Evaluating VueScan with HP 3300c?

[Marjolein Katsma]

I just downloaded and installed the latest driver from HP for my 3300c
flatbed scanner. I also downloaded and installed the latest eval version
of VueScan, hoping to be able to at least *evaluate* it with my
scanner... no joy.

I'm really just exploring for now, intending to buy a film/slide scanner
(most likely Nikon Coolscan V) and possibly use VueScan with that - but
for now all I have is the HP 3300c. And yes, I know that VueScan does
not support that particular model - at least not "officially".

What I was hoping was that I could somehow convince it that I actually
have a different (but similar) HP scanner so I could get an impression
of working with VueScan - even if I expected that to be limited. (And if
that would work, maybe getting round some of the limitations of the HP
software for the 3300c.)

Alas, on starting up the software immediately tells me it cannot see any
scanner, and suggests starting the scanner and restarting VS. So I tried
"starting" the scanner by starting HP's own scanner software and then
retrying VueScan, but it still doesn't see the scanner at all. So I
can't see much more than the menu and the tabs on its UI, not actually
"do" anything with it.

Would there be some sort of workaround to convince VueScan that I
actually do have a scanner attached to that computer, or should I give
up all hope until I actually have a film scanner?

Any insights appreciated.

 

[Don]

I'm really just exploring for now, intending to buy a film/slide scanner
(most likely Nikon Coolscan V) and possibly use VueScan with that

Nikon Scan is far superior (definitely more reliable!) and it will
produce everything you need. Furthermore, you have ICE which is
several orders of magnitude better than Vuescan's buggy "IR cleaning".

What specifically are you hoping Vuescan will do that NikonScan can't?

The reason I ask is that, usually, this ends up being some image
editing feature which is better done in a proper editor afterwards and
- strictly speaking - doesn't even belong in scanner software.

Alas, on starting up the software immediately tells me it cannot see any
scanner, and suggests starting the scanner and restarting VS. So I tried
"starting" the scanner by starting HP's own scanner software and then
retrying VueScan, but it still doesn't see the scanner at all.

The reasons is that Vuescan access to scanners is direct and it
bypasses all standard interfaces (like TWAIN). The only way to fool
Vuescan would be to have the scanner send a different identification
code. And even that would probably not work, because there are bound
to be differences in actual command codes which Vuescan sends
subsequently.

Would there be some sort of workaround to convince VueScan that I
actually do have a scanner attached to that computer, or should I give
up all hope until I actually have a film scanner?

You can "test" Vuescan up to a point by using the misleadingly
misnamed "scan from disk" function. This will take any graphics file
and pretend it's coming from the scanner.

In reality that "function" just applies various image editing. You can
(and should!) do all that in a proper image editor. But it can give
you a feel for the program so you can "enjoy" it's so-called "user
interface". As you can see I don't think very much of it! ;o)

BTW, if you do that, the source graphics you are using should be in
gamma 1.0.

Of course, that doesn't actually test the scanning portion which is
key. Vuescan is very buggy and unreliable and most of these bugs and
other problems manifest only when you attach a scanner.

It all really depends on your requirements. If you want a quick and
dirty scan for the web using highly compressed JPG then even Vuescan
will do. But if you want high quality scans and examine the image at
100% magnification (or care for reliable software) than Vuescan comes
way short.

On the other hand if you don't plan to do any editing afterwards and
just want "one size fits all" scanner software then Vuescan may do but
then the question is why waste money on a dedicated film scanner. If
your threshold is that low then you may just as well scan film on your
flatbed. Especially if you only have a set number of film you want to
scan (for example digitize your film collection) and will not be using
the scanner on a continuous basis.

Don.

 

[Don]

Pedantic P.S.

you a feel for the program so you can "enjoy" it's so-called "user

That should be "its", of course. (fx: smacks hand with a ruler)
Ouch!

100% magnification (or care for reliable software) than Vuescan comes

And here it should be "then". (fx: goes and stands in the corner)
Grumble, grumble...

NOTE to self: First thing to do in the morning (before anything else)
is apply coffee! ;o)

Don.

 

[Marjolein Katsma]

Don ([email protected]) wrote in

Nikon Scan is far superior (definitely more reliable!) and it will
produce everything you need. Furthermore, you have ICE which is
several orders of magnitude better than Vuescan's buggy "IR cleaning".

What specifically are you hoping Vuescan will do that NikonScan can't?

All theory for now - I've just been reading. I got the idea that in the
case of bad negatives VueScan might help by getting higher exposure in
one or more separate channels (pardon if I don't get the terminology
right) and that the Nikon couldn't do this by itself. I may have
understood incorrectly.

Anyway, I always like to have several options so if one approach doesn't
yield what I hope I could try another. Without actually having touched
the film scanner I can't be more specific than that, really.

The reason I ask is that, usually, this ends up being some image
editing feature which is better done in a proper editor afterwards and
- strictly speaking - doesn't even belong in scanner software.

I do understand that most scanner software settings are indeed image
editing - but is exposure?

The reasons is that Vuescan access to scanners is direct and it
bypasses all standard interfaces (like TWAIN). The only way to fool
Vuescan would be to have the scanner send a different identification
code.

Doesn't seem possible ;-)

And even that would probably not work, because there are bound
to be differences in actual command codes which Vuescan sends
subsequently.

I thought the 3300c was a derivative of another scanner (2200?) so my
guess was that it might have the same or a not a very different set of
command codes. I was hoping I could try that out.

You can "test" Vuescan up to a point by using the misleadingly
misnamed "scan from disk" function. This will take any graphics file
and pretend it's coming from the scanner.

In reality that "function" just applies various image editing. You can
(and should!) do all that in a proper image editor. But it can give
you a feel for the program so you can "enjoy" it's so-called "user
interface".

