AVS › AVS Forum › Display Devices › Display Calibration › Yay finally!!!
New Posts  All Forums:Forum Nav:

Yay finally!!! - Page 2

post #31 of 68
4/21/12 at 5:03pm
Quote:
Originally Posted by kanti123 View Post

So, you're saying dE94 is good for measuring Gamut?

definitely, I wouldn't use any other formula for gamut

the LCH breakdown is an extension of dE94 in CalMAN
Reply
Reply

AVS Top Picks

post #32 of 68
4/21/12 at 5:05pm
Thread Starter 
Quote:
Originally Posted by PlasmaPZ80U View Post

definitely, I wouldn't use any other formula for gamut

the LCH breakdown is an extension of dE94 in CalMAN

ah okay, so dEUV for gray scale, dE94 for Gamut. Nice, I'm going to write down my current setting and bump up my contrast to mids 90s like you suggested.
Reply
Reply
post #33 of 68
4/21/12 at 5:07pm
Quote:
Originally Posted by kanti123 View Post

ah okay, so dEUV for gray scale, dE94 for Gamut. Nice, I'm going to write down my current setting and bump up my contrast to mids 90s like you suggested.

the quickview DDC workflow in CalMANv4 already uses dE94 for gamut
Reply
Reply
post #34 of 68
4/21/12 at 5:11pm
Thread Starter 
I have never use Quickview DDC in Calman4.
Reply
Reply
post #35 of 68
4/21/12 at 5:11pm
Quote:
Originally Posted by PlasmaPZ80U View Post

also, with dEUV the better the RGB percentages the lower the dE and all grayscale points are treated equally since the L portion of the formula is removed

By default CalMAN doesn't use L for dE76 or dE94 on grayscale readings, but at least the measured luminance level is used to determine the L*a*b* values.

The point of any dE formula is to report a consistent value that correlates to a visible difference. So a dE of 2 between to shades of blue is a similiar difference for two shades for yellow that are seperated by a dE of 2.

In that regard deUV is a total failure, it's values aren't consistent from either one shade to the next and is even worse when comparing targets at different luminance levels.
Reply
Reply
post #36 of 68
4/21/12 at 5:19pm
Thread Starter 
Quote:
Originally Posted by sotti View Post

By default CalMAN doesn't use L for dE76 or dE94 on grayscale readings, but at least the measured luminance level is used to determine the L*a*b* values.

The point of any dE formula is to report a consistent value that correlates to a visible difference. So a dE of 2 between to shades of blue is a similiar difference for two shades for yellow that are seperated by a dE of 2.

In that regard deUV is a total failure, it's values aren't consistent from either one shade to the next and is even worse when comparing targets at different luminance levels.

So, in your point of view. You're saying when measuring Gamut and Grayscale, I should just use dE94 across the board would be more effective and efficient?
Reply
Reply
post #37 of 68
4/21/12 at 5:25pm
Quote:
Originally Posted by kanti123 View Post

So, in your point of view. You're saying when measuring Gamut and Grayscale, I should just use dE94 across the board would be more effective and efficient?

Looking at dE94 is going to give you a better idea of how much error you'll actually percieve across the board.

When correctly calibrated you should be able to get dE94 values <2 for grayscale.
Reply
Reply
post #38 of 68
4/21/12 at 5:25pm
Quote:
Originally Posted by sotti View Post

By default CalMAN doesn't use L for dE76 or dE94 on grayscale readings, but at least the measured luminance level is used to determine the L*a*b* values.

The point of any dE formula is to report a consistent value that correlates to a visible difference. So a dE of 2 between to shades of blue is a similiar difference for two shades for yellow that are seperated by a dE of 2.

In that regard deUV is a total failure, it's values aren't consistent from either one shade to the next and is even worse when comparing targets at different luminance levels.

what about dELUV? is the UV part the problem or that the measured luminance level is not used to determine the L*U*V* values?
Reply
Reply
post #39 of 68
4/21/12 at 5:26pm
Thread Starter 
Quote:
Originally Posted by sotti View Post

Looking at dE94 is going to give you a better idea of how much error you'll actually percieve across the board.

