Seems to me that if they realy wanted to show this thing off, they'd have posted some high ISO images. I see one 400 and a couple of 200s. Everything else is 100.
I think they're going to have to do better to impress people.
They did this before, and posted some wonderful images. But I have seen wonderful images from all sorts of cameras. For this kinda money, the camera BETTER be capable of wonderful images. But they are selling on how much better their chip is, so show me...why not post image comparisons of Sigma vs Canon vs Nikon? Show 100% crops, etc...
I first heard of Foveon a few years ago in "Discovery" magazine; they sounded as though they were going to be a revolution in digital photography.
Consequently I put off replacing my Canon A-1 35mm, and used a PS for a few years, thinking that the CCD and CMOS chips would become obsolete very quicklly.
Time hasn't verified this; Canon, Nikon, Pentax, Sony et al are way ahead of Sigma/Foveon in sales. You can fool some of the people all of the time . . . . . . .
"There is nothing that some man cannot make a little worse and sell a little cheaper, and he who considers price only is that man’s lawful prey". John Ruskin 1819 - 1900
I also wonder if by the time Sigma gets this thing to market that Canon/Nikon/Etc will have CCD/CMOS sensors that are high enough resolution and with decent enough Bayer interpolation that the Foveon sensors will be outpaced pixel for pixel when the higher resolution sensors are shrunk down to Sigma sizes...
Foveon sensors have some interesting qualities, but they also have intrinsic problems that may not have simple solutions.
The way Foveon sensors work is to exploit one of the qualities of Silicon itself; that Silicon absorbes wavelengths of light at different rates according to depth within the Silicon wafer. While this seems like an ideal solution to the problem of seperating the constituent colors (RGB), it also complicates some of the practicality of the design.
For instance, how do you stack photosensor sites (essentially pixels) without the upper sites masking the lower sites?
All modern imaging sensors, including Foveon, use micro-lenses to concentrate the light falling on the chip into the photo-sites. How do you design the lenses to efficiently work with the different depths of the Foveon design?
The Silicon wafer absorbs light as the light penetrates the wafer. How do you compensate for the light loss encountered by the deeper layers versus the shallow layers?
None of these design issues have an easy solution, and a couple of the problems are greatly complicated by pixel density, effectively limiting ultimate resolution at a particular imager size (for now, anyway).
But seriously, I concur that micro lenses may challenge the stacked-pixel sensor designer where refraction is concerned. As for the penetrating of different light wavelengths, I suspect that it isn't nearly such an issue. The blue pixels will be on top and transmit everything else. The green pixels are right below and transmit the red. The red pixels are furthest down, of course, but my guess is that they're only 1/2 to 3/4 into the silicon.
Any near IR passes all the way through the bottom (a handy feature I used to inspect chips from the "backside" with an IR microscope, when I worked at Intel in the far distant past).
Anyway, I tend to wonder a lot more about the effect of those microlenses than anything else, and how Foveon manages to compensate for that.
Foveon sensors have some interesting qualities, but they also have intrinsic problems that may not have simple solutions.
The way Foveon sensors work is to exploit one of the qualities of Silicon itself; that Silicon absorbes wavelengths of light at different rates according to depth within the Silicon wafer. While this seems like an ideal solution to the problem of seperating the constituent colors (RGB), it also complicates some of the practicality of the design.
For instance, how do you stack photosensor sites (essentially pixels) without the upper sites masking the lower sites?
All modern imaging sensors, including Foveon, use micro-lenses to concentrate the light falling on the chip into the photo-sites. How do you design the lenses to efficiently work with the different depths of the Foveon design?
The Silicon wafer absorbs light as the light penetrates the wafer. How do you compensate for the light loss encountered by the deeper layers versus the shallow layers?
None of these design issues have an easy solution, and a couple of the problems are greatly complicated by pixel density, effectively limiting ultimate resolution at a particular imager size (for now, anyway).
Unfortunately for Sigma and Foveon, the glass might not even be half full yet.
But that's just my opinion - somewhat based on the fact that I've been in several well stocked camera stores and haven't seen one.
I'm open minded, but where are they?
"There is nothing that some man cannot make a little worse and sell a little cheaper, and he who considers price only is that man’s lawful prey". John Ruskin 1819 - 1900
I think everyone is missing the boat...who is going to sink money into SA mount glass??? Only those unfortunate enough to purchase SD9's and SD10's and I doubt most of THEM invested in good glass. This is a who cares side show of wasted R&D. Sigma should spend it's money on refining its reverse engineering of ETTL and 3D Matrix metering, putting HSM on EVERY lens, and perfecting their OS system. Either that or license the F or EOS mount for their body. Spending money on good SA glass is just like throwing out the window on the interstate.
