Three new SD9 pics - all for now

  • Thread starter Thread starter Phil Askey
  • Start date Start date
exactly what England is. I know it is part of the country represented in the United Nations as the United Kingdom (or Great Britain) which includes England, Scotland, Wales and Northern Ireland. So why does England have its own World Cup team and Olympic athletes. Is England like New York and Scotland like New Jersey (God forbid) - they don't have their own World Cup teams. Why if you call someone from Scotland "British" there is a good chance you'll end up on your kiester? Why is it that "Braveheart" was so long if nothing was really accomplished? Is it true that the new world was actually started because everybody flunked geography and had to leave? But most importantly, why does there appear to be so much dust on the SD9 sensor when it is the only digital camera to use a dust protector?
 
Why does there
appear to be so much dust on the SD9 sensor when it is the only
digital camera to use a dust protector?
I've been wondering the same thing. Phil's samples are beautiful, and there isn't a lot of dust there but there is still some. The new samples on Steve's digicam's site are covered with dust as are most of the others we've seen.
Any comments on this issue Phil?
And thanks again for the fine samples.
--
Ian
6900
 
I've heard this but I don't see "a lot of noise", break the'00146'
image into red, green and blue channels and you'll find noise
levels are very low, especially in the red channel. Those skies
are as clear as I would expect from any digital camera.

Lastly most people think that a pure blue sky is a single swathe of
blue colour... It isn't.
First I would agree that some of the "sky noise" is natural, but there also appears to be a problem with the SD9 in dealing with CERTAIN flat colors.
Certain colors seem to have a lot of noise in them were other colors don't.

Green and Red in particular seem to have the most noise problems. Blue and Yellow seem to show little noise. If you look at a more Cyan colored (rather than dark blue colored) sky (such at you IMG00094), you will see that there is noise in the Green channel. What is interesting is that the seems to be some correlation with the noise between pictures suggesting a fixed pattern noise (it would be interesting to see two pictures with a clear "cyan" colored sky to see for sure if there is fixed patterning for sure).

Getting away from the variables of the sky, below is a crop of the Imaging Review color test chart. Notice the noise in the Green and the Red (under the Magental column). If you look at the Cyan patch in the green channel you will see green channel noise.

http://www.fototime.com/ {80ADB6C7-5B99-484D-B5D1-DEB56505E27F} picture.JPG
Phil Askey
Editor / Owner, dpreview.com
--
Karl
 
Why does there
appear to be so much dust on the SD9 sensor when it is the only
digital camera to use a dust protector?
I've been wondering the same thing. Phil's samples are beautiful,
and there isn't a lot of dust there but there is still some. The
new samples on Steve's digicam's site are covered with dust as are
most of the others we've seen.
Any comments on this issue Phil?
And thanks again for the fine samples.
--
Ian
6900
probably because the guy installed the dust cover was very dirty... hehe. But the softness mention ealier maybe due to DOF thing.
 
When you take a close up, the Depth of field gets very short unless the aperture is very small (high F-number).

On something like the shot of the leaf, the DoF was probably about 1 inch (2.54 cm).
Hi guys,

Well, the weather was fine this morning so we went out on a quick
shooting spree (good job because it turned cloudy again at lunch
time). Below are three samples saved as JPEG quality 10 directly
from Sigma Photo Pro software. All converted using default
settings with a slight boost of sharpness (0.4) and some highlight
adjustment (-0.7 on one, can't remember which).

This is all I'm providing for now, don't expect anything else
before the review. Please don't email me or post "Phil: " messages
on the forum because I won't respond and don't take special
requests for my reviews. The review will take approximately two
weeks (but don't start beating me up if I miss that date, the
weather here in London is very unpredictable).

So, here you go, three new samples from a production SD9 (originals
around 1.3 MB each).

http://www.dpreview.com/temp/IMG00094-001.jpg
Original:
http://www.dpreview.com/temp/IMG00094.jpg



Original:
http://www.dpreview.com/temp/IMG00129.jpg

http://www.dpreview.com/temp/IMG00146-001.jpg
Original:
http://www.dpreview.com/temp/IMG00146.jpg

--
Phil Askey
Editor / Owner, dpreview.com
--
Karl
 
First of all, thanks to Phil and the others who have made an extra effort to get images for us to look at, and I mean all the way back to Köln.

The only conclusion I can truly draw after looking at all of these images and comparing them to many others from Sony, Canon, Nikon, and Fuji cameras is that a comparison of the Foveon images with the others is extremely difficult and soon disintegrates from the sublime into the ridiculous. I know we have gone through just about every issue known to digital photography in trying to rip these things apart. What I see at the end of the day is what was there all along.

1. Clean pixels: Yes, there may be spots on the images. Yes, there may be residual this or that. But the pixels themselves are clean. By that I mean distinct from each other with clear edges. For what it’s worth, among other things that means they can be manipulated (resampled) more easily, if I want.

