Sigma SD10 VS Kodak 14N Image quality?

Started Aug 9, 2005 | Discussions
macaroon New Member • Posts: 17
Re: Are you kidding man? Wohow...

Adam Kozlowski wrote:

SD10 is NOT 10mp. It's 3mp.

That happens to output 13.72MP images.

halidedream Forum Member • Posts: 66
Re: Are you kidding man? Wohow...

I am not sure why I am even bothering to reply, but the silly comment "That happens to output 13.72MP images" actually got to me. There is a reason that no major stock agency would accept a sigma file. There is also a reason that you don't see any serious photographer with one. I truly hope that foveon will come out with a 6 or 8 mp sensor. It would rival the original 1ds, the kodak and the nikon d2x at lower iso's, but they haven't. Strange people that suddenly want/need their camera to be the best are missing the point, but they did in the film days too. No, even a leica did not compare at 16x20 to medium or large format. The sigma is fine and comparable up to 8x10 for most images, after that there really isn't any competition. Unless you don't have an eye for photography and you only see size rather than detail and substance. There are a lot of people like that and I am sorry that they miss so much of the beauty of an image.

all the best,
Gregory

Petros35 Senior Member • Posts: 1,630
Re: Are you kidding man? Wohow...

Suzie hasn't learned that it doesn't "output" 13.72 MP files. You have to do it later on with special tools. It can be done, however, which is why the Sigma can easily be used for magazine work. 2xA4 layout? No problem.

(Hey, we can't all be hanging huge stuff in galleries, can we? )

-- hide signature --

Peter Argyropoulos
http://petros.pl

macaroon New Member • Posts: 17
Re: Are you kidding man? Wohow...

Petros35 wrote:

Suzie hasn't learned that it doesn't "output" 13.72 MP files. You
have to do it later on with special tools.

False, SPP outputs 13.72MP interpolated image files as built-in option, 4536 x 3024.

macaroon New Member • Posts: 17
Re: Are you kidding man? Wohow...

halidedream wrote:

I am not sure why I am even bothering to reply, but the silly
comment "That happens to output 13.72MP images" actually got to me.

Why, I wonder? Can't other cameras interpolate to 14MP too?

There is more optical R+G+B information gathered by the Foveon sensor than the Kodak's sensor.

There is a reason that no major stock agency would accept a sigma
file. There is also a reason that you don't see any serious
photographer with one.

I agree that most people simply dismiss it based on group think alone.

I truly hope that foveon will come out with
a 6 or 8 mp sensor.

They do make a smaller version of the sensor.

It would rival the original 1ds, the kodak and
the nikon d2x at lower iso's, but they haven't. Strange people that
suddenly want/need their camera to be the best are missing the
point, but they did in the film days too. No, even a leica did not
compare at 16x20 to medium or large format. The sigma is fine and
comparable up to 8x10 for most images, after that there really
isn't any competition. Unless you don't have an eye for photography
and you only see size rather than detail and substance. There are a
lot of people like that and I am sorry that they miss so much of
the beauty of an image.

all the best,
Gregory

macaroon New Member • Posts: 17
Re: For your information..

Quentin (UK) wrote:

Can he / she / it be banned?

You would like that. It is easier than expanding your mind to understand that 3.43-full-color-MPs can rival or exceed 13.5-monochrome MPs.

bigbackyard New Member • Posts: 8
Ok, here is an SD9 sample @ 14MP

Gabriele Sartori wrote:

Let's do this, find for me the best SIgma photo you can find,
upscale to 13.5MP

http://www.pbase.com/sd9_14mp/image/47697434/original

I'll do the same with the Kodak

Your turn.

halidedream Forum Member • Posts: 66
Re: Are you kidding man? Wohow...

"Independent comparisons of quality tend to rate the Foveon X3 sensor as equivalent to a Bayer sensor of double the photosite count.

... advantage is largely overcome by the math required to reconstruct color information from the Foveon sensor, which results in a Foveon sensor with very large photosites often being unable to equal the low light performance of more conventional sensors with half the photosite area.

...the question of whether to count each stack of three photosensors as a pixel, versus counting each individual single-color photosensor as a pixel as is done in Bayer-mosaic sensors, has been a point of controversy for Foveon X3 sensors and for the specifications of cameras that use them. For example, the Sigma SD10, which produces a native RAW file size of 3.4 Million RGB pixels, is advertised as a 10.2 MP camera, sometimes with the clarification 3.4 MP Red + 3.4 MP Green + 3.4 MP Blue; an 8 MP Bayer-mosaic camera would similarly be clarified to be 2 MP Red + 4 MP Green + 2 MP Blue"(source wikipedia.org)

Marketing is a wonderful tool. If you feel the sigma resolution is enough for your needs, then it is. I feel that it is a wonderful 6mp camera and a great look toward what is possible, but not applicable for my needs. I too often get a request for a large image. I was just asked to quote a bid for some product shots and they were very wary of even 14 mpix image quality as being insufficient for trade show posters. For some prints for a studio show that I am having in October, I am printing images at 24x36 of dancers painted with henna. I need the detail to show and the Kodak does a great job of that. But even it is being pushed to its limits. To each their need, to each their solution. For me the sigma is not the optimal quality, but I hope that foveon can get their act together and get a better sensor out with one more generation.

all the best,
Gregory

bigbackyard New Member • Posts: 8
Re: Are you kidding man? Wohow...

halidedream wrote:

"Independent comparisons of quality tend to rate the Foveon X3
sensor as equivalent to a Bayer sensor of double the photosite
count.

