RE: filmscanners: Minolta DiMAGE Scan & Dimage 7 camera

["Frank Nichols" <frank@theNichols.net>]

I wonder if it would be posisble to create a "randomized" pattern of sensors
on a CCD/CMOS chip?

/fn

-----Original Message-----
From: owner-filmscanners@halftone.co.uk
[mailto:owner-filmscanners@halftone.co.uk]On Behalf Of Walter Bushell
Sent: Friday, June 29, 2001 11:42 PM
To: filmscanners@halftone.co.uk
Subject: Re: filmscanners: Minolta DiMAGE Scan & Dimage 7 camera


<SNIP>

> Any filtering of this nature would not be done at the lens level.  A
> lens is an optical device, and the best thing it can do is accurately
> translate everything it sees to the sensitive/recording layer.  This is
> what all lenses strive toward.  If any type of resolution lowering were
> to be required it would be done via electronic means.  Keep in mind film
> also has a "sampling rate", although it is somewhat more randomized and
> right now, still finer than most electronic sensors made available for
> mere mortals.
>

No, once you have aliased info in digital form it is indistingusible
from real data. Consider stripes that are .6 of the sensor frequency,
they will alias to .1 of said frequency and they cannot be distinguished
from such a pattern.

It is precisely the randomized nature of film that alaising does not
occur. There is no grid, so there is nothing to beat against, so to
speak.
> Art