filmscanners: VueScan 6.4.6 Available

[EdHamrick]

I just released VueScan 6.4.6 for Windows, Mac OS and Linux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.4.6

  * Added support for manual focus on Canon FS2700/FS2710

  * Added support for RGB exposure on Canon FS2710
(can expose CCD at 1x, 2x, 3x, 4x, and 6x times)

  * Added support for UMAX PowerLook 3000

  * Added support for UMAX Vista S6

  * Added support for AGFA Saphir Ultra 2

  * Added support for AGFA DuoScan (original model)

  * Added support for double optical resolution on
some UMAX scanners (i.e. PowerLook 3000, Mirage)

  * Color profiles are no longer written to B&W files

  * Fixed several buffer size problems that caused
aborted scans with some scanners at high resolution

Regards,
Ed Hamrick

Re: filmscanners: Sprintscan 120 now on B+H web site ...

[EdHamrick]

In a message dated 1/8/2001 11:55:37 PM EST, [EMAIL PROTECTED] writes:

> The 3.9 dynamic range sounds unbelievable. I wonder how they achieve that?

3.9 just means 13 bits of dynamic range.  They're using a 14-bit A/D
converter, which most vendors convert to a dynamic range of 4.2.
I suspect Polaroid is just being conservative.

Note that the Nikon 4000 ED and 8000 ED (I love these names )
have a 14-bit A/D converter, and Nikon says the dynamic range is 4.2.
The CoolScan IV ED has a 12-bit A/D converter, and Nikon says the
dynamic range is 3.6 (log10(2^12) = 3.6).

In summary, dynamic range is just another way of saying how
many bits the A/D converter uses:

10 bits = 3.0
12 bits = 3.6
14 bits = 4.2

Regards,
Ed Hamrick

Re: filmscanners: New scanners from Nikon

[EdHamrick]

In a message dated 1/9/2001 3:58:44 AM EST, [EMAIL PROTECTED] writes:

>  Nikon Coolscan IV ED, LS-40
>   Nikon Super Coolscan 4000 ED, LS-4000
>   Nikon Super Coolscan 8000 ED, LS-8000
>  
>4000 dpi and 14 bit/channel

The CoolScan IV ED (LS-40) is a 2900 dpi scanner
with 12 bits/channel.  All 12 bits are read by NikonScan 3.0.

Also, the LS-40 has a USB interface and the LS-4000
and LS-8000 have Firewire (IEEE 1394) interfaces.

The LS-4000 is for 35mm and APS, while the LS-8000
adds support for MF film.

All three scanners have a separate infrared channel
and use LED's for their light source.

There's also a new roll film adapter for the LS-4000
called the SA-30 which takes a whole roll of uncut
film.

The new ED scanners (love that name) use the same
film adapters as the LS-2000 series (including the
SF-200 slide film adapter).  It looks like the firmware
in the LS-40 doesn't support the SF-200 slide feeder
(just like the LS-30 doesn't).

Regards,
Ed Hamrick

Re: filmscanners: Vuescan

[EdHamrick]

In a message dated 1/8/2001 5:15:41 PM EST, [EMAIL PROTECTED] writes:

> How does the "Clean" option work in Vuescan with the Minolta Scan Dual II,
>  HP S20 and other scanners without Digital ICE.

Unfortunately, the Clean option just doesn't work very well without
the infrared channel of the Minolta Scan Elite, Nikon LS-30 and
Nikon LS-2000.  I hope to improve this, but it's hard to clean
dust spots without an infrared channel without also losing image detail.

Regards,
Ed Hamrick

filmscanners: Kodak RFS 3600 Information

[EdHamrick]

I've been doing a bit of research on the Kodak RFS 3600
film scanner, and found out a few interesting things.

It's actually made by Pacific Image Electronics, the
same company that makes the PrimeFilm 1800u film
scanner.  The RFS 3600 is apparently just the next
generation of the PrimeFilm 1800u (note that 2 x 1800
is 3600 dpi).  If you compare the physical dimensions
of these two scanners as well as the film feeding, you'll
see the similarities (Kodak apparently added more
plastic to the case to make it look fancier).

When I first learned of the RFS 3600 I wondered why
it was a 3600 dpi scanner and not a 4000 dpi scanner.
It turns out that most film scanners use CCD chips that
were originally designed for flatbed scanners (which is
a much bigger market than film scanners), and most
flatbed scanners are 8.5 inches wide.  A 300 dpi
flatbed scanner needs a 2550 pixel CCD, a 600 dpi
flatbed needs a 5100 pixel CCD, and a 1200 dpi
flatbed needs a 10200 pixel CCD.

Most 2700 dpi film scanners use the 2550 pixel CCD's,
since this lets you scan 24mm wide film.  The Polaroid
4000 dpi scanner uses a 5100 pixel CCD (they could
have designed it to go up to 5400 dpi, but I suspect they
didn't because they didn't want to build a 5400 dpi
stepper motor).  It appears that the Polaroid SprintScan 120
uses a 10200 pixel CCD (10200 pixels over 60 mm is about
4000 dpi).

It appears that the PrimeFilm 1800u uses a 2550 pixel CCD,
but this has to scan a 36mm width (because of the way the
film is fed from left to right) but the scan head travels
in the 24mm direction.  This means that the maximum
dpi of the 2550 pixel CCD over a 36mm width is 1800 dpi.
Similarly, the 5100 pixel CCD in the Kodak RFS 3600
spread over a 36mm width results in 3600 dpi.
(5100 * 25.4 / 36 = 3598 dpi).

The PrimeFilm 1800 uses parallel-port scanner firmware
internally, and converts USB to parallel port commands
internally (it's a huge kludge).  I suspect (but don't know for
sure) that the Kodak RFS 3600 uses the same internal
firmware as one of the PIE SCSI flatbed scanners, with
an optional USB->SCSI converter chip.

In summary, there appear to be mainly 3 sizes of CCD chips
in use in film scanners today.  Most use 2550 pixel CCD's,
the SprintScan 4000 uses a 5100 pixel CCD, and the
SprintScan 120 uses a 10200 pixel CCD.

Regards,
Ed Hamrick

http://www.hamrick.com/vsm.html

filmscanners: VueScan 6.4 Available

[EdHamrick]

I just released VueScan 6.4 for Windows, Linux and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.4

  * Added buttons for Preview, Scan, Abort

  * Added options to specify folders for TIFF, JPEG,
Index, Raw and Disk files (in File menu)

  * Added option to control preview resolution

  * Added support for AGFA Arcus 1200

  * Fixed problem with SnapScan 1212U

  * Turned off blinking crop box on Mac OS by default
(it makes the updating of controls not work right)

Regards,
Ed Hamrick

Re: filmscanners: Vuescan scratch removal

[EdHamrick]

In a message dated 12/16/2000 2:56:12 AM EST, [EMAIL PROTECTED] writes:

> What's peoples opinion of the Vuescan scratch removal on scanner without the
>  IR channel?

It doesn't work especially well without the IR channel.

>  Anyone know what method it uses for this?

It's a nonlinear filter, with the weight of each pixel proportional
to it's brightness.

>  Better than PS or not?

I suspect Photoshop's filter is better, and if someone can
send me a technical description of Photoshop's filter, I'll
consider adding it to VueScan.

Regards,
Ed Hamrick

Re: filmscanners: Ed's VueScan site

[EdHamrick]

In a message dated 12/15/2000 12:26:08 PM EST, [EMAIL PROTECTED] writes:

> Is anyone else having problems getting on to ed's site?
>  
>  He told me a couple of days ago he had changed the server, but I still 
can't
>  on...

I changed to a new server, and it looks like the traffic has shifted to
the new server successfully.

One or two people per hour are still accessing the old IP address,
and this appears to be caused by messed up domain name
servers or possible proxy servers.  I'm not sure what to do other
than recommending being patient (or directly using the
IP address - http://161.58.152.246/vsm.html

Regards,
Ed Hamrick

Re: filmscanners: RE: Film Scanners and what they see.

[EdHamrick]

In a message dated 12/14/2000 5:27:42 AM EST, [EMAIL PROTECTED] writes:

> The Minolta Elite seems to apply ICE in different ways.  If you have 
>  done a preview of a slide with ICE on, the final scan is done in a 
>  single pass.  If you have done a preview with ICE off but then switch 
>  it back on for the final scan, it does two passes. 

I've traced the Minolta software's scsi commands, and all it's
doing is another pass to compute the CCD exposure time.

The log of the scsi commands clearly shows that the infrared
and rgb channels are sampled all in one pass.  The log of the
scsi commands on the ScanWit 2740 clearly shows two
different scan passes, one collecting only the infrared channel
and one collecting the rgb channels.  The visible light is also
reportedly turned off during the infrared scan pass.

>  Interestingly, negs don't clean up as well when scanned as negs.  But 
>  when scanned as slides and inverted/colour mask removed using Vuescan, 
>  they come up very well.  So Minolta obviously implements ICE 
>  differently for slides and negs.

