Re: LCD color calibration
Re: LCD color calibration
- Subject: Re: LCD color calibration
- From: Florian Höch via colorsync-users <email@hidden>
- Date: Mon, 08 Jul 2019 00:28:22 +0200
Am 07.07.2019 um 16:33 schrieb THOMAS A LIANZA via colorsync-users:
> Communication is possible to nearly all displays using some level of
> the DDC 2.0 specifications.
Not always very reliable though, and the DDC specification(s) lack
consistent implementations, which seems to prevent general applicability
(i.e. to me it looks like calibration software vendors are mostly forced
to support DDC capable displays individually and often cannot rely on a
common set of capabilities or even instruction sets).
> A wide gamut panel will necessarily have an input lookup table, a
> matrix or 3D out, followed by a display look up table.
While a matrix or 3D LUT is not a requirement, it's surely nice-to-have
for supporting things like an sRGB mode, and/or "advanced" calibration,
inside the display itself.
> In general, use of the graphic card LUT is not recommended at all
> in most modern calibration environments.
Well, we'd need to define "most". My definition of "most" with regards
to (computer) display calibration would encompass typical desktop
systems which either run a version of Windows (very common), macOS (less
common) or an open source desktop (e.g. GNOME/KDE, even less common)
paired with a run-of-the-mill LCD monitor.
Unless Microsoft (same as Apple) decides to no longer support setting
graphics card 1D LUTs ("videoLUT") via OS APIs, I don't see this
functionality going away anytime soon, and it is still the main
functionality that most calibration packages are relying upon (due to
widespread support across different platforms).
> If a display has an HDMI interface, the graphic card LUT is normally
> inaccessible because there are a number of assumptions about the
> display color management.
Hmm. My own experience thus far has been that the actual physical
connection or even interface seems to be irrelevant regarding whether or
not typical videoLUT functionality will be available.
The only requirement really is that support for the respective
generalized OS APIs is implemented at the graphics driver level (and
there are some USB3 external "graphics cards" which indeed seem to not
implement it). Whether that means actual hardware LUTs, or emulated in
software, is up to the implementation.
> The current processes generally calibrate the displays to a given
> color space and use a very simple ICC profile to describe the
> display setup.
This assumes rather high-end display hardware though, that not everyone
owning a calibration device will have access to (certainly more common
in professional or semi-professional environments).
> As the industry moves towards HDR (high dynamic range) and Wide Gamut,
> calibration occurs in the display, not on the driving platform.
It may be a bit too early to anticipate, but FreeSync2 HDR which is
(imho) quite likely to see some adoption (with even major LCD TV vendors
onboard) seems to move in the opposite direction, with most or all HDR
processing delegated to the graphics hardware in the connected system(s).
> The ICC profile is used as a container for other information in HDR
> systems in Windows 10.
AFAIK, videoLUT still applies in HDR mode, although personally I
currently have no ability to check, lacking a HDR-capable display.
Cheers,
Florian Höch
_______________________________________________
Do not post admin requests to the list. They will be ignored.
colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden