Re: [Lcms-user] Display gamma

[Graeme G. <gr...@ar...>]

Richard Hughes wrote:
> Yes, an interesting read, thanks. It doesn't seem to address the fact
> that LCD displays can't show absolute black -- so the nice graphs
> showing a power curve that goes through (0,0) is wildly different to
> reality.

Yes. The only reference to this I'm aware of in industry literature
is BT.1886, but I haven't really gone searching. There are several
papers around on characterising CRT responses which may have dealt
with it to some degree, although it can be swept under the carpet
with CRT's due to the ability to get low black levels.

> dim surround. If it is displayed without correction in a bright
> ambient, it will appear contrasty. In this circumstance you should
> apply a power function with an exponent of about 1⁄ 1.1 or 1⁄ 1.2 to
> correct for your bright surround." -- there must be more maths here.

Formerly, no, there isn't a whole lot of maths, since it's all
based on Psychometric measurements. See Giorgianni & Madden's
"Digital Color Management, Encoding Solutions" to get the
flavour of this, while Mark Fairchild's "Color Appearance Models"
has a more current representation of the basis on which this
this type of adjustment can be made. I'm sure there's
lot of underlying data in various books by Hunt too.

[I'd imagine the original TV adjustment was made during the development
 of the technology and standards, "by eye" based on viewing tests. The
 Color Science basis for the adjustment wouldn't have had to been
 known by any involved, although the general principles may well have
 been known for a long time. It appears that the quantification of
 various such affects was done in the mid 60's by Hunt, Stevens,
 Bartleson, Breneman etc.]

In Argyll's dispcal I've made use of the Psychometrics encoded
in CIECAM02 to do the gamma adjustment. It's all kind on moot
for color managed applications though.

Graeme Gill.