I suspect they messed up the correction, doing the inversion of the
negative color before the ICE algorithm.

Regards,
Ed Hamrick

Re: filmscanners: RE: Film Scanners and what they see.

[EdHamrick]

In a message dated 12/13/2000 3:18:02 PM EST, [EMAIL PROTECTED] writes:

> > The ScanWit 2740 is a special case, since it does an entire separate
>  > scan pass just to get the infrared channel.
>  
>  Interesting. Do they just flip an IR filter in front of the CCD and use 
the 
> red channel output then?

I haven't actually seen one, but I have a copy of the scsi log files.
I was told that the main light goes out during the infrared pass.
I suspect it turns on an infrared LED during the infrared pass.

Regards,
Ed Hamrick

Re: filmscanners: RE: Film Scanners and what they see.

[EdHamrick]

In a message dated 12/13/2000 6:41:32 AM EST, [EMAIL PROTECTED] writes:

> Bad luck for most of the people outside this list!  I could not be more 
>  enthusiastic about a single feature of a scanner.

I have to agree - the infrared dust removal is a really nice feature
in a film scanner.  It only takes a long time with the ICE algorithm,
since ASF hasn't optimized it very well.  The algorithm in VueScan
doesn't add much to the total scan time.

>  >I've just put a stopwatch to it, and ICE makes scans over 6 times slower 
>  >on the 2740.

The ScanWit 2740 is a special case, since it does an entire separate
scan pass just to get the infrared channel.  I assume Acer fixed the
multi-pass registration problem of the ScanWit 2720, since the
dust removal won't work right if the infrared and color passes aren't
perfectly aligned.

Regards,
Ed Hamrick

Re: filmscanners: IR & dust removal (was VueScan 6.3.19 Available)

[EdHamrick]

In a message dated 12/12/2000 0:41:36 AM EST, [EMAIL PROTECTED] writes:

> And after giving some thought, why should even the Nikon software
>  soften the image?

It shouldn't, but it does.

>  If the IR channel is "flat" ... that is, indicates
>  no "defects", why would any algorthym be applied??

It turns out that the digital ICE algorithm has a color shift
when it's applied, and if it were applied in some areas and
not others there would be a noticable color change between
these areas.

>  Can I assume this version of Vuescan acquires the IR channel no
>  differently than past versions?

That's correct - it's just the raw infrared data that's acquired.

Regards,
Ed Hamrick

Re: filmscanners: VueScan 6.3.19 Available

[EdHamrick]

In a message dated 12/11/2000 8:19:34 PM EST, [EMAIL PROTECTED] writes:

> Now I'm confused (not hard to do).  Please explain to me "where" ICE
>  is.is it software or hardware?  Or a combination of both to get that
>  capability?

ICE is the marketing name for a dust removal technique invented
by Albert D. Edgar while he was working at IBM (he's at Applied
Science Fiction now).

The patent can be read at:

  http://www.delphion.com/details?&pn10=US05266805

It's U.S. Patent 5,266,805

This patent only describes the way the infrared channel
is used to correct the image, but it has several problems.

The first problem is that it assumes the infrared channel
doesn't show any image data, but in reality the infrared
channel isn't flat (especially for Kodachrome).  The
second problem is that the infrared and color channels
aren't perfectly aligned, which causes the edges of dust
spots to not disappear when the algorithm in this patent
is used.

VueScan uses an entirely different (and I think better)
approach to using the infrared channel to remove the
dust spots.  It doesn't result in any color shift, it works
with Kodachrome, and it doesn't apply a softening filter
in areas where there's no dust spot.

Regards,
Ed Hamrick

Re: filmscanners: VueScan 6.3.19 Available

[EdHamrick]

In a message dated 12/11/2000 12:34:38 PM EST, [EMAIL PROTECTED] writes:

> Imagine being able to select only certain
>  regions where dust is noticeable, and apply your
>  filter plugin selectively(?!)  This would go a long 
>  way for those who choose not to use it because it
>  will soften detail.

Interestingly, the cleaning algorithm in VueScan (which is
completely different than ICE) doesn't soften the image
at all, except in the area around actual dust spots.  The
ICE algorithm softens the image throughout when used.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.19 Available

[EdHamrick]

I just released VueScan 6.3.19 for Windows, Mac OS and Linux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.19

  * Significantly improved image cleaning on scanners
with an infrared channel (Scan Elite, LS-30/LS-2000)

  * Separated filter option into clean and sharpen options

  * Fixed small problem with blue channel calibration on
some Minolta Scan Speed and Scan Elite scanners

Regards,
Ed Hamrick

Re: filmscanners: Re: Umax scanners

[EdHamrick]

In a message dated 12/9/2000 8:13:22 AM EST, [EMAIL PROTECTED] writes:

> Ed, does vuescan work with any USB scanners in Windows and if so, which
>  ones?

This list is on:

  http://www.hamrick.com/vsm.html

>  Also, do you (or does anyone on the list) know whether Plustek Optic Pro
>  scanners are rebadged Umax or Microtek scanners?

No, Plustek scanners use their own unique commands.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.18 Available

[EdHamrick]

I just released VueScan 6.3.18 for Windows, Mac OS and LInux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.18

  * Fixed problems with some UMAX scanners

  * Fixed problem with Epson Perfection 1640

  * Added support for LinoType scanners that
are manufactured by UMAX

Regards,
Ed Hamrick

Re: filmscanners: words of wisdom from St Timo snipped from NG

[EdHamrick]

Most of Timo's mistaken impressions about scanning negatives
come from his trying to use a UMAX PowerLook III scanner
to scan negatives.  He was spectacularly unsuccessful doing
this, and he concluded that it was therefore impossible to
scan negatives with consumer film scanners.

I sent him an e-mail earlier today suggesting that he try
VueScan with his PowerLook III.  VueScan does a nice job
scanning negatives with the PowerLook III, and is an
existance proof that it's possible.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.17 Available

[EdHamrick]

I just released VueScan 6.3.17 for Windows, Mac OS and Linux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.17

  * Added support for AGFA SnapScan and Acer scanners

  * Fixed problem with UMAX 1200S

  * Fixed Macintosh problems with some scanners
(stack overflow problems)

Regards,
Ed Hamrick

filmscanners: VueScan Plans

[EdHamrick]

I just thought I'd let the group know what my short-term
plans are for VueScan development, in case people
would like to make suggestions about this.

I just finished adding UMAX support to VueScan, and
I'm now working on adding support for Acer flatbeds and
Acer-rebadged scanners (this includes the AGFA SnapScan
scanners).  In particular, I'm adding support for the
SnapScan e50 which has a strip film adapter.  These
changes will be in 6.3.17.  It's almost working now, since
it only involves small adaptations to the ScanWit code.

Once this is done, I hope to begin implementing the
user interface improvements that people have been
asking for.  These will be in 6.4.

Regards,
Ed Hamrick

Re: filmscanners: Vuescan support of Umax Astra 3400?

[EdHamrick]

In a message dated 12/3/2000 4:22:04 PM EST, [EMAIL PROTECTED] writes:

> Has anyone had any luck with Vuescan and the Umax Astra series
>  scanners?  I may order an Astra 3400 (USB connection, not SCSI) and am
>  curious if you believe Vuescan would support this scanner or someone has
>  verified that it does indeed work.

I bought an Astra 3450 and decoded the USB commands that it uses.
It doesn't use the same command set as the SCSI UMAX scanners, but
instead uses the National Semiconductor scanner-on-a-chip command
set.  VueScan doesn't (and probably never will) support this scanner.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.15 Available

[EdHamrick]

I just released VueScan 6.3.15 for Windows, Mac OS and Linux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.15

  * Added support for SCSI UMAX and rebadged UMAX scanners

  * Fixed problem with Microtek X6 transparency adapter

It turns out that there are a lot of types of UMAX scanners in
use, and this version has only been tested with a few of these.
types of UMAX scanners.  It should work with most UMAX
scanners and most rebadged UMAX scanners (including
LinoType-Hell scanners, Nikon AX-210 and Epson
Perfection 600).  In case it doesn't work, if you'd like me to
diagnose the problem:

1) Delete vuescan.ini
2) Run VueScan
3) Turn on "Files|Output log file"
4) Exit VueScan
5) Run VueScan
6) Choose "Preview|Device"
7) Choose "Scan|Device"
8) Exit VueScan
9) zip and e-mail vuescan.log to me

Several people have told me it works with the UMAX 1200S
and the UMAX PowerLook III, so I'd also appreciate getting
e-mails confirming which other UMAX scanners work ok.

Thanks,
Ed Hamrick

filmscanners: UMAX Test Version of VueScan

[EdHamrick]

I've got a UMAX test version of VueScan that's available for
testing (Windows version only).  It's been successfully
tested on several UMAX scanners, and I'd appreciate
it if people could test it out with a wider range of
UMAX scanners (scsi models only).