2. The images for the most part look great straight out of the camera at normal (100%) resolution. The colors ring true; the depth of shadow detail looks about right; and the edges seem natural. The images taken under extreme conditions – and by that I mean for this camera, i.e. poorly lit, very high dynamic range situations – reveal problems that seem very natural. As a case in point, what film handles poor-lighting shots well by giving us deep shadow detail and holding on to the hot spots and not getting grainy (noisy)? None that I ever used. Why would this sensor respond differently when it gets out of its stated range (ISO 100-400)?

3. The images taken under highly controlled situations by some pros (most likely) are outstanding. Printing one of those at A3+ on a humble Epson 2100 is a lesson in patience that pays off.

There is not much else I can say at the moment. I am sure that Phil’s review will reveal much and set off yet another firestorm of discussion, which will be good, but only underscore my point: These images are different and require a different perspective to be understood and accepted.
 
First of all, thanks to Phil and the others who have made an extra
effort to get images for us to look at, and I mean all the way back
to Köln.

The only conclusion I can truly draw after looking at all of these
images and comparing them to many others from Sony, Canon, Nikon,
and Fuji cameras is that a comparison of the Foveon images with the
others is extremely difficult and soon disintegrates from the
sublime into the ridiculous. I know we have gone through just about
every issue known to digital photography in trying to rip these
things apart. What I see at the end of the day is what was there
all along.

1. Clean pixels: Yes, there may be spots on the images. Yes, there
may be residual this or that. But the pixels themselves are clean.
By that I mean distinct from each other with clear edges. For what
it’s worth, among other things that means they can be
manipulated (resampled) more easily, if I want.

2. The images for the most part look great straight out of the
camera at normal (100%) resolution. The colors ring true; the depth
of shadow detail looks about right; and the edges seem natural. The
images taken under extreme conditions – and by that I mean
for this camera, i.e. poorly lit, very high dynamic range
situations – reveal problems that seem very natural. As a
case in point, what film handles poor-lighting shots well by giving
us deep shadow detail and holding on to the hot spots and not
getting grainy (noisy)? None that I ever used. Why would this
sensor respond differently when it gets out of its stated range
(ISO 100-400)?

3. The images taken under highly controlled situations by some pros
(most likely) are outstanding. Printing one of those at A3+ on a
humble Epson 2100 is a lesson in patience that pays off.

There is not much else I can say at the moment. I am sure that
Phil’s review will reveal much and set off yet another
firestorm of discussion, which will be good, but only underscore my
point: These images are different and require a different
perspective to be understood and accepted.
 
The only conclusion I can truly draw after looking at all of these
images and comparing them to many others from Sony, Canon, Nikon,
and Fuji cameras is that a comparison of the Foveon images with the
others is extremely difficult and soon disintegrates from the
sublime into the ridiculous.
So maybe now that we have established all 4 of the DSLR (Sigma,
Nikon, Fuji, and Canon) are somewhat bunched together in
terms of performance, maybe we can have a nice price war to
settle the matter ?

I'd take any one of these 4 cameras, but I'd much rather pay
$1100 (for the SLR body only) rather than the current prices.

Research expenses aside, I wonder how much it would cost any
one of these companies to run their assemby lines a few
more months in order to make a few extra cameras for the rest
of us.

I am not greedy. I promise I won't be asking for a 15 million
pixel Bayer, or a 6 million Foveon... I just want one of these
existing DSLRs for $1100.

Maybe in 18 months ?

No doubt there will be new companies in the DSLR market soon
enough. I hope those companies cause trouble (i.e. competition)
as well. Maybe I am too optimistic ?
 
exactly what England is. I know it is part of the country
represented in the United Nations as the United Kingdom (or Great
Britain) which includes England, Scotland, Wales and Northern
Ireland. So why does England have its own World Cup team and
Olympic athletes. Is England like New York and Scotland like New
Jersey (God forbid) - they don't have their own World Cup teams.
Why if you call someone from Scotland "British" there is a good
chance you'll end up on your kiester? Why is it that "Braveheart"
was so long if nothing was really accomplished? Is it true that
the new world was actually started because everybody flunked
geography and had to leave? But most importantly, why does there
appear to be so much dust on the SD9 sensor when it is the only
digital camera to use a dust protector?
Why assume that dust has found its way behind the protector? Could simply be a maintenance issue when changing lenses. Could be we'll all be packing an ionization gun and an anti-static brush with this camera.

These are the most beautiful digital images I've seen from a DSLR to date; rich, natural color and detail in shadow areas without blowing highlights. Hope Ritz gets mine soon!

Michael OHara / WetPlanet / Honolulu
http://www.DiveSlates.com
 
Hello,

It's clear that the Foveon, at its best, is very** good.

See this crop here: Notice the 1 pixel wide spires (the 3 lower ones): The Foveon actually reproduces details down to the pixel: A Bayer would smear this out to several pixels. Yes, I have seen the problems with the Foveon (or is it the conversion software), but I think these problems eventually will be overcome in the same way as CMOS sensor problems where solved by Canon :-)



Another crop reveals that there is an Airplane in the sky! I have upsampled this 200% using bicubic sampling to make it easier to see: It looks like a Boeing 747: There's a dome at the top. See here:



Geir Ove
 
Hello,

It's clear that the Foveon, at its best, is very** good.