Whether you like the user interface or not - a user interface it is. ;-)
I'll probably try that to see a little more of the user interface at
least.

Evaluating the user interface was another reason for wanting to try it;
though I admit I was secretly hoping I could actually use it (however
limited) with the 3300c because I'm not very happy with HP's driver: it
has a rather nice ranges of settings but it often tries to "straighten"
an image when I'd rather it would leave it a lone and let me do any
rotation (if needed at all!) in PSP - but there's no way to tell it
*not* to do it. Apart fom that, the scanner really is good enough for
what I bought it for - which is definitely not photographic work.

As you can see I don't think very much of it! ;o)

Hehe, I already knew that... I've been reading quite a bit here. Thanks
for your helpful reply in spite of what you think of the program.

It all really depends on your requirements. If you want a quick and
dirty scan for the web using highly compressed JPG then even Vuescan
will do.

Mostly I want to scan in my archive of travel photography - many rolls
of negative film. And yes, in general I want it to be good quality
scans, but a quick and dirty scan to make a preselection for instance
might come in handy (I think...). I want to put the best ones online,
may also try to sell some of my work.

Then there are some "pictures of text" where all I want is the actual
text - doing that directly with VueScan's OCR function sounds
attractive, if it works well enough. But I wonder what it (or any OCR
software for that matter) would be able to work with text with a
perspective distortion - I'll have to experiment.

I had a cheapo PrimeFilm scanner before (though it was a lot more
expensive then than it is now), but it broke. I used it just enough to
learn that I did want to digitize all my films but that I definitely
needed something of a much better quality and more reliable.

I'm just back from another trip with 45 rolls of film - and I'm
frustrated I cannot do what I planned with it yet. :-( But I'll need a
new computer as well - it'll be a while before I will actually be able
to start practicing with a new scanner. (sigh)

 

[Don]

All theory for now - I've just been reading. I got the idea that in the
case of bad negatives VueScan might help by getting higher exposure in
one or more separate channels (pardon if I don't get the terminology
right) and that the Nikon couldn't do this by itself. I may have
understood incorrectly.

This is something I've been wresting with quite a bit (dark Kodachrome
slides). It's a bit hard to explain, especially in theory.

Nikon Scan exposure is limited to +/-2 EV (exposure values) per
channel for a cumulative maximum of +/- 4 EV. This is *in addition* to
auto exposure so for all practical purposes you will never need more.
In other words, Nikon Scan determines the optimal exposure and then
enables you to go +/- 4 EV from there.

Vuescan exposure is open ended but there are many problems. First,
it's non-standard using a "multiplier" instead of EV. Second, it's not
as clear as EV because it start with 1. This means for negative
exposures you only have values from 0 to 1, while for positive
exposure it goes from 1 to (theoretically) infinity. To double the
exposure you double the value. In case of EV you just increase by 1.

But the biggest problem with Vuescan exposure is it's buggy! Once you
go over a certain value Vuescan exposure becomes erratic and the
display does not correspond anymore to the actual values used.

Anyway, I always like to have several options so if one approach doesn't
yield what I hope I could try another. Without actually having touched
the film scanner I can't be more specific than that, really.


I do understand that most scanner software settings are indeed image
editing - but is exposure?

No, you're right. Exposure is a hardware function as is ICE (automatic
cleaning of dust and scratches). This is done by using an infrared
light source in addition to normal RGB LEDs.

Vuescan can get this infrared data too but the difference is what is
done with this data afterwards. Nikon uses ICE software while Vuescan
has its own so-called "infrared cleaning" which is far inferior and
there are constant reports of bugs and problems. ICE is several orders
of magnitude superior as well as much more reliable.

Doesn't seem possible ;-)

Well, there are ways but it gets very complicated and messy. You need
to "wedge in" a software routine to intercept USB communications and
"spoof" a scanner. In a word, it's not worth the trouble.

I thought the 3300c was a derivative of another scanner (2200?) so my
guess was that it might have the same or a not a very different set of
command codes. I was hoping I could try that out.

Yes, in theory, most commands should be the same but there are always
differences. However, the problem is getting Vuescan to acknowledge
the scanner in the first place.

Whether you like the user interface or not - a user interface it is. ;-)

When a user interface gets in the way of the user it's hardy a user
interface. More like user "interference". Hey, I like that! ;o)
"Vuescan's user interference"!

Seriously though, using it in the "scan from disk" mode will let you
get the feel for the user interface. One other hint. Vuescan gets very
easily "confused" and if that happens (you start getting weird
results) delete Vuescan.ini and start again.

BTW, if you want to compare, download the Nikon Scan User Guide as
PDF and flip through it. It's not the same as actually using it, but
there are a lot of graphics to give you a hint of how things are done.

Mostly I want to scan in my archive of travel photography - many rolls
of negative film. And yes, in general I want it to be good quality
scans, but a quick and dirty scan to make a preselection for instance
might come in handy (I think...). I want to put the best ones online,
may also try to sell some of my work.

Then you definitely want to get the most out of them which means
scanning "raw". There have been many messages about this, but
basically it means *not* using any of the scanner software editing
features and only using scanner software to control the hardware.

A raw scan is basically maximum resolution and maximum color depth.
That is then saved as your "digital negative" and you do all your
editing in an external editor (e.g. Photoshop) on a copy.

In that case you should really be very weary of Vuescan! Not only is
it far too buggy and unreliable for such precise work but its vastly
inferior "color management" is more trouble than it's worth.

Then there are some "pictures of text" where all I want is the actual
text - doing that directly with VueScan's OCR function sounds
attractive, if it works well enough. But I wonder what it (or any OCR
software for that matter) would be able to work with text with a
perspective distortion - I'll have to experiment.

I have no idea how good or bad Vuescan's OCR is. I can only go by
"circumstantial evidence" i.e. judging by all the other features. And
if Vuescan OCR is like the rest of Vuescan then it's pretty bad.

