or Connect
AVS › AVS Forum › Display Devices › Display Calibration › Do digital displays devices have a "native" color space?....
New Posts  All Forums:Forum Nav:

Do digital displays devices have a "native" color space?....

post #1 of 13
Thread Starter 
Hi,

I'm a bit confused as I consider the way modern CMS systems function, so I would appreciate any insight.

I used to use CRT front projectors which have a native color space. Newer CRT FP's corrected the color space with lens filters or colored coolant fluid and corrective lenses.

There were also "Color" and "Tint" controls which would adjust the electrical signal to correct the color decoding problems.

Now,... I use a SXRD projector and it has a built in wonky CMS. Also,... it's default settings are off the charts calibration wise with Blue in particular over stated.

Anyways,... I know you can leave this projector at (insane) defaults and use an external video processor and you will get perfect graphs and a calibrated image.

However,... the internal settings of the pj are off to an amazing amount....

I believe such a solution could be resulting in clipping or compressing of the incoming video signal.

It seems to me there is a balance or competition between the native response of the display and calibration.... but,... admittedly,... i'm still a bit confused.

Do display devices have a native color space?... I imagine that they all do. And does adjustment of Gamut lead to compression or degradation of the video?... seems calibration should take precedence over any other notion unless it leads to observable noise.

=Brian
Edited by Brian Hampton - 4/25/13 at 1:42pm
post #2 of 13
Quote:
Originally Posted by Brian Hampton View Post

Hi,

I'm a bit confused as I consider the way modern CMS systems function, so I would appreciate any insight.

I used to use CRT front projectors which have a native color space. Newer CRT FP's corrected the color space with lens filters or colored coolant fluid and corrective lenses.

There were also "Color" and "Tint" controls which would adjust the electrical signal to correct the color temperature problems.

Now,... I use a SXRD projector and it has a built in wonky CMS. Also,... it's default settings are off the charts calibration wise with Blue in particular over stated.

Anyways,... I know you can leave this projector at (insane) defaults and use an external video processor and you will get perfect graphs and a calibrated image.

However,... the internal settings of the pj are off to an amazing amount....

I believe such a solution could be resulting in clipping or compressing of the incoming video signal.

It seems to me there is a balance or competition between the native response of the display and calibration.... but,... admittedly,... i'm still a bit confused.

Do display devices have a native color space?... I imagine that they all do. And does adjustment of Gamut lead to compression or degradation of the video?... seems calibration should take precedence over any other notion unless it leads to observable noise.

=Brian

well, I know that the 'Native' color space setting on Samsung TVs shows the native gamut of the display, as should 'Custom' at its defaults... 'Auto' is an example of a preset where video processing is applied to tame the native gamut of the display and adjusting the CMS in 'Custom' has a similar effect to using 'Auto'
Edited by PlasmaPZ80U - 4/25/13 at 9:39am
post #3 of 13
^ I think the physical primaries of the display determine the native gamut (and it has do with the display tech too, which I believe matters more than whether the display is digital or analog)

for example, for LCDs the color filters used in the LCD panel plus the backlight source determine the native primaries/gamut

the number of bits used in the video processing part of the TV (the 3DLUT or whatever it's called) will determine how large of a correction you can apply to a CMS without introducing artifacts to the picture/video... the more bits the better... I think Doug Blackburn mentioned Samsung TVs use a 18 bit data processing path and that allows you to make large corrections without side effects.
Edited by PlasmaPZ80U - 4/25/13 at 9:51am
post #4 of 13
Thread Starter 
Been thinking about this and I guess I see it this way.

Chances are my projector does have a Native color gamut. And,... it's not necessarily either of the built in presets.

It certainly has a native gamma response but that doesn't really matter since it's possible to correct the gamma response using the built in utility.

Similarly,.. if there's a way to improve the color decoding response then that's all that matters.

In my case the pj has only a 2D CMS and I can nail xy but the Delta E's don't get any better.... So,... I tended to focus on Grey and Gamma and just go with a preset color decoding option.

Maybe,... I can lower the Delta E of some of the gamut and if I can't ... it's pretty good looking anyway.

Certainly my next projector will have better built in CMS.

-Brian
post #5 of 13
Quote:
Originally Posted by Brian Hampton View Post

Been thinking about this and I guess I see it this way.

Chances are my projector does have a Native color gamut. And,... it's not necessarily either of the built in presets.

It certainly has a native gamma response but that doesn't really matter since it's possible to correct the gamma response using the built in utility.

Similarly,.. if there's a way to improve the color decoding response then that's all that matters.

In my case the pj has only a 2D CMS and I can nail xy but the Delta E's don't get any better.... So,... I tended to focus on Grey and Gamma and just go with a preset color decoding option.

Maybe,... I can lower the Delta E of some of the gamut and if I can't ... it's pretty good looking anyway.

Certainly my next projector will have better built in CMS.

