix than a bunch of laypeople think it is.

Maybe it's a design flaw. Maybe it can never be fixed without a new Quattron panel. Maybe it can't be fixed with a Quattron panel at all. That's a bunch of speculation that I'd rather not get into. Let's see where Sharp goes from here.

will be interesting to see if it gets fixed on the new Sharp Elite sets
post #106 of 284
Quote:
Originally Posted by zoyd View Post

When you desaturate a color you add the other two colors, but the other two colors contain bits of the color you are desaturating so you end up with a too high luminance at lower saturations. 1 + 1 != 2

Yes, but they seem to do the 3D color corrections in the YCC space which is Luma+Chroma (why else would these displays blur the chroma even from true RGB 4:4:4 inputs if not because they convert everything back to YCC 4:2:2 ; and why else would they do that if not for color processing in a luma+chroma space which is theoretically more convenient than doing it in RGB space...? - One convenience for sure that they can drop half of the chroma pixels, and it's also convenient for the motion interpolation techniques...), so this should not be a problem. (But I guess they manage to make it problematic, somehow... -> I guess it's correlated with the gamma corrections. More precisely with the gamma "issues"...)



Quote:
Originally Posted by D-Nice View Post

Then, you really have not measured a lot of "current" displays out there. A really simple pattern to use for checking color decoding problems with be the 709/601 color ramp chart on DVE. If color decoding is correct and the display has a viable RGB only mode (or you have a very good filter like what is provided with the Disney WOW disc) you should not see any problems with that pattern.

This won't tell you if you face a color decoding or a color processing problem! Only that you have some problem with the colors.


My opinion is that these displays have no problem with decoding the colors (either we feed them with YCC or RGB input).
They struggle a bit with the color processing, when they try to do gamut mapping between native device space and standard color spaces (the over-saturated uncorrected "shop mode" colors and Rec709...).


I did measure some displays (check the above PDF how) and this is what I make out of the data.


But I don't think we necessarily talk about completely different things. I guess it's just a misunderstanding over the terminology (color decoding, color processing, gamut mapping, etc - proper or incorrect words for the same thing and different people prefer different terms...).

------------

And please, let's forget those filters!

If somebody could create a simple RGB filter set which is compatible with the wide range of display spectra you can find out there then we would not need to create correction matrices for the cheap colorimeters like the i1d3.

Which plastic RGB filter set will beat the filters of an i1d3 colorimeter and when will an i1d3 provide accurate measurements without profiling it to a spectrophotometer?
Or to put it to other words: How do you profile your slip of paper and plastic to your spectro?


And the other problem with the built-in RGB mode...

Who says it completely shuts down the other two sub-pixels for sure, instead of emulating this separation at software level (probably in the YCC space)?
- If it does forcefully shut down those sub-pixels then it destroys some of the 3D color corrections made in the YCC stage (which -ideally- also involves luminance corrections to get the luminance right after the chroma correction).
- If it emulates this behavior in the YCC space then it's also buggy like the main color processing in the YCC space.

----------

Measuring more than one (preferably 8+) points on every color gradients and also some random mixed colors (not pure primaries or secondaries) with an accurate(-enough) meter is the way to tell what happens with the colors.

And even then, you only see the symptoms and you create ideas about the problem.
You know nothing for sure because you don't have access to the full source code of the TV's software to see what happens (or should happen) inside the black box which creates the false output.


And I did measure various Samsung and Panasonic PDPs this way.
More over, I put them side-by-side with a real reference monitor (which has it's own built-in color sensor for true 3DLUT based hardware calibration).
And I played a lot with the settings (both in the normal user and the hidden engineering menus).

And I have some idea about how different 3D color spaces work (CIE XYZ ; CIE Lab ; Rec709 ; sRGB)
Rec709 decoding is a simple matrix multiplication. The true, proper gamut mapping (like the emulation of the Rec709 space in an over-saturated native device gamut) is an art when compared to that!


And I don't dare to say I fully understand everything about these coor problems but I guess I have an idea.
post #107 of 284
Quote:
Originally Posted by zoyd View Post

By current I meant digital displays, it's amazing if manufacturers are still getting the 709/601 issue wrong.

