x-Rite i1 display pro - newbie questions - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 
Thread Tools
post #1 of 14 Old 05-11-2019, 01:42 PM - Thread Starter
Member
 
Join Date: Jan 2017
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 1
x-Rite i1 display pro - newbie questions

Hello all,


I'm completely new to calibration, today I received my x-rite i1 display pro and I started playing with it.

I have two Dell Ultrasharp U2515H monitors that are configured with factory settings. I positionned the i1 on the second screen and used i1 profile from x-rite to start the calibration. I took all the default settings to start with and I waited until the process was over. After calibration my screen looked much darker than my other screen and my subjective feeling was that colors were less accurate. As I wanted to measure the effect of calibration, I tried HCFR in the same conditions to measure primary color and here is what I got:


Screen 1 (unaffected, only factory settings)
Red 97.4%
Green 100.2%
Blue 105.5%
dE 2.9


Screen 2 (after calibration)
Red 96.4%
Green 101.4%
Blue 96.5%
dE 4.3


Both screens were turned on for more then 30 minutes

As I was unhappy about the results, I tried to modify manually my screen settings through the screen menu and I got the following measure


Screen 2 (after calibration and manual adaptation)
Red 100.3%
Green 99.9%
Blue 100.2%
dE 0.3



I tried the other types of measurements with additional colors and greyscales and the dE shifted a little bit to 0.5



My newbee questions:
- is my approach the correct one to have accurate colors?
- is there a reason why i1 profiler is leading me to a higher dE?
- can I assume that my second monitor is correctly calibrated now or should I do something else?



Thanks
Laurent

Owner of a Xgimi H1
LaurentW is offline  
Sponsored Links
Advertisement
 
post #2 of 14 Old 05-11-2019, 06:10 PM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,200
Mentioned: 114 Post(s)
Tagged: 0 Thread(s)
Quoted: 4308 Post(s)
Liked: 1549
Quote:
Originally Posted by LaurentW View Post
As I wanted to measure the effect of calibration, I tried HCFR in the same conditions to measure primary color and here is what I got:
<snip>
Screen 2 (after calibration)
Red 96.4%
Green 101.4%
Blue 96.5%
dE 4.3
I'm not sure what these percentages refer to. Do they correspond to RGB measurements of the white patch?

Quote:
My newbee questions:
- is my approach the correct one to have accurate colors?
- is there a reason why i1 profiler is leading me to a higher dE?
- can I assume that my second monitor is correctly calibrated now or should I do something else?
In general, you should not make any manual adjustments after calibration. The i1Profiler should give you an option to manually adjust R/G/B during calibration.

Also, are you using the correct display type in HCFR?
Dominic Chan is offline  
post #3 of 14 Old 05-11-2019, 06:58 PM
Senior Member
 
Vishwa Somayaji's Avatar
 
Join Date: Jun 2001
Location: Edmonton,Alberta Canada
Posts: 457
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 55 Post(s)
Liked: 55
In iProfiler use adc button. It will take care of everything and will do a good job. Also, don't forget to enter the brightness(in nits) you want in the monitor. By default, it will be 100.
Vishwa Somayaji is offline  
Sponsored Links
Advertisement
 
post #4 of 14 Old 05-12-2019, 01:20 AM - Thread Starter
Member
 
Join Date: Jan 2017
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 1
Quote:
Originally Posted by Dominic Chan View Post
I'm not sure what these percentages refer to. Do they correspond to RGB measurements of the white patch?


In general, you should not make any manual adjustments after calibration. The i1Profiler should give you an option to manually adjust R/G/B during calibration.

Also, are you using the correct display type in HCFR?
Hi Dominic, Vishwa,


Thanks



Those values indeed come from the RGB+white measurements with HCFR, I also ran a longer measure with gray scales and more colors to end up with more or less the same result.


I did the automatic calibration with i1 Profiler to start with, it resulted in a darker screen. I then did it manually and adapted brightness, contrast and colors as instructed and again I got a too dark screen. I found a page with instructions on Dell's site where they instruct to adapt the luminance, I did this but again ended up with the same result, the dE even went up.

For HCFR I did the following:
- select generator : automatic (the other choice is DVD)
- select the i1 sensor + do not use correction file
- in config I selected the i1 sensor, LCD Phosphor (where Dell Ultrasharp is mentionned), reading type display, observer type default

- I also tried to change the display to "refresh display"

Thanks

Owner of a Xgimi H1
LaurentW is offline  
post #5 of 14 Old 05-12-2019, 03:03 AM
Advanced Member
 
Anderegg's Avatar
 
Join Date: Jul 2012
Posts: 656
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 424 Post(s)
Liked: 227
Refresh is for CRT's.

