Digital vs Scan, Copies of Copies

[P3D Lew Clayman/K Szafran <kandlew@xxxxxxxxx>]

A word from the 'original poster' on scan-vs-digital.  First of all, I want
to thank the many folks who have jumped into this conversation.  I was
concerned, as a first-time poster, that I might be opening a thread of
little interest to the list.  My mind is at ease on that point.  

Second of all, yes I am "Clearly...hung up on digital," in the sense that my
whole aim is to post quality 3D images on the Web.  Scanning and direct
digital photography are the only two ways (known to me) of achieving that.
{Also practice would help, but that's another issue.}  And yes, I have
budget concerns.  Don't we all?

Consider a 35mm print, scanned.  If we assume a 4x6 inch glossy (10.16 x
15.24 cm) scanned at 1200 dpi (472.44 dpcm) the image field is then
4800x7200 pixels.  Citing figures posted by Gregory Wageman:  

        >A good chrome has a resolution of about 150 line-pairs/mm.  The
Kodak has 3060 pixels across the 34mm 
        >imaging area, or about 90 pixels/mm., or about 1/3 to 1/2 the
resolution of film. 

Thus the Kodak (I think we're discussing a DCS50, correct me if I'm wrong)
has about 42% of the horizontal resolution of scanned film.  My instinct is
that this difference is neither negligible nor "infinite," but somewhere in
between (call it "appreciable"???).  However, the effective resolution of
the scanned print is not simply the scanner resolution of 7200 horizontal
pixels.  Pulling in another thread on this list, Greg Ercker, responding to
George Themelis, writes: 

        >  The resolution (or sharpness) of the original slide is a function
        >of the resolution of the camera and the resolution of the film. I.E.
        >lower than the lowest of the two.
        >
        >  The resolution of the copy will be a function of the resolution
        >of the original, the resolution of the copy lens, and the resolution
        >of the copy film.  I.E. lower than the lowest of the three.
        >
        >  So unless the copy lens and copy films have way more resolution
        >than the original slide the copy slide will have lower resolution
        >(or sharpness).
        >
        >  Since the sharpness of the edges is a large part of seeing
        >the 3D in a slide, the copy will not have the original's snap.

I believe the same reasoning applies to scanners also.  That is, the
resolution will be somewhat lower than the lowest resolution of the camera,
the film, the enlarger, the printstock, and the scanner.  I think we can
take it as read that the lowest of these is the scanner, but the point is
that the effective resolution is not as high as the nominal dpi.  It follows
that the comparison of 7200 vs 3060 horizontal is not quite valid: the
effective ratio is less extreme.  {Does anyone have resolution figures for
filmscanners?}

There are many good stereo pairs posted out on the Web, and I presume that
many of the photographers involved read this list.  Has anyone posted a
digital pair, and if so, can you share what you used?  The proof, as they
say, is in the pudding - perhaps some "pudding" is already out there for
viewing.

Thanks

Lew Clayman


------------------------------