Re: Epson canned profiles
Re: Epson canned profiles
- Subject: Re: Epson canned profiles
- From: ben <email@hidden>
- Date: Tue, 17 Apr 2018 10:22:54 -0700
- Sun-java-system-smtp-warning: Lines longer than SMTP allows found and truncated.
On Apr 17, 2018, at 9:38 AM, Andrew Rodney <email@hidden> wrote:
> It's invisible to the Standard Observer.
Andrew, why do you do this to yourself?
Fire up your favorite image editor.
Fill half the screen with L*a*b* 6, 1, 4. Fill the other half with 7, 1, 4.
Overlay a couple patches of 6, 0, 0 and 7, 0, 0.
By your very own math, the two background halves differ by 0.85 dE and are as
close as your monitor is going to get to what you claim you're getting out of
your printer. The two overlays will be as close to the neutral axis as your
monitor will get; they'll also be less than 0.85 dE from the two halves.
Anybody here can do this and instantly confirm that the differences are _not_
invisible, that the one half is visibly darker than the other, that the
overlays are quite noticeably cooler than the background halves, etc., etc.,
etc. (Unless, of course, your display profile is abysmal.)
So how on Earth can somebody simultaneously take pride in calling himself a
color professional _and_ be so proud of being colorblind?
(And the Standard Observer isn't either standard nor an observer. There're at
least a half-dozen CIE observers, from the popular one from 1931 to J&V in
1978, ones preferred by the textile industry rather than the ones graphic arts
likes, etc, etc., etc. And they're not observers; they're small spreadsheets
that give the subjective S, M, and L responses to monochromatic samples at
typically either 10- or 5-nm intervals. Nothing in any of the observers even
remotely addresses the question of color resolution; they simply characterize
the relative spectral sensitivities of the differently-pigmented photoreceptors
in the eye. They don't even specify sensitivity to different light levels; only
color.)
> You think that the vast majority of printers send more than 8-bits per color
> to the print driver?
The Canon iPF line has sent 16-bit data via the Photoshop plugin for at least a
decade or so. The same time they retired the plugin, they updated the driver to
natively send 16-bit data. As you're so fond of noting, I'm not an Epson user,
so I'd have to look up when or if they support higher color resolution.
(There's a close interplay between spatial and color resolution, in that you
can trade the one for the other -- how halftoning creates the appearance of
grays despite there only being black ink and white paper, or how antialising
creates the appearance of smooth text despite a display being a mere 100 ppi or
so. Printers tend to have very fine spatial resolution and so you can use some
of that to effectively increase the color resolution. At 3K DPI you can afford
to work with lower color resolution.)
> As such, it’s best to talk about encoding as having a potential to define
> millions or billions of numbers, DEVICE VALUES, that could be associated to
> a color value and thus color, if we can see them. If you can't seem them,
> they are not colors.
Erm...Andrew...gamut and encoding resolution are entirely different, completely
unrelated. ProPhoto mathematically encodes colors that simply don't exist,
tristimulus values that the nervous system doesn't create. Think, "bluer than
blue," or the sorts of things people describe after an acid trip. (It's not
even a case of wavelengths outside of the visible spectrum; it's spectra that
don't even exist in the first place. You'd need "light" that makes your optic
nerve have less-than-no response.) That has nothing to do with the bit depth of
the image. It would be the case if you encode ProPhoto with four bits or forty
bits.
Conversely, you could throw as many bits as you want at a grayscale encoding
and you _will_ see each and every color it encodes. It might encapsulate more
subtlety than you'll be able to distinguish, but you'll have no trouble seeing
all the colors it encodes.
You're confusing resolution with volume? You don't know what the Standard
Observer actually is? And you're accusing others of a lack of education!?
b&
_______________________________________________
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