or Connect
AVS › AVS Forum › Display Devices › Display Calibration › Basic Guide to Color Calibration using a CMS (updated and enhanced)
New Posts  All Forums:Forum Nav:

Basic Guide to Color Calibration using a CMS (updated and enhanced) - Page 62

post #1831 of 1936
Quote:
Originally Posted by damag0r View Post

I just finished my first 10 point grayscale and primary color calibration with pretty good results. (TC-P50ST60) But I have a possibly dumb question (See the very bottom of this post).


I wish my colour space was as good as this. The best my screen seems to go is 3.55 since red, blue and cyan cannot meet the 100% targets.
post #1832 of 1936
I just did my first calibration on my samsung 55 es8000 ts01 panel (movie mode). i used a colormunki display, avshd 709, and disney wow discs. i would appreciate all insights or any comments regarding my measurements. thank you.

GreyscaleandPrimary-Secondary.jpg 90k .jpg file
LuminanceGraph.jpg 98k .jpg file
GammaGraph.jpg 139k .jpg file
RGBLevelsGraph.jpg 126k .jpg file
ColorTempreatureGraph.jpg 109k .jpg file
CIEDiagramGrays.jpg 140k .jpg file
CIEDiagramSaturations.jpg 145k .jpg file
post #1833 of 1936
if you guys don't want to download them, here they are:
Greyscale and Primary/Secondary Colors

Luminance

Gamma

RGB Levels

Color Temperature

CIE Diagram Grays

CIE Diagram Saturations


EDIT: changed greyscale and primary secondary colors to show xyY
Edited by prme19 - 11/21/13 at 6:58pm
post #1834 of 1936
I don't think anyone can get 0 greyscale right bang in the centre, right?
post #1835 of 1936
Quote:
Originally Posted by xvfx View Post

I don't think anyone can get 0 greyscale right bang in the centre, right?

i've tried several times but i couldn't bring 0 IRE to fall in the center with the others. so, it must be that my colorimeter is limited at that end or it's just plain too dark for it to measure accurately.
post #1836 of 1936
That's right, it's too dark to measure chromaticity reliably, don't worry about it. Nice calibration, I would recommend trying to hit the BT.1886 curve rather than where you have it just under 2.3 (if you have a 10 pt control) You are probably losing some shadow detail where it is now.
post #1837 of 1936
Quote:
Originally Posted by zoyd View Post

That's right, it's too dark to measure chromaticity reliably, don't worry about it. Nice calibration, I would recommend trying to hit the BT.1886 curve rather than where you have it just under 2.3 (if you have a 10 pt control) You are probably losing some shadow detail where it is now.

thanks for your insight zoyd.

pardon my ignorance, but what's BT.1886 curve? is it the gamma at 2.2? and yes, i do have 10 pt control. also, can i use gamma control option to correct this?
post #1838 of 1936
BT.1886 gamma is not a fixed display gamma, it is dynamic and varies depending on the video level / stimulus. It is calculated based on your black and white measurements. Generally it's higher at the top-end of the grayscale then gradually slopes down to black.
Edited by rahzel - 11/21/13 at 7:22pm
post #1839 of 1936
Quote:
pardon my ignorance, but what's BT.1886 curve? is it the gamma at 2.2? and yes, i do have 10 pt control. also, can i use gamma control option to correct this?

It's that light dotted line in your gamma graph. If you use your 10 pt controls to match Y = gamma Y on the measurements page you'll hit that line
post #1840 of 1936
Quote:
Originally Posted by prme19 View Post

i've tried several times but i couldn't bring 0 IRE to fall in the center with the others. so, it must be that my colorimeter is limited at that end or it's just plain too dark for it to measure accurately.

Even if it could be done how could it be rectified… 30% greyscale would become too red?
post #1841 of 1936
Quote:
Originally Posted by zoyd View Post

It's that light dotted line in your gamma graph. If you use your 10 pt controls to match Y = gamma Y on the measurements page you'll hit that line

Ok, i'll try to do that.. Match Y with gamma Y.. Thanks zoyd
post #1842 of 1936
Quote:
Originally Posted by xvfx View Post

Even if it could be done how could it be rectified… 30% greyscale would become too red?

