Re: perceptual differences in Lab deltaE
Re: perceptual differences in Lab deltaE
- Subject: Re: perceptual differences in Lab deltaE
- From: Wire ~ via colorsync-users <email@hidden>
- Date: Tue, 7 Jan 2020 14:56:19 -0800
Whoops, Roger, this Dell a UP2516 not a 2916. I think it's interfaces
support 10 bpc but MacOS doesn't drive it as such. I need to look into this.
Per your query the low-order bits are truncated and the high order sent to
your display. They're there internally and used to inform the least
significant bit (LSB), Using image adjustments, the information in those
hidden bits may be dragged up into the passband of your image or display
(so to speak), where you can take advantage of whatever information is
stored in them as part of the rendering. Photoshop calculates them. What do
these offer? If you are coming from raw, you get access to latitude that
would otherwise have been clipped to reach 8 bpc JPEG. In synthetic data
and image adjustments, you get better control of quantization noise.
As an aside, something I am coming to terms with is how display-centric our
view of color is. The reason I got these WCG displays is I want to
experiment with my presuppositions about the importance of color to
image quality.
I have come to a conclusion that I think may be generally interesting:
sRGB will not produce a color that looks unnatural. Like if you could do
anything you wanted with a format, and you were concerned about being able
to render all common scenes faithfully, but also concerned about
uncontrolled use creating freakish output, sRGB color limits are an
excellent compromise of these concerns. WGC readily makes colors that are
unnatural. In an increasingly hyper-real world of media and industrial
objects, the time has certainly arrived for wider gamuts. And I experience
my adaptation to color changing with ready access to WCG,
My doctrinal axe to grind is that while sRGB (and 709) has a poor
reputation of being an impoverished space, its design merits are
underappreciated. It's very interesting to watch the consumer space adapt,
with surprise that what seems to me to be unnaturally hot color is become
part of a new generation's view of the world.
On Tue, Jan 7, 2020 at 2:04 PM Roger Breton via colorsync-users <
email@hidden> wrote:
> Andrew,
>
> I'll start "humble", I'm only interested at this point to tell what type of
> "information reduction" is at play between pixel values in a file and
> measured pixel values on screen. For sure, we all *know* it's not a 1:1
> relationship. Like, this is crazy, but I confess I never stopped to think
> what happens in Photoshop when working in 16bit modes? Suppose my video
> card
> can only honor 8-bits: can you imagine the huge "reduction" going behind
> the
> scene between the image in the file and the image on screen? Now what
> happens in the case of a 10-bit video card scenario? Assuming a 10-bit
> capable display like your 271Q and my 271W, and Wire's UP2916? Surely the
> reduction must not never be as severe. But I'd like to know,
> quantitatively,
> what kind of "compression" are we talking about. For now, I'm only
> interested in looking at R=G=B values, that will give plenty of data to
> graph to play with. But I don't think I'll expand this to the whole 16.7
> million (let alone 1.04 billion) colors.
>
> / Roger
>
> -----Original Message-----
> From: colorsync-users
> <colorsync-users-bounces+graxx=email@hidden> On Behalf Of
> Andrew Rodney via colorsync-users
> Sent: Tuesday, January 7, 2020 3:25 PM
> To: Roger Breton via colorsync-users <email@hidden>
> Subject: Re: perceptual differences in Lab deltaE
>
> So if (ugh that be hard) you're going to use your 16 million color test
> file, what's going to happen if it's in Adobe RGB vs. sRGB vs. ProPhoto RGB
> vs. the gamut and calibration of your display?
>
> Andrew Rodney
> http://www.digitaldog.net/ <http://www.digitaldog.net/>
>
> > On Jan 7, 2020, at 1:21 PM, Roger Breton via colorsync-users
> <email@hidden> wrote:
> >
> > Ilah,
> >
> > Glad you brought this up because that's exactly one my current "pet
> > projects". My code is not finished yet but basically, I'm going to be
> > sweeping the entire "1024" scale with my instrument, to determine,
> > exactly, what is what. I'll be glad to report as soon as I'm done --
> can't
> wait!
> >
> > / Roger
>
> _______________________________________________
> 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
>
_______________________________________________
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