Re: Epson7000: Absorption/ReflectanceReadings
Re: Epson7000: Absorption/ReflectanceReadings
- Subject: Re: Epson7000: Absorption/ReflectanceReadings
- From: Chris Murphy <email@hidden>
- Date: Fri, 29 Dec 2000 11:57:25 -0700
>
1. Even though this sheet is coated, is there a possibility that the
>
inks would absorb "more" in these more saturated areas? I built a curve
>
to try to alter this and it's worked pretty well. I tried to isolate
>
only the 90-100 areas.
Sounds like maybe there is a lack of linearization before making the
profile. That is, perhaps there's just too much ink being laid down on
the paper above 90% coverage. I'm not exactly sure what the tecnical term
for "moosh" is :) so I can't say with certainty this is the problem.
>
2. Why do profiles have both a name and a description? It seems that
>
sometimes a profile is known by one or the other in a pull-down menu.
>
What's the purpose in this two-pronged approach? It seems only to risk
>
confusing matters.
Probably as a way to deal with how short file names are, and especially
were at the time of the adoption of the ICC. I don't know what the limit
is, but it's a lot more than 33 characters.
>
3. I also added Digital Swatchbook to the mix this week. After setting
>
it up and calibrating it, I notice that even though my 100% black chip
>
in the 21-step wedge LOOKS totally black, the reading from the
>
spectrophometer reads out at about 31/32/35 RGB.
Try L*a*b* which should get your black closer to a L* of around 2-4. I
don't even know what RGB "space" is being used to base those numbers on
(maybe CIE RGB?). They are high for RGB numbers, but that's not an ideal
way to report color from a spectrophotometer. L*a*b* is something that's
used in a variety of applications such as Photoshop, and is device
indepedent, unlike RGB.
>
I would think it would
>
be definitely 0/0/0, because the black is so saturated.
Well black can't be saturated, that's why it's black. A perfect black is
perfectly desaturated (well so is any perfect gray for that matter).
>
This seems to devalue all the other reading
>
that we got along the wedge, although the upper values seem to coincide
>
more with the RGB value from the InfoPallette.
Comparing RGB values between the info palette and a measurement device is
like comparing giraffes and oranges. You not even in the "they're both
fruit" category. I'd look into using Lab instead.
Chris Murphy