I understood that. They are still getting it wrong on digital displays! Amazing, isn't it

Quote:


The issue we were discussing with luminance and/or xy tracking I was postulating is a color processing issue based on correlated primaries, with the elite being a prime example of said problem.

Understood, but there are other LED LCDs out there that are just as bad and do not use a RGBY pixel structure.
post #108 of 284
Quote:
Originally Posted by janos666 View Post

This won't tell you if you face a color decoding or a color processing problem! Only that you have some problem with the colors.

Why are you separating decoding from processng as if decoding is completely different than processing. They are one and the same as decoding is taking the input signal encoding it into a display's output medium which is RGB. The encoding is the processing of color information.

Quote:


My opinion is that these displays have no problem with decoding the colors (either we feed them with YCC or RGB input).

Based on what? 2011 Panasonic displays, NA versions, cannot properly decode the inpout signal and map it to their RGB output. Are you not basing your comments on EU models? EU and NA models are not the same.

Quote:


They struggle a bit with the color processing, when they try to do gamut mapping between native device space and standard color spaces (the over-saturated uncorrected "shop mode" colors and Rec709...).

I would not define wide color gamut as "native device space". Native means it is indigenous which I'm sure you do not actually believe.


Quote:


I did measure some displays (check the above PDF how) and this is what I make out of the data.

Your PDF is of only 1 Samsung. I don't know, nor care about EU models. Only NA models. Do you have data on NA models?


Quote:


But I don't think we necessarily talk about completely different things. I guess it's just a misunderstanding over the terminology (color decoding, color processing, gamut mapping, etc - proper or incorrect words for the same thing and different people prefer different terms...).

I agree, however, I am aware of the terminolgies.

Quote:


And please, let's forget those filters!

If somebody could create a simple RGB filter set which is compatible with the wide range of display spectra you can find out there then we would not need to create correction matrices for the cheap colorimeters like the i1d3.

Which plastic RGB filter set will beat the filters of an i1d3 colorimeter and when will an i1d3 provide accurate measurements without profiling it to a spectrophotometer?
Or to put it to other words: How do you profile your slip of paper and plastic to your spectro?

My referencing "filters" was not what you have translated that point into. I was simply referencing a medium of which individuals who do not have access to 5nm and greater spectrometers can follow along in this conversation.


Quote:


And the other problem with the built-in RGB mode...

Who says it completely shuts down the other two sub-pixels for sure, instead of emulating this separation at software level (probably in the YCC space)?
- If it does forcefully shut down those sub-pixels then it destroys some of the 3D color corrections made in the YCC stage (which also involves luminance corrections to get the luminance right after the chroma correction).
- If it emulates this behavior in the YCC space then it's also buggy like the main color processing in the YCC space.

I take it you missed the word "viable" in my post. Viable means it actually works causing the items you pointed out to be irrelevant.

Quote:


And I don't dare to say I fully understand everything about these coor problems but I guess I have an idea.

Its facinating that you imply that I would say such. Here is one right back at you....

Don't YOU ever say I don't know and/or understand these items.
post #109 of 284
Quote:
Originally Posted by D-Nice View Post


Your PDF is of only 1 Samsung. I don't know, nor care about EU models. Only NA models. Do you have data on NA models?

Here are a couple from a NA D8000.

1. Calibrated to 100% saturation/100% luminance standard windows. link

2. Calibrated to 100% saturation/55% luminance custom APL w/ 1% windows and then tweaked some more (red/magenta luminance adjustments) to get final result. link

note the charts have the wrong instrument specified, it was an i1pro not a display.
post #110 of 284
Thanks. I don't have access to dropbox on this computer. I will review the pdfs later today when I get home.

Quote:
Originally Posted by zoyd View Post

Here are a couple from a NA D8000.

1. Calibrated to 100% saturation/100% luminance standard windows. link

2. Calibrated to 100% saturation/55% luminance custom APL w/ 1% windows and then tweaked some more (red/magenta luminance adjustments) to get final result. link

note the charts have the wrong instrument specified, it was an i1pro not a display.
post #111 of 284
BTW, I am with you guys regarding calibrators, and especially magazine reviewers, needing to be doing more measurements that what is in the current, canned, workflows in calibration software suites.
post #112 of 284
Quote:
Originally Posted by D-Nice View Post

Then, you really have not measured a lot of "current" displays out there. A really simple pattern to use for checking color decoding problems with be the 709/601 color ramp chart on DVE. If color decoding is correct and the display has a viable RGB only mode (or you have a very good filter like what is provided with the Disney WOW disc) you should not see any problems with that pattern.

