Vuescan Review at Photo-i

[drietow]

There is a new review of Vuescan at Photo-i for anyone who has
questions about it.

http://www.photo-i.co.uk/Reviews/software/Vuescan/page-1.htm

Aloha

 

[Don]

There is a new review of Vuescan at ....spam deleted...
for anyone who has questions about it.

Ed, instead of spamming the group with your regular "announcements"
you better go back to working on all those many VueScan bugs... :-/

Bugs are the only questions anyone has about VueScan!

Just read...

Don.

 

[Jeff Randall]

Don: Instead of shooting from the lip, I'd suggest you read the entire
Photo-i review of VueScan. I doubt seriously that Ed would spam the
group with this...

Jeff Randall

 

[Don]

Don: Instead of shooting from the lip, I'd suggest you read the entire
Photo-i review of VueScan. I doubt seriously that Ed would spam the
group with this...

Point taken! Will do...

In my (feeble) defense, the "announcements" I was referring to, do
seem to be quite regular posted by "users" gushing about some new
feature. Only (to use Bart's measuring stick...) these "users" have
never posted before, and never post again. Hmmm...!? But I seem to
have picked a wrong post to comment on.

Nevertheless - justified reflex or not - you're absolutely right!
Should check it out before blurting it out.

Don.

 

[Don]

Don: Instead of shooting from the lip, I'd suggest you read the entire
Photo-i review of VueScan. I doubt seriously that Ed would spam the
group with this...

OK, I've done my homework and read the review.

I don't know... Even though some VueScan flaws are pointed out, there
are many "unusual" conclusions papering over major failings, such as:

"Fortunately, the interface constantly changes"

Fortunately? FORTUNATELY? The reviewer is obviously unfamiliar with
basic user interface guidelines.

He also seems all too willing to gloss over many shortcomings but,
most of all, he totally failed to mention the key VueScan problem:
Bugs!!

However, he's right about one thing regarding VueScan:

"you get the feeling that you are paying to become a Beta tester"

I've been saying that for ages! ;o)

Don.

 

[Jeff Randall]

Don: Maybe the reviewer did not find any bugs related to his test
scanner -- I do not think he was obligated to test VueScan with all
supported scanners. It has been my observation over the years that
most of the VueScan bugs reported to this newsgroup are isolated to
only a few of the now 400 supported scanners. While the statement that
VueScan is buggy may be true in the literal, but narrow sense, it does
not reflect the fact that VueScan really is 400 programs contained in
one wrapper. As an example of this, I recall that you are a Nikon
Coolscan user (btw thanks for the separate R, G, B exposures) -- what
Coolscan bugs, especially those affecting the quality of the output,
are you aware of? I just finished scanning some Kodachrome slides and
Supra color negatives with my LS-4000 today and didn't encounter any
bugs that affected the high quality of the output.

Jeff

 

[Ralf R. Radermacher]

I just finished scanning some Kodachrome slides and
Supra color negatives with my LS-4000 today and didn't encounter any
bugs that affected the high quality of the output.

Now, as the troll calling himself Don has stated repeatedly on this
forum, that's only because we have no idea what decent quality looks
like.

Ralf

 

[Jumm]

And I have yet to see any of Don's "quality" image work.
Jumm

 

[Don]

Don: Maybe the reviewer did not find any bugs related to his test
scanner -- I do not think he was obligated to test VueScan with all
supported scanners.

If he was truly testing VueScan, then yes.
If he was *only* testing VueScan on scanner X, then no.

In other words, he should have explicitly stated that his "test" was
limited to a single scanner. Even though this was vaguely implicit
from the article, his "recommendation" should have been quite clear in
that it was based on a single scanner model (such a statement cuts
both ways, BTW!) as well as indicating the implications of that fact
for owners of all other scanners! (Need I say "Minolta"?)

Using a single scanner is not much of a test, anyway. It's more akin
to a message in a forum like this from a single user. Since this loud
"400 scanner" boast is one of the main VueScan's selling points - also
repeated in the review - a reputable reviewer would have, at the very
least (!), done a few spot checks to challenge this claim.

It has been my observation over the years that
most of the VueScan bugs reported to this newsgroup are isolated to
only a few of the now 400 supported scanners.

That's not really correct. There are many VueScan bugs which span all
scanners. The perennial preview bug (not corresponding to scan),
regularly recurring "can't scan at all", not to mention all the bugs
related to data corruption which affects all images regardless of
source, etc.