That is true - based exclusively on imaging pure B&W resolution targets, where 3 layer full color photosites yield little to no advantage.

This is from a full color resolution test, also from an "independent review." Would you rate the SD9 equal to the 6MP 10D?

Canon10D:

Sigma SD9:

... advantage is largely overcome by the math required to
reconstruct color information from the Foveon sensor, which results
in a Foveon sensor with very large photosites often being unable to
equal the low light performance of more conventional sensors with
half the photosite area.

Larger pixels/photosites is a distinct advantage of layering.

You might think the Full Frame 14n has more sensor area, but the SD9 actually has 130% of the 14n's sensing area, all concentrated under the sharpest parts of the lens.

...the question of whether to count each stack of three
photosensors as a pixel, versus counting each individual
single-color photosensor as a pixel as is done in Bayer-mosaic
sensors, has been a point of controversy for Foveon X3 sensors and
for the specifications of cameras that use them. For example, the
Sigma SD10, which produces a native RAW file size of 3.4 Million
RGB pixels, is advertised as a 10.2 MP camera, sometimes with the
clarification 3.4 MP Red + 3.4 MP Green + 3.4 MP Blue; an 8 MP
Bayer-mosaic camera would similarly be clarified to be 2 MP Red + 4
MP Green + 2 MP Blue"(source wikipedia.org)

Agreed. And that is the most correct answer. Obviously a camera's color performance is only as good as the individual, overlayed RGB exposures, not all three RGB exposures added together--that measures greyscale performance. Another excellent way to standardize advertising would be to specify whether the MPs are monochrome or color:

Kodak 14n - 13.5 (monochrome) MPs
Sigma SD9 - 3.43 (color) MPs

Marketing is a wonderful tool. If you feel the sigma resolution is
enough for your needs, then it is. I feel that it is a wonderful
6mp camera and a great look toward what is possible, but not
applicable for my needs. I too often get a request for a large
image.

You might be surprised, the SD9/10 enlarges very well with all the full color data embedded within the picture frame. It captures more full color data than the 14n/c, for example.

I was just asked to quote a bid for some product shots and
they were very wary of even 14 mpix image quality as being
insufficient for trade show posters. For some prints for a studio
show that I am having in October, I am printing images at 24x36 of
dancers painted with henna. I need the detail to show and the Kodak
does a great job of that. But even it is being pushed to its
limits. To each their need, to each their solution. For me the
sigma is not the optimal quality, but I hope that foveon can get
their act together and get a better sensor out with one more
generation.

I hope Foveon can keep going too. I kind of doubt it, the concepts are complicated to get accross and it will never be in a Nikon or Canon, so it's out for pro agencies that really do need to make safe buying decisions.

andrea buso Veteran Member • Posts: 3,319
Re: Are you kidding man? Wohow...

Although the sd9 color chart shows an outstanding resolution, the eos 10d cart it is shot completely out of focus.

I do not think that the 10d chart look so blurred simply because it is a color chart and that are color lines represented by a 10d, also the edges of the Greg Mcbeth chart are completely blurred.

Maybe another try on the 10d, without gaussian blurr. Or, we can still blame the notorius back focus of the 10d.

-- hide signature --
macaroon New Member • Posts: 17
I doubt Uwe gaussian blurred it

andrea buso wrote:

Maybe another try on the 10d, without gaussian blurr. Or, we can
still blame the notorius back focus of the 10d.

http://www.outbackphoto.com/artofraw/raw_05/essay.html

Gabriele Sartori Veteran Member • Posts: 4,433
This is a joke right?

bigbackyard wrote:

Gabriele Sartori wrote:

Let's do this, find for me the best SIgma photo you can find,
upscale to 13.5MP

http://www.pbase.com/sd9_14mp/image/47697434/original

I'll do the same with the Kodak

Your turn.

I promise not to write on this thread anymore but I understand that you guys are very weak with math. (and beside only a fool would use this photo for resolution determination)

Don't reply, I will not.

