Vuescan's new histograms (with sliders)

T

ThomasH

We probably all agree that the addition of the sliders
simplifies adjustment of white and black point during
preview. What puzzles me though is the indication of pixel
information *outside* of black or white point when these
are 0.0%. I would assume that if White Point (WP) is 0.0,
the right slider points to max. intensity and the entire
pixel information must be to the left of that point. An
example with Blue channel being "greater than WP 0.0%":

http://www.pbase.com/phototalk_thh/vue_hist_sliders


My impression is that the sliders have two other problems:

The slider drag-and-drop indication malfunction with WinME
and W2000. After the drop the new position of the slider is
not being indicated until the next preview is finished.
This has a substantial problem: Once you drop the slider
by mistake, you must run a preview to discover where it is!

In many cases the information in shadows or in highlights
is so crowded that a move of the slider even by a single
pixel causes quite massive changes in value of black or
white point. Back than Vuescan used to plot the histograms
over the entire width of the preview window. I wonder if
we could now return to this mode. This would allow finer
adjustments of the sliders.

Thomas
 
W

Wilfred

ThomasH said:
We probably all agree that the addition of the sliders
simplifies adjustment of white and black point during
preview. What puzzles me though is the indication of pixel
information *outside* of black or white point when these
are 0.0%. I would assume that if White Point (WP) is 0.0,
the right slider points to max. intensity and the entire
pixel information must be to the left of that point. An
example with Blue channel being "greater than WP 0.0%":

http://www.pbase.com/phototalk_thh/vue_hist_sliders

This looks familiar to mee. I have been thinking of putting this on the
Web, too. Thanks for saving me the time;-)
I think Vuescan displays the top histograms (before scanning)
inaccurately. As far as I understood, the bottom histograms serve to
show the expected outcome after scanning, and you can see that they
neatly correspond to what you would expect for 'whitepoint = 0%',
provided that the end of the mountain corresponds to RGB = 255,255,255.
My impression is that the sliders have two other problems:

The slider drag-and-drop indication malfunction with WinME
and W2000. After the drop the new position of the slider is
not being indicated until the next preview is finished.
This has a substantial problem: Once you drop the slider
by mistake, you must run a preview to discover where it is!

Same problem with Mac OS X here.
In many cases the information in shadows or in highlights
is so crowded that a move of the slider even by a single
pixel causes quite massive changes in value of black or
white point. Back than Vuescan used to plot the histograms
over the entire width of the preview window. I wonder if
we could now return to this mode. This would allow finer
adjustments of the sliders.

I agree.
 
B

Bart van der Wolf

SNIP
What puzzles me though is the indication of pixel
information *outside* of black or white point when these
are 0.0%.

The single Black/White point marker points to the Luminance (black line)
limits. The luminance is a weighted average of R+G+B, and blue weighs in the
least. When you lock the image color (on the input tab), you will get access
to the individual R/G/B markers.

SNIP
My impression is that the sliders have two other problems:

The slider drag-and-drop indication malfunction with WinME
and W2000. After the drop the new position of the slider is
not being indicated until the next preview is finished.
This has a substantial problem: Once you drop the slider
by mistake, you must run a preview to discover where it is!

More likely a driver issue. I also used to have some odd behavior with WinME
when zooming in, but since upgrading to WinXP no such issue exists with the
same hardware.

Bart
 
T

ThomasH

Bart said:
SNIP

The single Black/White point marker points to the Luminance (black line)
limits. The luminance is a weighted average of R+G+B, and blue weighs in the
least. When you lock the image color (on the input tab), you will get access
to the individual R/G/B markers.

This explains, its easy. Thanks.
SNIP

More likely a driver issue. I also used to have some odd behavior with WinME
when zooming in, but since upgrading to WinXP no such issue exists with the
same hardware.

Well, this means that Vuescan has a portability problem.
And let me guess: Ed will say get yourself WinXP and all
trouble will go away.

He did so regarding system resource use :-( One Vuescan
started on Win ME takes more from the OS pool than a large
CAD program + a Browser + Image processor like Paint-Shop Pro
taken together... Hey, if you Vuescan, the system is dead
untill you stop it, and than the resource meter indicator
jumps back to green.

Thomas
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top