It can be downloaded from:

  http://www.hamrick.com/files/umaxtest.zip

Unzip this file into c:\vuescan and then run vuescan.exe
from there.

If you find problems, please:

1) Delete vuescan.ini
2) Run VueScan
3) Turn on "Files|Output log file"
4) Exit VueScan
5) Run VueScan
6) Choose "Preview|Device"
7) Choose "Scan|Device"
8) Exit VueScan
9) zip and e-mail vuescan.log to me

Thanks,
Ed Hamrick

Re: filmscanners: Vuescan ~ manual exposure

[EdHamrick]

In a message dated 11/29/2000 1:39:44 PM EST, [EMAIL PROTECTED] writes:

> I was playing with the newest distribution of Vuescan last night,
>  trying to determine how and when VS determined the exposure.  I had a
>  difficult time with consistent determinations ... it would seemingly
>  determine the exposure (RGB and IR) only if I asked for 'scan|device'.

Yes, that's correct.  The reason for this is that doing batch scanning
doesn't do the color correction for the preview (to save time).  In
addition, sometimes people move the crop box, but don't do a
"Preview|Memory" command to refresh the preview.  It was easier
to implement (and more reliable) the way it is now.

Regards,
Ed Hamrick

Re: filmscanners: Vusecan problem

[EdHamrick]

In a message dated 11/29/2000 5:06:52 PM EST, [EMAIL PROTECTED] writes:

> I am sure that in the past, when I have scanned slides, I have done so at 48
>  bit and then ended up with a Tiff file which is 36 bit in PS (maximum bit
>  depth on the scanner).
>  
>  Now, when I scan, even set at 48bit in vuescan, all I get is a 24 bit file.

You need to set both the "Device|Bits per pixel" option and
the "Files|TIFF file type" option.  I suspect you didn't change
the option in the Files tab.

Regards,
Ed Hamrick

Re: filmscanners: saving rawfiles in Vuescan

[EdHamrick]

In a message dated 11/29/2000 3:30:12 AM EST, [EMAIL PROTECTED] 
writes:

> I'm working with HP S20 and Vuescan
>  I need to save rawfiles (scan0001.tif+ ...) of both color and BW negs for
>  further work. Do I really have to save in 48bits format?

I suspect you could save the BW negs in 16 bit format.  You
should definitely save the color raw files in 48-bit format.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.14 Available

[EdHamrick]

I just released VueScan 6.3.14 for Windows, Mac OS and Linux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.14

  * Significantly increased scan speed with Epson scanners

  * Added support for Konica Qscan (same as Epson FS200)

  * Added support for older Microtek scanners and rebadged
versions of these scanners (AGFA Arcus II, AGFA Duoscan,
Polaroid SprintScan 45)

  * Fixed problem starting VueScan with SnapScan connected

  * Improved support for multiple scanners of the same
type attached at the same time

  * Added preliminary (i.e. untested) support for UMAX and
rebadged UMAX scanners.

If you have any problems with a Microtek/Polaroid scanner
that's newly supported in this version, or problems with
a UMAX scanner (and this certainly will have problems
since it hasn't been tested at all), it will help me diagnose the
problems if you:

1) Delete vuescan.ini
2) Run VueScan
3) Turn on "Files|Output log file"
4) Exit VueScan
5) Run VueScan
6) Choose "Preview|Device"
7) Choose "Scan|Device"
8) Exit VueScan
9) zip and e-mail vuescan.log to me

Thanks,
Ed Hamrick

Re: filmscanners: Minolta Scan Dual II for Mac

[EdHamrick]

In a message dated 11/26/2000 12:34:01 PM EST, [EMAIL PROTECTED] writes:

> This is one of the scanners I am considering.  Question for Ed Hamrick:
>  What is the status of Vuescan support for this scanner?

VueScan doesn't work with USB scanners on the Mac.  I haven't
figured out how to read and write USB endpoints from a user
program, and unless someone sends me some sample code (and
not just a link to web pages on Apple's site), I won't be able to
make any progress with this.  I've spent more than 40 hours on
this already, and I can't justify spending the 3 months of solid work
it looks like it would take to write a device driver.

Regards,
Ed Hamrick

Re: filmscanners:

[EdHamrick]

In a message dated 11/24/2000 1:24:01 PM EST, [EMAIL PROTECTED] writes:

> I also see from the Help file that "Auto mask"
>  should also be turned off for uniform scans, but I'm wondering about the
>  "Media/Mask red", "Mask green", and "Mask blue" settings.
>  
>  What do those values represent?

These are the colors of the orange mask of the film, in units of 0..1,
where 1 is the maximum pixel value of the raw scan.

> What space are they in?

Raw CCD color space.

> Do they interact
>  with the "Options/RGB exposure" setting at all?

Yes - if you increase the RGB exposure, the mask values
will increase.

> Does scanning a clear
>  negative with "Auto Mask" on always produce the best possible value for
>  that film type?

Yes.

Regards,
Ed Hamrick

Re: filmscanners: Max Sharpness?

[EdHamrick]

In a message dated 11/25/2000 10:17:50 AM EST, [EMAIL PROTECTED] writes:

> I use a Nikon LS30, mostly with Vuescan. Is there a difference in the way
>  how Nikon software uses the ICE feature?

Yes, VueScan and NikonScan use the infrared channel
differently to remove dust spots.

> Which software achieves more
>  sharpness when using ICE?

VueScan doesn't smooth the pixels outside the area where
there are severe dust spots (using the Clean setting), while
NikonScan does.

NikonScan applies a sharpen filter after smoothing the pixel
data, and VueScan doesn't (you can sharpen in Photoshop).

Regards,
Ed Hamrick

Re: filmscanners: Max Sharpness?

[EdHamrick]

In a message dated 11/25/2000 6:34:04 AM EST, [EMAIL PROTECTED] writes:

> Vuescan
>  - 48-bit files, scanned at 64-RGBI
>  - Scour setting (does not seem to effect sharpness at all??)

The Scour setting significantly blurs the image.  You should
only use it for really dirty slides.

Try using the Clean setting.

Regards,
Ed Hamrick

filmscanners: Umax support in VueScan

[EdHamrick]

I just spoke with someone at Umax who agreed to
loan me a Powerlook III scanner with a transparency
adapter some time next week.  For all of you who
have asked me to add support for Umax scanners,
it should take another week or so after that to
release a version of VueScan with Umax support.

If you'd like to beta test this version when it's
ready, please send me an e-mail (but don't post
your request to this mailing list).

Regards,
Ed Hamrick

Re: filmscanners: Got my Epson 1640su photo - work with vuescan?

[EdHamrick]

In a message dated 11/24/2000 7:30:56 AM EST, [EMAIL PROTECTED] writes:

> BTW, Ed, does Vuescan work correctly with this? I could only select 1600 
dpi 
>  from the options page (using 6.3.9 at the moment), but the scanner will do 
>  1600x3200. I did a preview scan and that all worked fine.

It's only 1600 dpi optical, so that's all VueScan supports.

Also, you might want to use VueScan 6.3.13 instead of
6.3.9 - there are some improvements in Epson support.

>  BTW-2. Is it possible for me to get a scsi-1 to scsi-2 cable so I could 
plug 
>  the epson into the same scsi chain as my microtek 4000?

Yes, this should work fine.

Regards,
Ed Hamrick

Re: filmscanners: VueScan 6.3.12 Available

[EdHamrick]

>  If the epson 1600 has 2 focus positions,  i would expect only 2 values, one
>  for glass plate level, the other for transparancy holder level, e.g. 1 or 
0,.
>  After 4 differently focused scans of a reflective sample, i found that: 0 =
>  glassplate level, 1 seems to be transparancy holder level, and 0.5 gives 
the
>  same result as 1, 0.25 gives the same result as 0.

Yes, that's how the scanner works internally.  It takes an 8-bit
number and transforms this to a scanner-dependant focus position.
VueScan maps 0..1 to 0..255 and sends this to the scanner.

Internally, VueScan uses 0.349 for transparency mode and
0.25 for flatbed mode.  If you look at the Expression 1600
document on http://www.ercipd.com/isv/edr_docs.htm
you'll see that it says focus values 00H <= i < 4CH are for
the surface of the document glass and focus values
4DH <= i <= FFH are for 2.5mm above the document glass.

>  the preview on the epson 1600 with vuescan is very slow, as slow as a full
>  res scan, while the epson twain driver uses a fast preview mode which 
indeed
>  is very fast. Also i noticed no difference between preview with final scan
>  resolution set at 1600 dpi or at 200 dpi.
>  Could you make a fast preview available on vuescan with the epson 1600?

I suspect it's using 200 dpi for preview mode, since this is 1/8 of
the max optical resolution.  The Epson software is probably using
100 dpi or 50 dpi.

