Small addition to Ron's e-book w/permission

Differences can arise if Gray Gamma 1.8 is used in (PS CS) Color Settings. I struggled with this for a while then changed this setting to Gray Gamma 2.2 and found that my readings in the Info Palette matched exactly those stated in the eBook. I am using a Mac by the way, but I imagine the same differences would arise if Grey Gamma 1.8 was used on a PC.

Paul
I tested in Photoshop 6 and 8 (CS) on the PC. The values were the
same, but both were different than those shown on the file as
measured in Pshop CS on the MAC.

Ron
Thanks Ron.

There should be no difference in [value] between Mac & PC in value?!?

R100 B100 G100 is grey, yes?! R255 G255 B 0 is yellow.

I receive the same value in PS7 and Corel Paint even though the
perceived color is completely different.
--
Ron Reznick
http://digital-images.net
http://trapagon.com
 
There is no such a thing as color space in strict meaning of the word - we have filter bands and TRC. Color space would be somehow determined by averaging greens and adding red and blue values to a pixel formed from a cluster of four. Obviously, that is very limited approach. Interpolation scheme also plays major role here.
--
no text
 
There is no such a thing as color space in strict meaning of the
word - we have filter bands and TRC. Color space would be somehow
determined by averaging greens and adding red and blue values to a
pixel formed from a cluster of four. Obviously, that is very
limited approach. Interpolation scheme also plays major role here.
I don't agree that the interpolation scheme plays a role in this. Suppose
you capture a large surface of constant color (and illumination). All R,G,B
pixels have the same value (forget noise for a while). Obviously, the
interpolated image will be the one that have these R,G,B values on
each pixel, regardless the exact way the interpolation was done.
So, at least, there is an a priori way of determining how the camera
translates a given color into "unprocessed" RGB values (without WB
and logarithmic tone curves etc...). The exact interpolation algorithm
will determine how well this is done, and have influence on
sharpness and color fringing on edges and so on, but it should not
cause any color change on surfaces of constant color.
It may be stretching the concept a little bit, but one could say that
the recorded values in a raw file are in a certain color space...

--
Vtie
http://www.pbase.com/vtie
 
There is no such a thing as color space in strict meaning of the
word - we have filter bands and TRC. Color space would be somehow
determined by averaging greens and adding red and blue values to a
pixel formed from a cluster of four. Obviously, that is very
limited approach. Interpolation scheme also plays major role here.
I don't agree that the interpolation scheme plays a role in this.
Suppose
you capture a large surface of constant color (and illumination).
All R,G,B
pixels have the same value (forget noise for a while).
and uneven pixel/filter properties, too. Do we have a proof they play minor role here? AFAIK just the contrary...
Obviously, the
interpolated image will be the one that have these R,G,B values on
each pixel, regardless the exact way the interpolation was done.
So, at least, there is an a priori way of determining how the camera
translates a given color into "unprocessed" RGB values (without WB
and logarithmic tone curves etc...). The exact interpolation algorithm
will determine how well this is done, and have influence on
sharpness and color fringing on edges and so on, but it should not
cause any color change on surfaces of constant color.
If you deal with color transitions, it is a bit more complicated. Interpolation schemes take into account neighboring pixels, and different interpolation schemes render transitions in a different manner. Trying to use a color profile not suitable for interpolation is the way to gain color artifacts.
It may be stretching the concept a little bit, but one could say that
the recorded values in a raw file are in a certain color space...
Really depends on definition of color space IMHO
--
no text
 
That makes sense...

Good thinking.

Ron
Paul
I tested in Photoshop 6 and 8 (CS) on the PC. The values were the
same, but both were different than those shown on the file as
measured in Pshop CS on the MAC.

Ron
Thanks Ron.

There should be no difference in [value] between Mac & PC in value?!?

R100 B100 G100 is grey, yes?! R255 G255 B 0 is yellow.

I receive the same value in PS7 and Corel Paint even though the
perceived color is completely different.
--
Ron Reznick
http://digital-images.net
http://trapagon.com
--
Ron Reznick
http://digital-images.net
http://trapagon.com
 

Keyboard shortcuts

Back
Top