At 2:08 PM -0800 2/9/00, msvphoto@xxxxxxxxxxx wrote:
In a message dated 2/9/2000 Skip Williams wrote:
I understand your issue and argument on Olympus' strategy, but I'm not
convinced. I want to see some hard facts on this whole issue before I'll
turn over and yield. I'm sure there are more issues than I'm aware of
regarding lenses for Digital and Film, but I haven't seen the one that
describes why existing, state-of-the-art lenses won't produce optimal
results on a CCD.
I defer this to the EE and Optical Engineer types. I would like to hear
some hard facts too. In my life outside photography I am an electronic
technician. Among the things I am trained to work on are video
cameras...since before the days CCD imagers were used in them. I admit I
only know a little about the design of imagers, but I do know that the
individual RGB "buckets" are not in exactly the same physical location.
Think about the video world. All RGB. All red, green and blue dots next
to one another, _not_ stacked on top of each other (like emulsion layers
of film are). This is why it kinda makes sense to me that maybe there
_could be_ optical optimization differences between film and digital
lenses. No doubt someone more experienced than I can weigh in with
better information. My take is that Apochromatic lenses bring all colors
of light to the same location at the same time and that is the ideal for
film. Perhaps there is a better way for getting the individual primary
colors to land on their respective buckets? Hence the optimized for
digital optics thing. I can buy it...but I am also happy to be proven
wrong.
This all sounds pretty screwy to me too. although anyone who has access to
a disassemblable digicam should be able to answer this. The current reports
of lens focal length and chip sizes suggests that the chips subtend similar
angles wrt the lens as does 35mm film, unless the lens designs are massively
retrofocused (which would indeed be a way to get the light
more "parallel" going into the little buckets/plates.
The thing about screwing with the rear elements of lense sounds weird (unless
it's just a way to make the image fit a smaller sensing area, which is short
sighted). And I don't think that anyone has makes cameras to the kinds of
tolerances you would need to reliably focus light of different colors at
adjacent pixels using something on the lens. (Unless you could recalibrate
on the fly which of your pixels was R, G or B) What seems more likely (and
might be the correction plate the Oly guy derided) is a diffraction-grating
component on the sensor surface that directs the light to appropriate pixels...
paul
Paul Wallich pw@xxxxxxxxx
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|