While the statement that
VueScan is buggy may be true in the literal, but narrow sense, it does
not reflect the fact that VueScan really is 400 programs contained in
one wrapper.

That's not really correct either. The individual functions are all the
same (that's why their bugs affect all scanners). What's different is
the scanner commands (capabilities) which trigger these functions. So,
once this structure is in place, adding another scanner is simply a
case of interpreting its commands/capabilities and branching to an
appropriate routine.

Conceptually, it's no different to, say, editing different image
formats with the single program. I mean, you don't say Photoshop is 50
programs in one just because it can read 50 different image formats?

As an example of this, I recall that you are a Nikon
Coolscan user (btw thanks for the separate R, G, B exposures)

You're most welcome, and I appreciate the acknowledgment.

-- what
Coolscan bugs, especially those affecting the quality of the output,
are you aware of? I just finished scanning some Kodachrome slides and
Supra color negatives with my LS-4000 today and didn't encounter any
bugs that affected the high quality of the output.

That's an elastic question because first we need to define "quality".
For someone (and I don't mean you, but in general) who scans for web
quality means something completely different than to someone else who
scans for archival purposes.

Specifically, there were many problems (still outlined in the
archives). For example, unpredictable interaction of various (secret)
options. Do note that today - over two years later (!) - there are
currently two threads dealing with this very same thing!! Then there
were problems with sharpening (medium was sharper than low) constant
corruption of the INI file, dust and scratches corruptions, etc. As I
say, it's all in the archives.

In a nutshell, VueScan always insists on "doing things" to the image
even when everything is turned off. The reason is clearly that data is
in a very poor shape to start with (see noise in review, the infamous
Minolta stripes, etc) so VueScan is forced to try to mask this
extremely poor quality with various "adjustments".

Be that as it may, the key point which I keep repeating (and it keeps
being overlooked) is that if people don't care for such level of
detail or quality and are happy with what they get, that's fine.
What's not fine is making unsubstantiated statements about VueScan
"quality" based on what their scan looks on a web page.

Don.

 

[Ralf R. Radermacher]

if people don't care for such level of
detail or quality and are happy with what they get, that's fine.

Told you, Jeff. )

Ralf

 

[UrbanVoyeur]

Told you, Jeff. )

Ralf

Right as usual Ralf.

I guess these past five years of extremely high quality output from
Vuescan on 5 different scanners is just an illusion. Or at best, just
low exepctations. Guess that holds true for Silverfast and Photoshop,
since I used them too.

I wonder why, if Vuescan is so bad, that it works so well for me and so
many others. Maybe none of us knows what quality is?

 

[Jeff Randall]

It has been my observation over the years that

most of the VueScan bugs reported to this newsgroup are isolated to
only a few of the now 400 supported scanners.

You wrote: That's not really correct. There are many VueScan bugs
which span all
scanners. The perennial preview bug (not corresponding to scan),
regularly recurring "can't scan at all", not to mention all the bugs
related to data corruption which affects all images regardless of
source, etc.

All scanners? Not on my LS-4000 or N1240U (LiDE30). Yes on my LS-4000
the preview can be different from the final scan because the first
preview is done at an exposure = 1. The final image will be done at
the exposure calculated in the preview step. If Lock exposure is check
and preview is done again, then the preview will look basically like
the final scanned image because both will be done at the same exposure.
Moreover, I have not experienced the recurring "can't scan at all", or
any bugs related to data corruption which affects all images.

Coolscan bugs, especially those affecting the quality of the output,
are you aware of? I just finished scanning some Kodachrome slides and

Supra color negatives with my LS-4000 today and didn't encounter any
bugs that affected the high quality of the output.

You wrote: That's an elastic question because first we need to define
"quality".
For someone (and I don't mean you, but in general) who scans for web
quality means something completely different than to someone else who
scans for archival purposes.

I was asking you to list bugs you have experienced using VueScan and
your Coolscan scanner. For each "bug" you list, please fully define
your image quality criteria that you use.


You wrote: Specifically, there were many problems (still outlined in
the
archives). For example, unpredictable interaction of various (secret)
options. Do note that today - over two years later (!) - there are
currently two threads dealing with this very same thing!! Then there
were problems with sharpening (medium was sharper than low) constant
corruption of the INI file, dust and scratches corruptions, etc. As I
say, it's all in the archives.

