Vuescan Elite 5400 Problem

  • Thread starter Thread starter Jan
  • Start date Start date
J

Jan

Hi,

I know there has been a lot discussion about Vuescan and Elite5400.
However, I encountered a problem that I have not found before.

I use Vuescan 8.1.3
When scanning dark slides, I get artefacts in form of lines parallel
to the long side of the slide.
An example is here:
http://www.open-canoe.de/stripes_example.jpg

The problem seems to appear with different slide films and only in
dark areas.
I have to use Vuescan since I am running Linux. Does anyone have a
solution for this problem?

Thanks, Jan
 
Jan said:
I know there has been a lot discussion about Vuescan and Elite5400.
However, I encountered a problem that I have not found before.

I use Vuescan 8.1.3
When scanning dark slides, I get artefacts in form of lines parallel
to the long side of the slide.
Click to expand...

This was mentioned before and is commonly referred to as 'streaking'.

Paul
 
Hi,

I know there has been a lot discussion about Vuescan and Elite5400.
However, I encountered a problem that I have not found before.

I use Vuescan 8.1.3
When scanning dark slides, I get artefacts in form of lines parallel
to the long side of the slide.
An example is here:
http://www.open-canoe.de/stripes_example.jpg

The problem seems to appear with different slide films and only in
dark areas.
I have to use Vuescan since I am running Linux. Does anyone have a
solution for this problem?
Click to expand...
Vuescan and the Elite don't work, which is hardly surprising for
software that is effectively a rolling beta.
 
Paul Simons said:
This was mentioned before and is commonly referred to as 'streaking'.
Click to expand...

I have noticed my Elite F-2900 does that too, but most noticeable with the
blue channel (discovered that when 8x12" prints show yellow lines). The
Minolta software does not show any streaking, but Vuescan only. The "line"
is always at the exact same location. I suspect it was a hardware issue, and
I removed and cleaned the mirror inside the scanner. There was a "spot" on
the mirror and won't come off. I installed the mirror back and tested again.
Guess what? The line is still there, but on the other side of the picture
now. That means it was that "spot" on the mirror has been causing the
streaking, and I turned it 180 degree when I installed it back. The Minolta
software must be taking care for any hardware defects while Vuescan does
not.
 
Jan said:
Hi,

I know there has been a lot discussion about Vuescan and Elite5400.
However, I encountered a problem that I have not found before.

I use Vuescan 8.1.3
When scanning dark slides, I get artefacts in form of lines parallel
to the long side of the slide.
An example is here:
http://www.open-canoe.de/stripes_example.jpg

The problem seems to appear with different slide films and only in
dark areas.
I have to use Vuescan since I am running Linux. Does anyone have a
solution for this problem?
Click to expand...

Do you have the latest version of the firmware installed in your
scanner? To check this and/or to install it you'll need an Windows or
Mac OS computer with the latest version of the Dimage Scan Utility, I'm
afraid.
 
I use Vuescan 8.1.3
When scanning dark slides, I get artefacts in form of lines parallel
to the long side of the slide.
An example is here:
http://www.open-canoe.de/stripes_example.jpg
Click to expand...

Sorry Jan, this is a typical Vuescan problem with some Minolta
scanners (Scan Elite 5400, Scan Multi Pro, Scan Dual III/IV, maybe
other models too).
It has probably to do with issues about the type of calibration that
Vuescan performs. Probably it's different from the calibration done by
Minolta software and Silverfast.
There are no known solutions at the moment (I even try the latest
8.1.5. I have the latest firmware 1.10). I even tried initializing
with Minolta software and then switching to Vuescan. I tried Vuescan
under Linux and Windows, with Firewire and USB, with entirely
different machines, too. All the same.

You are welcome to file a Bug Report to Ed Hamrick: the more users
report those old, yet unsolved problems, the higher possibility that
Ed Hamrick tries harder to fix them. :)

Fernando
 
Hecate said:
Vuescan and the Elite don't work, which is hardly surprising for
software that is effectively a rolling beta.
Click to expand...