If you want me to diagnose this a bit more, turn on the
"Options|Output log file" option and do a preview.  Then
e-mail me the vuescan.log file.  If you want to look in the
log file yourself, the resolution is set with the "ESC R"
command, and the 4 bytes are the X and Y resolution
(they're displayed in hex, two bytes per resolution word).

Regards,
Ed Hamrick

Re: filmscanners: Need advice on focusing the Scan Dual II using Minolta's so...

[EdHamrick]

In a message dated 11/22/2000 12:26:46 PM EST, [EMAIL PROTECTED] writes:

> Although I have found Vuescan much superior for negatives I am now somewhat 
>  preferring Minolta's own software for slides.

I suspect the Minolta software is applying a sharpening filter
by default, and VueScan isn't.  Try setting VueScan's
"Options|Filter" to "Sharpen".

Auto focus on preview should be just fine, although it's possible
for dark slides that doing the autofocus on scan will be better
because more contrast can be seen with a longer CCD exposure.

>  Also are you putting them into the holder emulsion side up (reversed) as 
>  specified in the manual.  Does this have any effect on the ability to 
focus?

This shouldn't matter at all, since the auto focus just finds the
focus position where the contrast is maximized.

Regards,
Ed Hamrick

Re: filmscanners: Binuscan for SS4000

[EdHamrick]

In a message dated 11/22/2000 12:20:30 PM EST, [EMAIL PROTECTED] writes:

> Ed, I'm not sure what you mean by "it's the same as 2.2/gamma."

Contrast in the latest versions of VueScan is equal to 2.2/gamma
in previous versions.

For instance, a gamma of 4.4 in previous versions of VueScan
is the same as contrast of 2.2/4.4 = 0.5 in current versions of
VueScan.

>  is it better to make
>  this adjustment at the scanning stage in Vuescan, or in the post scan stage
>  in Photoshop?

It doesn't really matter.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.13 Available

[EdHamrick]

I just released VueScan 6.3.13 for Windows, Linux and
Mac OS.  It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.13

  * CCD exposure time computed from cropped area.  This
helps scan dark parts of slides and overexposed
parts of negatives with scanners that support exposure

  * Fixed problem with MF calibration on Minolta Scan Multi

Regards,
Ed Hamrick

Re: filmscanners: VueScan 6.3.12 Available

[EdHamrick]

In a message dated 11/22/2000 4:27:28 AM EST, [EMAIL PROTECTED] writes:

> How do i set that focus manually?

Turn off the "Options|Auto exposure" field and experiment
with the "RGB exposure" option.

Useful values range from 0 to 1 on the Epson 1600.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.12 Available

[EdHamrick]

I just released VueScan 6.3.12 for Windows, Linux, and
Mac OS.  It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.12:

  * Added support for 2500 dpi Microtek/AGFA scanners

  * Improved film color correction for SprintScan 4000,
PhotoSmart S20, Acer ScanWit, and SprintScan 35/LE

  * Fixed problem with focus on Epson Expression scanners
and add support for manual focus

  * Fixed problem with MF calibration on Minolta Scan Multi

  * Improved memory utilization on Linux version of VueScan

Regards,
Ed Hamrick

Re: filmscanners: Binuscan for SS4000

[EdHamrick]

In a message dated 11/20/2000 2:01:27 PM EST, [EMAIL PROTECTED] writes:

> I'm looking for something that will accept 48-bit raw scans I have made
>  with the SS4000 and do a reasonable job of compressing the tonal range
>  without muddying the shadows or blowing out the highlights, and
>  generate 24-bit files my client can look at and decide which images are
>  really worth looking at more closely.

VueScan will do this nicely.  Just experiment with the
"Color|Image contrast" option (it's the same as 2.2/gamma).

Regards,
Ed Hamrick

Re: filmscanners: Minolta Scan Multi

[EdHamrick]

In a message dated 11/18/2000 4:13:26 AM EST, 
[EMAIL PROTECTED] writes:

> Am I right in believing that the new 120 film holder, from Minolta enables
>  one to scan, at 2820dpi,  the entire surface if a 120 neg, by making 3
>  passes over the film , then stitching the three scans, together ?

I've gotten e-mail from someone who bought one of these (for
an exorbitant price).  You can indeed scan a 120 negative in
three strips with some contortions, but you have to stitch these
scans together with software you get separately.

>  Will this holder work with the VeiwScan software

Yes, it works fine with VueScan.

Regards,
Ed Hamrick

filmscanners: VueScan TODO Vote Results

[EdHamrick]

I've summarized the number of votes I got for the
various items on my VueScan TODO list.  I
combined a few items that were related.  There
were about 115 items voted for, and the items
with the number of votes they got are attached.

I'll try to prioritize the changes in VueScan 6.4
using this list.  Thanks to all of you who responded
with their priorities - it's been very helpful to me.

(Please don't respond with any off-topic comments
related to voting in the U.S.  This mailing list
is most useful when it stays on-topic.)

Regards,
Ed Hamrick

17 - Dialog boxes for file names & default directories
11 - Cropping improvements
 9 - Buttons, toolbar, hot keys
 9 - Make it possible to zoom into preview and scan
 8 - Make exposure time computation work only with cropped area
 6 - Align multi-scans in memory (for SS4000)
 5 - Give warning on all versions if tiff writing gets error
 4 - Add option to both sharpen and clean
 4 - Look into scanning speed for Canon (it's in the Sane source code)
 3 - Think about only saving settings if someone asks for this
 3 - Add menu for color linearity, scan b&w film to determine linearity
 3 - Implement "Scan exposure" on FS2710 by using a real prescan
 3 - Add tooltips
 2 - Allow text entry for comments field
 2 - Put non-default settings into .tif and .jpg comments field
 2 - Add APS setting for FS2710
 2 - Canon reportedly does an unnecessary extra calibration for each scan pass
 2 - Look into an undo capability for scan0001.tif, crop.tif, 
crop.jpg, indx.bmp
 2 - Look into GEM/ROC algorithms
 2 - Fix MacOS version so directory spec can be specified and have 
0001+.tif work

Re: Sv: filmscanners: VueScan TODO list

[EdHamrick]

Thanks!

Ed Hamrick

filmscanners: VueScan 6.3.11 Available

[EdHamrick]

I just released VueScan 6.3.11 for Windows, Linux and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.11:

  * Fixed problem with infrared channel on the
Minolta Scan Elite

  * Fixed problems with some scanners on Mac OS
(especially loading/ejecting film adapters)

Regards,
Ed Hamrick

filmscanners: VueScan TODO list

[EdHamrick]

I've attached a rough list of the things I'm thinking about
doing to VueScan, and I'd appreciate it if people would
send me e-mail (not posting to this list, but e-mailing
directly to [EMAIL PROTECTED]) with their top 5 priority
list.  This will help me prioritize the things people want.

I'll summarize people's requests and post a summary to
this e-mail list, but again I'd request that people not post
their top 5 list to this e-mail list, since it already gets too
much off-topic e-mail.

Regards,
Ed Hamrick

Device driver changes:

  Implement manual/auto focus on Epson driver
  Handle open door on Scan Dual II
  Work on Scan Multi MF calibration problem
  Fix Microtek X12USL calibration
  Reportedly crashes when Agfa SnapScan 310 SCSI on same system
  Implement "Scan exposure" on FS2710 by using a real prescan
  Add APS setting for FS2710
  Canon reportedly does an unnecessary extra calibration for each scan pass
  Look into scanning speed for Canon (it's in the Sane source code).

Disk files:

  Put non-default settings into .tif and .jpg comments field
  Allow text entry for comments field
  Make sure tiff rescanning is working on Mac version
  Look into an undo capability for scan0001.tif, crop.tif, crop.jpg, 
indx.bmp
  Give warning on all versions if tiff writing gets error
  Look into adding a "Redisplay memory" option (no file outputs)
  Add command for default output directory
  Fix MacOS version so directory spec can be specified and have 0001+.tif 
work
  Reportedly putting VueScan folder in non-desktop location on Mac doesn't 
write crop files right
  Update file name conventions on Mac version
  Use file dialog boxes to specify file names (any file name)
  Make sure .ini file is stored in same dir as .exe
even if HOMEPATH is set to something else

Efficiency:

  Optimize memory allocation of large memory blocks on Linux
  Repaint taking too long on Macintosh (use fixed-size preview image)

Functionality:

  Make it possible to zoom into preview and scan
  Look into why cropping is different in 5.9t53 and 6.3
  4-frame cropping on PhotoSmart doesn't always work well
  Add night calibration back in
  Cropping often has dark borders
  Make exposure time computation work only with cropped area
  Look into GEM/ROC algorithms
  Add option to both sharpen and clean
  Align multi-scans in memory (for SS4000)
  Add menu for color linearity, scan b&w film to determine linearity

Ease of use:

  Make rotations specified with frame numbers relative to global rotation
  Add buttons for Preview, Scan, Abort, Feed, etc.
  Add helper message to Preview and Scan tabs when empty
  Look into toolbar & hot keys
  Add tooltips
  Think about only saving settings if someone asks for this (or option
to save settings on exit)
  Look into making auto-preview an option on Scan Dual
or always do preview

Bugs:

  Reportedly using 1F for frame number doesn't work right
  Reportedly modifying Y spacing doesn't cause auto preview memory
  Reportedly Preview|Memory (or changing Y offset) resets Y spacing

Re: filmscanners: Profiling, Ilford XP2 and Vuescan.