I would use the R G and B controls on the monitor to try and get your gamma and greyscale as best as you can prior to running iProfiler auto ICC creation. I am assuming that your monitors R G B controls act simply as a TV's R G and B Gain controls. For TV's it's suggested to do the Gains at 80%, maybe 100% if you feel lucky, but whatever gets you the best pre calibration readings. Use HCFR to do your pre iProfiler R G B manual adjustments.

Paul

Sony X900F SDR and HDR Calibrations https://www.avsforum.com/forum/166-l...l#post57551552
Anderegg is offline  
post #6 of 14 Old 05-12-2019, 03:28 AM - Thread Starter
Member
 
Join Date: Jan 2017
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 1
Quote:
Originally Posted by Anderegg View Post
Refresh is for CRT's.

I would use the R G and B controls on the monitor to try and get your gamma and greyscale as best as you can prior to running iProfiler auto ICC creation. I am assuming that your monitors R G B controls act simply as a TV's R G and B Gain controls. For TV's it's suggested to do the Gains at 80%, maybe 100% if you feel lucky, but whatever gets you the best pre calibration readings. Use HCFR to do your pre iProfiler R G B manual adjustments.

Paul
ok, thanks, I'll try it that way.



Are the readings I mentioned earlier a good way to evaluate if calibration has succeeded? (dE and deviation for each color)

Owner of a Xgimi H1
LaurentW is offline  
post #7 of 14 Old 05-12-2019, 04:30 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,200
Mentioned: 114 Post(s)
Tagged: 0 Thread(s)
Quoted: 4308 Post(s)
Liked: 1549
Quote:
Originally Posted by LaurentW View Post
Those values indeed come from the RGB+white measurements with HCFR, I also ran a longer measure with gray scales and more colors to end up with more or less the same result.
I still don’t understand what you mean by “measure the primary color... Red 100.3%, Green 99.9%, Blue 100.2%”. Are these for the white patch?

Quote:
I then did it manually and adapted brightness, contrast and colors as instructed and again I got a too dark screen.
This is really confusing. i1Profiler should not make the screen darker than what you set manually. When you adjusted the brightness manually, did you find the screen too dark at that moment?
Dominic Chan is offline  
post #8 of 14 Old 05-12-2019, 04:42 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,200
Mentioned: 114 Post(s)
Tagged: 0 Thread(s)
Quoted: 4308 Post(s)
Liked: 1549
Quote:
Originally Posted by Anderegg View Post
For TV's it's suggested to do the Gains at 80%, maybe 100% if you feel lucky, but whatever gets you the best pre calibration readings.
When calibrating computer monitors, always use 100% instead of 70% or 80% . i1Profiler will create a “1D LUT” to correct the grey scale at lower input points.
Dominic Chan is offline  
post #9 of 14 Old 05-12-2019, 05:17 AM - Thread Starter
Member
 
Join Date: Jan 2017
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 1
Quote:
Originally Posted by Dominic Chan View Post
I still don’t understand what you mean by “measure the primary color... Red 100.3%, Green 99.9%, Blue 100.2%”. Are these for the white patch?



This is really confusing. i1Profiler should not make the screen darker than what you set manually. When you adjusted the brightness manually, did you find the screen too dark at that moment?
The values I used in this post are coming from HCFR in the RGB level section (left side of the screen). I have done measures with white (values indicated above) but also for red, blue and green.
If I click on red for example, I see that I have a dE of 4 while I have 0.8 for green and 4.0 for blue.



I'm basically trying to find out how to measure success (or failure)



When using i1 profiler, I used the automatic mode for screen calibration (ADC) and it automatically adjusted the screen brightness around 30 while it was set to 50 with factory defaults. I brought my screen back to factory defaults and did the whole process manually but I ended up with the same result. I have two Dell screens (exactly the same model), one is still in factory settings and the other one calibrated by i1, the one calibrated with i1 has a much darker image and when I measure white with HCFR, I get a higher dE than my screen with factory settings.

Owner of a Xgimi H1
LaurentW is offline  
post #10 of 14 Old 05-12-2019, 05:55 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,200
Mentioned: 114 Post(s)
Tagged: 0 Thread(s)
Quoted: 4308 Post(s)
Liked: 1549
Quote:
Originally Posted by LaurentW View Post
The values I used in this post are coming from HCFR in the RGB level section (left side of the screen). I have done measures with white (values indicated above) but also for red, blue and green.
That answers my question. The original post was not clear to me.

Quote:
When using i1 profiler, I used the automatic mode for screen calibration (ADC) and it automatically adjusted the screen brightness around 30 while it was set to 50 with factory defaults.
That setting by itself does not mean much. What defines the brightness is the actual luminance (cd/m2, or “nits”) that you get. Typically people calibrated monitors to 100 nits, but in bright environments people calibrate to mich higher values. If you set it manually (ADC Off), i1Profiler should not change it.
Dominic Chan is offline  
post #11 of 14 Old 05-12-2019, 08:57 AM - Thread Starter
Member
 
Join Date: Jan 2017
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 1
Quote:
Originally Posted by Dominic Chan View Post
That answers my question. The original post was not clear to me.