I think everyone is missing the boat...who is going to sink money into SA mount glass??? Only those unfortunate enough to purchase SD9's and SD10's and I doubt most of THEM invested in good glass. This is a who cares side show of wasted R&D. Sigma should spend it's money on refining its reverse engineering of ETTL and 3D Matrix metering, putting HSM on EVERY lens, and perfecting their OS system. Either that or license the F or EOS mount for their body. Spending money on good SA glass is just like throwing out the window on the interstate.
... and yet, there are some accomplished Sigma shooters who would disagree:
Admittably, the Sigma system is not very popular in the US. In the UK and western Europe, the Sigma film and digital SLRs are very popular and well thought of.
As the SD14 is much closer to a general purpose dSLR, I think it has found a fairly large base of loyal followers, just not me.
Historically there have been many superior designs that did not become commercial sucesses or the dominant product... (especially in the electronics).
The concept is excellent, but the R&D budget is not on par with the competitors, who are happy selling a product that they have improved to the point of doing a great job.
I think the bigger problem of the Sigma dslr system is that you have to go to their lens system & forego prior investments in canon, nikon, etc. & also there are no third party lenses for a sigma.
I take lots of pictures--sometimes I make a photograph.
Mihihi, who would've guessed Canon would become a 3rd party lens manufacturer
“To consult the rules of composition before making a picture is a little like consulting the law of gravitation before going for a walk.”
― Edward Weston
Found an article with a detailed comparison between the SD14 and the 5D, including head-to-head sample images. Good article, and it demonstrates some of the major issues with Bayer interpolation.
Comments
I think they're going to have to do better to impress people.
www.ackersphotography.com
Consequently I put off replacing my Canon A-1 35mm, and used a PS for a few years, thinking that the CCD and CMOS chips would become obsolete very quicklly.
Time hasn't verified this; Canon, Nikon, Pentax, Sony et al are way ahead of Sigma/Foveon in sales. You can fool some of the people all of the time . . . . . . .
It would be an interesting test, to be sure
The way Foveon sensors work is to exploit one of the qualities of Silicon itself; that Silicon absorbes wavelengths of light at different rates according to depth within the Silicon wafer. While this seems like an ideal solution to the problem of seperating the constituent colors (RGB), it also complicates some of the practicality of the design.
For instance, how do you stack photosensor sites (essentially pixels) without the upper sites masking the lower sites?
All modern imaging sensors, including Foveon, use micro-lenses to concentrate the light falling on the chip into the photo-sites. How do you design the lenses to efficiently work with the different depths of the Foveon design?
The Silicon wafer absorbs light as the light penetrates the wafer. How do you compensate for the light loss encountered by the deeper layers versus the shallow layers?
None of these design issues have an easy solution, and a couple of the problems are greatly complicated by pixel density, effectively limiting ultimate resolution at a particular imager size (for now, anyway).
Moderator of the Cameras and Accessories forums
But seriously, I concur that micro lenses may challenge the stacked-pixel sensor designer where refraction is concerned. As for the penetrating of different light wavelengths, I suspect that it isn't nearly such an issue. The blue pixels will be on top and transmit everything else. The green pixels are right below and transmit the red. The red pixels are furthest down, of course, but my guess is that they're only 1/2 to 3/4 into the silicon.
Any near IR passes all the way through the bottom (a handy feature I used to inspect chips from the "backside" with an IR microscope, when I worked at Intel in the far distant past).
Anyway, I tend to wonder a lot more about the effect of those microlenses than anything else, and how Foveon manages to compensate for that.
But that's just my opinion - somewhat based on the fact that I've been in several well stocked camera stores and haven't seen one.
I'm open minded, but where are they?
... and yet, there are some accomplished Sigma shooters who would disagree:
http://www.wildaboutkent.co.uk/Equipment/my_choice.htm (Ian Andrews)
http://www.mophotos.co.uk/ (Mike Otley)
http://www.visionsofnature.co.uk/index.php (Steve Hooper-Lawrie)
Admittably, the Sigma system is not very popular in the US. In the UK and western Europe, the Sigma film and digital SLRs are very popular and well thought of.
As the SD14 is much closer to a general purpose dSLR, I think it has found a fairly large base of loyal followers, just not me.
Moderator of the Cameras and Accessories forums
The concept is excellent, but the R&D budget is not on par with the competitors, who are happy selling a product that they have improved to the point of doing a great job.
I think the bigger problem of the Sigma dslr system is that you have to go to their lens system & forego prior investments in canon, nikon, etc. & also there are no third party lenses for a sigma.
http://leeper.smugmug.com/
http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00Bs8p&tag=
Moderator of the Cameras and Accessories forums
A nice controlled shot can be taken by just about any camera now days.
Without a comparison of another camera taking the same shot, there lacks a point of reference for me.
The foveion chip is suppose to give better color renditions though.
Mihihi, who would've guessed Canon would become a 3rd party lens manufacturer
― Edward Weston
http://www.ddisoftware.com/sd14-5d/