Rolling beta is a fitting name for Vuescan and its users are called head
bangers.
 
That means it was that "spot" on the mirror has been causing the
streaking, and I turned it 180 degree when I installed it back. The Minolta
software must be taking care for any hardware defects while Vuescan does
not.
Click to expand...

I agree about software performance in removing hardware defects. Look at
my tests in total darkness here (images at the bottom of the page) :

http://christian.tsotras.free.fr/scanner_reset/

These scans are taken with a slide that passes no light. And what I see is
the CCD amplificated response.

For me it is CCD defects that Minolta software compensates while Vuescan
don't.


This said, I think Vuescan is a good software. I use it now on a Nikon
Coolscan 5000ED and I am fully satisfied with the results.
 
Christian Tsotras said:
This said, I think Vuescan is a good software. I use it now on a Nikon
Coolscan 5000ED and I am fully satisfied with the results.
Click to expand...

Just wondering, does Nikon have the streaking issue as well, particularly
the 5000ED?
 
Just wondering, does Nikon have the streaking issue as well, particularly
the 5000ED?
Click to expand...

I tried the same "aluminium sandwich" slide tests with the LS5000.

But it refused to work, maybe because it tries an IR preview before a
scan, and with this slide *nothing* can go through it. So I tried the same
tests with a slide with two unexposed slides stacked on one slide cache.

The scanner produced a very few CCD defects, almost unnoticeable, but
nothing comparable with my Minolta Elite tests. I had to push all the
levels to max on VS, and the same with post retouch (Gimp) to enhance
these few defects.

Then, I scanned a very unexposed slide taken by night. The results were
very satisfying, as the LS5000 did not produce any streaking, where the
Minolta Elite had a big problem.

In the real life, after six months of heavy use, I never noticed streaking
with the LS5000, whereas Minolta's defects were noticeable on dark parts
of even normal slides.

I'm very happy with my LS5000.
My personal advice is to not buy Minolta filmscanners. One day you'll
reach its limits.
 
I tried the same "aluminium sandwich" slide tests with the LS5000.
Click to expand...

I forgot to say that I made some tests with the Nikon, but didn't do a
lot.

When I saw the results were better that my Elite scanner, although not
perfect, I stopped the tests because tests are just tests and I didn't
want to be very disappointed just due to some tests that are not real life.

After six months of LS5000 I don't care about these tests any more, as my
heavy use of it proved me that it has no problem.

Of course, it's expensive. But this price is not only for avoiding
streaking. I am also very satisfied with the sharpness of the scans.
 
Hi Christian,

Since you had the Elite too, I was wondering if the 5000ED was able to
penetrate and reproduce the shadow details of slides (compared to Elite
F-2900). I am not quite satisfied with the almost completely dark shadows
scans from the Elite (and Photoshop can only do so much). Does 16bit (vs
12bit) really make any significant difference in practice? Also, is the
Nikon software about to produce colour negative scans w/o cropping both ends
of the histogram (one thing I am very unhappy with the Minolta software)? It
is great to know that you used to use the Elite so you may compare them in a
more meaningful ways.
 
Christian said:
I tried the same "aluminium sandwich" slide tests with the LS5000.

But it refused to work, maybe because it tries an IR preview before a
scan, and with this slide *nothing* can go through it. So I tried the same
tests with a slide with two unexposed slides stacked on one slide cache.

The scanner produced a very few CCD defects, almost unnoticeable, but
nothing comparable with my Minolta Elite tests. I had to push all the
levels to max on VS, and the same with post retouch (Gimp) to enhance
these few defects.

Then, I scanned a very unexposed slide taken by night. The results were
very satisfying, as the LS5000 did not produce any streaking, where the
Minolta Elite had a big problem.

In the real life, after six months of heavy use, I never noticed streaking
with the LS5000, whereas Minolta's defects were noticeable on dark parts
of even normal slides.

I'm very happy with my LS5000.
My personal advice is to not buy Minolta filmscanners. One day you'll
reach its limits.
Click to expand...

