Camera Labs

Wow, thanks for the very detailed reply. You have a nice lens collection.

Yes I promptly forgot about the ability to use third party lenses via adaptors, sorry about that, but you will lose autofocus and this is a limitation most people don't want to be stuck with. I get why you've gone down the route that you have but I think you know what I mean when I say that what you've chosen to do requires a bit more commitment than most people would be willing to give, me included.

_________________
Graham
Flickr

+1 to nice lens collection DSG, but I agree in that fast autofocus is one of the main selling points for SLR photography for most people. It was certainly why I changed from rangefinders to SLRs.

_________________
Nikon D5000 and D300 with 12-24mm, 35mm f1.8, 50mm f1.4, 60mm f2.8 Macro, 18-200mm, 17-55mm f2.8 (all Nikon)

Manual lenses are'nt for everyone I agree but I much prefer them to the Sigma AF kit lenses I got with my SD10.
I find I can focus quicker manually, and to exactly where I want, instead of wiggling the lens around whilst its hunting for something to lock onto.
BTW, there is another AF lens option if your not keen on Sigma lenses or want to be able to use at least some of you existing lenses on the SD1 as Canon EF lenses can be converted to SA mount whilst retaining all functions except IS, for those lenses that have it.
This is because EF and SA share the same communication protocalls between body and lens and they have the same 44mm registration distance too. The EF electrical connector block is slightly different to the SA version, and its at a slightly different angle around the mount, but of course if your swapping whole mounts that does'nt matter. There is a ribbon cable to desolder from the connection block on the lens...Swap the mounting plates then solder this cable onto the SA connection block in the same pin configuration, screw the block onto the SA plate and your in business.

Wow, okay. I knew the mounting to flange to sensor distance was the same with the EF mount but I didn't realise that you maintained most electronic functions. Pretty cool.

_________________
Graham
Flickr

Wow... I can still get decent shots out of a 3MP cam, 46 may I say, is absurd unless you need to make an enormous print on the side of a building. Plus, I can only imagine the super high price tag! I'll stick to a Nikon DSLR, thank you!

_________________
-Evan

Gear: 7 Nikon Nikkor AI-S and AF-S lenses, SB-700 flash, Nikon D7000, Nikon FM, variety of accessories

"There are no rules for good photographs, there are only good photographs."
- Ansel Adams

Congratulations on missing the point! This is what I'd consider a true 15MP sensor, unlike almost every other camera on the market which gets their count by using processing tricks. Best way to look at it is this should be pretty close to an ideal 15MP in sharpness and colour position accuracy, instead of the blurry output current sensors give at 100% unless you give it a big dose of processing.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

EvanK, the Foveon sensors can detect all three colours for every pixel, so each of the 15 million pixels has a sensor layer for each colour.

Every other sensor, has 1/2 green pixels, 1/4 blue and 1/4 red and they user Bayer interpolation so that each pixel on the sensor maps out onto a pixel in the final image that can display the full range of colour. Each pixel is effectively taking readings from all the surrounding pixels. Foveon sensors don't, each pixel is self sufficient you could say. This means that an image viewed at 100% should look perfectly sharp. Have a look at some sample images, they are stunning.

Sigma's 46mp claim is taken by multiplying the true 15mp count by 3, one for each colour channel. I think this is odd because it is using the same principle that bayer sensors use and the Foveon sensor claims to be superior for avoiding the whole interpolation thing. I think this is quite ironic but the fact is that, like Popo says, this is the only sensor that will actually give you 15 million pixels.

_________________
Graham
Flickr

A 10MP CMOS/CCD camera has 5 million sensor sites that captures ONLY green light, 2.5 million that capture ONLY red and 2.5 million that capture ONLY blue.

Following this logic that everyone seems to accept, what other choice does Sigma have?

They have 15 million sensor sites capturing ONLY green light
They have 15 million sensor sites capturing ONLY red light
They have 15 million sensor sites capturing ONLY blue light

Using the EXACT same method as everyone else, SIGMA SD1 is EXACTLY a 45 MP camera.

Just like with CCD and CMOS cameras, you can use a SETTING on the camera where the OUTPUT (I.e. image file) is X number of mega-pixels..say 12, 8 and 4.

When my Nikon D300 outputs 4MP, does it become a 4MP Camera? No, of course not. Everyone would say that it's silly to say so.

When the SD1 outputs an image-file that is 15 MP, does that make it a 15MP Camera?

I am personally at a complete loss at trying to understand why this has caused so much controversy...lol. Sigma is doing EXACTLY the same thing as everyone else...!

