Re: Profiling the ip5000
Re: Profiling the ip5000
- Subject: Re: Profiling the ip5000
- From: Marco Ugolini <email@hidden>
- Date: Mon, 15 May 2006 01:57:06 -0700
In a message dated 5/14/06 8:52 PM, Dana Rasmussen wrote:
> Here is my thinking. Maybe I am completely wrong.
> The test target generation algorithms, such as the one used by Profilemaker
> measuretool, are based on a printer using CMYK, and maybe lC, lM.
Hi Dana.
Testchart-generation algorithms, as you call them, have no direct relation
to the color *gamut* itself of the device being profiled. Other than for the
mode in which they operate (RGB or CMYK), testcharts are not
device-specific. They just contain a list of color *numbers* (like for
example R 35, G 106, B 210, and such) which are sent to the printing device
through the printer driver, hopefully with no additional color management
being done on the color values by the driver itself.
> Can we assume that these algorithms are going to work to effectively profile a
> printer that has all the above, plus RGB? Do the target sets include colors
> that simply cannot be printed without RGB inks?
Let me say it this way: the same color numbers in the testchart produce
different results on different devices. In devices with a small gamut the
results have a small gamut; in devices with a larger gamut, the results have
a larger gamut.
There is one thing having to do with testcharts that may improve the
profiling procedure, though: it's probably desirable to use testcharts with
a higher number of patches when profiling devices with a larger gamut, since
the color range has a wider span, and more sampling points provide a more
accurate picture of all those intermediate points that would otherwise have
to be interpolated, i.e., guessed at by the profiling software.
It takes a while to wrap one's head around this aspect of color management,
but it becomes clearer in time.
> One would think that these extra colors would allow a much larger gamut than
> a CMY inkset would allow, but viewing the profiles I have made so far with
> Profilemaker, and comparing Epson 4800 profiles for the same paper, I am
> seeing some extension, but certainly not what I was expecting.
This is what I think is happening: you feel disappointed because the results
are coming short of your expectations. But perhaps your expectations need to
be redefined. Or perhaps you are right to expect more, and there is
something in the way the testchart is being printed that is limiting the
gamut of the output.
Are you sure that the printer driver is not adding its own color management?
Or that *your* color management choices are correctly set? (For example, in
"Print with Preview", leave file untagged and choose "Same as source" in
Photoshop CS, or "No Color Management" in Photoshop CS2.)
Regards.
--------------
Marco Ugolini
Mill Valley, CA
_______________________________________________
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