Re: : Dumb question - Printing Colorchecker
Re: : Dumb question - Printing Colorchecker
- Subject: Re: : Dumb question - Printing Colorchecker
- From: <email@hidden>
- Date: Wed, 26 Oct 2005 12:11:01 -0400
Edmund,
On Wed, 26 Oct 2005 16:21:38 +0200
edmund ronald <email@hidden> wrote:
Hi Danny, thanks for your support.
1. I used i1share 1.4 to make the measurements. Which
version does what re. conversion?
If I recall correctly, version 1.3 does Rel. Col and
version 1.4 Abs. Col.
2. I exported the data into a Photoshop palette in
AdobeRGB (probably
my mistake here, I'll try saving into LAB).
That is the solution. The L*a*b* values obtained with
MeasureTool are also good.
I then pasted the values
by filling the various squares, into the AdobeRGB
version of your
Colorchecker. I'm starting to understand this may have
been a mistake, and will do some checking myself.
You could have replaced the RGB of any of the RGB charts I
provide since they do not have embedded profiles; however,
the chart frame is labeled with the proper color space
title. Also, when you open the chart in Photoshop, you
should assign it the RGB space for which the values were
computed.
Using RGB data gives you somewhat more precision than
L*a*b* in Photoshop since all L*, a* and b* values are
rounded to integers in Photoshop. However, you need good
RGB values to start with (you can use BabelColor to
measure them one by one, using the Eye-One, by selecting
"Adobe RGB" and "L*a*b*/L*u*v* input" in the main screen).
If you use L*a*b*, check that the entered value is the one
displayed since sometimes Photoshop plays some trick and
shows a higher or lower value. Also be sure to use L*a*b*
D50 data (not D65) in Photoshop.
3. I'm appending the various data -spectral and palette
- to the copy
of this email that is being sent directly to you. If
somebody wants to
write a macro to automate this stuff ...a macro to
compute the errors
from the print would also be welcome ...
Thanks. I will give you personal feedback on these.
4. My paper settings were the same as used for
profiling, and the
paper profile was applied, for the print (not for the
testchart). The
relative colometric print was actually pretty good,
optically. However
I had issues with the red and skin color, which made me
measure the
plastic colorchecker sample I was using, which seems
different from my
memory of my previous cardboard chart. A friend made the
same remark
abut a memory color difference.
5. As far as I can see, the red is well within the
profile's gamut for
this glossy paper, I checked with Colorthink.
6. I really would like to have a standardized procdedure
for
measure-print-test for the colorchecker documented
somewhere, with
some macros, maybe to do it faster.
The exported i1 Share RGB data is likely the main issue
here.
The comparison between your just-purchased ColorChecker
and the others should clarify the "memory colors" effect.
But... you mentioned "plastic" chart, did you make your
measurements on a ColorChecker SG (Semi-Gloss)?
If yes, then almost all colors which represent the
traditional ColorChecker on this new chart are different
from the original ColorChecker and should not be compared
with these. Are they different enough to trigger memory
effects?
Danny Pascale
email@hidden
www.BabelColor.com
---------------------------
On 10/26/05, email@hidden
<email@hidden> wrote:
Hello Edmund,
Hope you had a good vacation where there is no storms,
no
earthquakes, no tsunamis, no ozone holes, no
uncalibrated
printers, no...
You say:
>I have an image of the Colorchecker
> from Danny Pascale's site, updated with my own
sample's
>values.
This is the L*a*b* version (I suspect!)?
If not, then have you used i1 Share to get your RGB
data?
i1 Share gives you RGB in Abs. Colorimeteric (the older
versions of i1 Share gave Rel Col).
From Photoshop (File/Print with Preview...):
Let Photoshop determine colors
Printer profile: your calibrated printer profile
Relative colorimetric, black point compensation
Have you set your printer color management to "OFF" and
selected the same paper as the one used for profiling?
You may just have forgot that one after your vacation!
Hope this helps.
Oh!, since you have a new ColorChecker, could you send
me
a file of the spectrum data? (Any format, ASCII, CXF,
etc)
I would compare it with the average shown on the web
site.
(Thanks!)
Danny Pascale
email@hidden
www.BabelColor.com
_______________________________________________
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