You'd be far better off getting a proper OCR program. I've done some
OCR but very little. I used an old version of ABBYY Fine Reader and it
produced some amazing results. It's never 100% but it comes very
close. It even recognized columns and images etc. The recognition
improves quite a bit with a good dictionary. It's still advisable to
run it through a spelling/grammar checker afterwards.

I had a cheapo PrimeFilm scanner before (though it was a lot more
expensive then than it is now), but it broke. I used it just enough to
learn that I did want to digitize all my films but that I definitely
needed something of a much better quality and more reliable.

I'm just back from another trip with 45 rolls of film - and I'm
frustrated I cannot do what I planned with it yet. :-( But I'll need a
new computer as well - it'll be a while before I will actually be able
to start practicing with a new scanner. (sigh)

I know... Same here. I don't shoot analog film anymore, but I can
certainly do with a new computer. And there are always other things
which get in the way. I've been wrestling with my Kodakchromes for
about 3 years now. Every time I start making progress, life gets in
the way... :-( Anyway, just finished my last roll of KC so it's on to
Ektachromes next. And then there are still the negatives! I haven't
even began to tackle them so more fun ahead...! ;o)

Don.

 

[Marjolein Katsma]

Don ([email protected]) wrote in

This is something I've been wresting with quite a bit (dark Kodachrome
slides). It's a bit hard to explain, especially in theory.

Nikon Scan exposure is limited to +/-2 EV (exposure values) per
channel for a cumulative maximum of +/- 4 EV. This is *in addition* to
auto exposure so for all practical purposes you will never need more.
In other words, Nikon Scan determines the optimal exposure and then
enables you to go +/- 4 EV from there.

Vuescan exposure is open ended but there are many problems. First,
it's non-standard using a "multiplier" instead of EV. Second, it's not
as clear as EV because it start with 1. This means for negative
exposures you only have values from 0 to 1, while for positive
exposure it goes from 1 to (theoretically) infinity. To double the
exposure you double the value. In case of EV you just increase by 1.

I understand what you're saying - theoretically. But I'd have to *see*
it to really understand! (And I can't do that until I get my hands on a
film scanner.)

But the biggest problem with Vuescan exposure is it's buggy! Once you
go over a certain value Vuescan exposure becomes erratic and the
display does not correspond anymore to the actual values used.

Well, it sounds reasonable to me that this type of correction will only
work within a certain "range" of values.

Well, there are ways but it gets very complicated and messy. You need
to "wedge in" a software routine to intercept USB communications and
"spoof" a scanner. In a word, it's not worth the trouble.

OK, correction - I meant it doesn't seem "feasible". I did understand
that technically, theoretically it could be done but while I enjoy
programming I don't do *that* type of programming - and I don't intend
to learn either. ;-)

When a user interface gets in the way of the user it's hardy a user
interface. More like user "interference". Hey, I like that! ;o)
"Vuescan's user interference"!

Seriously though, using it in the "scan from disk" mode will let you
get the feel for the user interface.

I tried that with two scans I made off of a contact sheet. ;-) At least
I got to see all elements of the not-so-bad "interference" though it
takes some getting used to; but most of that is the underlying concepts
(although all the reading I've done the last few days certainly helps).
At least I could get to all "advanced" settings in the trial, which is
good.

BTW, if you want to compare, download the Nikon Scan User Guide as
PDF and flip through it. It's not the same as actually using it, but
there are a lot of graphics to give you a hint of how things are done.

Ah, good tip. I'll do that!

Then you definitely want to get the most out of them which means
scanning "raw". There have been many messages about this, but
basically it means *not* using any of the scanner software editing
features and only using scanner software to control the hardware.

Including ICE, of course.

A raw scan is basically maximum resolution and maximum color depth.
That is then saved as your "digital negative" and you do all your
editing in an external editor (e.g. Photoshop) on a copy.

In that case you should really be very weary of Vuescan! Not only is
it far too buggy and unreliable for such precise work but its vastly
inferior "color management" is more trouble than it's worth.

Still, it might be helpful in *some* (non-standard) cases. I wouldn't
use it if what the Nikon does with its own software is good enough, but
in hard cases I just might try both. I'm keeping an open mind for now.

I have no idea how good or bad Vuescan's OCR is. I can only go by
"circumstantial evidence" i.e. judging by all the other features. And
if Vuescan OCR is like the rest of Vuescan then it's pretty bad.

Well, my reasoning is *if* it works, "pretty bad" might just be "good
enough". I experimented a bit with *very* bad images though and got
practically nowhere - but then no OCR software actually might be able to
do much with an image like that, even though I did some clearing up and
sharpening. But they were scans from a contact print - not exactly an
ideal starting position.

Of course, if I need to do too much work to get a an "OCRable" image, I
might as well put the scanned image on screen and type out the text. ;-)

You'd be far better off getting a proper OCR program.

I have one (and got pretty good results scanning in a newspaper
clipping) - much less work than retyping the text *and* reformatting to
get the same layout (which I needed in that case).
The attraction of VueScan was that (theoretically) it would be just one
process step. But that advantage disappears if I need to do things like
perspective correction to make the image OCRable in the first place.


Anyway, without access to a film scanner for now I'm undecided about
VueScan. I can live with the user "interference" if it actually helps,
even if only for a few exceptional cases. And that's what I can't judge
without a film scanner... I'll leave it for now.

I know... Same here. I don't shoot analog film anymore, but I can
certainly do with a new computer. And there are always other things
which get in the way. I've been wrestling with my Kodakchromes for
about 3 years now.

(At least I've never used Kodakchrome - I don't need that extra headache
when scanning!)