Uh, no. If you have ever measured multiple displays at difference stimuli levels, you would see where this is false. All displays do not perform the same way. Have a gander at the Sharp Elite as a reference

You know, after reading all this techincal stuff I have two conclusions:

1) I'm really ignorant about how all this stuff works. I guess I thought this was a much more benign and well-sorted technology than it really is.

2) After living with my 65GT30 for about a month now I will likely never spend more than $2K for a TV (approx what I spent on the GT30) if the supposed "reference" sets like the old and new Elites have their own set of problems. I get the whole "no TV is perfect" thing but I would have expected the manufacturers to be pretty close to perfection now at that (what, about $7K) price point.

Sometimes ignorance is bliss, and honestly at this point the less I know beyond what I already do about my new favorite TV the better. I'm just gonna sit back, pop open a cold one, and stare at the purty moving pictures.
post #113 of 284
Quote:
Originally Posted by D-Nice View Post

Based on what? 2011 Panasonic displays, NA versions, cannot properly decode the inpout signal and map it to their RGB output. Are you not basing your comments on EU models? EU and NA models are not the same.

Yes, I work with EU models. But I saw some data (not too much though, far from enough to make sure) from US models and they seem to be similar (at least with Samsung where I saw more data from the US models).

Quote:
Originally Posted by D-Nice View Post

I would not define wide color gamut as "native device space". Native means it is indigenous which I'm sure you do not actually believe.

In my boook,
Native device color space is the color space defined by the real chromaticity of the true device channels (the red chromaticity which you can see and measure when only the red sub-pixel emits light, and so on with G,B and possibly Y sub-pixels...).
I am sure it's a valid term (at least in the world of monitors if not HDTVs).


Quote:
Originally Posted by D-Nice View Post

Your PDF is of only 1 Samsung. I don't know, nor care about EU models. Only NA models. Do you have data on NA models?

No. But I would like to see that data as I mentioned this earlier. (Don't worry, not necessarily from you. Don't read this like that. It's an open forum...)

Quote:
Originally Posted by D-Nice View Post

My referencing "filters" was not what you have translated that point into. I was simply referencing a medium of which individuals who do not have access to 5nm and greater spectrometers can follow along in this conversation.

Ok. I still don't like those but yes, I agree it's a good example for them to get an idea.

Quote:
Originally Posted by D-Nice View Post

I take it you missed the word "viable" in my post. Viable means it actually works causing the items you pointed out to be irrelevant.

Why are those irrelevant?
I tried to express why I think it's very complicated, if not impossible to offer "viable" R/G/B-only mode.
But OK, example for the end-users again. I take it...

Quote:
Originally Posted by D-Nice View Post

Don't YOU ever say I don't know and/or understand these items.

Sorry if it goes through that way.

I simply don't get it why you (not only you in person but many people) talk about color decoding issues when it seems like a processing issue to me.
(And it's hard to believe it's a strict decoding issue which is a simple matrix calculation, unlike other complex processing steps).

I wish to know why exactly you (again, not you in person) call it a decoding issue.
What's the proof this happens during the decoding and not later in the chain...?

But I think I start to get an idea about that after reading your above post to zoyd. Thanks.
post #114 of 284
Quote:
Originally Posted by janos666 View Post

In my boook,
Native device color space is the color space defined by the real chromaticity of the true device channels (the red chromaticity which you can see and measure when only the red sub-pixel emits light, and so on with G,B and possibly Y sub-pixels...).
I am sure it's a valid term (at least in the world of monitors if not HDTVs).

Right, but are you, or I, to say what the native color space is for... say the Samsung D8000 without having access to every piece of code that drives that display?

Quote:


Why are those irrelevant?
I tried to express why I think it's very complicated, if not impossible to offer "viable" R/G/B-only mode.
But OK, example for the end-users again. I take it...

I understand your point, but if there were a viable RGB only mode on a display, would not your comments be irrelevant? That was the point I was attempting to make.

Quote:


Sorry if it goes through that way.

