Re: Media Testing for maclife.de
Re: Media Testing for maclife.de
- Subject: Re: Media Testing for maclife.de
- From: Eric Chan <email@hidden>
- Date: Sun, 14 Sep 2008 22:30:09 -0700 (PDT)
> I'm a bit at odds with the
> scene-referred/output-referred distinction
> for theoretical reasons and prefer to think in process
> chains. For my
> evaluation, I'm taking the "earliest" result
> a user can get hold of in
> a specific process chain.
It's a very important distinction, because it comes down to what you're trying to evaluate.
If you're doing scene-referred colorimetric evaluation, then this is much more in line with "what the camera sees" because everything is still linear. The evaluation needs to be done on the linear data, which is rarely done by the home user because it can be very difficult (sometimes impossible) to get the linear data out of the raw converter. (Most of the difficulty comes from the fact that, if such an option exists, it's not documented; Camera Raw falls into this camp.)
In my view, doing output-referred evaluation is more practical in terms of studying whether the output is pleasing or useful to the user, but is much farther removed from the camera because it is subject to non-linearities introduced by the raw converter (almost always a strong brightening tone curve, and a bit of an S curve).
> Anyway, I use images of a Digital ColorChecker SG chart
> shot under
> seven different lighting conditions from bright sunlight to
>
> fluorescent and halogen light, and calculate mean/max delta
> E (1976)
> for each of them, then average the result for all 7
> lighting conditions.
> The mininum delta E would be most "metrologically
> exact".
Where did you get the reference patch values for the SG chart?
_______________________________________________
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