Every time I start making progress, life gets in the way... :-(

Ah, yes, life.
I've been back for three weeks (it feels like last week) but half my
head is still in Asia. The brain needs time to process it all, and the
trip was pretty intensive. Half of the rest is reading up on hardware to
spec out a new computer. Half of the rest programming for my travel
blog... Then I still need to write up a lot of stories...
Eerr, life??


Thanks a lot for your comments, Don. They were quite helpful!

 

[Don]

Well, it sounds reasonable to me that this type of correction will only
work within a certain "range" of values.

No, no, that's not what I mean. What I'm talking about is a(nother)
Vuescan bug: The actual exposure which Vuescan uses to perform the
scan is not the same as the exposure displayed on the screen.

This is not a real example but, say, you enter +2 EV. However the scan
is performed with +1.5 EV! That's bad enough but after the scan the
display then changes to +1 EV!? Vuescan does the equivalent of that
(above a certain exposure value).

I tried that with two scans I made off of a contact sheet. ;-) At least
I got to see all elements of the not-so-bad "interference" though it
takes some getting used to; but most of that is the underlying concepts
(although all the reading I've done the last few days certainly helps).
At least I could get to all "advanced" settings in the trial, which is
good.

It's essential to display all settings and not "hide" them! This is
because of the Vuescan's user "interference". The author apparently
never heard of ergonomics or "muscle memory", and so on. There is no
underlying concept and that's just one of the problems.

This is manifested once you actually start using it on a regular basis
(not just a quick test). Specifically, the "secret" interaction of
various options on different tabs. Be on a lookout for that!

Actually, the author himself has totally lost track which is why his
panacea advice was to "delete Vuescan.ini" (which is the equivalent of
reinstalling) or "set the gray point" and ignore all the many options.

Including ICE, of course.

Yes, except as noted Vuescan doesn't do ICE. You can still scan raw
with Nikon Scan (using ICE) and then feed that raw file to Vuescan
using scan-from-disk. But (to me) that's a bit silly because it means
using Vuescan as an image editor and there are far better image
editors out there.

Still, it might be helpful in *some* (non-standard) cases. I wouldn't
use it if what the Nikon does with its own software is good enough, but
in hard cases I just might try both. I'm keeping an open mind for now.

Oh absolutely! I'm not saying you shouldn't use Vuescan. I'm just
pointing out problems to watch out for, to spare you all the nasty
surprises later. You know, something looks very good in the shop, but
it's a whole different story when you bring it home.

Of course, if I need to do too much work to get a an "OCRable" image, I
might as well put the scanned image on screen and type out the text. ;-)

That's the thing! I mean OCR specs may look very good on paper, say,
98% accuracy. But in real life that means 2 out of every 100
characters are wrong! And that's a lot.

....
The attraction of VueScan was that (theoretically) it would be just one
process step. But that advantage disappears if I need to do things like
perspective correction to make the image OCRable in the first place.

As I say, I haven't done a lot of OCR but I think that at least some
of the programs are capable of taking input directly from a TWAIN
source.

(At least I've never used Kodakchrome - I don't need that extra headache
when scanning!)

Like the saying goes: If I knew then, what I know now... ;o)

Ah, yes, life.
I've been back for three weeks (it feels like last week) but half my
head is still in Asia. The brain needs time to process it all, and the
trip was pretty intensive. Half of the rest is reading up on hardware to
spec out a new computer. Half of the rest programming for my travel
blog... Then I still need to write up a lot of stories...
Eerr, life??

It's pretty hectic over here too. I'm on the verge of moving. Maybe...
Not just moving house, but possibly moving to another continent. But
it's all still up in the air which makes it hard to concentrate on all
the other things going on.

Thanks a lot for your comments, Don. They were quite helpful!

Graag gedaan! ;o)

(I lived in the Netherlands for a few years, on and off.)

Anyway, let us know what scanner you do get in the end. I have the
Coolscan V (you are considering) and I'm quite happy with it (aside
from Kodachromes).

Don.

 

[John]

It's pretty hectic over here too. I'm on the verge of moving. Maybe...
Not just moving house, but possibly moving to another continent.

So ..... Mr Hamrick has found out where you live

 

[Marjolein Katsma]

Don ([email protected]) wrote in

No, no, that's not what I mean. What I'm talking about is a(nother)
Vuescan bug: The actual exposure which Vuescan uses to perform the
scan is not the same as the exposure displayed on the screen.

This is not a real example but, say, you enter +2 EV. However the scan
is performed with +1.5 EV! That's bad enough but after the scan the
display then changes to +1 EV!? Vuescan does the equivalent of that
(above a certain exposure value).

He, now that could be rather confusing ... ;-)

It's essential to display all settings and not "hide" them! This is
because of the Vuescan's user "interference". The author apparently
never heard of ergonomics or "muscle memory", and so on.

Maybe he thinks M$ is a good example?

I always turn off such options and immediately go for full display of
menus, options, etc. In *every* program. (Even when installing something
I invariably choose "Custom" so I know what the options are - even if I
accept all of them.)

There is no underlying concept and that's just one of the problems.

I was referring to the terminology used in the UI - some of that still
escapes me but I'll learn.

This is manifested once you actually start using it on a regular basis
(not just a quick test). Specifically, the "secret" interaction of
various options on different tabs. Be on a lookout for that!

Hmm - I think I noted some of that while experimenting; I was clicking
around though, not quite sure. But I remember going - Eh?! *Something*
was a little odd (apart from me).

Actually, the author himself has totally lost track which is why his
panacea advice was to "delete Vuescan.ini" (which is the equivalent of
reinstalling) or "set the gray point" and ignore all the many options.

Well, one could create some different ini files as "presets" for
different types of work, and then use a little batch file to copy the
chosen one to Vuescan.ini and start the program. That way you can't
completely mess up your settings once you have a satisfactory set.

That's the thing! I mean OCR specs may look very good on paper, say,
98% accuracy. But in real life that means 2 out of every 100
characters are wrong! And that's a lot.

