RE: filmscanners: RE: filmscanners: VueScan clipping & flat images

["Jawed Ashraf" <Jawed@cupidity.force9.co.uk>]

> -----Original Message-----
> From: owner-filmscanners@halftone.co.uk
> [mailto:owner-filmscanners@halftone.co.uk]On Behalf Of Rob Geraghty
> Sent: 24 November 2001 12:00
> To: filmscanners@halftone.co.uk
> Subject: Re: filmscanners: RE: filmscanners: VueScan clipping & flat
> images
>
>
> "Jawed Ashraf" <Jawed@cupidity.force9.co.uk> wrote:
> [so much I can't begin to respond to much of it...]

When I eventually run out of film to scan I'll fall silent here...

> > > adjusting levels actually does.  But presumably the fewer
> bits from the
> > > A/D, and the smaller the input range, the poorer the result would be.
> > I dare say you're correct.  If the scanner is, internally, 8-bit, then
> some
> > data values have gone "unused" in your example.
>
> It was an artificial example, but I was trying to demonstrate how
> important
> the bit depth of the A/D is, even when the film is a neg, because
> if the A/D
> is set the return a fixed range of numbers based on a certain
> fixed voltage
> range, you're going to lose a lot of the potential resolution of the A/D
> when you scan colour negs.  If there's not enough bits in the A/D and the
> density rang eof the neg is very narow, presumably there's a risk of
> posterisation.

I spent some time experimenting scanning negs as positives with Nikon Scan,
and building curves *in Nikon Scan* (to both invert the negative and to
linearise/de-gamma) in combination with analog gain adjustments for each
channel.

Fascinating and frustrating.  Anyway, I learnt a lot.

An 8-bit A/D really would struggle. If you think that there's, say, 11 stops
of dynamic range in the raw image captured in the negative film (max, say)
and less than 8-bits of dynamic range in the film itself (7.x), then those
7.x bits of scanned data are gonna be stretched a bit thin by the time the
image has been "decoded".  Even more so when you start doing Levels/contrast
adjustments or curves.

>
> > BUT, because of exposure control in scanners, the scanner and
> > its controlling software in your workstation *may* manipulate the
> > exposure (brightness of the light or duration of exposure per scan
> > step) and/or manipulate the gain of the electronics (like the volume
> > control on your hifi).  So the end result is, frankly "there's no
> predicting"!
>
> But in practice I don't know of any consumer CCD scanner which does adjust
> the gain of the input.  AFAIK the only thing which is done is to
> adjust the
> exposure (integration) time.  Adjusting the integration time won't expand
> the input range, although it may shift the input to a level further above
> the noise floor.

There's also a question of how "straight" or "s-shaped" is the relationship
between illumination/gain and data coming out of the A/D.  An S-shape will
tend to cause varying ranges of output data, depending on "gain" or
"integration time".  Whether this effect is ever significant is something I
don't know.  Really needs someone who's implemented a scanner at this level
of detail to comment.

>
> > What happens when the image "naturally" contains more than 8 stops
> > of dynamic range (how does the 8-bit scanner and associated software
> > cope)?
>
> And my point being what about when that 8 stop range is compressed into a
> fraction of the possible values the CCD can read?  A normal
> colour neg is a
> long way from the range of a slide in terms of the transmitted light.  A
> slide like Provia 100F gets pretty close to the range from no transmission
> (black) to total transmission (clear or white).  But a colour neg has its
> sensitivity compressed into a much more narrow range.  If you don't have
> enough bits, you won't get the subtle tones whichare in the film.

Agreed.  Though this does depend on how far from the desired result the scan
is.  If it's close, then further manipulation won't be noticeable (I'd say).
But anyway this is that hypothetical 8-bit A/D which was current technology
way back when sometime?  Wasn't it?  Ah, the good old days.  (I first came
across a Nikon filmscanner at work way back when, about 1990/1 - boy was
that a groovy concept, and expensive!  Wonder how good its A/D was...)

>
> > So this experiment shows that on my monitor 6-bits are all that
> are needed
> > to satisfy my perception.
>
> Provided you can get the right 6 bits.  For instance, my experience with
> Nikonscan has been that 8 bits are useless with a high contrast
> silver based
> B&W neg.  It just doesn't handle the range of density.

When you scan a true b+w film, it seems the scanner loses a very significant
amount of dynamic range, a bit or more (I don't know what the mechanism for
this is - is it solely because scanning is not done across the R, G and B
channels - but only on, say, G?).  Somebody mentioned this effect recently
on here.  There's some name for this effect that I can't remember (can't
find the posting now).

>
> > The LS40 and LS4000 (used with Nikon Scan) do.  It's what
> happens when the
> > auto-exposure kicks in (I believe) which changes the brightness of the
> > "lamp" (there's logic for why I could be wrong - I'll let somebody else
> > argue the point).
>
> No, they don't.  As Ed explained, they change the exposure time not the
> brightness of the LEDs.

I believe you're misquoting Ed.  He responded to the opposite effect, in
response to someone who suggested what you're saying.  I'm fairly sure he
was quite specific that the LED brightness changes.  The manual says the
brightness changes, too (apparently - can't be bothered trying to find the
manual).  (Can't find this message from Ed, either - damn - must try
harder.)

Unless you mean that there is no connection between Nikon Scan's
auto-exposure algorithm and LED brightness.  I have to say if that was the
case, it would seem to be missing out on a key feature of the scanner
(variable lamp brightness!).  Oh well.

Jawed