> Computers > Peripherals > Scanners
Various Topics Home | Disclaimer | Report Adult Posts

Various Topics on Scanners



Scanners - "Vuescan bug with updating info box?" in Peripherals


Old 05-29-2006   #1
..ger.. ..ail.c..
 
Default Vuescan bug with updating info box?

In the info box at the bottom of the app, even though scan0003.tif (raw
file) was being currently scanned in and displayed (as well as other
raw image files in the batch list), the info box kept stating
"scan0001.tif" was being scanned in!

Vuescan-83 & Gentoo Linux

 
Old 05-29-2006   #2
..n
 
Default Re: Vuescan bug with updating info box?

On 28 May 2006 15:45:18 -0700, "rogerxx@gmail.com" <rogerxx@gmail.com>
wrote:

>In the info box at the bottom of the app, even though scan0003.tif (raw
>file) was being currently scanned in and displayed (as well as other
>raw image files in the batch list), the info box kept stating
>"scan0001.tif" was being scanned in!
>
>Vuescan-83 & Gentoo Linux


Annoying as that may be but that's really cosmetic and the least of
your worries...

As you've discovered and do***ented in parallel messages (Auto Setting
Anomaly, Seg fault on vuescan, invalid pointer, etc...) VueScan has
much bigger problems! At best, it may only be suitable for low quality
quick-and-dirty highly compressed Web JPGs or tiny prints. VueScan is
bug-ridden and notoriously unreliable causing m***ive corruption in
image data even when it pretends to "work".

I realize you already know all that and on Linux your options are
fairly limited but it helps to put things into perspective and not
expect too much from VueScan. Quite the contrary!

Now, I gather from your other messages that you know your way around
the system so - even though it's a long shot - but have you considered
rolling your own? At the most fundamental level a scanner program is
really simple data acquisition and not very difficult to write. For
what it's worth I put together a scanner program in less than a week
(albeit different OS and different scanner). So, it's not as daunting
as one may think as long as you leave out all the image editing stuff
which, actually, doesn't belong in a scanner program anyway. Doubly so
for Linux/Unix systems which are inherent modular from the ground up
i.e. why repeat Gimp functionality in a scanner program.

Anyway, I haven't really done any scanner programming on Linux but
there's the SANE library which, I believe, should make things easier.
In the long run, and depending on how much scanning you have to do,
it's far less time consuming to write your own and be done with it
instead of permanently chasing the never ending VueScan bugs.

Don.
 
Old 05-29-2006   #3
..ger.. ..ail.c..
 
Default Re: Vuescan bug with updating info box?

Yes. Thanks for the verification. As good as things are or as bad as
they are, I rarely believe my own eyes at times and rely on 2nd
opinions.

As far as the info box, it doesn't help to know I'm still scanning
scan0001 and I'm really scanning image=image+1 and things are working
properly. The application is still lying to me! :-)

You've really peaked my interest on this one. If I were in school for
Programming, I would readily pick-up this task. Sane needs better
support for batch scanning negatives & photos.

Until you mentioned, I wasn't realizing most of the programming is
already done and all one would need to do is impliment the already
written routines!

Sane only seemed to be missing the color profiles for negatives along
with batch scanning and raw output -- for which, is odd.

 
Old 05-29-2006   #4
.... .. ....
 
Default Re: Vuescan bug with updating info box?

Please ignore Don: he is a troll...

With kind regards,
Henk de Jong

--
http://www.hsdejong.nl/
Nepal and Burma (Myanmar) - Photo Galleries


 
Old 05-30-2006   #5
..n
 
Default Re: Vuescan bug with updating info box?

On 29 May 2006 10:00:49 -0700, "rogerxx@gmail.com" <rogerxx@gmail.com>
wrote:

>Yes. Thanks for the verification. As good as things are or as bad as
>they are, I rarely believe my own eyes at times and rely on 2nd
>opinions.


I run lots of tests because data doesn't lie and our eyes lie all the
time. We may misinterpret the data but that's a different story. ;o)

>As far as the info box, it doesn't help to know I'm still scanning
>scan0001 and I'm really scanning image=image+1 and things are working
>properly. The application is still lying to me! :-)


I know the feeling! :-) I just don't like it when applications do that
because I can't trust them after that. I mean, if they manage to screw
up something so elementary what about all the other (complicated)
things going on under the hood?

>You've really peaked my interest on this one. If I were in school for
>Programming, I would readily pick-up this task. Sane needs better
>support for batch scanning negatives & photos.
>
>Until you mentioned, I wasn't realizing most of the programming is
>already done and all one would need to do is impliment the already
>written routines!


As I say, I haven't really done any scanner stuff on Linux but I do
know that SANE already has a whole bunch of routines.

>Sane only seemed to be missing the color profiles for negatives along
>with batch scanning and raw output -- for which, is odd.


In theory you could do batch with a script. That's where Unix shines.

Raw is really nothing special. It's simply a scan with all the editing
turned off. You can take any scanner program, turn off curves,
brightness, etc. or leave them in a neutral or "flat" state and you'll
get a raw scan. Technically, you should also use linear gamma but most
people don't bother with that since their monitors aren't linear.

The scanner profiles are of very limited use because they often go
against the final edit. Also, you can always ***ign some profiles
later. I personally prefer to scan negatives as positive and do
everything manually i.e. remove the orange mask and stretch the
dynamic range plus add some contrast enhancement with an "S" curve.

Don.
 

Thread Tools
Display Modes





Powered by vBulletin®
Copyright ©2000 - 2009, Jelsoft Enterprises Ltd.
Search Engine Friendly URLs by vBSEO 3.3.0