-- hide signature --

Regards
Gabriele
California, CA
--------------------
Equipment list in profile

 Gabriele Sartori's gear list:Gabriele Sartori's gear list
Leica D-Lux (Typ 109) Nikon D800E Leica M Typ 240 Panasonic Lumix DMC-GH3 Nikon 1 J1 +3 more
bigbackyard New Member • Posts: 8
Another full color resolution test

Here's another test against the 10D using a full color res chart.

But the most interesting thing isn't the SD9 displaying 130% of the 6MP Bayer's red & blue lph (ROT = you need double the MPs to increase by 25%). The most interesting thing is that the paper is flat, yet the Bayer blurs equally sharp lines on that flat plane differently. People spend thousands to buy perfect lenses, but then they accept a sensor that blurs colors very differently on the same focal plane. That this fundametal optical flaw is routinely swept under the rug is astonishing.

10D:

SD9:

macaroon New Member • Posts: 17
I count 4500 x 3000 pixels, what do you see?

Gabriele Sartori wrote:

http://www.pbase.com/sd9_14mp/image/47697434/original

I promise not to write on this thread anymore but I understand that
you guys are very weak with math. (and beside only a fool would use
this photo for resolution determination)

Don't reply, I will not.

Warren Sarle Veteran Member • Posts: 3,676
The photo has been switched

The photo at that link earlier was about 3MP, as I recall. It was a nice photo, aside from being small.

Anyhow, you can't seriously think this photo is anywhere close to the image quality of a 14n, can you?

Adam Kozlowski Senior Member • Posts: 1,108
That's a cheap trick...

...to make you Sigma users happy. Better use Shortcut Photozoom. Ok, let me rephrase - Sigma has one of the best lines of lenses out there - some are unique, some are better than Canon and Nikon counterparts - it's really awesome. I love it. I'm gonna use it more soon. However the SD9 and SD10 is nothing more than an experiment, which i'm surprised Sigma decided for, cuz i might understand Nikon or Canon do it, cuz they can afford such a mistake.

Actually D1X at 10mp is better than the interpolated Sigma, cuz it will interpolate only in one direction, which then can be deblurred easily in that direction and you will have a great 10mp file of uncomparable quality to the Sigma files at 10mp or wherever close.

-- hide signature --

Adam Kozlowski Photography - http://www.spooky.pl
Creative and fun outlet - http://spooky79.deviantart.com

Adam Kozlowski Senior Member • Posts: 1,108
Exactly my thought

If Sigma is enough for your needs - i'm happy for you I almost bought a Kodak, but i think 1Ds (old) is a better compromise for my needs. And now the 5D? Sweet. But each of us has different needs and expectations, so if you're happy with what you have - then so be it.
--
Adam Kozlowski Photography - http://www.spooky.pl
Creative and fun outlet - http://spooky79.deviantart.com

Adam Kozlowski Senior Member • Posts: 1,108
This test is weird.

Put that 10D RAW file in PS CS2 and you will arrive with very different results i believe. Plus, to be honest - this is a Sigma oriented test It's not very realistic. And, also, try the same test with Kodak, i'm curious of the results.
--
Adam Kozlowski Photography - http://www.spooky.pl
Creative and fun outlet - http://spooky79.deviantart.com

Quentin (UK)
Quentin (UK) Senior Member • Posts: 2,437
Re: For your information..

No, I just prefer to deal with someone who does not parade around using multiple online personalities in an attempt to make it seem like their whacky ideas have more support than they have.

Macaroon? Fuitcake more like...

macaroon wrote:

Quentin (UK) wrote:

Can he / she / it be banned?

You would like that. It is easier than expanding your mind to
understand that 3.43-full-color-MPs can rival or exceed
13.5-monochrome MPs.

-- hide signature --
halidedream Forum Member • Posts: 66
Re: I doubt Uwe gaussian blurred it

Hello Macaroon,

Please take a look at Andrea's website. Not meaning to discredit Uwe, but I I think you will find he knows a little more about photography than Uwe does. He was also poking fun at the horrible test, because anyone who shoots for a living, or even someone who has used these cameras knows that test is flawed. It is funny, because I really wanted the Foveon chip to be an amazing breakthrough. In one sense it was showing that 3 mpix's could rival a 6mpix bayer filter cmos. The problem was that the world moved on and foveon didn't. If you get a chance, go to a camera store that carries both systems and try them out for yourself. I think that you will find, as most of us who have, that the Kodak is really, really good in the resolution and color department. (image wise) it is in the league of the D2x and the 1DSII, the sigma is not. The web is a wonderful place for information, and all of those graphs are great fun, but nothing compares to using the camera in the way you shoot and then looking at the results for yourself. What has been really special about the Kodak forum is that many people, despite their camera choices, were able to come here and get good information without the chest pounding need to prove that brand x was better than brand y. Many of these people shoot for a living and trust me, they test their equipment more than any reviewer, because their livlihood, reputation and passion all rely on it.

all the best,
Gregory

Keyboard shortcuts:
FForum MMy threads