The main problem here is that manufacturers use dpi or ppi to refer to
the number of physical elements or density on the CCD, it tells us
nothing about what happens along the route and afterward in terms of
real output. Typically an lpi rating is a result of actual output from
a resolution chart, which tells you a lot more about what is going on
within the scanner.
Scanners have optical paths including lenses, mirror and filters, and
then electronics all of which will subtract from that absolute pixel per
inch number based upon the physical number of sensor elements in the
CCD.
If a manufacturer was to list an lpi speck using the same method they
list dpi specks, they would simply be i/2 the dpi or ppi number. I do
not believe any CCD scanner can take this theoretical and bring it into
reality.
That also explains why the two Nikons (for example) show little
difference in resolution when tested at output, in spite of about a 35%
increase in the number of sensor elements per inch. There appear to be
other matters than the sensor density in terms of reaching limitations
in desktop scanners.
Art
Jawed Ashraf wrote:
>
> A comparison of the LS4000 and some Imacon, a while back, seemed to indicate
> that the LS4000 has resolution roughly equal to the Imacon's 3200dpi (some
> parts of a test image were better, others worse).
>
> Does anyone have a thorough understanding of this? Have I got the wrong end
> of the stick?
>
> Can someone explain how to go from ppi to lpi?
>
> Jawed
>
> > -----Original Message-----
> > From: owner-filmscanners@halftone.co.uk
> > [mailto:owner-filmscanners@halftone.co.uk]On Behalf Of BeckettJB@aol.com
> > Sent: 29 July 2001 20:03
> > To: filmscanners@halftone.co.uk
> > Subject: Re: filmscanners: 35mm filmscanner choice
> >
> >
> > I appreciate all the feedback to date. It seems that the SS4000
> > is the "tried
> > and true" choice, a scanner with a track record. Roger's point
> > about software
> > being no substitute for resolution/dpi is well taken. Does the stated
> > manufacturer's dpi spec. correlate exactly with how many lpi you
> > get in the
> > final scan? (I noted in a Pop Photo review the IV ED's "optical
> > resolution"
> > was 53.3 lpi while the nikon 4000 was 60 lpi, so despite having
> > roughly 75%
> > of the 4000's stated dpi it achieved roughly 90% of its optical resoving
> > power...hmmmm, is this a meaningful test?) The SS4000, not being
> > a recently
> > released scanner was unfortunately not included in their testing.
> > ()
> >
> > What would really be nice at this point would be a head to head
> > comparison
> > between the new Canon 4000 and the SS4000...At this point I am
> > leaning toward
> > the Canon because of its FARE feature. But, it has only been out
> > for a few
> > months so has had little time to be evaluated by many users.. I'm
> > a little
> > nervous about being an "early adopter"
> >
> > James
> >