Many users have reported two sources of streaks on a Minolta 5400.
Minolta's newest driver release fixes the problem when scanning with
Minolta's utility (must confirm scanner's firmware is updated after
installing the newest driver). When using Vuescan, streaks still exist
even after installing the newest driver and firmware. It is a known
Vuescan bug.

A better advice is not to buy Vuescan. Minolta 5400 at half the cost is
a great value compared to the Nikon LS5000.

All the above can be found in this ng archive. Search for author
fernando.
 
Hello, Celebrate.

I'm very happy with my LS5000.
My personal advice is to not buy Minolta filmscanners. One day you'll
reach its limits.
Click to expand...
Many users have reported two sources of streaks on a Minolta 5400. [...]
A better advice is not to buy Vuescan. Minolta 5400 at half the cost is
a great value compared to the Nikon LS5000.
Click to expand...

Another advice would be to let people test this scanner themselves, as
some people may push the scanner to its limits and others not. On Usenet,
I found half people that didn't see any streaking on Minolta scanners, and
half other that did.

I think, and it is my personal opinion, that Minolta made some debatable
choices when they tried to reduce the price of their scanners.
 
Hello, Alan.

Since you had the Elite too, I was wondering if the 5000ED was able to
penetrate and reproduce the shadow details of slides (compared to Elite
F-2900). I am not quite satisfied with the almost completely dark
shadows scans from the Elite (and Photoshop can only do so much). Does
16bit (vs 12bit) really make any significant difference in practice?
Click to expand...

After having read many articles on Usenet and the web, my opinion is that
the quality of the CCD (signal to noise ratio, "leaking" pixels on high
contrasts, quality of construction) is more important for me than its
number of bits per pixel. But these detailed informations are never
advertised.

A good 12bit CCD can give better results than a bad noisy 16bits CCD. You
have to test by yourself, or analyze scans given on serious reviews
websites.

Another point is the software you use, and your capacity to master it.
Personally, I needed many months to learn the subtelties of my filmscanner
(with Vuescan and Gimp on Linux), especially for negatives.

About shadow detail, I was satisfied with LS5000. Better than the Elite,
but I can't say I saw an enormous difference. What I liked more is the
sharp lens of the 5000ED. When observed at 400% zoom, my scans with the
5000ED show nice pixel details where my old Elite scans were a little
blurry.
Also, is the Nikon software about to produce colour negative scans w/o
cropping both ends of the histogram (one thing I am very unhappy with
the Minolta software)? It is great to know that you used to use the
Elite so you may compare them in a more meaningful ways.
Click to expand...

I don't have Windows, so I didn't test Nikon software. I always used
Vuescan on Linux. A long time ago I used Minolta software on win98 to make
some tests, but I prefer Vuescan for the full control I have over the
scanning process.
 
Christian Tsotras said:
Hello, Alan.



After having read many articles on Usenet and the web, my opinion is that
the quality of the CCD (signal to noise ratio, "leaking" pixels on high
contrasts, quality of construction) is more important for me than its
number of bits per pixel. But these detailed informations are never
advertised.

A good 12bit CCD can give better results than a bad noisy 16bits CCD. You
have to test by yourself, or analyze scans given on serious reviews
websites.

Another point is the software you use, and your capacity to master it.
Personally, I needed many months to learn the subtelties of my filmscanner
(with Vuescan and Gimp on Linux), especially for negatives.

About shadow detail, I was satisfied with LS5000. Better than the Elite,
but I can't say I saw an enormous difference. What I liked more is the
sharp lens of the 5000ED. When observed at 400% zoom, my scans with the
5000ED show nice pixel details where my old Elite scans were a little
blurry.
Click to expand...

Alas, Nikon software also can have a tendency to clip histograms. Not always
but it does happen

It is great to know that you used to use the
 
Could it be that you have IR activated with the Minolta Software and
not with Vuescan. That could be why Minolta Software is able to
correct for the spot in you mirror.
 
Back
Top