Cheers

_________________
Nikon stuff and other stuff, but not nearly as much stuff as I really want!

Sorry, LahLahSr, popo

I'm new in this Foveon X3 sensor tinging.
i really confess like LahLahSr say that it really 45mp.
but popo say it not 45mp, it just 15mp with ad say 45mp.
because, every is a pro here and i never see or use real simga dslr before.
can anyone share some light.

_________________
Canon EOS 7D + Red Canon IXUS 300HS.
Canon EF 17-40mm f/4L,Canon EF 100mm F/2.8L IS USM.
Sigma 70-200mm F/2.8 APO EX DG OS HSM, Sigma 50mm F/1.4 HSM
Canon Speedlite 480EX II + Tripod+ B+W MRC (2x77+67), hoya HD UV (77) B+W MRC 77mm(CPL slim+10 ND).

I see, thanks for clearing it up guys. Sorry, I've never realized that.

None the less, for making normal sized prints I would never need that, I guess it's geared more at photographers looking at making huge poster prints.

_________________
-Evan

Gear: 7 Nikon Nikkor AI-S and AF-S lenses, SB-700 flash, Nikon D7000, Nikon FM, variety of accessories

"There are no rules for good photographs, there are only good photographs."
- Ansel Adams

KSK, a 15mp bayer sensor has 7.5mil green photosites, 3.25mil blue photosites and 3.25mil red photosites.

A 15mp foveon sensor has 15mil photosites that can each detect green, blue and red light. This is because each photosite has 3 photodiodes that each detect one colour. The 46mp figure is achieved counting the total number of photodiodes.

LahLahSr, I do not think that Foveon is wrong to claim that their cameras have X number of pixels. However, I'm not sure why you say that they have 15 million sensor sites that detect ONLY.... etc. Isn't the main distinction between a Foveon camera and a Bayer sensor the fact that the Foveon has 15 million sites that detect red, green and blue?

_________________
Graham
Flickr

@grahamnp
Thank grahamnp for your kind explained.
because, I never see any real life sigma DSRL,before.
but there are ton of sigma lens (that a good part).

_________________
Canon EOS 7D + Red Canon IXUS 300HS.
Canon EF 17-40mm f/4L,Canon EF 100mm F/2.8L IS USM.
Sigma 70-200mm F/2.8 APO EX DG OS HSM, Sigma 50mm F/1.4 HSM
Canon Speedlite 480EX II + Tripod+ B+W MRC (2x77+67), hoya HD UV (77) B+W MRC 77mm(CPL slim+10 ND).

Hi grahamnp,

As I understand the FOVEON sensor, it has three layers. Each layer has it's own sensor sites capturing it's distinct color.

I guess that when these discussions happen - and they do in many photo forums - it's because of the difference between the "resolution" at the capture of the light vs. the resolution of the image-file.

While the FOVEON has 3X15 million points of capture, it produces 15MP worth of pixels each with their own RGB information, without any algorithmic interpolation/processing to do so.

A CMS/CCD may produce 15MP pixels in their image files, but the number of capture points is significantly smaller - thus requiring processing magic to arrive at the same result.

It is probably fair to discuss if the most common definition behind what has been the industry-standard for classifying/describing the "resolution" of cameras, should be re-defined with the advent of the FOVEON sensor.
Perhaps it's time now to define it as R= X mp, G = Y mp and B = Z mp?

Technical enthusiasts like us may be interested in the underlying technology behind it all, but I suppose we should really all focus more on what we like and prefer in terms of the actual output..lol.

Cheers

_________________
Nikon stuff and other stuff, but not nearly as much stuff as I really want!

Thanks for clearing that up LahLahSr

_________________
Graham
Flickr

LahLahSr wrote:
"Following this logic that everyone seems to accept, what other choice does Sigma have?

They have 15 million sensor sites capturing ONLY green light
They have 15 million sensor sites capturing ONLY red light
They have 15 million sensor sites capturing ONLY blue light"

Its not quite as simple as that because there is actually some overlap between the layers...In other words some of the blue photons make it through to the green layer and red layer and of the some of the green photons make it through to the red layer...This problem causes a real headache for Foveons software developers as it has to try and subtract this unwanted mixed colour data from the wanted pure stuff...The colour response has been changed and improved with each subsequent SD model. The SD9 was known was giving very unatural looking overly blue skys. Some actually like that effect but Sigma did'nt and so the skys were much more natural looking with the SD10. The colour response of the SD14 is similar to the SD10 but not identical and the SD15 is much better at capturing natural reds than any of the previous models.
The SD1 should build on the experience gained with their best model to date, the SD15, with improvments unique to the new larger 1.5x crop sensor and so it should have the most accurate colour response yet.