I've tried to make RGB levels flat from 10 IRE to 70 IRE by using 10 pt white balance, but it kept changing the levels for the next IRE (blue was all over the place for me) and even came to a point it wont change anymore despite adjusting the 10 pt all the way.. I've also tried going backwards, but that didnt make sense since i'll be changing my 100 IRE reference white.. So, i just tried to make it all even from 10 to 30 IRE instead
post #1843 of 1936
What were the CRT TV's like in the 0 - 30% greyscale. I honestly have no memory of it now. Probably because it was something I never looked for.
post #1844 of 1936
Quote:
Originally Posted by prme19 View Post

I've tried to make RGB levels flat from 10 IRE to 70 IRE by using 10 pt white balance, but it kept changing the levels for the next IRE (blue was all over the place for me) and even came to a point it wont change anymore despite adjusting the 10 pt all the way.. I've also tried going backwards, but that didnt make sense since i'll be changing my 100 IRE reference white.. So, i just tried to make it all even from 10 to 30 IRE instead

This is an indication that the controls are not lined up at the corresponding video levels which on Samsungs usually means your contrast is set below 90, what is it set at?
post #1845 of 1936
Quote:
Originally Posted by zoyd View Post

This is an indication that the controls are not lined up at the corresponding video levels which on Samsungs usually means your contrast is set below 90, what is it set at?

My contrast is at 83, to put it in the higher end of 30-40 ftL
post #1846 of 1936
Quote:
Originally Posted by prme19 View Post

My contrast is at 83, to put it in the higher end of 30-40 ftL

Set it to 95 and use cell light to get the output you want, that will align the 10pt controls to the proper levels.
post #1847 of 1936
Quote:
Originally Posted by zoyd View Post

Set it to 95 and use cell light to get the output you want, that will align the 10pt controls to the proper levels.

Do you mean the backlight?.. I will try that then.. Thanks
post #1848 of 1936
ok so i followed zoyd's advice to adjust backlight and leave contrast at 95 when adjusting greyscale, and man he was right! never thought i could get any lower delta Es. Here are the new calibration charts:

Grey Scale and Primary/Secondary Colors


Gamma


RGB Levels


CIE Diagram Grays and Saturations


a few questions though:
1. when adjusting 10 pt. white balance, can i redo 100 IRE or just leave it alone, it being my reference white? i just did 90 IRE all the way back to 10 IRE.
2. does anybody know why my gamma reference curves like this? It didn't look like that during calibration. it was a straight line on 2.2, with my display average just hovering above it.
post #1849 of 1936
Quote:
Originally Posted by prme19 View Post


a few questions though:
1. when adjusting 10 pt. white balance, can i redo 100 IRE or just leave it alone, it being my reference white? i just did 90 IRE all the way back to 10 IRE.
2. does anybody know why my gamma reference curves like this? It didn't look like that during calibration. it was a straight line on 2.2, with my display average just hovering above it.

with the delta e at 1.5 you won't see a difference. if you are a perfectionist I could not hurt to adjust it. your gamma looks like it is following closely to the bt spec, change it in preferences and see how close it tracks.

 

I don't own one of these panels, but the 0% Y value of .093 seems pretty high, what meter are you measuring with? my D series measures .009 with my i1display 3.

post #1850 of 1936
Quote:
Originally Posted by vega509 View Post

with the delta e at 1.5 you won't see a difference. if you are a perfectionist I could not hurt to adjust it. your gamma looks like it is following closely to the bt spec, change it in preferences and see how close it tracks.