[EdHamrick]

In a message dated 11/14/2000 5:50:54 PM EST, 
[EMAIL PROTECTED] writes:

> Am I being naive here or does the settings for colour film take into
>  account, neutralizing the orange mask on the film base

Scanning black/white film with VueScan is very similar to
scanning color negative film.  VueScan finds the color of
the base in order to set the origin of the sensitometric
curve.

If you use one of the black/white film settings in VueScan,
the final scan will have the red, green, and blue channels
identical.  They're made identical in the film processing
algorithm.

>  I understand that XP2 IS a C41 film, but isn't the base fog grey ?

Yes.

Regards,
Ed Hamrick

Re: filmscanners: Minolta Scan Multi II experiment with stitching (long)

[EdHamrick]

In a message dated 11/10/2000 1:26:21 PM EST, [EMAIL PROTECTED] 
writes:

> I had a problem with the color balance between two of the strips, but I was
>  able to stitch the images together into a 240Mb file!

The VueScan help file describes how to set the options so you
can stitch multiple images together.  It's aimed at people scanning
panoramas that they want to stitch together, but it works for
this application also.

It's in the "Advanced Workflow Suggestions" section.

The main things you have to turn off are Auto exposure,
Auto mask, Auto white point, and Auto black point.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.10 Available

[EdHamrick]

I just released VueScan 6.3.10 for Windows, Linux and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.10:

  * Added "Device|Frame offset (mm)" option for Nikon
LS-30 and LS-2000 scanners.  It's useful if you have
some leader on a film strip, or if you're scanning
panoramic film on these scanners.

  * Added support for slide feeder on LS-1000

  * Added check for 50 MByte free memory block on Mac OS

Regards,
Ed Hamrick

Re: filmscanners: settings suggestions for epson 1600 pro?

[EdHamrick]

In a message dated 11/4/2000 9:33:09 AM EST, [EMAIL PROTECTED] writes:

> 1. How do i set the focus for using the filmstrip holders with transparancy
>  scanning?
>  The epson seems to have two focus settings, one for glassplate level and 
one
>  for the holders, which seems to be 2.5 mm difference.. the scan i just made
>  was definitly not in focus.)

It automatically uses the 2.5mm setting when doing transparency
scanning, and the glassplate setting for normal scanning.

If you want me to verify this, turn on the "Files|Output log file"
option and do a scan.  Then zip and e-mail vuescan.log to
me.  You can look yourself:

1) VueScan sends "ESC p" command
2) Scanner returns ACK (0x06)
3) VueScan sends focus position (0x00-0x4c is glass surface, 0x4d-0xff is 
2.5mm above)
4) Scanner returns ACK (0x06)

VueScan uses 0x59 for transparency mode, and 0x40 for flatbed mode.

>  2. Is it possible to restrict the preview and analysis(exposure, B-W pooint
>  settings) to the small area of each of the images in the filmstrip holder,
>  which contains four strips of six images ? Is there or can you make an
>  option to save these settings, so they can be called anytime, without going
>  through the hassle of resetting everything ?

You might be able to do batch scanning with careful setting of the
Crop options.  You can set the number of frames across and down,
as well as the spacing between frames across and down.

Regards,
Ed Hamrick

filmscanners: Re: Vuescan and Nikon Slide Feeder?

[EdHamrick]

In a message dated 11/4/2000 8:23:00 AM EST, [EMAIL PROTECTED] writes:

> Just want to make sure before I plunk down many dollars
>  for the slide feeder for the LS-2000:  Will Vuescan
>  just rip thru the stack of slides unattended (as long
>  as the feeder itself doesn't jam)?

Yes, this should work.  I borrowed one from Nikon some
time ago, and it worked fine.  I haven't gotten any
e-mail from anyone saying it doesn't work.

Regards,
Ed Hamrick

filmscanners: VueScan 6.3.9 Available

[EdHamrick]

I just released VueScan 6.3.9 for Windows, Linux, and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.9:

  * Fixed problem using transparency adapter on some
newer Epson scanners (including Perfection 1640)

  * Improved color correction for color films
on most scanners, especially images with flowers
or other saturated colors

  * Fixed problem with scanning from disk file with
mode set to Epson

  * Added preliminary support for SprintScan 45 Ultra
and SprintScan 120

Regards,
Ed Hamrick

Going on Vacation

[EdHamrick]

I'm unsubscribing from this list for a week and am off to
Egypt for a week-long vacation.

I know this is off-topic, but this is why I won't be answering
much e-mail for the next week.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Scanner < $1500 selection ?

[EdHamrick]

In a message dated 10/19/2000 8:54:18 AM EST, [EMAIL PROTECTED] writes:

>  - ICE software [I have a lot of water damaged
>  negatives] or something like it

The infrared channel in the Nikon or Minolta Scan Elite won't
help you in cleaning B&W negatives.  These negatives are
just as transparent at red, green, blue, and infrared wavelengths.

If you need 4000 dpi, then the Polaroid SprintScan 4000
will do a nice job for you.  If you need less resolution, then
the Minolta Scan Dual II is probably good enough.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: VueScan & Epson 1200U for Transparency

[EdHamrick]

In a message dated 10/18/2000 11:01:26 PM EST, [EMAIL PROTECTED] writes:

> I'm having a problem scanning large (eg A4) areas at 1200dpi and I've
>  reported this to Ed.  I'm using W98SE with 256MB RAM.  I can scan an A4 
area
>  at 600dpi without any problems.  It may be that this is associated with the
>  way that the (in my case) 1200U can't scan this size area under the Epson
>  Twain software at 1200dpi.

This is a fairly simple problem.  A 1200 dpi scan of the whole scanning
area is 8.5 x 1200 * 11.7 * 1200 * 24 / 8 = 430 MBytes.

VueScan allocates this 430 MByte area from the paging file.
Go into the "Advanced" tab of the "System properties"
on Windows 2000 and click "Performance Options".
Increase the size of virtual memory to at least 600 MBytes.

On the Mac, you need to make sure the "About this computer"
command shows a "Largest unused block" large enough to
hold the entire image in memory.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan 6.3.5 Available

[EdHamrick]

I just released VueScan 6.3.5 for Windows, Linux, and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.5:

  * Added support for mouse cropping on Mac OS

  * Added support for Epson FilmScan 200

  * Cursors now indicate adjustments to crop box

  * Fixed a small rounding error when adjusting crop box

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan feature request

[EdHamrick]

In a message dated 10/16/2000 6:10:38 PM EST, [EMAIL PROTECTED] writes:

> When batch scanning, would it be too much to ask Vuescan to stop
>  if the disk runs full?

I definitely need to do this - I just haven't gotten to it yet.

A lot of people have asked for this, so I'll move it to the
top of my list.   Hopefully I can release a version with this
fix this week.

Regards,
Ed Hamrick


Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: LS30 BIOS was Re: Negative film scanning - Polaroid SS4000+SilverFast vs....

[EdHamrick]

In a message dated 10/16/2000 12:27:23 PM EST, [EMAIL PROTECTED] writes:

> It occurs to me to wonder whether one
>  could flash the LS30 with the LS2000 upgrade BIOS - given the cost of 
> replacing the LS30, I haven't tried *that* either. :)

I tried this about a year ago, and finally determined that the
LS-30 firmware prevents non-LS-30 firmware from being loaded.
I suppose I could have done binary patches on the firmware,
but it really wasn't worth the trouble.

>  PS There's a device from Japan for flashing the BIOS from any PC with it's
>  upgrade - I wonder if it might work with the LS30 BIOS chip?  I also have
>  to flash the BIOS on my PC motherboard... *shudder*

It's a surface-mounted chip (not removable).

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: IR exposure (was VueScan help)

[EdHamrick]

In a message dated 10/16/2000 9:22:42 AM EST, [EMAIL PROTECTED] writes:

> I do hope you haven't ruled out histograms for VueScan but will
>  consider adding them to the wishlist

I certainly haven't ruled them out.  When people ask for new
features or support for new scanners, I tend to resist for a
while. This helps me understand why people want things, and
helps filter out those things that aren't as important.

In the end, the things that the most people ask for are the
things I add.  For instance, I didn't think supporting Epson
scanners was a useful thing to do, but so many people asked
for it that I finally just added it.  Since I've done so, I've
gotten lots of feedback from people about how useful it is.