Yeah, I think I did take your comment the wrong way. It seems we are on the same page, just bad translations

Quote:


I simply don't get it why you (not only you in person but many people) talk about color decoding issues when it seems like a processing issue to me.
(And it's hard to believe it's a strict decoding issue which is a simple matrix calculation, unlike other complex processing steps).

I wish to know why exactly you (again, not you in person) call it a decoding issue.
What's the proof this happens during the decoding and not later in the chain...?

Is there proof it is not actually happening at the decoding level? There is proof that digital displays are not properly decoding 709 and 601. Not trying to be combative here. I just want to understand your position.
post #115 of 284
Quote:
Originally Posted by C63 View Post


2) After living with my 65GT30 for about a month now I will likely never spend more than $2K for a TV (approx what I spent on the GT30) if the supposed "reference" sets like the old and new Elites have their own set of problems. I get the whole "no TV is perfect" thing but I would have expected the manufacturers to be pretty close to perfection now at that (what, about $7K) price point.

I actually think you hit on a critical point here. The value in buying up is vanishingly small, not just for you but for most people. When the extra dollars are spent on a car, people get something that impresses their neighbors, even if they never get to really use all that extra engine power in an AMG or an M5. And those "investments" are mostly "foolish" financially, which is why top-end performance autos like those are bought by very, very few people who are most concerned with standing out and spending money than much else.

The TV business has failed to differentiate "up" in a way that delivers enough consistent value to ordinary people. And, worse, when it tries to deliver it to videophiles, it fails with too many bugs, problem sets, etc.

Part of the reason the business is so dreadful is that it can't sell enough higher margin product. It's only part of the problem, but it's a real part nevertheless.
post #116 of 284
Quote:
Originally Posted by rogo View Post

I actually think you hit on a critical point here. The value in buying up is vanishingly small, not just for you but for most people. When the extra dollars are spent on a car, people get something that impresses their neighbors, even if they never get to really use all that extra engine power in an AMG or an M5. And those "investments" are mostly "foolish" financially, which is why top-end performance autos like those are bought by very, very few people who are most concerned with standing out and spending money than much else.

The TV business has failed to differentiate "up" in a way that delivers enough consistent value to ordinary people. And, worse, when it tries to deliver it to videophiles, it fails with too many bugs, problem sets, etc.

Part of the reason the business is so dreadful is that it can't sell enough higher margin product. It's only part of the problem, but it's a real part nevertheless.

It's particularly funny that you use the AMG/M5 analogy since as you might guess from my forum name I drive a C63 AMG. I'd like to believe that I don't fall into the profile you mentioned ( I generally don't give a crap what people think about me and I just happen to really love cars), but who knows. I will say that I think the average person could more easily discern the difference between a car with 450hp and one with 300hp than they could between ST/GT/VT ranges, but I guess it all comes down to priorities, personal taste, etc. But for sure the point of diminishing returns/value is well at play in the luxury car/audio/video/watch markets that men love to spend money on.
post #117 of 284
Quote:
Originally Posted by D-Nice View Post

Is there proof it is not actually happening at the decoding level? There is proof that digital displays are not properly decoding 709 and 601. Not trying to be combative here. I just want to understand your position.

I don't have solid proof. But here is an example:

If I measure the primary color gradients at 8 points on a Samsung D550 (EU) when it's set to Native color space settings (Of course, this is just a "preset" like the others. I do know it has some non-zero/non-peak values stored in the main-borad software. I can even modify them if I wish to build a new custom color space for myself and it will be still labelled as Native...) and I draw a chart from the chromaticity then I get a straight constant line.
If I measure the same eight points with the Auto color space settings then I get a seemingly-exponential curve (It was hard to believe when I saw this at the first time but it's repeatable and makes sense if I examine movie contents with over-saturated dark and under-saturated bright skin tones after a calibration with 75% stimulus patterns).


This let me think that the problem is the CMS code which incorrectly mixes the emulated colors from the native primaries.
And the same thing happens if I go to the service menu and I unlock the Custom color space (just like when I fire up a serial console to re-adjust either the Auto or the Native presets from the top debug menu).



