Re: Canon 5D stomps Sigma SD14 into the dust

On Mar 18, 4:51 am, "Robert Brace" <rlbr...@xxxxxxx> wrote:
<achilleaslazari...@xxxxxxxxxxx> wrote in message

news:1174169246.238158.156330@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx



On Mar 18, 12:09 am, "Robert Brace" <rlbr...@xxxxxxx> wrote:
<achilleaslazari...@xxxxxxxxxxx> wrote in message

You also appeared to miss my second paragraph. But never mind, carry
on replying to what you think is being said instead of what is
actually being said.

Since here is where you pointed out his non-impartiality, perhaps my
understanding of the obvious is better than your statement of whatever it
is
you thought you said.

Please explain how

"His arguments are perfectly fine and his conclusions may or may not
be
right (I suppose they're not as far wrong as most people here would
immediately think, but have no idea, not having used a non-CFA
camera), but he hardly seems like an impartial observer."

can be interpreted to mean that I disagree with or dislike his
conclusions. I was pointing out that he quite obviously started by
blatantly throwing out 3/4 of the available pixels in the canon. This
is hardly impartial; did he study colour accuracy in the sigma?
especially as the ISO goes up? Look at fig. 5 in
http://www.foveon.com/files/CIC10_Lyon_Hubel_FINAL.pdf
and tell me how easy it is to separate R, G and B. Or look at the
acrobatics needed to properly convert the sigma files (the R G and B
separation isn't very good so the off-diagonal elements in the colour
matrix are huge). Is the response of each separate channel linear? How
much does it deviate? These things can be worked out from papers put
out by Foveon themselves (with a bit of effort), and they are not very
pretty.

Look, I'd love for Foveon sensors to work, I find the low ISO
photographs excellent in terms of sharpness (but it is partly due to
the fact that they can get away with not having an AA filter because
aliasing is less visually objectionable than in CFA cameras). But they
do have their problems and it is easy to see why they have them (so it
is not just a problem of the particular implementation but something
that really does need more work). He said not a single thing about
these; instead, he talked only about sharpness. That tells me he was
out to make a point, not to test things. It doesn't mean that I will
ignore what he says, just that it rings a warning bell.

Just like all the people here who concentrate on signal/noise ratio to
the complete exclusion of resolution, ergonomics etc, or other
examples I can think of.

I am sure he did this partly on purpose to have fun with the reactions.

Of course he was out to "have fun with the reactions" and to point out the
obvious in a comparison of AA vs. non-AA captures. He also pointed out some
glaring firmware shortcomings as well. However, let's not forget he also
pointed out some obvious shortcomings in the production of the AA filter
equipped 5D's images.
My God he actually had the fortitude to stray into that hallowed ground of
Canon criticism and that, in my opinion and especially in this theatre, is
interesting in itself!!


In my opinion, whether or not the Foveon is ultimately successful will
depend upon whether or not the bodies can be made publicly acceptable (as in
the mass market) before the development funds are exhausted. No astonishing
news there.

I think they mainly have to do something about the colours and about
the noise. If they manage to get good consistent colours up to, say,
ISO 1600 (from what I've seen, things currently turn nasty much before
that; it has to do with what I wrote above, I think), and if they
manage to get a bit higher resolution (no matter what he says,
upsampled images from the SD10 I've looked at seem significantly worse
than from my D200; I don't know how different it is from the SD14,
don't forget that we can also sharpen which this guy didn't seem to),
then everything is fine.

I saw some pbase gallery of SD14 images today, very disappointing.
There were lots of ISO 800 images, and there was quite a bit of noise;
not so bad, it was high frequency so probably not disturbing in a
print, and I don't mind high-frequency noise anyway. But the problem
was that it was there at lower ISOs too, even at 100! I assume the
photographer lifted the shadows and that's why, but it's still
unacceptable.

I hope they manage to solve these problems, as an idea it's great
(mainly because you can avoid the AA filter since you don't exacerbate
aliasing/ringing artefacts by colour interpolation) but currently I
personally wouldn't pay so much for this camera. Well, plus I wouldn't
really like to be forced to use only Sigma lenses, and the camera
itself is not all that impressive, but that's another story.

.

Relevant Pages

  • Re: S6500FD/S6000FD owners report
    ... every camera you've ever commented upon. ... take for you to grow disenchanted with the FZ8's noise? ... batteries.Never required or will purchase a spare battery for the F20, ... greatest in the middle of the ISO range. ...
    (rec.photo.digital)
  • Re: A beginner question
    ... of calculating exposure times of astronomical objects ... ISO values with dim astronomical targets is rarely advantageous. ... because the important components of noise are amplified at the same rate ... When I looked at the question concerning a magnitude 5.8 object on a terrestrial camera, I had no scale to equate this to daytime photography. ...
    (sci.astro.amateur)
  • Re: Nikon D80/200 - Canon 30d
    ... than an el cheapo Point&Shoot camera with a tiny teeny-weeny CCD ... from a manufaturer that its CMOS sensor has less noise than a CCD ... just look at the images (not the ...
    (rec.photo.digital.slr-systems)
  • Re: More on Canon Rebel XT noise at high ISO - 2 main new data points
    ... > No it's YOUR fault.The firmware in the camera can't adapt, ... And, when exposure is very bad, I don't understand ... >>As to noise reduction, I don't currently own Neat Image, ... > is free and will batch process up to ten images as a time, ...
    (rec.photo.digital.slr-systems)
  • Re: Panasonic DMZ-FZ8 versus comparable dSLR
    ... ISO 200, which is the D70's lowest ISO setting, but the noise on the ... D70 at ISO 200 is essentially non-existent while on the FZ5, noise is ... I personally almost always shoot with my FZ8 at ISO 100, ... artifacts from in camera processing. ...
    (rec.photo.digital)