When correctly calibrated you should be able to get dE94 values <2 for grayscale.

Right, I will try this out now.
Reply
Reply
post #40 of 68
4/21/12 at 5:31pm
see attached...

low dEUV values contribute to even lower dE94 values

 

QuickView Grayscale dE94.pdf 50.05078125k . file

 

QuickView Grayscale.pdf 29.3037109375k . file
Reply
Reply
post #41 of 68
4/21/12 at 5:35pm
Thread Starter 
Ah, I see that now. Lest margin of error and more precise. BTW to use dE94, I go to Settings>Miscellaneous and change DeltaE Formular from deUV to de94 under Interative Target?
Reply
Reply
post #42 of 68
4/21/12 at 6:32pm
Quote:
Originally Posted by kanti123 View Post

BTW to use dE94, I go to Settings>Miscellaneous and change DeltaE Formular from deUV to de94 under Interative Target?

no, you right click the dE section in the workflow, select properties and then select the dE formula you want from the top right of the screen
Reply
Reply
post #43 of 68
4/21/12 at 6:58pm
Thread Starter 
okay, got it. Hey Adjusting Gamut is really annoying. Any tips on how should I approach this?
Reply
Reply
post #44 of 68
4/21/12 at 7:08pm
Quote:
Originally Posted by kanti123 View Post

okay, got it. Hey Adjusting Gamut is really annoying. Any tips on how should I approach this?

for starters, you can set color space to "auto" and set color to get red luminance as close as possible and the secondary hues a low as possible overall

later on, you can use "custom" color space to dial in LCH for each primary and secondary
Reply
Reply
post #45 of 68
4/21/12 at 7:13pm
Thread Starter 
Quote:
Originally Posted by PlasmaPZ80U View Post
for starters, you can set color space to "auto" and set color to get red luminance as close as possible and the secondary hues a low as possible overall

later on, you can use "custom" color space to dial in LCH for each primary and secondary
So, while I'm in auto for color space, I would adjust Color and Tint to achieve this correct?. BTW I have attached my new setting with contrast being at mid 90s. Its a little red, but that'll do for now.

 

QuickView Grayscale.pdf 58.2724609375k . file
Reply
Reply
post #46 of 68
4/21/12 at 7:18pm
Quote:
Originally Posted by kanti123 View Post

So, while I'm in auto for color space, I would adjust Color and Tint to achieve this correct?. BTW I have attached my new setting with contrast being at mid 90s. Its a little red, but that'll do for now.

looks good from 20% on up, looks like your meter can't read 10% properly
Reply
Reply
post #47 of 68
4/21/12 at 7:18pm
Quote:
Originally Posted by kanti123 View Post

So, while I'm in auto for color space, I would adjust Color and Tint to achieve this correct?

yes
Reply
Reply
post #48 of 68
4/21/12 at 7:48pm
Thread Starter 
Quote:
Originally Posted by PlasmaPZ80U View Post
yes
okay, this gamut adjustment is starting to piss me off LOL

 

Adjust Color Gamut.pdf 116.76171875k . file
Reply
Reply
post #49 of 68
4/21/12 at 9:57pm
Quote:
Originally Posted by kanti123 View Post

okay, this gamut adjustment is starting to piss me off LOL

you need to turn the color control up until the red luminance error is as close to zero as possible... it's way too low right now
Reply
Reply
post #50 of 68
4/22/12 at 5:31am
Thread Starter 
And that would be C chart? CalMAN said you have to get all the luminance to match each other. Thats how it throws me off
Reply
Reply
post #51 of 68
4/22/12 at 5:46am
Quote:
Originally Posted by kanti123 View Post

And that would be C chart? CalMAN said you have to get all the luminance to match each other. Thats how it throws me off

You do but you get Red as close as you can first.