I'm kicking myself for not doing it earlier, since it was actually
the easiest scanner I've ever supported (one set of Epson
commands can be used for most Epson scanners, even 5
year old scanners).

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: medium and large format film scanning

[EdHamrick]

In a message dated 10/16/2000 9:49:43 AM EST, [EMAIL PROTECTED] writes:

> The 1600 is a nice scanner but very marginal with
>  negative films using either the Epson Software or the Silverfast software.

It turns out that correcting color negative film is really hard to
do well.

>  While the Epson is no drum scanner, it produces very acceptable results at
>  1600dpi which is more than adequate for large prints from medium format and
>  large format negative.  It is also very fast.

The Expression 1600 is a very nice scanner.  It's capable of returning
the full 12-bit data from the A/D converter, and this helps make the
image histogram quite smooth.  However, even the Epson Perfection
series, which returns 12-bit data mapped through an 8-bit palette,
produces well-calibrated color that results in good negative color
correction with VueScan.

>  I have switched to Vuescan for my Microtek 4000, Nikon LS-1000, and now the
>  Epson.

Jeez - I need to charge more .

>  I can not recommend the software and especially the support too
>  highly.

Thanks for the kind words - it's always nice to hear.

I've been pleasantly surprised by the feedback I've been getting
about VueScan's Epson scanner support, especially the
negative processing.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Negative film scanning - Polaroid SS4000+SilverFast vs. NikonLS-2000

[EdHamrick]

In a message dated 10/15/2000 7:07:42 PM EST, 
[EMAIL PROTECTED] writes:

> Crazy!
>  It must cost Nikon money to cripple it!

No, this is a very sensible thing for Nikon to do.  This way
they can have a single hardware design that they sell to
multiple market segments.  The manufacturing cost of
a Nikon film scanner is probably much less than the cost
of an LS-30.

>  Do you know if the firmware's flashable?

It is flashable, but the firmware in the scanner prevents the
LS-2000 firmware from being loaded into the LS-30.

>  Do you know if the LS30's been like that from the start?

Yes, the LS-30 has been like this from the start.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Kodak RFS 3600

[EdHamrick]

In a message dated 10/14/2000 7:43:07 PM EST, [EMAIL PROTECTED] writes:

> > I am thinking now what is better to buy this scanner with probably not
>  > perfect software and 3600 resolution or Minolta Elite with well done
>  > software (for example Vuescan6.2 or Minolta software) and 2800dpi ?
>  
>  Maybe there would be a VueScan for this soon after it's release?

It costs a lot of money, so I don't really want to buy one.  If someone
loans me one I'll look into doing a VueScan driver for it.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan 6.3.2 Available

[EdHamrick]

I just released VueScan 6.3.2 for Windows, Linux and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3.2:

  * Added an option to the Window tab to beep when
scanning is done.  This is turned off by default.
Not yet implemented in Mac OS version of VueScan
due to lack of support for wxBeep() in wxWindows.

  * Fixed problem scanning on Epson Perfection 1200
using transparency unit with lamp that isn't warmed up

People have asked me to add the beep feature for
a long time, and I really should have added it months
ago (it was less than 10 lines of code).  If there are other
small things that people have been wanting for a long
time, please send your suggestions via e-mail.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: IR exposure (was VueScan help)

[EdHamrick]

In a message dated 10/13/2000 6:42:58 PM EST, [EMAIL PROTECTED] writes:

> Yes, but I would like to see this in VueScan.  This would let me do a
>  prescan and quickly check the levels instead of waiting until I have done a
>  full resolution 16x multiscan and imported it into PhotoShop.

Having a histogram of the preview wouldn't help, since the scan
uses the preview to recompute the CCD exposure.  The histogram
of the scan will be significantly different than the histogram of
the preview.

>  I thought that 16-bit Grey setting would give me a B&W image with no IR
>  info.  Will this file be just the IR channel or is it mixed with the RGB
>  info?  What do you get on scanners with no IR channel?

The 16-bit Grey setting is a kludge that I added to extract the IR
channel (if available).  If it isn't available, then the 16-bit Grey is
a mixture of the red, green, and blue channels
(306*R+601*G+117*B)/1024

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: IR exposure (was VueScan help)

[EdHamrick]

In a message dated 10/12/2000 3:08:58 PM EST, [EMAIL PROTECTED] writes:

> I would like to see histograms for each of the RGBI channels displayed
>  within VueScan.  And a count of the pixels clipped to either min or max
>  values.

You can do this with Photoshop.

>  I would also like to be able to save the Infrared image and later examine 
it
>  within Photoshop.

Just set the "Files|TIFF file type" to "16-bit Grey" and do a scan.
The infrared data will be put into the tiff file and you can view the
histogram in Photoshop.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan 6.3 Available

[EdHamrick]

I just released VueScan 6.3 for Windows, Linux, and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.3:

  * Added support for most Epson flatbed scanners

  * Changed vuescan.ini to be written to local directory
on Linux and Mac OS (previously written to home
directory or System Preferences folder).

If you have any problems, please let me know.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: VueScan beta for Epson Scanners

[EdHamrick]

In a message dated 10/12/2000 11:41:09 PM EST, [EMAIL PROTECTED] writes:

> Mmm, interestingly, it works with the 1200U  (W98SE, USB).  However, the
>  scanner needs to have powered up and initialised before starting VueScan 
and
>  then I find that if it is the only scanner available it appears 
pre-selected
>  in the 'Scan from' box.

I just released VueScan 6.3, and I added a fix for the problem you
noted.  There's one bit returned from a status command that says
that the scanner is warming up.  I had to add a wait look till the
lamp is warmed up.

You can download it from:

  http://www.hamrick.com/vsm.html

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan "mode"

[EdHamrick]

In a message dated 10/12/2000 1:16:48 PM EST, [EMAIL PROTECTED] writes:

> and otherwise cannot figure out what "mode"
>  is for except for making parameter assumptions(?)

That's mainly what it's for.  Use the media type option
to set up for slides vs. negatives.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: VueScan help

[EdHamrick]

In a message dated 10/12/2000 1:06:26 AM EST, [EMAIL PROTECTED] writes:

> I would be using VS much the same way you do, but this "help page"
>  would seem to imply "masks" is as much part of creating the raw files
>  as is "exposure" ... and I'm not too sure what "masks" are for(?)

The orange mask is simply the red,green,blue color of the orange
color of unexposed film.  The scale is 0..1.

The mask only affects the color processing, it doesn't change
the raw data.

>   Also regarding exposure, I noticed too that if I scan a blank part of
>  a film strip ... my IR channel exposure is quite different than if I
>  had scanned a "normal" negative.

That's correct - exposed film absorbs infrared light more than
unexposed film.

>  How does IR channel exposure affect
>  "cleaning" and "scrubbing"?

As long as it isn't overexposed, it doesn't really matter too much.

> For example, what are the symptoms of
>  an overly exposed IR channel?

If it saturates, then it won't clean right.

>  Should it be more or less than the RGB
>  exposure?

It's roughly proportional to RGB exposure.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan beta for Epson Scanners

[EdHamrick]

I added support for Epson scanners to VueScan 6.3 beta 1,
and I'd appreciate it if people could test it for me with a
variety of Epson scanners.

You can download it from:

  http://www.hamrick.com/vsm.html

If it doesn't work, could you:

1) Run VueScan
2) Turn on "Files|Output log file"
3) Exit VueScan
4) Run VueScan
5) Choose "Device|Preview"
6) Exit VueScan
7) E-mail me vuescan.log

I don't think it will work right with the FilmScan 200, but if
someone in the London area has one they can loan me
for a few days, I can add support for it as well.

Thanks,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Negative film scanning - Polaroid SS4000+SilverFast vs. NikonLS-2000

[EdHamrick]

In a message dated 10/11/2000 10:33:05 PM EST, 
[EMAIL PROTECTED] writes:

>  The LS30 has only a 10 bit A/D converter chip, and is a generation behind 
> the LS2000.

I've gotten e-mail from someone who took their LS-30 apart, and
it has exactly the same 12-bit A/D converter chip that the LS-2000
has.

It's _exactly_ the same hardware as the LS-2000.  The only
difference is the firmware, which only allows transferring 10-bits
of the 12-bit data from the A/D converter.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: questions

[EdHamrick]

In a message dated 10/10/2000 9:44:52 PM EST, [EMAIL PROTECTED] writes:

> There is a Vuescan "files" preference box which needs to be checked.
>  It is called "TIFF profile" ... and if check that and it still doesn't
>  work then send something direct to Ed.  I just today noticed this pref
>  is unchecked by default, so I'll not get to try it 'til tonight.

There was a bug in 6.2 that prevented the profile from being
embedded (I introduced it 5 minutes before releasing 6.2
when I was fixing a bug building the Mac OS version).  I'm
pretty sure it's fixed in 6.2.1.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan 6.2.1 Available

[EdHamrick]

I just released VueScan 6.2.1 for Windows, Linux, and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.2.1:

  * Added support for two more color spaces, Ekta Space
and ProPhoto RGB (Kodak ROMM)

  * Fixed problem embedding ICC profile in files

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: newbie question: what is scanner software needed for?

