Re: Total Dot, Number of Patches & "Reading" Devices
Re: Total Dot, Number of Patches & "Reading" Devices
- Subject: Re: Total Dot, Number of Patches & "Reading" Devices
- From: email@hidden
- Date: Fri, 26 Apr 2002 10:43:20 EDT
In a message dated 4/26/02 9:03:05 AM, email@hidden writes:
>
Measure patch data is used to is the construction of the profile grid.
>
Keep
>
in mind that a 17x17x17 grid would require something on the order of 5000
>
patches to construct a table directly form the data. Suffice it to say,
>
there is some calculation done at the grid-building stage. Once the grid
>
is
>
built interpolation is then used to describe the color values between the
>
grid points. More patches should yield a more accurate grid values and
>
more
>
grid points should give you better interpolations. For example, in
>
MonacoPROFILER4.0, a user can choose both the number of patches
>
(530,917,1379,2989) and the size of the grid (17x, 21x, 33x). Any MP4.0
>
output profile can be reloaded for settings evaluation and/or
>
reconfiguration - in other words, you can see all the settings for the
>
profile prior to processing. Tools such as Colorthink and ProfileViewerPro
>
are invaluable in understanding how your profile is working after it's
>
made.
Monaco Profiler is one of the more flexible solutions for controlling these
factors. Some users feel no need to control them, and find it confusing...
and for very linear, predictable devices the number of patches and the size
of the grid are both far less important. This is more or less the EyeOne
Match philosophy. But I find that for inkjet printers the optimal number of
patches is far above Match's two hundred and some odd patches... more in the
range of 800, with diminishing returns above a thousand. This is assuming
control of the grid size, or a fixed grid size large enough to utilize the
extra data.
That said, there are lots of other factors, and its easy to blame number of
patches or size of grid when in fact its the actual matching technique, or
method of mending holes in the profile, or perhaps the accursed ICC blues
shift, or optical brightener whitepoint issues that are at fault.
C. David Tobie
Design Cooperative
email@hidden
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.