Re: Gamma (Andrew Rodney)
Re: Gamma (Andrew Rodney)
- Subject: Re: Gamma (Andrew Rodney)
- From: "Farnau, Ryan" <email@hidden>
- Date: Wed, 25 Jul 2007 15:01:23 -0700
- Thread-topic: Gamma (Andrew Rodney)
Here's an example of how I try to teach my students about TRC (a rough idea
of a gamma response) in output. I created a "gray target" in Adobe RGB 1998
and in ProPhoto RGB, where there are 3 rows of 20 1/2" x 1/2" squares each
with RGB values increasing in 1, 2, and 4 level increments. So, the first
row goes from values of 0,0,0-20,20,20 the second 0-40, the third from 0-80.
While this only represents the "bottom end" response of the color spaces-
what we're interested in is representing the difference between our working
color spaces and the response of custom Epson icc profiles for specific
media types.
We then convert a duplicate of these targets to profile using several
different methods Perceptual, RC, AC, with BPC on and off...then look at the
shift in the numbers- and view the resulting print. I call it a "black
response test". What you see in the RGB numbers is a " bumpy curve" in that
the conversion algorithms are shifting the working color space values to
those of the given media and ink set profile. I actually have them "plot"
the RGB shifts of each square on a graph resembling the curves dialog in PS
- where black is 0 and the brightest value is 80. While this is only the
bottom end of possible tonal scale- it provides the students with "hard" (at
least practical evidence of the shifts that are occurring). They see that
the tonal response of a particular paper and ink set will shift dramatically
from one to the next - and they see it as practical RGB color data that they
can reference when imaging files in Adobe RGB or ProPhoto - to make
considerations as to where to "target" black point for output on a specific
media.
What we're also doing is just trying to find the point at which a
specific media, ink set combination (profile) will produce a black with
separation. I know it's a rough method, but it works.
Ryan Farnau
On 7/25/07 2:07 PM, "Roger Breton" <email@hidden> wrote:
>> So, could it be an irregularly-shaped, "bumpy" curve as well, as opposed to
>> the smooth- and even-looking curve one obtains from a straight power
>> function?
>>
>> Marco Ugolini
>
> Seems that way from my recollection of HP sRGB response "curve".
>
> Roger Breton
>
>
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> Colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
>
u
>
> This email sent to email@hidden
_______________________________________________
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