Re: Decimal precision in Photoshop's info palette
Re: Decimal precision in Photoshop's info palette
- Subject: Re: Decimal precision in Photoshop's info palette
- From: Marco Ugolini <email@hidden>
- Date: Thu, 18 May 2006 23:45:43 -0700
In a message dated 5/18/06 5:28 PM, CS Carl Stawicki (4211) wrote:
> Yeah, RGB gives you those high precision numbers, but I never understood why.
> To me, 8-bit should mean integers from 0-255. There's more going on under the
> hood than I understand.
Hi again, Carl.
Ah, look what you've done: now I can't stop plating with these scripts of
yours... :-)
There is something else that I have found out, and that makes me question
how precise Photoshop really is after all in performing its internal
calculations.
Please indulge me and try this:
1) Open a new CMYK document in Photoshop CS, of any size you wish;
2) Click on the color picker's foreground patch, and enter the following
color values in the Color Picker dialog box: C 14, M 89, Y 64, K 22;
3) Run your "get the foreground color" script; the results will be
{class:CMYK color, cyan:14.0, magenta:89.0, yellow:64.0, black:22.0}: just
like they should be. All good so far;
4) Now fill the CMYK document with this very same foreground color (using
either Option-Delete or Shift-Delete at 100% -- it doesn't matter which of
the two ways);
5) Now sample the color in the CMYK document with the eyedropper; then run
again the "get the foreground color" script;
6) The results this time will be {class:CMYK color, cyan:14.12,
magenta:89.02, yellow:63.92, black:21.96}. The colors have shifted, albeit
slightly, but nonetheless they are not exactly what they are supposed to be.
If you try this procedure in RGB, you will find that the numbers you enter
in Color Picker full are never full integers (except for 0 and 255,
apparently); on the other hand, they remain unchanged after filling the
document with the color and sampling it.
In Lab, the values entered by hand as integers are shown as integers when
retrieved with the script, but after the fill-resample routine, the L* value
shifts slightly, while a* and b* remain unchanged.
This all feels very puzzling and inconsistent to me. It may have something
to do with the "quantization" problems I have heard mentioned at times in
imaging circles, perhaps? But then, why the different behaviors in different
color modes?
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