You can be as snippish and haughty as you want, Derby, but your
comment about "Bayer demosaicing" didn't make any sense and had
nothing to do with the issue. It was nonsense speak: I had no idea
what you were talking about.
You've articulated what you intended to say much more clearly since I
responded. I'm sure Dave will appreciate the clearer explanation of
why the red channel might seem less sharp.
G
On Sep 27, 2008, at 7:36 PM, Derby Chang wrote:
I know you understand what I mean. Dave asked why his red channel
looked soft. I meant that looking at just one channel after
demosaicing was not looking at the full resolution of the sensor
I humbly apologise to your Holiness I was not clear.
D
Godfrey DiGiorgi wrote:
Then I don't understand your sentence below. Dave is looking at
individual channels in RGB which, as you say, will be softer than
the full three channel composite.
At this point in the process, Bayer demosaicing and gamma
interpolation happened long ago ...
So what does "The Bayer demosaicing will then be trying to
interpolate from what information it has" have to do with the fact
stated above?
G
On Sep 27, 2008, at 5:45 PM, Derby Chang wrote:
Yes, that's my point. If you are looking at a predominantly green
image, the detail is in the green. Looking at the red channel
means you are looking at rather less resolution than the full RGB
image
D
Godfrey DiGiorgi wrote:
When you're looking at an RGB channel file in Photoshop, the
Bayer demosaicing is long past.
Godfrey
On Sep 27, 2008, at 4:37 PM, Derby Chang wrote:
Don't forget, Dave, individual channels will be "softer" than
the full RGB image, especially if you are looking in one channel
and the most of the colours are in the other two. The Bayer
demosaicing will then be trying to interpolate from what
information it has.
D
David J Brooks wrote:
Good, its not my eyes.:-)
I tried adjusting sharpness on that channel but did not see any
change. I;'m sure i;m doing something wrong, or, the photos
from the
D200 ARE really soft.
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow
the directions.
