color shift between preview and scan tab in vuescan

  • Thread starter Thread starter philap
  • Start date Start date
P

philap

Hello,

Almost everything is in the title: I do a preview with my SDIII.
preview image and scan image are identical and good. Then I scan and
the color in the scan tab has shifted towards a light red. The saved
image is unfortunately identical to the scan tab image. What did I do
that was wrong?
Thanks
 
Maris V. Lidaka Sr. said:
You did nothing wrong - it's a Vuescan quirk/defect.
???

SNIP
Click to expand...

Try locking the exposure, redo the preview, and lock the image color.

Is the shift towards red extreme, or is it slightly more red?

If you use an auto-color balance mode, the scan (which is probably
made at a different resolution than the preview) may resolve enough
detail to arrive at a different balance choice. Locking the color will
prevent that from happening.

Bart
 
You mean yet another VueScan bug.

Please don't get me wrong, but I'm genuinely puzzled. Why do you
people put up with this?

I mean - honestly - with the constant flood of messages about VueScan
bugs I just can't conceive that native Minolta software could possibly
be any worse.

Don.
 
I only recently noticed this too. Sometimes my scan comes out
noticeably darker and contrastier than the preview, though most of the
time the difference seems slight.
 
Try locking the exposure, redo the preview, and lock the image color.
Click to expand...
Thanks; That sounds good . I am gonna try ASAP.
Is the shift towards red extreme, or is it slightly more red?
Click to expand...
It is a slight shift towards pale red (I mean lighter and a bit more
red almost like pink). It is not outrageous but clearly visible and
annoying.

Thank you again

Philap
 
You mean yet another VueScan bug.

Please don't get me wrong, but I'm genuinely puzzled. Why do you
people put up with this?

I mean - honestly - with the constant flood of messages about VueScan
bugs I just can't conceive that native Minolta software could possibly
be any worse.

Don.
Click to expand...

Don't like it? Simple solution=Don't use it. It's not a big issue -
it's only a slight change. I've seen worse in Nikon Scan. Seems to
depend on the particular image being scanned. I happen to like vuescan
and the results it gives me. I also like having a choice between
Vuescan and Nikon scan for particularly difficult negatives or slides.
I've purchased several copies of Silverfast, and it's ok, but I found
it really annoying having to pay for it every time I upgraded my
scanner. I simply quit using it.

Ed H. responds to user feedback and continually tweaks his program's
performance. As a program user, I very much appreciate that.

Dan
 
Because Vuescan gets all of the information and data available from the
film. I can deal with color casts and shifts, contrast, and everything else
in Photoshop. Vuescan gives me the digital data to do so.

Maris
 
Well, it is worse.
Click to expand...

That sounds really awful. :-(

There are many things I dislike about NikonScan but, on reflection,
maybe I shouldn't be so hard on it because once most of the "auto
things" are switched off (which is not as easy as it sounds!!!) it
does what it's told.

Don.
 
Well, that's not quite true for a number of reasons.

Postprocessing (whether in the scanner or afterwards) causes image
deterioration with every edit. That's why it's advisable to try and
limit the number of changes by, for example, combining them whenever
possible.

In addition, Photoshop only uses 15-bits which may cause further image
deterioration.

Finally, and this would be my main concern, by accepting VueScan's
color shifts and casts etc. you are actually *not* getting all the
information available on the film. Indeed, you are starting with
compromised data.

What's more, with the constant stream of problems, I would have very
low confidence in the data VueScan delivers (which is why I discarded
it). In other words, one may think everything is fine, only to read
later there is one bug or another which actually compromised the data.
And this seems to repeat periodically.

I would think that eventually a point would be reached when this
constant uncertainty would outweigh any perceived advantages.

Anyway, this is not really important but I was just curious because I
myself have much lower tolerance for such major program bugs which
affect the data. However, humans are creatures of habit, and once we
are familiar with something (e.g. software) we tend to stick with it,
warts and all. It's just that I myself don't consider familiarity a
(good) reason to stick with it. But we all have different
requirements.

Don.
 
Don't like it? Simple solution=Don't use it. It's not a big issue -
it's only a slight change. I've seen worse in Nikon Scan. Seems to
depend on the particular image being scanned. I happen to like vuescan
and the results it gives me. I also like having a choice between
Vuescan and Nikon scan for particularly difficult negatives or slides.
I've purchased several copies of Silverfast, and it's ok, but I found
it really annoying having to pay for it every time I upgraded my
scanner. I simply quit using it.
Click to expand...

Of course, to each his own. That goes without saying. If you're happy
with the software, that's all that matters.

I was just curious because my personal tolerance for bugs is much
lower. Also, and more importantly, I personally would have very low
confidence in the data when there is a continuous stream of problems
which would force me to constantly have to re-evaluate this data I
previously thought was fine.

But that goes back to the above point, each person's individual
requirements.
Ed H. responds to user feedback and continually tweaks his program's
performance. As a program user, I very much appreciate that.
Click to expand...

My experience with that differs quite a bit. Two examples: the preview
window and individual RGB controls.

The first was before my time but I read on good authority (Kennedy)
that apparently he vociferously refused to implement the preview
window for a very long time because "you don't need it". Ironically,
the preview window is now the cornerstone to his panacea advice ("just
set the gray point").

The individual RGB controls was a long saga I had a hand in. For
months he refused to acknowledge the importance of individual RGB
controls stating again "you don't need it". Following a protracted
thread and after I posted a number of images clearly illustrating the
problem, he grudgingly conceded and reluctantly implemented it.

So, I would say that his response to user feedback is, at best, very
selective which is further illustrated by the current "Minolta
stripes" problem.

Anyway, be that as it may, it all goes back to the first sentence
above...

Don.
 
Back
Top