I don't own one of these panels, but the 0% Y value of .093 seems pretty high, what meter are you measuring with? my D series measures .009 with my i1display 3.

hehe i guess I'm just being OC with the delta Es.
i'm using a colormunki display and i could never get my 0% luminance and xy values to fall in the D65. Zoyd said this is probably because it was already too dark for the meter to measure accurately. or maybe my meter is also limited in that low end.
post #1851 of 1936
Quote:
Originally Posted by prme19 View Post


hehe i guess I'm just being OC with the delta Es.
i'm using a colormunki display and i could never get my 0% luminance and xy values to fall in the D65. Zoyd said this is probably because it was already too dark for the meter to measure accurately. or maybe my meter is also limited in that low end.

 

the 0% will rarely be at D65 if ever on any panel unless there is absolutely 0 mll. You may want to check the colormunki specs and see where its low level lies. you can also put up a 0% field and take several measurements to see if the meter is repeatable. you can also try lowering the brightness and backlight controls to see if the 0% drops as your contrast is only at 95 you still have some room to keep the light output.


Edited by vega509 - 11/23/13 at 9:44am
post #1852 of 1936
Quote:
Originally Posted by prme19 View Post

o
2. does anybody know why my gamma reference curves like this? It didn't look like that during calibration. it was a straight line on 2.2, with my display average just hovering above it.

It's hard to see which curve is which in the that plot, try changing the plot colors (right click to customize). The light curved line is your target, BT.1886 since you have that selected in preferences. The blue line is your average and I think your measured is close to the average, meaning flat. With a 0% reading of 0.09 cd/m^2 (I guess this is an LCD) you should lighten your 10, 20, 30% levels to the Gamma Y values, that will bend your measured gamma down to the target BT.1886 line.
post #1853 of 1936
Quote:
Originally Posted by zoyd View Post

With a 0% reading of 0.09 cd/m^2 (I guess this is an LCD)

Whats the best LCD will go?
post #1854 of 1936
Quote:
Originally Posted by xvfx View Post


Whats the best LCD will go?


that depends, different panels and different back lighting technologies will make vast differences, FALD if implemented properly should be almost 0.

post #1855 of 1936
My CalMAN 0% level reading is x243 y217 z530 and 0.057 cd/m^2. Is it a decent black level for a 40" Samsung CCFL LCD from 2009 for $700 (back in 2009)?
Edited by MonarchX - 11/23/13 at 2:06pm
post #1856 of 1936
Quote:
Originally Posted by zoyd View Post

It's hard to see which curve is which in the that plot, try changing the plot colors (right click to customize). The light curved line is your target, BT.1886 since you have that selected in preferences. The blue line is your average and I think your measured is close to the average, meaning flat. With a 0% reading of 0.09 cd/m^2 (I guess this is an LCD) you should lighten your 10, 20, 30% levels to the Gamma Y values, that will bend your measured gamma down to the target BT.1886 line.

here's the gamma graph with a darker background


the white dotted line is supposed to be the display reference but i don't know why it curves like that. it wasn't like that when i was calibrating earlier.
the blue line is my display average, it's close to 2.2.
post #1857 of 1936
^Your measurements are not on the graph, right click and select "Luminance" to display a yellow line by default. You want the yellow line to match the white line if you have selected BT.1886 as your reference.
post #1858 of 1936
Quote:
Originally Posted by zoyd View Post

^Your measurements are not on the graph, right click and select "Luminance" to display a yellow line by default. You want the yellow line to match the white line if you have selected BT.1886 as your reference.

oh ok.. here it is.. yellow line is a little higher than the reference measure, at least on the low end.. is that why it curves like that?
post #1859 of 1936
The white line is your target, it curves to compensate for your elevated black level. Raise the 10pt controls at 10%, 20% and 30% until the yellow line overlaps the white line. Ignore the blue line.
post #1860 of 1936
Quote:
Originally Posted by zoyd View Post

The white line is your target, it curves to compensate for your elevated black level. Raise the 10pt controls at 10%, 20% and 30% until the yellow line overlaps the white line. Ignore the blue line.

got it.. back to calibrating tonight.. thanks zoyd
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › Basic Guide to Color Calibration using a CMS (updated and enhanced)