L = Luma Y
C = Saturation y
H = Hue x

Primaries get Luma matched, Secondaries get Hue matched... and then have fun balancing back and forth for hours until it's all as perfect as can be (at this point a little OCD is required ).

Jason
Reply
Reply
post #52 of 68
4/22/12 at 7:13am
Thread Starter 
Quote:
Originally Posted by DaGamePimp View Post
You do but you get Red as close as you can first.

L = Luma Y
C = Saturation y
H = Hue x

Primaries get Luma matched, Secondaries get Hue matched... and then have fun balancing back and forth for hours until it's all as perfect as can be (at this point a little OCD is required ).

Jason
The OCD part I have it get primary Luma matched like this? and ignore everything else? Had to turn up color to 80 to get red to 0 in L chart. Weird


Update: I think I got it down now, not perfect but close to 0 enough.

 

Adjust Color Gamut.pdf 116.76171875k . file

 

Gamut After.pdf 115.986328125k . file
Reply
Reply
post #53 of 68
4/22/12 at 10:20am
Quote:
Originally Posted by kanti123 View Post

The OCD part I have it get primary Luma matched like this? and ignore everything else? Had to turn up color to 80 to get red to 0 in L chart. Weird


Update: I think I got it down now, not perfect but close to 0 enough.

is the first attachment auto color space and the second custom?
Reply
Reply
post #54 of 68
4/22/12 at 10:32am
Thread Starter 
No both custom, I think I have my color too high, it's over saturated to me. My settings for color is 79 tint 50/50

Color space,


R G B
45 1 0 Red
0 46 0 Green
0 2 78 Blue
45 50 1 yellow
0 48 50 cyan
42 0 37 magenta
Reply
Reply
post #55 of 68
4/22/12 at 11:25am
Quote:
Originally Posted by kanti123 View Post

No both custom, I think I have my color too high, it's over saturated to me. My settings for color is 79 tint 50/50

Color space,


R G B
45 1 0 Red
0 46 0 Green
0 2 78 Blue
45 50 1 yellow
0 48 50 cyan
42 0 37 magenta

for the after result, can you lower cyan and magenta luminance?
Reply
Reply
post #56 of 68
4/22/12 at 11:52am
Thread Starter 
I think I can, right now my picture looks too saturated. I also notice my blue luminance is very low. By adding green 1 - 2 ticks it adds luma significantly.

Update: I got every color nice and tweaked, just blue. For some reason Delta C is over 15 while other are at close to zero. Any tips on this? This setting make my picture looks better now, no more overly saturation.

 

Enhanced Gamut .pdf 116.4111328125k . file
Reply
Reply
post #57 of 68
4/23/12 at 5:24pm
Thread Starter 
Can any one take a look at my Delta C chart? my blue is waayyy high. Already max out my blue in color space (100) had to turn up green and red.

 

BlueDeltaC-Problem.pdf 116.3623046875k . file
Reply
Reply
post #58 of 68
4/24/12 at 9:41am
Quote:
Originally Posted by kanti123 View Post

Can any one take a look at my Delta C chart? my blue is waayyy high. Already max out my blue in color space (100) had to turn up green and red.

do not turn up green and red in the cms section for blue, it will desaturate the blue primary further and make things a lot worse

also, use dE94 in the datagrid, not dE76
Reply
Reply
post #59 of 68
4/24/12 at 10:43am
Thread Starter 
I can't change that for some reason. I click on de74 and go to property. There's no option in there.
Reply
Reply
post #60 of 68
4/24/12 at 11:44am
Quote:
Originally Posted by kanti123 View Post

I can't change that for some reason. I click on de74 and go to property. There's no option in there.

you right click the datagrid and select the add row option and choose dE94 then you highlight dE76 and select remove row
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration

AVS Top Picks

AVS › AVS Forum › Display Devices › Display Calibration › Yay finally!!!