If the gamut mapping is correct then I think I should see constant chromaticity across the points of the color gradients, no matter what color space settings I use (the actual values may change but remain equal between the stimulus levels - but it's not the case...).
If it's only the color decoding to blame then I guess I should see similarly incorrect behaviors with all color space settings (including a factory default Native preset).
But what I see is that the Native color space looks consistent (the chromaticity is constant across the whole gradient) and the emulated gamuts (no matter if it's a factory preset or a custom calibration) are obviously wrong.


But, of course, this does not necessarily mean there is no problem on the decoding side (may be it's just hard to see when masked by other factors - for example, if the Native is set to be very saturated than it clips a curve to a straight line as you can't get further than your native primaries. So the decoding step can theoretically produce a curve which would exceed the native gamut and the CMS can clip this back to the borders, leaving you with a constant chromaticity. This is -while seems less probable to me- looks theoretically possible.).




You are right. We can never be sure if we see the true primaries.
Especially not with the Panasonics. On their 2011 TVs, the primaries also move if I make a huge adjustment with the white point controls (for example, R,G,B-DRV=FF,FF,FF gives me a very narrow gamut, even though this should produce a state closer to the "native" panel characteristics).


By the way, I usually push the color controls to their upper limit in the user menu or use the built-in test pattern generators in the service menus when I profile my colorimeter.
I measure R,G,B only and I keep the white for the validation of the colorimeter correction matrix.

I don't say this gives you the highest possible saturation but I think it should because the marketing guys could always kill for even more colorful displays, so a "shop mode" (which is not magic, just another preset...) should be able to achieve full saturation. (Or the poor software engineer may never sleep well at night but dream with mad marketing guys. )





About the original Panasonic color issues...

Does that one happen with every panel brightness settings?
Our EU models do not have that control and I guess the ~25fl peak brightness equals with the peak brightness of the Low setting in the US.
And I see no obvious problem with the colors.

In that case (I am only assuming it's linked with the Panel Brightness) how the Panel Brightness setting may affect the color decoding?
Shouldn't that control make it's work well after the decoding step (at the end of the chain, as it's a panel driver setting)?
post #118 of 284
After a lot of off-topic, I have an question about the VT50

Did anybody ever check if the VT/GT50s really preserve the full chroma map resolution (not only in 1080p24 and/or True Cinema but with 1080p60 and ISF/Prof modes)?

I know it's not really important or even useful for most of the people but I would appreciate this feature.
post #119 of 284
Quote:
Originally Posted by C63 View Post

It's particularly funny that you use the AMG/M5 analogy since as you might guess from my forum name I drive a C63 AMG. I'd like to believe that I don't fall into the profile you mentioned ( I generally don't give a crap what people think about me and I just happen to really love cars), but who knows. I will say that I think the average person could more easily discern the difference between a car with 450hp and one with 300hp than they could between ST/GT/VT ranges, but I guess it all comes down to priorities, personal taste, etc. But for sure the point of diminishing returns/value is well at play in the luxury car/audio/video/watch markets that men love to spend money on.

OK, I use the AMG/M5 analogies here often, and the way I did it this time, I regret.

I don't mean to imply that buyers of said cars are ridiculous fools in any way. In fact, I live in a very affluent area where a lot of those vehicles sell and there's a C63 in my neighbor's driveway.

I actually agree with you that most people will be able to tell the difference between the vehicles (and more so than the TVs). There is something entirely rational about buying the best, highest performance car money can buy and also something very irrational about spending the extra 10s of thousands to get that incremental performance. That said, I'm a huge fan of performance vehicles because they have an amazing amount of trickle down on the economy (they get detailed more, they get serviced more, etc.)

High end TVs deliver a similar experience to high-end cars. Unfortunately, they deliver it for a somewhat similarly size audience.

As an AMG owner, you're in an elite -- and small group. The VT series, the Samsung ES series, et al. are supposed to be in an elite -- and much larger -- group. But they don't tend to deliver enough to generate those volumes even though they deliver "AMG performance". And part of the problem is they do so with 1980s Jaguar quirks and reliability far too often.
post #120 of 284
Quote:
Originally Posted by HearingImpaired View Post

I still say next year is the year the Panasonic VT series will overcome the Kuro as the best tv ever.

Next year?? Early reviews are saying this year.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Plasma Flat Panel Displays
AVS › AVS Forum › Display Devices › Plasma Flat Panel Displays › First VT50 review