Only if there are a lot of characters! On a short text, having to
correct a dozen or so might end up being a lot faster than retyping the
whole text. With some practice you get to be able to estimate whether
OCR is worth it before you start - but I've only worked with flat scans
from paper and magazine clippings so far. Photographed "billboards" with
text, perspective distortion and maybe lighting problems is a whole
different world. And usualy there *is* perspective distortion because
that's the only way to photograph a text. I'll see...

As I say, I haven't done a lot of OCR but I think that at least some
of the programs are capable of taking input directly from a TWAIN
source.

True - but the limiting factor in my case would be the tolerance of an
OCR program (any OCR program) for various distortions and uneven
lighting. To correct those (and I'll probably need to, I've that much
already) you're already editing a scan before it is fed to the OCR
program.

With one of my two test scans (the worst case) it took a lot of tweaking
to even get readable text on screen - but by that time I could just
type, since it was still not good enough for OCR while it *was* good
enough for a human (the original scan definitely wasn't, and just
looking at the contact sheet - even with a magnifying glass - it wasn't
completely readable either).

Anyway, let us know what scanner you do get in the end. I have the
Coolscan V (you are considering) and I'm quite happy with it (aside
from Kodachromes).

It's top of my list for now though the new Minolta model is looking
better than their first take. But I'll (try to) concentrate on a getting
a new computer first. Old Grace (which I'm typing this on) is still
quite trustworthy but at over 5 years old too weak to handle high-res
scanning and image editing. The monitor is sickly, too. But that's
fodder for another NG and/or forum...

I'll keep watching here, and certainly report back when I get a scanner
(and I may need help then!) This is quite a nice newsgroup. Glad I found
it!

 

[Don]

So ..... Mr Hamrick has found out where you live

ROTFL! Very good!

No danger of that , though. Hiding from hordes of furious Vuescan
customers he's too petrified to come out in the open. ;o)

Don.

 

[Don]

....
I always turn off such options and immediately go for full display of
menus, options, etc. In *every* program. (Even when installing something
I invariably choose "Custom" so I know what the options are - even if I
accept all of them.)

Exactly!!! I do the very same thing because I like to know what's
going on. I absolutely hate it when programs "hide" things from me or
try to "do me a favor" I didn't ask for.

At computer shows I always stump the exhibitors ("exhibitionists" as I
like to call them.. ;o)). They usually go through an elaborate
explanation about some feature or other, that will change my life and
do everything for me, etc... I listen patiently and when they're done
I go: Fantastic! Now... How do I turn it off? ;o)

Hmm - I think I noted some of that while experimenting; I was clicking
around though, not quite sure. But I remember going - Eh?! *Something*
was a little odd (apart from me).

Being an eccentric at heart I like odd very much, but Vuescan is more
than a little "odd"! ;o)

Only if there are a lot of characters! On a short text, having to
correct a dozen or so might end up being a lot faster than retyping the
whole text. With some practice you get to be able to estimate whether
OCR is worth it before you start - but I've only worked with flat scans
from paper and magazine clippings so far. Photographed "billboards" with
text, perspective distortion and maybe lighting problems is a whole
different world. And usualy there *is* perspective distortion because
that's the only way to photograph a text. I'll see...

When I finally get to it, I only plan to OCR newspaper articles and
printouts so I should be OK. My few initial test produced pretty good
results especially with a dictionary and a spell check afterwards.
Since this is only for personal use it's not that critical and if
anything is still wrong I can usually figure it out from the context.

With one of my two test scans (the worst case) it took a lot of tweaking
to even get readable text on screen - but by that time I could just
type, since it was still not good enough for OCR while it *was* good
enough for a human (the original scan definitely wasn't, and just
looking at the contact sheet - even with a magnifying glass - it wasn't
completely readable either).

Yes, perspective and other distortions are definitely a problem. I was
pleasantly surprised, though, that when I ran the initial tests my OCR
program was even able to handle different fonts on the same page.

It's top of my list for now though the new Minolta model is looking
better than their first take.

Make sure it's the Mark II model, not the early one. That's because
5400 Mark II has an LED light source. The difference is that Nikon has
3 LEDs (R, G & B) which offers better color control and color purity
while Minolta has a single white LED. Still, with an LED light you
don't have to worry about it fading over time or burning out because
LEDs last "forever".

In addition to higher resolution (5400 vs Nikon's 4000) Minolta also
has single-pass multi-scanning which may come in handy (although I
personally prefer twin scanning). The only thing I worry about Minolta
is reliability. This is all circumstantial evidence and not definitive
but *some* people reported problems after about a year of extensive
use. Again, this is not a firm fact but just something check out.

But I'll (try to) concentrate on a getting
a new computer first. Old Grace (which I'm typing this on) is still
quite trustworthy but at over 5 years old too weak to handle high-res
scanning and image editing. The monitor is sickly, too. But that's
fodder for another NG and/or forum...

This is a small group so we do digress quite often... ;o)

I'm doing all this on a ancient SONY Vaio so I'm way overdue for an
upgrade too. But lots of other things going on right now so all my
upgrade plans are on hold...

I'll keep watching here, and certainly report back when I get a scanner
(and I may need help then!) This is quite a nice newsgroup. Glad I found
it!

Glad you like it! I myself learned a whole lot. Before I joined I
wasted about 6 months with Nikon's so-called "support" (even sending
in a slide) and the whole thing was explained here in a couple of
messages!

Don.

 

[Philip Tobias]

Vuescan is very buggy and unreliable and most of these bugs and
other problems manifest only when you attach a scanner.

Vuescan is capable of very nice scans, in the right hands. I've used
Vuescan quite a bit with a variety of film and flatbed scanners. Good
results.

