Sigma SD14 Tri-color pixels Debuts Photokina 2006

If one mentions photo sites that record RGB channels in separate layers, the smart bet would be the coming or the next 1D flagship. Most of us remember the promise of Foveon brilliantly innovative technology in the SD 9 and SD 10 cameras. However, they are not significant market fighters in the arena of DSLR's.

As of now, Fuji is hardly mentioned but Sony is. Why, because it is alive, growing and eats into the Canon market share in overall digital camera sales.

So now Sigma, the great lens maker, presents the SD14. It appears to share the same essential body design as the SD10 from the incomplete images Sigma has released to build up interest and demand. Since the nomenclature of previous bodies referred to a multiple of the photo sites by 3 (for the RGB layers in each cell), the SD14 likely has 14/3 photo sites, or 4.66 MP photo sites. However, the previous SD series was reported to have effective resolution double that of that of other cameras with the same number of photosets, but where each photo site is a different color. So this camera likely is not full frame (if it was, it would be trumpeted).

What purpose the camera serves depends on its focus speed and frame rate.

If it works fast and furious, it’s a superb sports camera priced perhaps about $1800 and selling for say $1,200.

So which cameras will it compete against?

I think it may turn out to be an excellent portrait, medical, macro and product camera where the use of the camera is one purpose and with perhaps one lens. Or else it may be a niche camera for those who just will love the color and dynamic range like those devoted to the Leica back, the Contax ND and the Fuji.

Asher

Apparently, from what I've read, 2640x1760 pixels = 4.65 Mp. Times 3 colors, makes 13.9 M RGB samples, with R/G/B being co-located. Lens mount seems to be a Sigma mount.

Formal introduction apparently is planned on Sept. 26th at the Photokina 2006 (http://www.sigma-sd14.com/).

If it's anything like its predecessors, an issue would be the lack of an anti-aliasing filter, and its noise at higher ISO settings (one can only collect so many photons in the space of a single pixel divided by 3). At least the aliasing artifacts were not of the colored kind.

In the preceding models there was/is also a notable lack of color accuracy, e.g. resulting in yellow skin issues. It was caused by the absence of a filter needed to match the sensor array's response to the eye sensitivity curves. That was both a cost saving, as well as a sensitivity preserving choice.

Things like out of camera JPEGs were important omissions for other (potential) users.

One can only hope that the major technical issues have been addressed, because competition is good for progress in technical innovation.

Bart

Bart_van_der_Wolf said:

.......In the preceding models there was/is also a notable lack of color accuracy, e.g. resulting in yellow skin issues. It was caused by the absence of a filter needed to match the sensor array's response to the eye sensitivity curves. That was both a cost saving, as well as a sensitivity preserving choice.

Things like out of camera JPEGs were important omissions for other (potential) users....
Bart

Click to expand...

Hi Bart!

This is exciting. I want people to expect more from Nikon, Sony, Panasonic and Canon! The more challenge from below the better.

I'd not be too concenred about the color accuracy as one can make custom profiles for your camera to be used in RAW processing.

For professional work, like product and portraits, science or medicine, I see no issues that a good workflow wouldn't not attend to in the first two weeks maximum.

Think about the plus side, we will be giving work to color consultants. Not much, but a lot of us can benefit from calibrated and profiled equipment.

I doubt that for the uses I mentioned, that noise is likely to be an important issue. How did this affect the SD10?

Asher

Asher Kelman said:

I doubt that for the uses I mentioned, that noise is likely to be an important issue. How did this affect the SD10?

Click to expand...

First I want to remind everybody that most cameras, in the hands of skilled people, can produce stunning quality, in both a technical and creative sense.

I have read and participated in many discussions between those pro and contra the Foveon sensor results from the SD-9 and 10. It is exiting technology, but with some debatable implementation choices. Personally I didn't like the misleading spin from its inventor, not only a great mind but also a gifted speacher (although it appeared he turned to the dark side of integrity, guess it was a money thing).

As far as the noise is concerned, comments from users indicated that ISO 400 produced too much noise to be of use. One can check e.g. the relevant sections of Pbase for examples, although while keeping in mind that some (or most) of those images were made by unskilled photographers.

However, since results from the past don't necessarily predict the future, I'm keeping an open mind to what will be presented on the Photokina. Whatever stirs up competition ...

Bart

Bart_van_der_Wolf said:

As far as the noise is concerned, comments from users indicated that ISO 400 produced too much noise to be of use. One can check e.g. the relevant sections of Pbase for examples, although while keeping in mind that some (or most) of those images were made by unskilled photographers.

However, since results from the past don't necessarily predict the future, I'm keeping an open mind to what will be presented on the Photokina. Whatever stirs up competition ...

Bart

Click to expand...

Good points, Bart, (or bad if one includes the "Dark Side", of which I know nothing in this case).

Remember I said, " it may turn out to be an excellent portrait, medical, macro and product camera where the use of the camera is one purpose and with perhaps one lens."

In that case it could be used at ISO 100, and how much issue from noise was there with the previous SD series?

Asher

Asher Kelman said:

Remember I said, " it may turn out to be an excellent portrait, medical, macro and product camera where the use of the camera is one purpose and with perhaps one lens."

In that case it could be used at ISO 100, and how much issue from noise was there with the previous SD series?

Click to expand...

Yes at ISO 100 it seems to be okay if aliasing could be avoided. For higher ISOs it's hard to say, there seem to be very few images available other than ISO 100 ones. Also remember that those cameras only produce Raw images. The accompanying software was quite capable, so maybe noise reduction was part of it.

I never used the camera, but in his evaluation of the improved SD10, Phil Askey mentioned at: http://www.dpreview.com/reviews/sigmasd10/page19.asp
that he "did have a couple of niggles, increasing sensitivity has the side effect of producing softer images and can lead to changes in saturation and increased levels of color 'bleed' (which can appear as a hue shift at very high sensitvities), additionally long exposures for night shots were pretty disappointing".

To me it sounds that the Raw converter software had some role in all this, and it is strange that higher ISO settings caused desaturation. But then, that was March 2004 so we'll have to see what happened between then and now.

Bart

although i really no longer shoot DSLR (D-MF is addicting...) , the sigma announcement got my attention...i had the SD-9 and i thought it was a great camera...8x10 from the SD-9 blew away comparable canons and nikons of the time..at low asa, raw only...the crispness of the files is something that canon and nikon still don't have...no sharpening can replace a file that is just crisp from the start...
color was fine...i think that all DSLRs just don't have great color..it can be fixed to look good, but that is one of the advantages of a D-MF back...great, natural color to start with...
i am really hoping the SD14 will be a little more successful then the SD9&10...the fovenon imager is definitely a promise for the future....a 45x60mm 3 layer 16mpix imager would be amazing (yes i am dreaming, but i think that this is the way the industry will go..canon has been trying to build a 3layer chip for years)....
but we all know the power of marketing...the 14sensors or 5mpix or whatever will not be enough for people who are asking for 7mpix cellphones...that there is so much more to it then just mpix does not count...i am just wondering what people do with all these huge prints? do i really want all that detail blown up that big? or do i want better quality detail?
anyway...i am excited to see someone else (other then the big boys) come out with real innovation...go sigma!