Precalibrating A Reference Display (without using LUTs) Part 1

Day 21: 24 Insights in 24 Days – 2018 New Year Marathon

Using CalMAN Part 6: How to calibrate without using Look-Up Tables (Part 1)

In this CalMAN Studio series you’ve learned how to generate the most accurate calibration possible, using 1D and 3D Look-Up Tables (LUTs). But using Display Calibration LUTs requires one of these two things:

• Your reference display must support internal LUTs
• You have an external LUT box to upload your Calibration LUTs into, and have that output feeding into your reference display

The problem is: LUT boxes are a specialty item and good ones cost money. And most brands of reference displays don’t allow you to override their internal LUTs.

You can calibrate your reference display without using Calibration LUTs

This type of calibration performs best with extremely ‘linear’ displays where R, G and B are adjusted separately. Ideally, the R, G, and B channels are also ‘decoupled’. If your RGB channels are decoupled then adjustments to individual channels don’t effect the other two channels.

If you want a linear and decoupled reference display then you want to buy an OLED-based display. They are the current benchmark for this type of highly adjustable and predictable televisions.

In the next two Insights you learn how to calibrate an OLED reference display without the use of Look Up Tables.

First, you learn how to precalibrate your OLED, using test patterns

And I’m not talking about color bars. The traditional color bar test pattern you’re probably familiar with were designed for Cathode Ray Tubes (CRTs). Since none of the displays you’ve been able to buy for that last 10 years use phosphors illuminated by a stream of electrons, that test pattern doesn’t provide you with useful calibration information.

Instead, we’re going to look at three different test patterns to help us set your display for proper calibration. What are the three questions that these three test patterns help us answer?

• Data vs Video Levels: Is our software sending the same digital signal that our reference display is expecting to see? If not, the first test pattern clearly informs us if there’s a mismatch between our software and the reference display.
• Black Levels: Using the knobs on the front of your reference display, the next test pattern helps us set the overall brightness of the display by looking at the darkest portions of the image.
• Peak White Levels: Using the knobs on the front of your reference display, a third test pattern helps you set the peak white values by looking at the brightest portions of the image.

Essentially, before using a colorimeter we ‘eyball’ the initial settings on the reference display. Then, we go forward and do the actual calibration.

You can download the test patterns I’m using in this Insight from Light Illusion’s website

These test patterns (and more) are available in CalMAN Studio – but not if you’re using Davinci Resolve as the pattern generator (which is what I’m doing). Instead, I downloaded a set of modern test patterns from Light Illusion’s website and loaded them into a timeline in Davinci Resolve. For the low, low cost of an email address you can download these same test patterns here:

• Light Illusion’s Download Page: In the ‘Download Selection’ pull-down menu, select ‘Calibration Images’. A download link will be sent to the email address you provide.
  https://www.lightillusion.com/downloads.html

In Part 2, you learn how to perform the actual calibration

We wrap this Insight once we’ve completed the pre-calibration. In Part 2, we do the grayscale calibration using the display’s knobs to dial in the white-point adjustment and then run a series of checks to make sure the display is in alignment.

Questions? Ask them in the Comments below

Be sure to ask your questions in the comments. I’ll be doing an Insight just answering thos comments, once we’ve completed our initial look at CalMAN Studio.

-pi

Comments

Homepage › Forums › Precalibrating A Reference Display (without using LUTs) Part 1

• Creator
  Discussion
• January 21, 2018 at 2:28 pm #33045
  
  Patrick Inhofer

  Keymaster

  This is the discussion thread for the Insight: Precalibrating A Reference Display (without using LUTs) Part 1

• Creator
  Discussion

• 
  Willian Aleman

  Guest

  January 22, 2018 at 9:58 am

  Patrick, thanks a lot for the details in this part 1 calibration insight. I can not wait for part 2.
  Doesn the method you are discribing here only work with OLED monitors?
• 
  Patrick Inhofer

  Guest

  January 22, 2018 at 10:53 am

  On LCDs you don’t want to do a lift on the dark patches, the way I discuss on the OLED. Even the first code value out of black should be discernible, due to the nature of LCD technology. Also – I’ve had some offline discussions about these test patterns and will be digging into them more deeply (about how to use them properly, and how not to use them properly).
• 
  Patrick Inhofer

  Guest

  January 22, 2018 at 10:57 am

  Also – the 2-point Grayscale calibration is less effective on LCDs. And the less ‘reference quality’ the LCD the more the 2-point Grayscale is ineffective. But CalMAN will show that to you.
• 
  Willian Aleman

  Guest

  January 22, 2018 at 11:27 am

  Patrick, thanks for prompt and the detailed response.
• 
  Orash Rahnema

  Member

  January 27, 2018 at 12:45 pm

  Yes, Please do it!
  When i tried to calibrate an eizo with lightspace, following theyr guide, this patterns have been the first major obstacle that i encounter.
  I’m stuck there :/
• 
  Patrick Inhofer

  Guest

  January 27, 2018 at 12:53 pm

  ‘First Major Obstacle’ – Can you describe the nature of this obstacle, in detail?
• 
  Orash Rahnema

  Member

  January 27, 2018 at 7:54 pm

  Well, i’m talking about an eizo calibration
  i pre calibrated with colornavigator to get the min and max brightnes, then used the charts to check it and fine tune it, but It looked wrong whaterver i was doing, mostly i’m talking about the video vs data chart.
  It Just confused me in what signal i was feeding and what It was getting It.
  The eizo monitor can’t switch between video and data, so im not sure which One is expectin and the chart never looked correct
• 
  Roel Videler

  Member

  March 23, 2018 at 6:01 am

  Maybe it’s useful to define what eizo you have, mine does have those settings, but they’re labeled as limited, 109% and full I think. I’m on vacation so I can’t check specifically.
• 
  Patrick Inhofer

  Guest

  March 23, 2018 at 7:52 am

  Thanks Roel for commenting. I didn’t realize I missed this… take a look at Part 2 of this Insight. The general rule… if your display doesn’t have a video / data level setting it’s probably video level. But the easiest thing to do is call the company and get the word directly from them.

  Also – I’m not clear what ‘charts’ you’re using and what ‘fine tuning’ encapsulates?
• 
  Ron Illingworth

  Guest

  May 7, 2018 at 1:14 pm

  Patrick, do you offer a download of test patterns and chip charts, by chance?
• 
  Patrick Inhofer

  Guest

  May 7, 2018 at 1:32 pm

  Hi Ron – I’m using the patterns from LightSpace, “Light Illusion’s Download Page: In the ‘Download Selection’ pull-down menu, select ‘Calibration Images’. A download link will be sent to the email address you provide. https://www.lightillusion.com/downloads.html “
• 
  Ron Illingworth

  Guest

  May 7, 2018 at 2:10 pm

  That’s great, thanks! Much appreciated!
• Author
  Replies

Log in to reply.

• Public
• All Members
• Only Me

• Public
• All Members
• Only Me

• Public
• All Members
• Only Me

• Public
• All Members
• Only Me

• Public
• All Members
• Only Me

• Public
• All Members
• Only Me