Re: New EyeOne ruler and soft case
Re: New EyeOne ruler and soft case
- Subject: Re: New EyeOne ruler and soft case
- From: Andrew Rodney <email@hidden>
- Date: Mon, 31 Oct 2005 19:27:25 -0700
- Thread-topic: New EyeOne ruler and soft case
On 10/31/05 7:21 PM, "Marco Ugolini" wrote:
> By the way, isn't it also risky to edit a profile based on what looks
> "right" in any one particular image?
Perhaps. I've tried to do it numerically too. Output say a gray patch,
measure it with the Spectrophotometer, try get it closer to neutral "by the
numbers". That sometimes don't work either (don't ask me why). I also
suspect the rendering intent you're editing plays a role. Be real cool to be
able to edit by measuring or by eye but it doesn't always work as planned.
> How does one know that a specific edit
> that looks good on a certain image will not cause unwanted shifts on another
> that has a different color structure?
I use a test image with all kinds of different imagery and synthetic
elements. One thing I can say is this. People with profile editors often
start editing based on too small a sample of output and end up spinning
their wheels. I'd output a lot of images before even considering IF I need
to edit the profile and then where.
> There is a definite risk of arbitrariness in visually-based profile editing
> that may possibly end up defeating the very purpose of creating and using
> profiles in the first place.
Might explain why I've had a common experience of improving a profile
through editing and getting almost done, then spending hours trying to get
that last 1% of my goal, only to go several steps backwards. Man's got to
know his limitations. Man's got to know when to stop editing. Now if it's
OPM and you're paid by the day... Just kidding.
> If edits are to be done visually, shouldn't we make sure that they are
> effected using a group of standard images that contain as wide a range as
> possible of color situations (skin tones, neutrals, saturated colors, etc.),
> a bit like the images that are traditionally seen accompanying a press
> proof?
Absolutely!
Andrew Rodney
http://www.digitaldog.net/
_______________________________________________
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