David Townend wrote:

> Thomas Holm wrote :
> 
>> Something like an Eye one will do the job very nicely.
> 
> Do I need to get a special target for Small Gamut Inks with different
> Reference Data though ? Will not the colour targets supplied with the iOne
> Pro kit (9.1B RGB and 1.5 RGB) give strange results when attempting to make
> a profile for a  SG ink-set and paper combo ?

No you don't need special targets, the standard ones does a good job.

Essentially building profiles work like this:
1). You print some patches of known colour numbers, say 128,128,128. Nobody
knows which colour this value is on a random printer (and please forget the
notion that it should be a 50% grey, that only works on linear working
spaces like Adobe RGB, ColorMatch etc.).
2). Out of the printer comes a patch with a given colour, that was obtained
by printing the above mentioned value.
3). This patch is measured by a spectrophotometer - essentially making a
reading that objectively will tell a computer how we humans perceive this
particular colour in daylight.
4). Print the other 917 patches and measure them, effectively telling the
computer how your printer prints 918 different colour numbers, and where
they end up in CIELAB (lab color). These colours may be muted or vivid but
it doesn't matter, the profile tells the computer what colour you get when
you print a certain combination of RGB values. The values that were not
printed (all the ones in between the colour numbers we sent to the printer)
can be interpolated from nearby grid points.
5). Send an image (50% grey - RGb 128,128,128) in a known colour space to
your printer. In the background the colour numbers in the RGB file will be
translated/converted to a certain lab value. Then it is converted from Lab
to your printer profile.
Remember that the computer known that the 128,128,128 value you printed
originally corresponds to, say, L57 a-1 b+3 on your printer. However one of
the other patches (maybe 123,137,115) equals a lab value of L50, a0, b0. As
you want to print a 50% grey that value in Adobe RGB(1998) is thus converted
to the colour numbers your printer needs to obtain a colour that a human
observer will decipher as a 50% grey.

Whether the gamut of the printer is huge of small doesn't matter much - the
process is the same. A case could be made for generating a bunch of custom
patches that would all fall within the gamut of Small gamut inks, and
presumably this might improve the quality of the profile IF THE PRINTER IS
NOT (I repeat NOT) reasonably linear, as you would have more patches in the
in-gamut region and thus shorter between interpolation grid points. It does
thak A LOT of tinkering around to achieve this though... If the printer is
fairly linear it doesn't really matter though. I've profiled SG printers on
both targets 288/918 and the difference between the two is marginal, which
makes me believe that the inks are reasonably well balanced, and thus that
an 1160 with SG inks are reasonably well linearized.

Best Regards

Thomas Holm / Pixl ApS

- Photographer & Colour Management Consultant
- Adobe Certified Training Provider in Photoshop�
- Apple Solutions Expert - Colour Management
- Imacon Authorized Scanner Training Facility
- Remote Profiling Service (Output ICC profiles)
- Seminars speaker and tutor on CM and Digital Imaging etc.

- Home Page: www.pixl.dk � Email: th[AT]pixl.dk
-- 


===============================================================
GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE

Reply via email to