I've also had good experiences with the program's developer. Once, when
I was having difficulties with an obscure hardware setup, Ed Hamrick was
good enough to work with me personally. He helped me get going again.
I've never had that type of service from the scanner manufacturers.

It's too bad that these positive comments about Vuescan and its
developer sometimes get drowned out in a chorus of negativity. But often
that chorus seems to be just a voice of one. Many others are singing the
praises of Vuescan.

Oh well, that's the Internet for ya. ...pt

 

[Don]

Vuescan is capable of very nice scans, in the right hands. I've used
Vuescan quite a bit with a variety of film and flatbed scanners. Good
results.

That's totally meaningless. Without specifics or facts it's just one
person's subjective *feelings*. What's "good"? Web JPGs? How closely
have you examined the results? Can you provide objective (!)
*independently verifiable* (!) data? Etc. Etc.

Those who have, know just how buggy and unreliable Vuescan is.

I've also had good experiences with the program's developer. Once, when
I was having difficulties with an obscure hardware setup, Ed Hamrick was
good enough to work with me personally. He helped me get going again.
I've never had that type of service from the scanner manufacturers.

That's funny... In public he's bad tempered, spews obscenities and
sends petty "blacklist" emails to users who dare complain about bugs.

Oh, wait... It's you Ed, isn't it? ;o)

It wouldn't be the first time either for the Vuescan author to pretend
being a "satisfied Vuescan customer"... See below what *real*
customers think of Vuescan!

It's too bad that these positive comments about Vuescan and its
developer sometimes get drowned out in a chorus of negativity. But often
that chorus seems to be just a voice of one. Many others are singing the
praises of Vuescan.

Riiiight... Instead of shooting the messenger tell that to the
exasperated "VuescaM" users below!

Don.

--- arbitrary start ---

I had a similarly negative experience with VueScan:
I tried VueScan with the Minolta Dimage Scan MultiPRO and found it
unusable because of severe banding problems.


Unfortunately, to date VueScan is not capable of scanning the Raw data
with a linear gamma...

Yes that'll be one of the effects the VueScan D-max bug will cause.

But: being a novice in the
trade I could not determine for myself that what was claimed: Vuescan
supports Minolta Scan Dual IV, wasn't true.


Tried that. It doesn't help.
I tried all the avenues that Vuescan allowed and no combination of features
provided a good scan. Your suggestion even made things worse.

So this bug has survived through two subsequent versions to 8.1.13,
rendering Vuescan more-or-less useless, if you use scan-from-disk
workflow.

ICE manages to clean my problematic slides very well, doing a much more
complete job, and much more "seamlessly". Vuescan leaves so much, and
leaves obviously softened areas.

I'm really getting tired of even
trying new releases, it's a time consuming waste of time.

Somewhere around recent version .20 "something bad happened" to Vuescan
speed. Since then, several new version descriptions have promised
greatly improved speed etc. Atleast as of .23, my personal experience
is it's still very pokey.

About a couple of weeks ago I bought Vuescan to use with my brand new
Minolta. I was worried about reports of lines but was told that has
been fixed. IT HASN'T!! The damn lines are everywhere! Vuescan is total
CRAP! I wrote two emails but got no reply and I'm really fed up and
pissed off! I WANT MY MONEY BACK! What a ripoff! It's Vue-SCAM! That's
what it is!

I'm using VueScan with Canon FS4000US over SCSI connection. Just
upgraded from 8.1.32 to 8.1.36 and noticed a problem with "Preview"
command. In version 36 it takes forever, compared to version 32.
Apparently, version 36 does preview at full resolution (4000dpi) even
though the "Input | Preview resolution" is manually set to mere 500dpi.

I just updated to 8.2.03, and I'm getting "double" images side by side
of the SAME scans in the preview OR scan window..

Eddie Wiseman

...After I disabled batch mode and pressed
'Scan', VueScan went on to scanning all six frames in batch mode,
despite that fact that I explicitly asked it to scan only one frame.

What's going on with VueScan? Apparently, nobody is even trying to do
even the most basic testing of the new version before the release.

So it looks like a serious bug with the cropping system, as you suggest. Don
will say "told you so" -well, he did! Stick to your working version.
Upgrade at your peril!

Would if I could. I'm on his black list, since venting here regarding
Vuescan's undocumented feature of assigning icc profiles to raw file if
scan-from-disk outputting new raw file is done "at save".

Do you know for certain that you are blacklisted and that it is a result of
your posts to this NG? I don't remember any of your posts regarding the
above as being particularly damning.

The thread titled:

"Vuescan raw files saved "at save" have altered color balance"

prompted Ed Hamrick to email me with the good news. He mentioned his
action was in light of my recent Usenet posting.

Well I just tried the latest version, 8.3.03, and I'm seeing the same
exact effects. ....
the appearance of dust spots and imperfections actually look WORSE on
the "light" setting than they do with "none". And since the "medium"
setting begins to blur, it basically makes the dust removal almost
unusable.

--- no end... ---

 

[Marjolein Katsma]

Philip Tobias ([email protected]) wrote in (e-mail address removed):

It's too bad that these positive comments about Vuescan and its
developer sometimes get drowned out in a chorus of negativity. But
often that chorus seems to be just a voice of one. Many others are
singing the praises of Vuescan.

IMO they're only getting "drowned out" if you're reading selectively.

I was reading up on scanners, saw a reference to a separate scanner
program, dug "vuescan" out of my memory and went googling. That's how I
found this NG. I proceeded to read here what I could find about VS and
saw both positive and negative comments, as well as useful information
related to both. Saved what I found useful (a lot) before I even made my
first post here.

No non-trivial program is bug-free -- and a scanner program necessarily
is non-trivial. As a software developer, and ex-product manager, I know
about the "house of cards" phenomenon, and that non-trivial software is
particularly vulnerable to that after a long development period with
many updates. All that said, none of that means that a piece of software
is useless, or bad. Even a buggy piece of software can be very useful if
you learn about its limitations and strengths and use it wisely.