-Brian

as long as the changes you make to the display's picture controls don't add any unwanted side effects like banding/contouring, you should make whatever changes result in the most accurate picture... just focus on errors you can easily see over those that are purely technical (below the threshold of visibility)

I believe for gamut, big Y is more important than xy.
post #6 of 13
Thread Starter 
Thanks.
post #7 of 13
Every video display in the world has a "native" color space. It would be impossible to create a video display that does NOT have a "native" color space. You can't have a car without an onboard means of propulsion of some kind... well, you could, but it wouldn't be worth anything. A video display without images wouldn't be worth anything either. Some TVs even have a color space menu selection and sometimes there's a "Native" choice in the list.

But there are BAD native color spaces and GOOD native color spaces. A BAD native color space would have one or more primary or complimentary colors undersaturated... if that happens, you can't fix it with TV controls or video processor controls. Once a color is at 100% saturation with the display's controls, and the color is still undersaturated, there are no controls or processor that can make the saturation 110% or 150% just so the right color is achieved. Likewise, if you have made one of the primary or complementary colors as bright as it can be made with the TV controls and it is still not bright enough, there's no way to correct that in the TVor with an external video processor.

Rec 601 is a document that described standard definition color space.

Rec 709 is a document that describes high definition color space.

You have to use the right color space for the content you are viewing or you end up with errors, probably visible errors.

If the NATIVE color space is larger than the Rec 709 color space, you should be able to make all the primary and complimentary colors (red, green, blue, cyan, magenta, yellow) accurate using the CMS controls in the TV (if they work well). If there are no CMS controls in the TV, you can make the primaries and complementaries accurate with an external video processor. But if the NATIVE color space is smaller (anywhere) than the Rec 709 color space, and the controls in the TV can't correct whatever color (or colors) is placed "inside" the Rec 709 triangle (undersaturated), there is NO way to fix it.

Fixing color problems typically WILL NOT create artifacts because video processing is done in 10-bit color (30 bits total) at least with Samsung displays having 18-bits. The displayed images are only 8-bits, so when you work in 10-bit (or larger) space when processing the images, there's a lot more "room" to do things and when you are done, you convert the 10-bit (or larger) video data back to 8-bits. That said, there are video displays that are MESSED UP internally. Some Toshiba displays I've reviewed, for example, have CMS controls with a range of +/- 10 and if you move any of the controls beyond +/- 3, you get a horrendous amount of posterization and block artifacts in the images... it's really obvious and really bad. A couple of years before that, their TVs' CMS controls had adjustment ranges of +/- 30 and you could not exceed +/- 10 without the same problems. But that's the ONLY time I've run into TV adjustments that cause image quality to tank in any obvious way.

All the SXRD displays I can remember seeing (all XBR series models) had Rec 601 and Rec 709 color space settings by resolution so you could set 480i & 480p to Rec 601 and set 720p, 1080i, and 1080p to Rec 709.

Color space is only the starting point for video performance though... there are many other image quality deciding/determining factors.
post #8 of 13
Thread Starter 
I use a SXRD front projector (vpl vw60) and have mainly done grey scale and gamma calibration,...ignoring gamut.

Sony's Real Color Processing is a 2D CMS and I will experiment with it further (in time) to see if its useful or useless.

When I used it previously, I can get xy nearly correct (+.002) but the delta E is no better so I assume it takes Y in the wrong directon.

I just need to find the time when I can tinker and see if its viable or not.
post #9 of 13
Quote:
Originally Posted by Brian Hampton View Post

I use a SXRD front projector (vpl vw60) and have mainly done grey scale and gamma calibration,...ignoring gamut.

Sony's Real Color Processing is a 2D CMS and I will experiment with it further (in time) to see if its useful or useless.

When I used it previously, I can get xy nearly correct (+.002) but the delta E is no better so I assume it takes Y in the wrong directon.

I just need to find the time when I can tinker and see if its viable or not.

are you using colorhcfr? because the deltaE formula used for gamut in that software is not dE1994 or dE2000, which are the ones you want
post #10 of 13
Thread Starter 
Yep, using HCFR.

Want to move into a new software though.
post #11 of 13
Quote:
Originally Posted by PlasmaPZ80U View Post

are you using colorhcfr? because the deltaE formula used for gamut in that software is not dE1994 or dE2000, which are the ones you want
Zoyd has changed the error calculating formula to dE94 for color gamut in his builds.

@Brian, use this build of HCFR.
post #12 of 13
Quote:
Originally Posted by Brian Hampton View Post

Yep, using HCFR.

Want to move into a new software though.

are you using the new version that has dE94?
Quote:
Originally Posted by rahzel View Post

Zoyd has changed the error calculating formula to dE94 for color gamut in his builds.

@Brian, use this build of HCFR.

is this a very recent update?

EDIT: I see it was added 2/8/2013
post #13 of 13
Thread Starter 
Hi,

I just downloaded the new HCFR but haven't checked it out yet.

Seems like a good time to check it out.

Thanks again,

Brian
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › Do digital displays devices have a "native" color space?....