Let us limit our discussion to CoolScan scanners since we both have
used them with VueScan. Based on my 4 years use of my LS-4000 and as
applicable my 3 years use of my N1240U (LiDE30):

1) I have never experienced unpredictable interaction of various
options. Yes the Users Guide is terse and sometimes cryptic and does
not include all the technical background, but what documentation is
great especially for software costing much less than $100?
2) I have never experienced the sharpening problem you list.
3) I have never had a corrupted ini file. Yes, I have had to delete
an old ini file when Ed add new features and start clean. Agreed a
hassle, but not due to corruption.
4) I have never had any problems with dust and scratches corruption
(if this is even applicable to the LS-4000)


You wrote: In a nutshell, VueScan always insists on "doing things" to
the image
even when everything is turned off. The reason is clearly that data is
in a very poor shape to start with (see noise in review, the infamous
Minolta stripes, etc) so VueScan is forced to try to mask this
extremely poor quality with various "adjustments".

Please list everything VueScan does to the Coolscan image when
everything is turned off that is not done by NikonScan.

I do not think the poor shape of the data that you list is applicable
to the Coolscan. Please list the details of the poor shape data coming
from your Coolscan (you define poor shape) and specifically how Vuescan
adjusts the data to make up for this poor quality (you define quality).



You wrote: Be that as it may, the key point which I keep repeating
(and it keeps
being overlooked) is that if people don't care for such level of
detail or quality and are happy with what they get, that's fine.
What's not fine is making unsubstantiated statements about VueScan
"quality" based on what their scan looks on a web page.

Please define "such level of detail or quality" that some people don't
care about (ie., substiantiate your definition of quality) based on
your experience with your Coolscan scanner and VueScan.

Jeff

 

[Ralf R. Radermacher]

Let us limit our discussion to CoolScan scanners since we both have
used them with VueScan.

Not quite, Jeff. You have, and so do I. Quite a lot for a number of
years, even. The troll calling himself Don has stated that he's been
thinking about using a Coolscan. Your scans have been on the web and so
are mine. Nothing from him.

Ralf

 

[UrbanVoyeur]

That has been my experience as well.

Let us limit our discussion to CoolScan scanners since we both have
used them with VueScan.

I have a Nikon LS-4000 and have been using it with Vuescan since 2001. I
have had no problems out of Vuescan at all. Every scan before
calibration was part of the program was good. Every scan since
calibration has been excellent. There were one or two minor issues with
the coolscan a 2 or 3 years ago that were fixed in a matter of days.

I currently use 2 other scanners with Vuescan, a Umax flat bed and a
Minolta medium format film scanner. It works perfectly with each of those.

I've also used Vuescan with a PIE film scanner and Microtek film
scanner. It worked perfectly with those scanners before I gave them
away, and continues to work perfectly for two photographers who have
them now.

Thousands of scans and many versions later, it is still a solid program.

On each and every one of these scanners, I have scanned color negatives
- including Supra and almost every color neg film Fuji makes, *every* 35
or 120 FujiChrome, Tri-X, all three T-max's, Efke, several KodaChrome
and ExktaChrome flavors, and most of Ilford's B&W products. I've also
scanned a number of the B&W films as reversal chromes (www.dr5.com).

The *only* issue I ever regularly encountered was "pepper grain" with
some FujiChrome emulsions scanner at 4000 dpi on the LS-4000. The
"pepper grain" is a shortcoming of the LS-4000, *not* the software used
with it and has been extensively documented on the net.


If Don can show where Vuescan fails on the LS-4000, Minolta Multi, Umax
2400s, PIE 1600u or Microtek 35t, I'd be eager to test it.

I have not encountered any of the problems he describes.

 

[Don]

You wrote: That's not really correct. There are many VueScan bugs
which span all scanners.

All scanners? Not on my LS-4000 or N1240U (LiDE30). Yes on my LS-4000
the preview can be different from the final scan because the first
preview is done at an exposure = 1.

That's not what I'm taking about. It's the sudden "false-color"
appearance of the preview image. It's a known bug acknowledged many
times.

I was asking you to list bugs you have experienced using VueScan and
your Coolscan scanner. For each "bug" you list, please fully define
your image quality criteria that you use.

I already listed the gist of some of the problems. If you really want
the details, check the archives as I suggested. You'll also find many
other reports of VueScan bugs occurring with predictable regularity!

In the meantime, here's one I just happen to remember. Set manual
exposure to any value over 100. Scan. The value will be "magically"
rolled back to 100. What exposure was used? Nobody knows...