That setting by itself does not mean much. What defines the brightness is the actual luminance (cd/m2, or “nits”) that you get. Typically people calibrated monitors to 100 nits, but in bright environments people calibrate to mich higher values. If you set it manually (ADC Off), i1Profiler should not change it.
Thanks !


I did it manually too but at one stage of the process i1 profiler asks to set the brightness to a specific level and that level is unfortunately too low.

Here is an example.


I can ignore it but afterwards the results measured in HCFR tend to have higher dE.

Am I right to try to have the lowest dE for each measured color or should I use a different indicator to see if my calibration is ok ?

Owner of a Xgimi H1
LaurentW is offline  
post #12 of 14 Old 05-12-2019, 09:31 AM
AVS Forum Special Member
 
Join Date: Dec 2014
Location: Mississauga, ON, Canada
Posts: 6,200
Mentioned: 114 Post(s)
Tagged: 0 Thread(s)
Quoted: 4308 Post(s)
Liked: 1549
Quote:
Originally Posted by LaurentW View Post
Thanks !


I did it manually too but at one stage of the process i1 profiler asks to set the brightness to a specific level and that level is unfortunately too low.

I can ignore it but afterwards the results measured in HCFR tend to have higher dE.

Am I right to try to have the lowest dE for each measured color or should I use a different indicator to see if my calibration is ok ?
What I’m trying to say is that, if you prefer a brighter display you should set the target luminance higher, and not change the luminance afterwards, which would invalidate the calibration.

The objective is indeed to get the lowest delta-E. However, most monitors do not provide independent calibrations for each colour, so you can only aim for the lowest average.
Dominic Chan is offline  
post #13 of 14 Old 05-12-2019, 09:40 AM - Thread Starter
Member
 
Join Date: Jan 2017
Posts: 29
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 1
Quote:
Originally Posted by Dominic Chan View Post
What I’m trying to say is that, if you prefer a brighter display you should set the target luminance higher, and not change the luminance afterwards, which would invalidate the calibration.

The objective is indeed to get the lowest delta-E. However, most monitors do not provide independent calibrations for each colour, so you can only aim for the lowest average.
That makes sense, I'll first set my preferred luminance and get the lowest values for dE afterwards.


Now, I have this:
- red dE 3.5
- green dE 0.7
- blue dE 4.1
- white dE 0.5
I suppose it's not bad but not yet perfect, I'll keep on trying.

Thanks for your time!

Owner of a Xgimi H1
LaurentW is offline  
post #14 of 14 Old 05-14-2019, 04:34 AM
Member
 
Join Date: Nov 2018
Posts: 15
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 1
Quote:
Originally Posted by LaurentW View Post
That makes sense, I'll first set my preferred luminance and get the lowest values for dE afterwards.


Now, I have this:
- red dE 3.5
- green dE 0.7
- blue dE 4.1

- white dE 0.5
I suppose it's not bad but not yet perfect, I'll keep on trying.

Thanks for your time!

AFAIK HCFR validates readings against some user configurable colorspace... lets say Rec709 gamma 2.2


Your display colorspace boundaries ARE NOT the same as Rec709/sRGB... they are slighty bigger although they intersect al 99% or whatever coverage manufacturer advertises.
That means that U2515H native gamut red is not going to match Rec709 red CIE xy coordinates.


i1Profiler or DisplayCAL are going to calibrate ("fix") white point, brightess, neutral grey and gamma... and nothing more. Then they are going to apply that 1DLUT calibration in graphics card and measure display: what is calibrated white, grey, and RGB primaries (and other things). This info plus graphics card calibration is going to be stored in a ICM profile. Your slightley oversaturated R, G and B are going to be the way they were... but now its actual xy coordinates are recorded.
GIMP or Photoshop are going to use that info to do not send red "255,0,0" when an sRGB image has sRGB "255,0,0" but another slighty modified 3numbers that have in YOUR display the same xy coordinates.


AFAIK HFCR is going to send reference RGB values (lets say "255,0,0") and it expects that display response will match reference colorspace "255,0,0".


So you dell calibration after i1Profiler or DisplayCAL may be not wrong... maybe you are measuring it in the wrong way.


If you want that your display behavior matches sRGB/Rec709 gamut boundaries closely you'll need one of the following:
-use 6axis controls if available in your display OSD (I'm not sure that U1515H has such feature, I would say "no", but check user manual)
-use an external LUT3D (not cheap comparect to your display cost)
-use applications with profile based color management (Firefox, GIMP, Photoshop.. etc) or applications with software LUT3D support like madVR, Reshade...etc (you can compute these LUT3Ds from ICM profile and reference colorspace)
Vicent is offline  
Sponsored Links
Advertisement
 
Reply Display Calibration

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