Vuescan long exposure pass

[SS]

I have tried this (scans once normal and once slowly i.e. more exposure,
then combines the two) with my Canon FS4000US and under exposed film, but
there seems to be no significant difference from a normal scan. How can I
best scan an underexposed film to get out the detail or can this be achieved
just as well through altering brightness and contrast in PSP/Pohotoshop?

 

[Roger S.]

The difference is in noise levels in the shadows, especially if you
will be brightening them with curves in Photoshop. Be careful, though-
look at the scan at 100% and make sure it's not blurry or out of
alignment. I've generally found that the long exposure pass is useful,
however.

Another route is to scan at 2 diffferent exposures and combine them in
the program photomatix (hdrsoft.com I think). This will get you more
shadow detail, but if the scan is so much as 1 pixel off it won't work.
You also have to manually pick 2 exposures in Vuescan. I recommend
the normal value and then significantly greater (maybe exp 8 or 11).

You should only need this for slide film as the FS4000US is good enough
for any negative film without any tweaking. Just use the advanced work
flow.

 

[SS]

So does selecting a different exposure setting then affect hardware (i.e
scan speed - effectively putting more light through the film) rather than
just adjust brightness via software?

 

[Roger S.]

Bingo. For difficult slides, I also recommend enabling pixel colors in
VueScan when doing slide scans as if you see that the whole image is
out of gamut, VueScan isn't scanning for long enough. Try manually
increasing exposure as much as you can without blowing out your
highlights (pixel colors will show you that too- you'll see lime green
dots).

If you just brighten shadows with curves you end up dredging up scanner
noise and not getting any more detail past a certain point.

 

[Bart van der Wolf]

So does selecting a different exposure setting then affect
hardware (i.e scan speed - effectively putting more light through
the film) rather than just adjust brightness via software?

Yes, the exposure time is adjusted. If I'm not mistaken, the FS4000
does so in a limited number of fixed steps.

Bart

 

[Roger S.]

That's right, Bart. You can easily see this when you look at scan
histograms. There is no difference between 1.9 and 2.0 but a huge
difference between 1.9 and 2.3 for example.

 

[Bruce Graham]

That's right, Bart. You can easily see this when you look at scan
histograms. There is no difference between 1.9 and 2.0 but a huge
difference between 1.9 and 2.3 for example.

yes, it just changes the speed of the film stepper motor. you can hear
it slow down by a big factor (2?) for each step.

 

[Don]

I have tried this (scans once normal and once slowly i.e. more exposure,
then combines the two) with my Canon FS4000US and under exposed film, but
there seems to be no significant difference from a normal scan. How can I
best scan an underexposed film to get out the detail or can this be achieved
just as well through altering brightness and contrast in PSP/Pohotoshop?

Here's a free program that will do that for you automatically:

http://www.ict.usc.edu/graphics/HDRShop/.

Also, download the plugin called "Reinhard HDR Tonemapping Plugin" to
export the combined image.

The program only works with 8-bit images but since your goal is
TV/Computer output that shouldn't be a problem.

Don.

 

[Don]

So does selecting a different exposure setting then affect hardware (i.e
scan speed - effectively putting more light through the film) rather than
just adjust brightness via software?

It's the same as taking a picture with your camera. Longer exposure
means the aperture stays opened longer.

When you scan you're essentially "taking a picture of a picture". The
major differences are that instead of film being exposed it's the
digital sensor, and while the camera exposes the whole frame at once,
the scanner does so one line at a time.

Don.

 

[Steven]

yes, it just changes the speed of the film stepper motor. you can hear
it slow down by a big factor (2?) for each step.

FYI,

The FS4000 has speeds of 1, 2, 3, 4, 5, 6, 8, 10, 12. The Vuescan speed
(1 to 6 valid) is rounded down to an integer and then multiplied by 2
and the result is sent to the scanner which uses it as a divisor for the
scan time clock. This means that VS speed 2 is twice the exposure of VS
speed 1 and VS speed 6 is 6 times the exposure of speed 1. So the total
exposure range is 6 times.

-- Steven

 

[SS]

Strangely having tried the full range of exposure (0-10) on the scanner in
Vuescan it seems to make no difference to the end result!

 

[Don]

Strangely having tried the full range of exposure (0-10) on the scanner in
Vuescan it seems to make no difference to the end result!

It's probably the Vuescan exposure bug...

When I tested Vuescan a couple of years ago it turns out if exposure
is boosted beyond a certain point it essentially becomes random.

To add insult to injury, the exposure display is secretly rolled back
i.e. changed, and the actual (and unknown!) exposure bore no relation
to this display.

That's Vuescan...

Don.

 

[Roger S.]

"I have some FS4000 software that will scan at different exposures if
you
are interested in seeing the results."

Can you explain how this differs from what Filmget or Vuescan do? Is
there a wider range of exposures or more increments?

 

[Steven]

"I have some FS4000 software that will scan at different exposures if
you
are interested in seeing the results."

Can you explain how this differs from what Filmget or Vuescan do? Is
there a wider range of exposures or more increments?

It is a scanning program that drives the FS4000 and has auto or manual
exposure modes. It doesn't do any colour correction so what you get is
just the FS4000 output as a TIFF file.

It differs from professional software in that it is much cruder and that
the documented source is supplied.

Vuescan allows exposures of 1 to 6 which become 2 to 12. The extra
possible speeds (1, 3, 5) are less useful so this is probably why Ed
didn't include them. Speed 1 is too fast as even a blank slide won't
saturate the CCD. Speed 3 has some value as it divides the full F-stop
difference between speeds 2 and 4. Speed 5 splits 4 and 6 but since the
difference is only 50% it is less valuable. Also, these odd speeds are
implemented by using the longer step interval (= slower scan) of the
higher even speed but reducing the exposure time (duty cycle) of each
step.

The URL my my software is <http://users.tpg.com.au/hys143>. It doesn't
do IR but if someone wants it I could try to add it to the TIFF file and
then you could get VS to process the RGBI file.

-- Steven