[EdHamrick]

In a message dated 10/10/2000 8:53:55 AM EST, [EMAIL PROTECTED] writes:

> > This is also why SilverFast is obsolete.  It was written with a
>  > distinct scanner model in mind - download curves to the scanner
>  > to map 10-bit or 12-bit data to 8-bit data inside the scanner.
>  > Nowdays, good scanners can return the full bit depth to the
>  > scanner program, and scanner programs can pass the full
>  > bit depth to Photoshop.
>  
>  OO Ed, you bitch!!! ;)



My point wasn't to slam SilverFast, but rather to point out the
fundamental assumption underlying it's design - that the
scanner does the mapping from 12-bits to 8-bits before
transferring the data over the scsi interface.

If a scanner transfers the whole 12-bits of data to the
application, then it makes sense to do as much of the
color processing as possible within Photoshop.
There's a lot of overlap between the functionality of
SilverFast and Photoshop, and with modern scanners
this functionality is only really needed in Photoshop.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: newbie question: what is scanner software needed for?

[EdHamrick]

In a message dated 10/9/2000 5:59:36 PM EST, 
[EMAIL PROTECTED] writes:

> Do you know, or need to know, the peak response of the RGB filters used in 
> the CCD and their passband characteristics, coupled with the lamp output, 
for 
> this Ed?

No, I don't use CCD specs in calibrating the scanners.  I just scan
a Q60 slide and derive the 3x3 matrix that produces the best fit
with the actual Q60 colors.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: newbie question: what is scanner software needed for?

[EdHamrick]

In a message dated 10/8/2000 3:36:35 PM EST, [EMAIL PROTECTED] writes:

> I think this scanning model is dead. Now that I can get raw 16 bit data into
>  Photoshop, scanning software must be better than Photoshop to be useful.

Exactly - this is why VueScan doesn't have (and doesn't need) curves.
It can easily be done in Photoshop with the full bit depth, so there's
no real reason to add Photoshop functions to VueScan.

This is also why SilverFast is obsolete.  It was written with a
distinct scanner model in mind - download curves to the scanner
to map 10-bit or 12-bit data to 8-bit data inside the scanner.
Nowdays, good scanners can return the full bit depth to the
scanner program, and scanner programs can pass the full
bit depth to Photoshop.

>  Create a 16 bit, full frame, raw scan at max resolution, and archive it. 
Now
>  there is no reason to ever scan the film again, unless you get a new 
scanner
>  with better performance (resolution, noise, etc.) You can crop, resize, and
>  tweak to your heart's delight in Photoshop, or any other of the many aps
>  that now accept 16 bit data.

The scanner software also needs to either transform the color data
to a known color space or embed the color space of the scanner
into the tiff file holding the raw scan data.

The other significant thing the scanner software needs to do is
crop the scan - select the part of the scan area to store in the
tiff file.

>  Now, what will happen to companies like Microtek/Umax/Agfa, and Lasersoft? 
I
>  spent $500 on Lasersoft Silverfast software, and no longer use it.

If one were to apply Occam's Razor to scanner software, a program like
VueScan could be further simplified.  One program would simply let you
select the area to crop and would write the raw scan data to a tiff file.
Another program would be a Photoshop plug-in for handling the colors
for color negative film and slide film.

However, since many people don't own Photoshop and many people
want to do batch scanning, VueScan combines the raw scan
acquisition and the color processing.  This way the raw scan files
don't have to be written to the hard drive.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan: custom color space (?)

[EdHamrick]

In a message dated 10/7/2000 3:11:24 PM EST, [EMAIL PROTECTED] writes:

> rather than continuing to add to what is
>  already an adequate list of color spaces, if you create an ability for
>  a user to define and use a custom color space, which would be defined
>  with the usual Photoshop ICM parameters (?)

I could do this, but I doubt that this would be a widely used
feature.  If there are one or two particular color spaces people
would like me to add, I can add them if they e-mail me
the .icc or .icm file (not a ColorMatch file).

I'll add a color space if three separate people e-mail me
an .icc or .icm file for the same color space.  (A
ColorMatch file isn't useful to me, and just the name
of a color space isn't useful to me).

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: PhotoCD RGB (?)

[EdHamrick]

In a message dated 10/7/2000 4:06:42 PM EST, [EMAIL PROTECTED] writes:

> > The fact that PhotoCD uses the same color primaries and white
>  > point as sRGB doesn't imply that they have the same gamut,
>  > since PhotoYCC can store negative components and sRGB
>  > can't.
>  
>  Your response confuses me

There's a lot more on this subject in the Color FAQ.  Search
for "Color FAQ" on the internet (the author is Charles Poynton),
and all this is explained in detail.

Regards,
E dHamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: PhotoCD RGB (?)

[EdHamrick]

In a message dated 10/7/2000 1:19:19 PM EST, [EMAIL PROTECTED] writes:

> That is rather amazing ... I would have thought the color space
>  chosen for reproducing color from film would have respected the
>  original gamut more than "sRGB".  But, I do suppose they were only
>  trying to make the colors look right on a typical monitor.

The fact that PhotoCD uses the same color primaries and white
point as sRGB doesn't imply that they have the same gamut,
since PhotoYCC can store negative components and sRGB
can't.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan6.2 and Minolta Dual Scan II

[EdHamrick]

In a message dated 10/6/2000 9:13:14 AM EST, [EMAIL PROTECTED] 
writes:

>  1. Why do multiple scans improve shadow detail- I assume that it is signal 
>  averaging. Does it effect resolution with image shifting between scans?

Multiple scans don't make as much difference with the Scan Dual II
since it has such a good A/D converter.

>  2. Why do the scans appear darker than what appears visually in the slide?

I don't know - you might check your monitor calibration.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: VueScan 6.2 Available

[EdHamrick]

In a message dated 10/6/2000 10:17:35 AM EST, [EMAIL PROTECTED] writes:

> > There are several models of Duoscan, and VueScan only
>  > works with the ones that are the same as the Microtek
>  > Scanmaker 4 and Scanmaker 5.
>  >
>  Does this include the AGFA Duoscan T1200?

The DuoScan T1200 is the same as the ScanMaker 4, so
it should work fine with VueScan.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan6.2 and Minolta Dual Scan II

[EdHamrick]

In a message dated 10/6/2000 10:25:41 AM EST, [EMAIL PROTECTED] writes:

> I found that the scan's brightness, for one particular negative, was
>  extremely sensitive to the whitepoint % setting.  I could vary it from
>  0.5% to 0.7% and the resulting scan would be noticeably brighter.

This happens when there's a tiny part of the negative that's
extremely bright.  If you don't want detail from this bright part,
increase the white point % setting. However, you
then lose detail in the bright parts.

Alternatively, decrease the "Color|Image contrast" option
(this increases the gamma).

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan6.2 and Minolta Dual Scan II

[EdHamrick]

In a message dated 10/6/2000 1:42:39 PM EST, [EMAIL PROTECTED] writes:

> I'm using a Multi II, while not a Dual II it pretty much has the same 
>  features.  When doing a multipass scan I can't notice any difference in 
the 
>  image quality even when doing a 16x mutlipass scan.

The only difference you might possibly see is in extremely dark
areas of slides.  However, the Scan Multi has very little noise
in it's scans, so this might not make much difference.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Agfa and Microtek Model Numbers

[EdHamrick]

I did some research, and I'm pretty sure I've found the
Agfa and Microtek scanner models that are basically
the same:

DuoScan = ScanMaker 5
DuoScan T1200 = ScanMaker 4
DuoScan T2500 = ArtixScan 2500

I'm positive that VueScan works with the ScanMaker 5, and
I'm pretty sure it works with the other two as well.  Could
anyone who's used VueScan on one of the other two
scanners confirm that it works?

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: PhotoCD RGB (?)

[EdHamrick]

In a message dated 10/6/2000 7:16:37 PM EST, [EMAIL PROTECTED] writes:

> did Ed remove this "color space" option?. 
>  My search for what defines PhotoCD space only yields 
>  the whitepoint (6500K), gamma (2.2), and the 
>  transformation equations.  

The PhotoCD color space is identical to the sRGB color
space (same Rec. 709 primaries, same D65 white point).
If you use sRGB in VueScan 6.2, it will produce identical
results to PhotoCD in previous versions of VueScan.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan6.2 and Minolta Dual Scan II

[EdHamrick]

In a message dated 10/6/2000 7:04:51 PM EST, [EMAIL PROTECTED] writes:

> Ed Hamrick the author of Vuscan has mentioned several time that
>  because of the very low noise level on the Sprintscan 4000 multiscanning is
>  of very marginal value. Despite this we will be adding multiscanning to our
>  new 4000 DPI medium format scanner. Why? It is perceived to be of real
>  value.

