Tony,
You're to be commended for bringing this problem to our attention.
I've mulled it over a bit and come to some conclusions. I could be
wrong however, so with that in mind, here are my opinions.
It seems to me from eyeball guessing that my LS-30 is resolving grain
in 100 ISO films at roughly 40-80% distortion, which looks pretty bad
on the monitor at 100% view. 800 speed color neg film does much
better at what I would guess to be roughly 25% distortion.
The silver lining to this cloud is aliasing distortion (with the
LS-30) looks worse on screen than print, IMO. When the ink hits the
page aliased grain looks more or less like analog grain, being worse
in certain tonal areas than others. Personally I don't find the
actual image degradation objectionable, even all that noticeable, in
prints of full frame negs up to 24x36 inches with slight interpolation
to reach that size. Perhaps this explains in part why this problem
hasn't assumed greater weight in the work a day world. Still, other
things being equal (they never are), I would prefer less aliasing
distortion. But I suspect that when aliasing distortion levels are
kept under about 100%, the effect in print will be *relatively*
negligible.
Dave King
PS - Just for funsies I recently printed a max res file from an Fuji
S1 digital camera to 24x36. While the image was remarkably good in
many respects, and completely grain (pixel) free in areas of even
tone, image resolution was far less than from even 800 speed neg film
scanned in the LS-30. For the time being at least I'll take the
grain, aliasing and all.
----- Original Message -----
From: Tony Sleep <TonySleep@halftone.co.uk>
To: <filmscanners@halftone.co.uk>
Sent: Sunday, April 08, 2001 1:18 PM
Subject: Re: filmscanners: Review of the Nikon CoolScan 4000
> On Sat, 07 Apr 2001 15:07:11 +0930 Mark T.
(markthom@camtech.net.au)
> wrote:
>
> > Eeek. I thought grain-aliasing and film resolution was covered in
> > either lesson 1 or 2 when you do Filmscanning 101..! :)
>
> When I first came across this, and began to suspect it was an
aliasing
> phenomenon, I was unable to find any references anywhere. Not one.
It
> didn't exist, and nobody had questioned why images which produce
> near-grainless prints should suddenly produce easily-visible grain
in
> scans. Nevertheless, it seemed to be that it was completely
intelligible
> as an aliasing artifact, so I wrote it up as a tentative
explanation.
>
> About a year later, Pete at Photoscientia noticed the same
phenomenon and
> did some research. Like me, he found no reference material, except
the
> material I had posted about it at my site. He contacted me and we
> discussed what we were both seeing and that we were not
hallucinating but
> it appeared that scanners were, which was reassuring for both of us.
We
> agreed that the fundamental mechanism was aliasing arising from
grain
> pattern interference with the matrix of pixel geometry. His
investigations
> resulted in the feature at http://www.photoscientia.co.uk/Grain.htm
which
> remains the most thorough attempt at an explanation - you still
won't
> find it in any text books AFAIK.
> Apart from Pete's Acer review at www.photoscientia.co.uk, I have
still not
> seen *any* other review of a scanner which mentions it, though many
> wrongly assert that 2700ppi is enough to image film grain even from
ISO100
> materials. I have even been contacted by a manufacturer rep and
asked if I
> could suggest any reason why a user was reporting massively
exaggerated
> grain with ISO400 film, so I don't think this problem is widely
correctly
> perceived, let alone understood - probably because many reviewers
and
> others within digital have minimal experience of film photography.
>
> It may be that the engineers who design scanners have a huge file on
the
> problem, and I would be astonished if they do not as aliasing is
very well
> understood and many techniques are being developed to deal with it,
> especially within digicams. But mfr's. mouthpieces, the marketeers,
are
> hardly going to tell us about it, as it devalues the sellable notion
of
> scanning as a near-perfect process.
>
> Regards
>
> Tony Sleep
> http://www.halftone.co.uk - Online portfolio & exhibit; + film
scanner
> info & comparisons
>