Digital BW, The Print

Claude,

> > response", that's another one of those things like "sharpness".  A
> two pixel
> > cameras is sharper than any photograph you can get!  But, that's
> > meaningless.
>
> The new X3 chip has 3 color layers just like film does.

I know how the Foveon chip is architected...  There are differences in the
Foveon and film, it's not "just like", it's "similar".

> The rest have
> holes in the data.

It depends on the camera, and how the microlense is arranged, as well as
what exactly you mean by "holes".  There are absolutely no holes in the data
as far as luminosity goes.

> I can't make it any clearer than this.

You don't need to make anything clearer to me, I know EXACTLY how the Foveon
chip works, as well as how the Bayer pattern sensors work.  I've designed
with Bayer pattern sensors, as well as many other types of imaging sensors,
for 25 years now.

> I have been
> shooting full color capture and selling prints that look
> indistinguishable from film.

That's purely opinion, and they may be indistinguishable to you, but that
does not mean they are to everyone.  I'd bet you I can tell the difference
in a 20 x 24 print.

> I sold my Hasselblads when I got the
> Foveon because the results were just as good in my market.

I don't know what your market is, but it doesn't have a very high
requirement if that's the case.  I'm not being mean, just being realistic to
your "hype".  The Foveon sensor does not offer near the resolution that a
Hasselblad offers, period.

> Now it's
> available for a song.

They aren't getting the highest they've ever gotten used, but they aren't
going for "a song".

> No one would buy a 2 pixel camera.

Of course not, but it's an example showing that "sharpness" is meaningless,
and the issue of sharpness is completely misunderstood and exaggerated by a
lot of people.

> As far as sharpness for the
> others, if the MTF of the lens matches the well site pitch of the
> chip at the critical aperture, you have a greater potential moire
> condition, at some sympathetic frequency in the subject since the
> blue and red color "grids" are 1/4 of the total well sites. 3/4 of
> the data is simply not there.

That's not true.  All the luminosity data is there, the color data is what's
reduced...and it's not necessarily necessary either.  If you understood how
TV worked, and how the eye works, you'd understand why.

> That's why monochip cameras have to
> pixel smear with AA filters or demoirize the file after the fact.

No, the new Foveon is a "monochip" camera...does it have this issue?  No.
The new Kodak and Canon cameras also have no anti-aliasing filter either, as
far as I know.  Anyway, this is really a "so what".

> ...the Foveon... It simply made the most "filmlike" color image I ever
had.

It absolutely does not have the detail that film has, as it doesn't have
near the resolution that film does.  Minox film, yes, but not high end 35mm
and especially not medium format.  There are many aspects of an image, and
just because one imaging technique "appears" sharper (or what ever) does not
mean it's an overall "better" image, or has better "fidelity".  Fidelity is
accurate reproduction...and that means of ALL aspects of an image, not just
one of them.

I don't believe you are understanding the implication of my example of a two
pixel camera being as sharp as it gets...yet the image is obviously not very
good...  It shows that sharpness is not a good metric for overall image
"fidelity".

> None of
> the other cameras could match it.

Well, could match WHAT specifically?  What other cameras?

The fact is, there are advantages and disadvantages to both medium, and for
some, digital suits their needs quite well, and for others, film.  To
blanketly say that the Foveon is better than (all) film is just simply not
true, as it is not.  It may be better than "some" film, but not all.

Austin