In my opinion (and I'm as wrong as often as anyone else ),
multiscanning isn't really useful with scanners that have low-noise,
high quality A/D converters.  The scanners that don't really need
multi-scanning include the SprintScan 4000 and the Scan Dual II.

I'd be the first to admit that I'm not totally convinced one way or
another, and since it's a trivial thing to add to scanner firmware,
I can't think of any reason for a vendor not to add it.  At a minimum,
it's useful for marketing reasons.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: VueScan 6.2 Available

[EdHamrick]

In a message dated 10/6/2000 3:07:01 AM EST, [EMAIL PROTECTED] writes:

> I just wonder if VueScan 6.2 works with AGFA Duoscan.

There are several models of Duoscan, and VueScan only
works with the ones that are the same as the Microtek
Scanmaker 4 and Scanmaker 5.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan 6.2 Available

[EdHamrick]

I just released VueScan 6.2 for Windows, Mac OS and Linux.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.2:

  * Added support for more color spaces and for adding
color profiles to crop tif and jpg files

  * Image gamma is now implicit in choice of color
space, and Image contrast is used instead to
change the overall image contrast.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan/Old Kodachrome Unable to Focus

[EdHamrick]

In a message dated 9/30/2000 6:51:08 PM EST, 
[EMAIL PROTECTED] writes:

> > I've graphed this function vs. focus position, and it seems to be
>  > a nice smooth curve with a single maximum position.
>  
>  My guess is that you did this test with a nice fine-grained slide, Ed.
>  I think you might get a different result if you did the test with a 400 
ISO 
> negative, or something else that showed the dreaded 'monster grain' effect.

What focus function would you recommend instead?

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Embed colour profiles in images

[EdHamrick]

I found the code I needed for embedding profiles in tiff
and jpeg files, and am now working on this.  There's a
fair chance I can get this released by Friday.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan/Old Kodachrome Unable to Focus

[EdHamrick]

In a message dated 9/29/2000 7:16:40 AM EST, [EMAIL PROTECTED] writes:

> Kodachromes are the most transparent to IR.

Actually, when you scan a Kodachrome slide with
infrared light and compare it to an Ektachrome slide
scanned with infrared light, you can see more image
detail (contrast) in the Kodakchrome scan.

You can test this in VueScan by setting the
"Files|TIFF file type" to "16-bit Grey".  This causes
the infrared channel to be written to the TIFF file
if the scanner has an infrared channel.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: VueScan 6.1.2 Available

[EdHamrick]

In a message dated 9/29/2000 7:06:54 AM EST, [EMAIL PROTECTED] writes:

> Any plans on supporting the LeafScan 35/45? ;-)

No, there aren't enough of these on the market for me
to make enough money at $40 per copy to make it
worthwhile.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Adjusting VueScan output

[EdHamrick]

In a message dated 9/28/2000 5:29:26 PM EST, [EMAIL PROTECTED] writes:

> For many scans, I find that an auto-levels in PS is all
>  that is necessary.  How can I tweak VueScan so that I can skip the levels
>  step for non-critical scans?

Set "Color|Color balance" to Neutral.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Negs and backlight

[EdHamrick]

In a message dated 9/28/2000 6:11:34 AM EST, [EMAIL PROTECTED] writes:

> > BTW ... I use a LS-2000 ... and I have no idea if these features are
>  > available if the software detects a LS-30 ... or even if your scanner
>  > comes with the "feeder" adapter(?)
>  
>  AFAIK the only missing settings are the single pass multiscanning
>  and the high bit setting.  I think everything else works.  Yes, the LS30
>  comes with all three film holders - slide holder, film strip holder, and
>  film strip feeder.

The one other difference is that the LS-30 firmware doesn't
support the batch slide feeder which holds 50 slides.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

VueScan 6.1.2 Available

[EdHamrick]

I just released VueScan 6.1.2 for Windows, Linux, and Mac OS.
It can be downloaded from:

  http://www.hamrick.com/vsm.html

What's new in version 6.1.2:

  * Added support for scanning MF at 2820 dpi
on Minolta Scan Multi (center 24 mm of film)
 
  * Added color calibration for ScanMaker 5

  * Fixed problem with LS-1000 and some scsi cards

  * Options in Files tab now aren't reset when
"File|Default options" is used.

  * Improved the help file

  * Slightly reduced size of window on Mac OS so it
will work on 640x480 displays.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Negs and backlight

[EdHamrick]

In a message dated 9/27/2000 9:06:21 PM EST, [EMAIL PROTECTED] writes:

> It would appear the default "automatic" is not
>  appropriate for backlit subjects.

The exposure of the scanner has nothing to do with the
existence of a backlit subject.

The exposure of the scanner for negative film is adjusted
to to map the color of the orange mask to roughly white.
This lengthens the green and blue exposure time relative
to the red exposure time.

If you look at the raw scan file (scan0001.tif) that gets
written when you turn on the "Files|Output raw file", you'll
see this effect.  Whether there's a backlit subject or
not is detail in the dark parts of the negative, not the
light parts.

The problem of a backlit subject is caused by the wider
exposure latitide of color negative film.  The intensity
range that's captured by color negative film is much
wider than that of slide film.

It's actually no problem at all if you take pictures with
a strong backlight, assuming you've set the camera
exposure for the light levels in the foreground.  The
color information for the people in the foreground
is still on the film - you just have to adjust the
"Color|Image brightness" option to emphasize
the foreground.  If you want to get the detail from the
backlit area, you can instead experiment with the
"Color|Image gamma" option (at the expense of making
the foreground image look flatter).

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan/Old Kodachrome Unable to Focus

[EdHamrick]

In a message dated 9/27/2000 8:35:02 PM EST, [EMAIL PROTECTED] writes:

> Somewhat related, does anybody know how the auto focus works in slide
>  scanners?  Does it try to maximize the high frequency content in a
>  particular region?  Does it even use the CCD sensor? Or some other measure
>  of defocus? Can the region be selected? Does the scanner return a number
>  indicating the amount of defocus that can be read by software?

I can't tell exactly how other programs do the auto focus, but the
way VueScan works is to read the CCD data at a single scan position.
It reads it for each focus position, and finds the focus position that
maximizes the sum of the absolute values of the differences between
adjacent pixels.

I've graphed this function vs. focus position, and it seems to be
a nice smooth curve with a single maximum position.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: SprintScan 4k is really SCSI I, not SCSI II as advertised...WAS => Scanne...

[EdHamrick]

In a message dated 9/27/2000 9:09:56 PM EST, [EMAIL PROTECTED] writes:

> But, I can't imagine why any 
>  SCSI designer for the past 6 years would make an asynch SCSI device 
>  either...

Every scsi scanner I've used (and I've used a lot) use async SCSI.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan/Old Kodachrome Unable to Focus

[EdHamrick]

In a message dated 9/27/2000 9:37:31 PM EST, [EMAIL PROTECTED] writes:

> Nikon specifically say *not* to use ICE on Kodachrome slides in the
>  NikonScan reference manual. I assume this would apply to VueScan for the
>  same reasons (regardless of what they are).

The dust removal in VueScan works just as well with Kodakchrome
as Ektachrome.  I use a different algorithm than the Digital ICE
algorithm.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Vuescan - native sharpening applied?

[EdHamrick]

In a message dated 9/27/2000 1:29:36 PM EST, [EMAIL PROTECTED] writes:

> In the last couple of days I have made some comparison scans using both
>  Vuescan (V. 6.1 beta 6) and Polaroid Insite (V. 4.5) on the same negs.

You might upgrade to VueScan 6.1.1, although there aren't
any sharpness differences from beta 6 (or any other beta).

>  When
>  comparing the scans in Photoshop viewed at 100%, it seemed that the Vuescan
>  scans were clearly sharper.  The Polaroid Insite ones needed USM at approx.
>  150/1.5/1  to achieve the same apparent sharpness.  I used both scanning
>  programs with no sharpening selected.

Are you sure you have focus turned on in Insight?  It uses basically
the same focusing commands as VueScan does, so there really
shouldn't be any difference in focus.

You should be able to hear when the focusing happens.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.

Re: Canon FS2710 - APS

[EdHamrick]

In a message dated 9/27/2000 8:45:27 PM EST, [EMAIL PROTECTED] writes:

> Has anyone tried Vuescan with the APS(IX240) adapter?  I cannot find any of
>  the normal settings in Vuescan.

I haven't added any special options for this adapter.  The problem
is that there isn't any way I can figure out to ask the scanner if it
has this adapter inserted.

All you need to do (I think) is set the "Crop|Crop size" to APS.

Regards,
Ed Hamrick


The filmscanners mailing list is hosted by http://www.halftone.co.uk
To resign,  with UNSUBSCRIBE FILMSCANNERS in the 
title, or UNSUBSCRIBE FILMSCANNERS_DIGEST if you are reading the Digest.