Display Calibration Question

[ksporry]

<p>Guys,<br>

So I got myself an x-rite i1DisplayPro calibration tool.<br>

Although I understand the need for calibrating monitors I'm not familiar with the background science, nor do I want to be.<br>

Using the i1DP did raise some questions at my end.<br>

First time I ran it I just ran the basic GUI, which ran through a basic version of the calibration process. It presented me with a small patch sample, which finishes fairly quickly. The calibration had quite a pronounced effect.<br>

I then decided to run it in advanced mode, selecting a Large patch sample. I was somewhat surprised to find that the result gave much more green in it.<br>

Today I tried running the advanced mode using a medium patch sample. The result in colour tone seemed to be similar to the Large mode, but significantly less bright.<br>

For comparison I tried running it through DisplayCal, an open source calibration software based on Argyl. This seemed to be brighter that the medium size i1DP software sample.</p>

<p>Taking the DisplayCal SW as a standard:<br>

1) The i1DP software at Basic seems to be similar but slightly darker and slightly more contrasty.<br>

2) The i1DP software at Advanced Medium patch sample seems much more subdued. Much less bright, warmer and less magenta (slightly more green)<br>

3) The i1DP software at Advanced Large patch sample seems much brighter than the medium sample. It's also slightly more green than the DisplayCal</p>

<p>I repeated the measurements with the x-rite software several times and its pretty consistent in the results it provides so Basic is always more magenta and bright, Medium is always more subdued and slightly greenish, and Large is always more bright and greenish.<br>

Honestly I'm not sure what profile to use. I didn't expect such a big difference.<br>

I was going to attached the screenshots, so you can see the difference, but obviously that won't work.<br>

Any thoughts/recommendation?</p>

[digitaldog]

<blockquote>

<p>Taking the DisplayCal SW as a standard:</p>

</blockquote>

<p>That's the first problem. Using the same hardware with differing software products set for the same calibration aim points (targets) will almost always produce different results visually! <br>

Pick a software product. Ideally one with as many options for setting the aim points as possible (any CCT value versus preset aim points, same for cd/m^2). The correct values are the one's that produce a visual match of display to print, assuming that's your calibration goal. <br>

<br>

Start here:<br>

<em>Why are my prints too dark?</em><br>

<em>A video update to a written piece on subject from 2013</em><br>

<em> In this 24 minute video, I'll cover:</em><br>

<em>Are your prints really too dark?</em><br>

<em><strong>Display calibration and WYSIWYG</strong></em><br>

<em>Proper print viewing conditions </em><br>

<em>Trouble shooting to get a match</em><br>

<em>Avoiding kludges that don't solve the problem</em><br>

<em> High resolution: http://digitaldog.net/files/Why_are_my_prints_too_dark.mp4</em><br>

<em>Low resolution: https://youtu.be/iS6sjZmxjY4</em><br>

</p>

[blumesan]

<p>Throughout your post you describe the "results" of your calibrations (darker, warmer, less contrasty, etc.). as the various calibrations affect the appearance of an image. You do not specify what image(s) you are using to make the comparisons.<br>

While you consistently use the term "calibration" to refer to your process, I assume you are familiar with the difference between calibrating a monitor and producing a profile for a monitor.</p>

<p> </p>

[ksporry]

<p>So,I assume you are all familiar with x-rite and it's profiler software that comes with the i1 Display Pro Calibration tool?<br>

Ignore the DisplayCal for a moment, why does the same software with the same calibration tool generate different effects?<br>

As for what image I use, I use the x-rite profiler software. It has some images that allow you to compare before and after results. My observations remain the same as I identified before:<br>

using the option for small sample of patches, medium sample of patches, and Large sample of patches, the tool generates different results. Why does it generate such significant different results when everything else (bar none) remains the same?</p>

 

[digitaldog]

<blockquote>

<p>why does the same software with the same calibration tool generate different effects?</p>

</blockquote>

<p>Because you're feeding it different data. In a perfect world, you would measure 16.7 million different 'colors' (device values) and build any profile. That would take a very, very long time and result in a profile bigger than many images you edit! So there's extrapolation used to take a much smaller sample of measurements to come up with a profile that has to define 16.7 million device values. <br>

Use the largest number of samples, more data is always a better option. <br>

Don't get caught up in the numbers, only the results visually! <br>

The before and after toggle is just a feel good option. Not very useful. Does the '<em>after</em>' produce a match to the print next to the display? Assuming that's the goal for the calibration aim points you've set. </p>

[ksporry]

I thought the point if a calibration tool and calibration software was that I wouldn't have to manually adjust the calibration

results to match a print? If I hate to do that I might as well not calibrate and adjust display settings by sight...

[digitaldog]

<blockquote>

<p>I thought the point if a calibration tool and calibration software was that I wouldn't have to manually adjust the calibration results to match a print?</p>

</blockquote>

<p>Why all the possible calibration settings for white point, cd/m^2, contrast ratio (when an option)? If there's <strong>one</strong> setting for calibration everyone would and should use, why not hardwire that into every product that calibrates the display? The answer is, <strong>if</strong> your goal is to provide a match to the print, calibration plays a huge role and no one setting is necessarily correct. <br>

<br>

Watch the video! </p>

[ellis_vener_photography]

"I thought the point if a calibration tool and calibration software was that I wouldn't have to manually adjust the calibration

results to match a print?"

 

Hi Kryn,

 

The point of calibrating is to choose a consistent set of values that you want the graphics system (gpu and display) to

follow. What profiling does is to process a target created using those calibration values through the graphics system,

have the actual onscreen output read by the colorimeter and the profiling software then compares the target patches numbers that make

up each color in the target to what is actually displayed and generate a table of algorithms to process the original input

values so that the output numbers either match or match as closely as possible the original numeric values you

specified.

 

The purpose of calibrating and profiling a display is so that you see an accurate version of the image you are working on.

 

Profiling a printer or using a profile for a specific printer, ink and print media combination works basically the same way: the

profile is there to ensure that the actual colors and tones in the print are reproduced neutrally - that is to remove any

inherent bias in the combination or paper, ink, and printer.

 

But matching what is on the display to the print is a bit like the idea of parallel universes in physics - the print and the computer are not "aware" of the other's existance.

 

In other words, the computer's display system has no way knowing the print exists or what it looks like.

 

To get around this we as operators have to tell the image processing program to use a printing profile to emulate what

the print will look like (AKA "soft proofing"). And even then there will be some differences due to the reality that on screen

we are looking at a virtual image Created by backlit projection while the print is a solid object illuminated by the light

falling on it and being absorbed and reflected. As I learned from John Paul Capnigro and R. Mac Holbert, the print is the

proof.

 

" If I have to do that I might as well not calibrate and adjust display settings by sight."

 

I agree! And using the ICC color profiling system eliminates the need to do that. But as Andrew Rodney points out in at

least one of the videos he posted a link for, setting the right calibration values that will be used for your display profile and

using the correct profile for your printer, ink, and media combination solves almost all of the technical problems. In the

end you have to be pleased with the way the print looks. Only you will be able to see any difference between the display

and the print and then only if you are also looking at the print very controlled lighting conditions..