ICDM At Work On Display Standards To Answer `Specsmanship’
Insight Media’s Chris Chinnock
One of the more frustrating issues surrounding the launch of new display technologies, like high dynamic range, wide color gamut, and color volume is the lack of unified standards to measure and monitor performance.
Thanks to the Society for Information Display (SID) a multi-company/multi-industry Committee called the International Committee for Display Metrology (ICDM) is developing exactly that. The ICDM is a standards committee within SID, which represents over 300 top display metrologists and specialists of related sciences.
Speaking at the recent QLED & HDR10 Summit sponsored by Samsung and Insight Media, Chris Chinnock, Insight Media principal, said the key to this organization is to develop accepted test procedures and to provide them as a resource for anybody in the industry who needs to test a display.
“It’s important to have some agreed upon test methods, so there isn’t market confusion or specsmanship – at least it can attempt to mitigate some of that by having accepted test procedures,” Chinnock said. “It also helps to have accepted procedures if you are buying a display so that you can compare the specifications of a spec sheet and know that what that spec sheet lists is truly representative of what you would see. That’s not always the case, but it’s a worthy goal, anyway.”
Of course, it also means test procedures are in development on the latest technologies for TV reviewers and calibrators.
Read more on the mission of the ICDM after the jump:
Chinnock said the ICDM also provides liaison work with other standards bodies, like ISO and IDC, and they also support organizations like the Ultra HD Alliance (UHDA), so, some of the tests that go into the UHDA Premium Certification programs, for example, can reference work done by the ICDM.
So far the ICDM has produced Display Metrology Version 1 , which Chinnock described as a very substantial publication of 563 pages that took many years to compile. It includes sets of procedures and methods, “but it doesn’t tell you to, for example, `Use test pattern A for such-and-such a display,’ ” he said. “It is the reviewer’s decision about which test method to use. These are the agreements for creating a spec sheet.”
In Version 1 are methods, evaluation details, reporting templates, as well as explanations, tutorials, diagnostics, etc.
Chinnock called this a useful tool that is supported by all of the display metrology companies, and most of the display manufacturers.
There are many people very active in the organization who are now working on Version 2, Chinnock said. The goal is to get this published by 2018.
“This is important because there are a bunch of new display technologies coming down the road that we have no display metrology for right now. High Dynamic Range (HDR) is one of the biggest ones out there,” said Chinnock. “There are also head-up displays, curved displays, transparent displays, color volume displays, flexible displays, free-form displays, VR displays, AR displays, automobile displays etc.”
People are developing displays and commercializing them, but every organization has its own test methodology. So, how do you compare like displays?
To accomplish its goals, the ICDM is comprised of sub-committees and working groups, Chinnock explained. Subcommittees work to update an existing Version 1 document. There are a number of these chapters that are already being reviewed and revised. This is primarily done to make sure that they are staying current with the latest technologies as evolution occurs within the various categories.
“A lot of this now has started to reflect the move from standard dynamic range (SDR) to HDR. A lot of the previous criteria and procedures were developed for an SDR world, so now we have to at least relook at all of these things as they related to HDR technology,” he said.
Also, the working groups now are dedicated to new metrology where there are not existing chapters in the Version 1 volume.
“There is a group on specification reform covering HDR file formats, color volume, etc. There is also work on a new glossary because we don’t all agree on what certain terms mean, and Florian Friedrich is working on the video test patterns,” Chinnock said.
“As a member of the HDR Working Group it is assumed you are going to be an active participant, have some ability to test displays or contribute to the metrology. Within the HDR Group is a sub group that is more of a monitoring group. We don’t expect you to be active, but just to monitor what is going on.”
The ICDM has face-to-face meetings. There was one in March in San Diego. These run over three days and cover a lot of the action taking place in the various working groups and committees to develop new specifications and standards. There was a brief meeting at SID a couple of weeks ago in LA, and another one is scheduled for September in Korea, Chinnock stated.
The focus of the new HDR metrology now is trying to understand how the native display performance is intertwined with the processing that goes on. So in many cases today we measure the whole system, but we’re also trying to understand that the display part of HDR is separate from the processing.
“From a philosophical point of view there is a matrix that this group is working on and they define a whole bunch of different categories,” said Chinnock. “But each of these categories — contrast for example — may have different tendencies. They have spatial dependence, temporal dependence, colorimetric dependence and metadata dependence.”
The members are trying to define this and bring it down into a set of a particular matrix that takes specific test methods to quantify and qualify the different variables.
One of the things that video test-pattern expert Florian Friedrich has been working on is developing dynamic test patterns instead of static ones. This alters the measurable criteria, particularly with HDR, because of the way TVs process the signal.
“Historically, the indusgtry has used static test patterns for peak luminance and black levels, and that’s a big area of measurement that has to change going forward,” said Chinnock.
Another committee is working on color volume and has generated a lot of participation. The current 5.31 section document in Vol. 1 was developed for SDR displays.
The primary new procedure was written based on CIELAB for testing and validation.
It has determined a 10×10 matrix cube is best (601 data points ~2% accurate), but 5×5 will be sufficient for practical use (~5% accurate); while 3×3 is unusable because it may create too much error.
Included is an option to use the Fairchild or modified Bradford Chromatic Adaptive Transforms to go from the XYZ data to a human color appearance model. So, white appears white regardless of the power spectrum of the illuminant.
Other work under consideration is drafting a possible new section (5.32) for color volume accuracy based on CIELAB and DeltaE2000.
Chinnock said there had been some concerns about the shortcomings of CIELAB as a color model and possible alternatives could include: CIECAM or ICtCp.
Meanwhile, the ICDM contrast committee is doing some round robin testing right now
on new procedures for analyzing HDR displays. This is focusing on gray scale and gamma of 17 levels or more luminance loading to determine a display’s sensitivity to white luminance percent of the total screen area (APL). The committee is also investigating starfield contrast to determine the display’s dimming characteristics.
By Greg Tarr
Have a question for the HD Guru? HD GURU|Email
Copyright ©2017 HD Guru Inc. All rights reserved. HD GURU is a registered trademark.