1) I have never experienced unpredictable interaction of various
options. Yes the Users Guide is terse and sometimes cryptic and does
not include all the technical background, but what documentation is
great especially for software costing much less than $100?
2) I have never experienced the sharpening problem you list.
3) I have never had a corrupted ini file. Yes, I have had to delete
an old ini file when Ed add new features and start clean. Agreed a
hassle, but not due to corruption.
4) I have never had any problems with dust and scratches corruption
(if this is even applicable to the LS-4000)

Those are all subjective statements not objective facts. "Works for
me" is not an objective fact. You need more than a "sample of 1" -
either way! And the best tool is, again, to use the archives. They are
*flooded* with reports of VueScan bugs, from mundane to obscure.

Besides, what does "works for me" mean? You seem very keen on
demanding specifics while not providing any yourself.

Please list everything VueScan does to the Coolscan image when
everything is turned off that is not done by NikonScan.

If you really do care for quality you can do your own tests. Scan raw
and compare using appropriate tools, such as 16-bit histogram, etc.

Indeed, anyone who cares for quality would have performed such tests
*before* using any program!

I do not think the poor shape of the data that you list is applicable
to the Coolscan. Please list the details of the poor shape data coming
from your Coolscan (you define poor shape) and specifically how Vuescan
adjusts the data to make up for this poor quality (you define quality).

Quality in this context is *uncorrupted* data. As simple as that.
VueScan virtually always corrupts data. If you *really* care for such
level of detail you can do (and, indeed, *should have* done!) your own
tests as outlined above. Not come at it after the fact.

Please define "such level of detail or quality" that some people don't
care about (ie., substiantiate your definition of quality) based on
your experience with your Coolscan scanner and VueScan.

I have! Here it is again:

Scanning for web and scanning for archiving.

And it relates not only to Coolscan but to *any* scanner! Anyone who
scans for web will be blind to even the most glaring VueScan
inadequacies patently obvious to anyone scanning for archival
purposes.


Again, you seem very keen on specifics without providing any yourself.
Besides, all the things you're asking are either available in the
archives or can easily be answered by running a few tests yourself.

Finally, with the flood of VueScan versions tripping over themselves
to keep up with the bugs any testing will only be limited to a
specific version which will probably be out of date and irrelevant by
the time the tests are complete because the new VueScan version will
appear with a whole new set of bugs.

The only thing which is consistent in all versions (as the archives
prove) is the never-ending flood of VueScan bugs. They may skip a
version or morph, but they are perennial.

As I already said, you may not care, or may not be affected, but that
does not change the objective fact that VueScan is riddled with bugs.

Don.

 

[Ralf R. Radermacher]

In the meantime, here's one I just happen to remember. Set manual
exposure to any value over 100. Scan. The value will be "magically"
rolled back to 100. What exposure was used? Nobody knows...

Which goes a long way to demonstrate just how much practical relevance
the nonsense uttered by our resident troll has. Is there anyone on this
list who has ever used exposure values of, say, 50 or more?

Ralf

 

[Ralf R. Radermacher]

It's a known bug ...

You'll also find many
other reports of VueScan bugs ...

*flooded* with reports of VueScan bugs...

VueScan virtually always corrupts data.

VueScan
inadequacies patently obvious ...

VueScan versions tripping over themselves
to keep up with the bugs ...

the new VueScan version will
appear with a whole new set of bugs....

the never-ending flood of VueScan bugs...

VueScan is riddled with bugs.

All that in one single message. Can you spell paranoia? Obsession?
Crusade?

Ralf

 

[UrbanVoyeur]

Which goes a long way to demonstrate just how much practical relevance
the nonsense uttered by our resident troll has. Is there anyone on this
list who has ever used exposure values of, say, 50 or more?

Ralf

What I don't understand is why expend so much energy on something he so
dislikes.

Why not spend the time doing something pleasant and gratifying, like
making photographs?

 

[Jumm]

You forgot "Jihad"! Heheh!

All that in one single message. Can you spell paranoia? Obsession?
Crusade?

Ralf

--
Ralf R. Radermacher - DL9KCG - Köln/Cologne, Germany
private homepage: http://www.fotoralf.de
manual cameras and photo galleries - updated Jan. 10, 2005
Contarex - Kiev 60 - Horizon 202 - P6 mount lenses

 

[Carsten Schurig]

All that in one single message. Can you spell paranoia? Obsession?
Crusade?

Well, it's just a long known bug of Don. And he doesn't seem to be
working on it. (Or is it a feature now?)

SCNR