RE: [Leica] Thoughts on digital and the impact on Leica

["Austin Franklin" <darkroom@ix.netcom.com>]

> At 3:22 PM -0400 9/29/02, Austin Franklin wrote:
> >  > .....I suppose which side of the fence you are on means you
> are going to
> >>  promote your team.
> >
> >Steve,
> >
> >You're certainly correct, but badly done testing (or biased,
> intentional or
> >not, reading of the results) is simply unacceptable.
> >
> >I don't mean to sound arrogant at all, but I'm not on any fence
> with respect
> >to this issue.  I have been designing digital imaging equipment
> for over 20
> >years, and I have done VERY extensive testing in this area, and
> I know the
> >facts first hand.  I KNOW that claim is simply wrong.  I've done hundreds
> >and hundreds of tests, and the conclusions drawn by this "article" are
> >simply flawed, or misleading at the least.
>
> So, in your hundreds and hundreds of tests of Canon 1Ds 11Mp cameras,

Oh Henning, I NEVER said I tested the Canon 1D...and I KNOW you know that.
I HAVE tested many Bayer pattern CCDs with the same, less and even more
resolution, as well as three pass cameras/backs, as well as very high end
one-shot backs.  I KNOW what this camera is capable of.  It can't make water
from whine...no matter how much people want it to ;-) (and yes, I said
"whine" purposely)

> this is the conclusion you've come to? :-) What we're seeing here
> with this camera is a new level of performance, and in spite of
> preconceptions and 'knowing' the right answer, we'll just have to
> wait until we get a chance to try it ourselves or have someone we
> trust do a test.

Why?  I KNOW what 11M pixels is, I KNOW what they look like, and I KNOW what
they can do compared to film.  It's a HARD FIXED number, and it is what it
is, no matter how much people want to believe it's more...it simply is not.

> >A VERY poorly exposed, developed and scanned, 10 year outdated high speed
> >film that was sitting on the dashboard of one's south facing car, in
> >Florida, for a week, before being developed, would probably be
> inferior to
> >an 11M pixel Bayer patterned camera.  Is that "fair" to use
> images from that
> >as the poster child for film?  Of course not.
>
> These assumptions, or rather implied assumptions are of the worst
> sort.

No, not at all.  It is a fact that an 11M pixel Bayer pattern CCD can NOT
give the same resolution as ALL 35mm film.  It's a simple fact.

> He scans on an
> Imacon, which may not be up there competing with $50,000 drum
> scanners, but is as good or better than what most of us will have
> access to.

Then he simply can not conclude that 11M pixels is better than ALL film.
It's a physical impossibility.  SOME film, yes, but it is misleading to say
"better than film" without specifying SOME film, or a particular film,
whatever.

The CCD in that camera is, first off, a Bayer pattern CCD, which means it is
NOT NOT NOT an 11 M PIXEL sensor.  It is an 11M SENSOR sensor.  It has only
%50 of the sensors green, %25 red and %25 blue.  That's a fact, and is
undisputable.  What happens is software in the camera INTERPOLATES, as in
MAKES UP FAKE DATA, though based on the real data, to get from 11M SENSORS
worth of data to 11M PIXELS worth of data.  It's a THREE TIMES "making up"
of data!  You can argue that all you want, but it simply isn't arguable.
Now, I completely understand that it does a good job, but these are simply
stated facts.

OK, so the CCD in this camera is 11M "pixels", and let's give it even the
highest benefit of the doubt that the "made up" data is perfect.  That's 4k
x 3k.  At 240PPI to the printer, you can make a print as large as 16" x 12".
Certainly that is very sufficient for many people...in fact for most people,
I am not in doubt about that.  But, what kind of enlargements can be had
from film?  That's what the real question is.  Of course you can "rez up"
the digital file, but that's fake data, once again...and, oh, BTW, you can
do the same to a scanned 35mm film as well...so we really don't need to
bring that into the equation.

Well, I have a scanner that can scan 35mm film at 5080...and it scans the
film with all three colors, so all ready the data from the film has higher
"integrity".  What that gives me is 5080 x 7620...and if the film is good
(and some is, and some isn't), and at 240PPI to the printer, I can get an
enlargement of  20" x 30".  Nearly twice the size of what the digital file
can give me...  How does that visually compare?  Well, I have made some very
nice 20" x 24" prints from 35mm low speed color film (even Delta 100 and
Plus-X B&W film, and let's not even get into how these digicams handle
B&W...)...

> I think that before we say that his 'tests' are flawed or misleading
> we need a lot more information. I would tend on first impressions to
> say that Michael Reichman is possibly onto something, in spite of my
> own experience and analysis of the digital vs. film situation at the
> present stage of development.

What can he be on to?  Facts are facts.  There is, of course, the issue of
"perception", and "subjectivity".  A LOT of people are drawn to digital
because of what they perceive to be sharper, cleaner, images...but they are
lacking in detail compared to film.  It's just a fact of the medium.

> >Also, sharpness has nothing to do with resolution, as I stated in another
> >post.  It's something that is commonly misperceived.
>
> Most of us know that quite well, Austin.

Well, no, Henning...most people actually don't.  They confuse the two, and
believe because an image is sharper, it has better resolution.

Now, in now way did I say that the images from the Canon didn't look great,
they do, I know that.  What I said is to claim this camera has higher
resolution than "film" is simply incorrect, and pure BS.

Regards,

Austin

- --
To unsubscribe, see http://mejac.palo-alto.ca.us/leica-users/unsub.html