Sigma SD-1 IQ vs 40MP MF

[mhecker*]

Sigma SD1 samples...

See http://www.sigma-sd.com/SD1/sample-photo/index.html

They look like crap compared to MF digital.

Luminosity about as good as a Canon 7D or Nikon D7000.
Color maybe a tad better.

I don't think I'll be trading in my 645D anytime soon.    

What do you fellows think?

[feppe]

We have a very different definition of "crap," (that's how the unedited OP claimed without the modifier of "compared to MF") as I'm seeing the grains of makeup on her eyelashes in the first picture. Quite impressive dynamic range in the Eiffel Tower shot as well. Some strange artifacting going on in the light starbursts, though - hard to say if it's compression, lens, or sensor, but I doubt it shows up in print. Not too impressed with the canyon shot's sharpness, but that seems to be the lens's fault as edges are soft.

None of them are crap by any stretch of the word. Whether it's worth 7500 EUR is questionable, though.

[Graeme Nattress]

Some really weird sharpening / edge artifacts in the Grand Canyon shot, and some strong chromatic aberrations too.

[uaiomex]

I don't think the sample pictures are crap either. I see incredible detail for aps-c and some of the cleaner shadows I've seen outside of medium format. Is it worth 97 one hundred bills?  I don't think so. More like likely 1/3 of that.
Eduardo

[nightfire]

These pictures aren't worth $9700 to me, and the disclaimer doesn't help either ("Some images may have been retouched by the photographer." - ok, I get it...)

[mhecker*]

I modified my 1st post.

I was in my mind comparing them to the MF digital files Sigma thinks they will compete with.

[telyt]

I'd have to do a side-by-side comparison with a MF camera under controlled conditions to decide if the image quality is crap or not.

[EricWHiss]

I looked first at the Grand Canyon image too and thought the image really lacked crispness and was trying to figure out the CA on the little bump of rock on the right side horizon that Graeme Nattress has already pointed out. Some of the images had great detail and convincing color and depth, but some just didn't.    The part about Foveon that is getting tiring is their claim that this equals a camera with 3x the pixels because theirs are not interpolated, but maybe they think by marketing this camera as a Medium Format equiv that no one will notice the lack of video.  There is something nice to some of the sample images in color depth but I think it will have a hard time competing in terms of detail.

[Christoph C. Feldhaim]

In their "Philosophy" part they claim

"After all, photography exists to reflect truth, and cameras exist to take photographs."

First this marketing speak is what it is - marketing speak, second their own way how to deal with truth is questionable.
Truth does also mean the x3 sensor does not have three times the megapixel information of its spatial resolution.
Truth also means true colors.
Truth does also mean a product (lens) matches its specs regularly and not by chance.
To me in this case the word "uncredible" appears more appropriate than the word "incredible" - they are simply trying to hide their own lies ...
/end rant

[donaldt]

if it is the sharpness you are after you dont need to buy a camera for it, you just need any program that does sharpness
on those portraits it is obviously over-sharpened, the first one is way too grainy
if one is to use any 40-50MP back, you should get the same sharpness with half the sharpening

and the pictures just seem flat to me

[Graeme Nattress]

The all look like they've gone through some noise reduction if you ask me.

[MichaelEzra]

Files obviously went through excessive post, but they do look impressive otherwise, although still overpriced.

[JeffKohn]

Definitely not impressed by the Canyonlands shot (couldn't find that Grand Canyon shot others were talking about ). Looks over-sharpened to me, unless that's just aliasing from the sensor.

Facial closeups are misleading IMHO, it's not difficult to render sharp eyelashes or skin pores at that kind of magnification.

I've always found it strange that the sample shots released with the press releases for various new camera models are so consistently mediocre. And I'm not just talking about Sigma, Nikon and Canon do it to (especially the latter).

It's also funny to me in the various chatter around the internet about this camera people seem to be assuming that it will have excellent color and DR, I guess because they assume a true RGB sensor must help in these regards. But in fact neither have been strengths of past foveon sensors; quite the contrary, color accuracy and noise (DR) have been weak points due to the inefficiency of the filters used and how little of the original light actually makes it to the bottom layers.