Oh well, that's the Internet for ya.

Well, the internet allowed me to find this NG - which is surpisingly
civilized and helpful compared to many NGs. I'll stick around.

 

[Marjolein Katsma]

Don ([email protected]) wrote in

At computer shows I always stump the exhibitors ("exhibitionists" as I
like to call them.. ;o)). They usually go through an elaborate
explanation about some feature or other, that will change my life and
do everything for me, etc... I listen patiently and when they're done
I go: Fantastic! Now... How do I turn it off? ;o)
LOL.



Make sure it's the Mark II model, not the early one. That's because
5400 Mark II has an LED light source. The difference is that Nikon has
3 LEDs (R, G & B) which offers better color control and color purity
while Minolta has a single white LED. Still, with an LED light you
don't have to worry about it fading over time or burning out because
LEDs last "forever".

Yes, I was referring to the "Mark II" model. I'm well aware already of
the advantage of LEDs: my old little PrimeFilm scanner was already hard
to use and the software frequently "losing" its settings, so when the
lamp burned out - and it didn't look like I could easily replace that
myself: I screwed the dratted thing open - I just gave up on it. I
already knew I needed something better anyway.

(Anyone want a broken PrimeFilm?)

In addition to higher resolution (5400 vs Nikon's 4000) Minolta also
has single-pass multi-scanning which may come in handy (although I
personally prefer twin scanning).

Yes, some boxes ticked there. But is the higher resolution really
relevant? I read opposing opinions about that - so that box is only
half-ticked.

The only thing I worry about Minolta is reliability. This is all
circumstantial evidence and not definitive but *some* people reported
problems after about a year of extensive use.

Anecdote: A friend of mine got one of the first machines (mark I) - a
show model but barely used - as a special deal. Within two months
something had broken (I forget what); they replaced it though, and since
then he's had no more problems. Could have been teething problems in
production (if so, they could re-appear with a new model).

Rumor: I heard it's made in China - and in a factory where QA isn't very
good. And I heard that from a Chinese. He told me *not* to buy a Minolta
scanner... ;-)
(Not that the Chinese aren't able to make good quality products - but
much depends on how a factory is run.)

Certainly after my experience with the PrimeFile reliability is high on
my list - especially with what this class of scanners (Nikon/Minolta)
costs here in the Netherlands. I see $600 quoted for the Minolta - but
that's in the US - here it's closer to $900-1000 but prices vary widely
and warranty (Dutch or "European") makes a difference as well. So, if
I'm going to spend that kind of money (a lot for me - and that's apart
from needing a enw computer for all this as well) I want to spend it
wisely. I'm not going to hurry.

 

[Don]

Yes, some boxes ticked there. But is the higher resolution really
relevant? I read opposing opinions about that - so that box is only
half-ticked.

Nominally a 4000 dpi scanner should get everything on the film. The
only exception may be a perfectly exposed shot taken with a tripod
and high quality film. The key word here is "may". It also depends on
scanner optics, etc.

However, not all grain is of the same size. Every film has some grain
which is very small. The consensus seems to be that in order to get
every piece of grain regardless of film rating is to have a scanner
with about 10,000 dpi.

But there's a catch. At this level of precision the scanner does not
see film as flat. The film also has a depth (i.e. thickness). That's
why grain is often called "grain clouds".

Given all that, 4000 is most likely to be enough resolution but 5400
doesn't hurt, although scanner optics or the shot may make the
difference hard to see. Finally, as I like to say, a scanned image is
really "a picture of a picture". When one scans film one is taking a
picture of this film, not of the original subject. (That's why I went
digital to avoid this intermediate step.)

Anecdote: A friend of mine got one of the first machines (mark I) - a
show model but barely used - as a special deal. Within two months
something had broken (I forget what); they replaced it though, and since
then he's had no more problems. Could have been teething problems in
production (if so, they could re-appear with a new model).

You're right. It may haven been just teething problems and if he
hasn't had any trouble since that's definitely a very good sign even
though it's the previous model.

Certainly after my experience with the PrimeFile reliability is high on
my list - especially with what this class of scanners (Nikon/Minolta)
costs here in the Netherlands. I see $600 quoted for the Minolta - but
that's in the US - here it's closer to $900-1000 but prices vary widely
and warranty (Dutch or "European") makes a difference as well. So, if
I'm going to spend that kind of money (a lot for me - and that's apart
from needing a enw computer for all this as well) I want to spend it
wisely. I'm not going to hurry.

That's why companies hate the Euro! They can't price their products
differently in different European countries anymore.

You may also consider ordering one from the States. These days most
electronic products come with a universal power supply rated 110 - 240
volts. When I lived in Germany the last few things I bought (an
external CD burner and a couple of USB drives) came with such a power
supply. All I need to use them in Canada is a plug converter. Even my
ancient Sony Vaio I bought in the US over 5 years ago has a universal
power supply.

Now, there are problems with ordering out of US from Europe because
the warranty is void. If something goes wrong it may be expensive to
have it fixed. However, in the UK, for example, it's very common for
traders to import stuff from the US (so-called "gray imports"). Again,
companies hate that and sometimes even refuse to repair such products.

Don.

 

[Don]

IMO they're only getting "drowned out" if you're reading selectively.

Exactly! There are also *reasonable* Vuescan users who read in full
and don't object because their specific requirements are met by
Vuescan. They don't perceive it as an "attack" if someone points out
negative facts about Vuescan. They just shrug and continue using it.

However, there is a small but vocal minority (so-called "Vuescan
Taliban") who have a huge chip on their shoulder and immediately lash
out with indiscriminate rabid attacks but, notably, without any facts.

As has been shown over time (in at least two documented cases) this
fury was really at Vuescan. They were just mad that Vuescan was so
buggy and - instead of taking the author to task - they would rather
shoot the messenger.

