I asked a few months ago about the difference between manually doing Levels
and using the Auto Levels command because when the program does it
automatically there are no holes in the histogram and whenever I did it
manually I got a lot of spikes with spaces around them. At the time I was
told -- and it sounded convincing -- that the program does it individually
on each of the channels separately and then merges the result which
produces
a continuous histogram.
WELL ... I've tried doing it channel-by-channel (R/G/B) manually about a
dozen times now and I still get a combined RGB histogram that shows many
missing levels, much as it did when I just leveled the whole image.
SO ... I'm asking again: what does the program do when applying Auto
Levels
that results in a continuous histogram that can't be duplicated manually?
Does it intentionally spread out the levels so that there are no holes?
TIA
Norm