See this crop here: Notice the 1 pixel wide spires (the 3 lower
ones): The Foveon actually reproduces details down to the pixel: A
Bayer would smear this out to several pixels.
The Phil's D60 Sample picture of the roughly the same scene shows the same spike on top of the spires. While it is true that the SD9 should be better than a 3.4MP Bayer, it is not clearly able to pick up more detail than today's 6MP Bayers. So you seem to be "Bayer bashing" without evidence.
Yes, I have seen the
problems with the Foveon (or is it the conversion software), but I
think these problems eventually will be overcome in the same way as
CMOS sensor problems where solved by Canon :-)
The problems may be fixed in the long run, but as of right now, the Bayer cameras have a lot of advantages. We will have to see which of the problems can be fixed in software, hardware redesign, or are inherent in the technology. But it would seem to be a big gamble at this time to hope that they are all fixable in software.

Karl


Another crop reveals that there is an Airplane in the sky! I have
upsampled this 200% using bicubic sampling to make it easier to
see: It looks like a Boeing 747: There's a dome at the top. See
here:



Geir Ove
--
Karl
 
It's clear that the Foveon, at its best, is very** good.

See this crop here: Notice the 1 pixel wide spires (the 3 lower
ones): The Foveon actually reproduces details down to the pixel: A
Bayer would smear this out to several pixels.
Why amazing? The competing Bayer cameras DO have the several pixels to work with. You can see the same small spires on the D60 and D100 photos (no S2 bridge photo?) even if you resize them to down to SD-9 dimensions.

We are finally beginning to see directly comparable SD-9 vs other DSLR shots. So far the SD-9 has been as good or almost in some and noticeably worse in others. There have been several times where I though "Aha, here the SD-9 is better" but then realized I had swapped the order of my side-by-side comparisons and had choosen the S2 or D60 photo after all. Sort of accidental blind testing.

--
Erik
 
Erik Magnuson wrote:
Oops. I forgot to paste in part of my posting:
It's clear that the Foveon, at its best, is very** good.
There are two questions to ask about any camera:
  • is it good enough for my style of photography
  • is there a better camera in my price range
I don't think that there is any dispute that for certain classes of shots, the answer to the 1st question probably yes.

It's the second question that is more troublesome (this is a review site after all.) Viewed on their own the photos are good. But so far side-by-side, it's not quite measuring up.

--
Erik
 
The 70-200 EX HSM 2.8 lens is as sharp as Canon's L lens. That lens and the 50mm both come in with the same kinds on numbers as the Canon L lenses. This should make the lens issue a nonissue and see what this sensor means for our pictures.

I am interested in how big you can make salable prints?

Will it function like a 9mp sensor for enlargements or will it function like a 3mp sensor for enlargements?

Can we regularly sell 20x30's from this sensor as a pro print?

Is the camera itself good enough to shoot weddings with?

Will the Sigma 500 flash with it's AF assist work with the SD9 to help in low lighting?

Is the auto focus at least as good as the Canon D30/D60?

Will the camera work with other flash units like an Auto Thyristor flash?

Peter Gregg
 
It's the second question that is more troublesome (this is a review
site after all.) Viewed on their own the photos are good. But so
far side-by-side, it's not quite measuring up.

--
Erik
How so? There is more to a pleasing photograph than absolute resolution at the single pixel level, that is if image making is why you're involved with photography. I find this new technology to be quite compelling and at this price point especially sweet.

--
Michael OHara / WetPlanet / Honolulu
http://www.DiveSlates.com
 
It all comes down to supply and demand.

As long as people are still snapping digital SLRs up so fast that there continues to be no backlog of supply, then the manufactureres would be cutting their own economic throats if they lowered the prices, just to be "nice guys".

They make cameras, yes, but their first order of business is to make a profit. They are a business, not a photographers advocate and friend.

That said, I am also hoping that prices will start to come down, but if the demand stays high, then I would not bet on it happeneing anytime too soon.
The only conclusion I can truly draw after looking at all of these
images and comparing them to many others from Sony, Canon, Nikon,
and Fuji cameras is that a comparison of the Foveon images with the
others is extremely difficult and soon disintegrates from the
sublime into the ridiculous.
So maybe now that we have established all 4 of the DSLR (Sigma,
Nikon, Fuji, and Canon) are somewhat bunched together in
terms of performance, maybe we can have a nice price war to
settle the matter ?

I'd take any one of these 4 cameras, but I'd much rather pay
$1100 (for the SLR body only) rather than the current prices.

Research expenses aside, I wonder how much it would cost any
one of these companies to run their assemby lines a few
more months in order to make a few extra cameras for the rest
of us.

I am not greedy. I promise I won't be asking for a 15 million
pixel Bayer, or a 6 million Foveon... I just want one of these
existing DSLRs for $1100.

Maybe in 18 months ?

No doubt there will be new companies in the DSLR market soon
enough. I hope those companies cause trouble (i.e. competition)
as well. Maybe I am too optimistic ?
--
http://www.sushicam.com
 

Keyboard shortcuts

Back
Top