[donaldt]

you did point out something important
with the magnification those eye lashes arent really all that impressive
I just went back to some pictures I done a while back, a chest-up tight portrait shot with a 210mm HC lens
after some minor sharpening with the Phocus, it was sharp like that without those unnatural grains (or sharper than that)
the DR is also surprisingly weak for a high end DSLR, I think even my D3 has better DR at about half the price
so who is this camera targeting?
very wealthy enthusiasts who can afford Leica S2 and want something less than impressive?
or true pro who knows its a inferior product charging too much?

Definitely not impressed by the Canyonlands shot (couldn't find that Grand Canyon shot others were talking about ). Looks over-sharpened to me, unless that's just aliasing from the sensor.

Facial closeups are misleading IMHO, it's not difficult to render sharp eyelashes or skin pores at that kind of magnification.

I've always found it strange that the sample shots released with the press releases for various new camera models are so consistently mediocre. And I'm not just talking about Sigma, Nikon and Canon do it to (especially the latter).

It's also funny to me in the various chatter around the internet about this camera people seem to be assuming that it will have excellent color and DR, I guess because they assume a true RGB sensor must help in these regards. But in fact neither have been strengths of past foveon sensors; quite the contrary, color accuracy and noise (DR) have been weak points due to the inefficiency of the filters used and how little of the original light actually makes it to the bottom layers.

[RobertJ]

Sigma = Lenses with chromatic abberation and sensors that produce images that look like over-processed, worn-out jpegs.

[ErikKaffehr]

Hi,

I'd assume that MF quality from a DX size sensor is pure nonsense. There is some advantage from the Foveon design, due to possible elimination of OLP (Optical Low Pass) filtering. Bart van der Wolf, who seems very knowledgeable in the area seems to estimate the improvement in resolution to about 6%.

On the other hand, a small sensor is a small sensor. That mean that it needs to have high resolution. MTF on lenses is essentially inversely proportional to resolution so a sensor four times larger would yield four times the MTF (like 80% instead of 20%) no technology in the world can compensate for that. It may be possible to build lenses that are diffraction limited a f/2, such a lens would deliver very high MTF even from a small format, at large apertures. I don't think present day Sigma lenses are even near.

Best regards
Erik

Sigma SD1 samples...

See http://www.sigma-sd.com/SD1/sample-photo/index.html

They look like crap compared to MF digital.

Luminosity about as good as a Canon 7D or Nikon D7000.
Color maybe a tad better.

I don't think I'll be trading in my 645D anytime soon.    

What do you fellows think?

[Bart_van_der_Wolf]

Bart van der Wolf, who seems very knowledgeable in the area seems to estimate the improvement in resolution to about 6%.

Hi Erik,

That number is used a bit out of it's original context. Many moons ago, I've done a simple test which determined the effect of demosaicing on Luminance resolution. With the demosaicing algorithm used, there was indeed only a loss of resolution of 6.4%. In addition there are specific color combinations that pose worst case scenarios for  demosaicing of Bayer CFAs, compared to situations where there is a luminance difference between sensels. While significant, the demosaicing loss is not the end of the world. Slight defocus, or some diffraction blur, or camera shake, or heaven forbid a combination, does more damage.

A much larger difference is caused by the use of an AA-filter on the majority of 35mm or cropped Bayer CFA sensor array cameras. However, on most MF backs that is not a factor of concern. In addition, as you said, there are differences in MTF due to magnification factor and lenses.

Furthermore, having a lot more sensels, i.e. independent spatially discrete samples, opens up a lot of resolution. Having 40 M Luminance (in a CFA arrangement) samples is something more useful than 15 M RGB samples for the same field of view, even if the chromatic resolution is somewhat limited compared to the luminance resolution.

Cheers,
Bart

[ndevlin]

My 645D is waiting in the schoolyard...let the rumble begin.

Web samples aren't worth much, but the detail on the dam shots looks quite impressive for APS-C. Trying them out side-by-side will be very interesting. 

- N.