In the end, those particular two came out and admitted how frustrated
they are with Vuescan, thereby implicitly eating their words and
taking back all the angry outburst and insults thrown at people who
calmly told them that in the past.

No non-trivial program is bug-free -- and a scanner program necessarily
is non-trivial. As a software developer, and ex-product manager, I know
about the "house of cards" phenomenon, and that non-trivial software is
particularly vulnerable to that after a long development period with
many updates. All that said, none of that means that a piece of software
is useless, or bad. Even a buggy piece of software can be very useful if
you learn about its limitations and strengths and use it wisely.

I myself have a couple of decades of software development under my
belt ranging from assembly programming to project management and, over
a wide variety of platforms ranging from IBM mainframes to PCs.

However, I must say that I have never seen a program as sloppy as
Vuescan. It really is in a league of its own as the quotes (by Vuescan
users *themselves*!) in the previous message clearly document.

Having said that, I still agree with you. It's quite possible for some
people to find Vuescan useful. If their requirements are low, and they
follow a few careful guidelines (don't upgrade automatically, keep old
versions, etc) they can have a relatively hassle-free time of it.

The problem is when some of them (like our friend Philip, here)
translate that very narrow paradigm into sweeping and utterly
meaningless statements such as "Vuescan is good" or "I like it",
neither of which carry any weight without supporting evidence. And
then, vague as all that is, use this irrelevant subjective feeling to
indiscriminately attack people who merely point out objective facts.

Well, the internet allowed me to find this NG - which is surpisingly
civilized and helpful compared to many NGs. I'll stick around.

Yes, the group is amazingly civilized. Probably because it's
relatively small. That also makes it easy to follow and it doesn't
take long to read. In return the knowledge gained is quite high.

Curiously, over the 3-4 years I've been reading the *only* person to
shout obscenities here was the Vuescan author himself! That alone
speaks volumes! There were a few near-expletives, here and there,
again voiced by Vuescan "supporters" which is also quite telling...

Don.

 

[Philip Tobias]

That's totally meaningless. Without specifics or facts it's just one
person's subjective *feelings*. What's "good"? Web JPGs? How closely
have you examined the results? Can you provide objective (!)
*independently verifiable* (!) data? Etc. Etc.

I have been producing professional photography, writing, and/or graphic
design work for over 30 years. That work appears in magazine ads,
brochures, books, flyers, posters, press kits, Web sites, and whatever
you can imagine. One color, two color, PMS, four color, five color,
varnish coats, die cuts and folds.

Today I picked up samples of two different four-color projects in which
my work recently appeared. The ad agency and printer should be commended
on the exceptional reproduction. Clean color, a full range of tones, nice.

Besides those credentials, I have personally used Vuescan to scan images
for all sorts of projects. It has performed well. I have also used
scanning software from various manufacturers and third-party suppliers.

If a fraction of the time spent whining about Vuescan were spent doing
something positive for humanity, this world would be a better place.

....pt

 

[Marjolein Katsma]

Don ([email protected]) wrote in

Exactly! There are also *reasonable* Vuescan users who read in full
and don't object because their specific requirements are met by
Vuescan. They don't perceive it as an "attack" if someone points out
negative facts about Vuescan. They just shrug and continue using it.
Right.

However, there is a small but vocal minority (so-called "Vuescan
Taliban") who have a huge chip on their shoulder and immediately lash
out with indiscriminate rabid attacks but, notably, without any facts.

Well, I'd shrug about that, too. And not archive such posts. ;-)

If their requirements are low, and they follow a few careful
guidelines (don't upgrade automatically, keep old versions, etc)

Having read how I install software, what'd you guess I do with old
versions? Or old beta versions, for that matter? ;-)

 

[Marjolein Katsma]

Don ([email protected]) wrote in

Finally, as I like to say, a scanned image is
really "a picture of a picture". When one scans film one is taking a
picture of this film, not of the original subject. (That's why I went
digital to avoid this intermediate step.)

Well, it's always better than scanning a print made from a negative:
that would be a picture of a picture of a picture.

But I'm not ready for digital - getting a camara with (by now)
equivalent resolution would be a huge investment, quite apart from the
fact that I'd need the film scanner and new computer anyway.

And I'm not quite convinced a high-resolution digital camera would by
itself be sufficient for (say) a 9-week trip across Asia. And that would
mean extra hardware to store the pics externally - and extra weight to
carry.

My Canon Eos 5 with one (Tamron) "travel lens" is doing just nicely for
me now. (And I have my little 2M camera phone as a backup, no need to
carry a spare camera any more, that saves some weight.)

You may also consider ordering one from the States. These days most
electronic products come with a universal power supply rated 110 - 240
volts. When I lived in Germany the last few things I bought (an
external CD burner and a couple of USB drives) came with such a power
supply. All I need to use them in Canada is a plug converter. Even my
ancient Sony Vaio I bought in the US over 5 years ago has a universal
power supply.

Now, there are problems with ordering out of US from Europe because
the warranty is void. If something goes wrong it may be expensive to
have it fixed.

I'm not worried about the power supply - but I *am* worried about
warranty. (I have brought hardware from the US with me, but smaller
stuff, and always a carefully calculated risk. In once case I could buy
extended international warranty, and the price difference was still
huge.)
Even here, where the Nikon scanner with "European" warranty is cheaper
than one with the Dutch warranty, I worry about that. I've also looked
at ordering in Germany where such hardware and electronics in general is
a lot cheaper than here (and that's not just the difference in VAT
percentage) - but one problem and just shipping costs for repair could
eat up the price difference.

I tend to think it may be better to accept a little higher price and
order from my local (pro) photostore. Official dealer and all, and
always excellent and personal service from the store. (And I could try
to haggle a bit. ;-))