Canon 10D versus Sigma 10 Megapixel Camera

[msvirick]

I see in the Sams Club a Sigma 10 Megapixel camera with 3 lenses for about $1400. The CCD is fovian, supposed to be very color correct.
How does the output and quaility compare with Canon 10D?
Any experience?

[CyberDyneSystems]

The Sigma is really 3.4 megapixels.

It has 3.3 (or is it 3.4) MP X 3 colors,. so there are about 10 million sensors points.. but the file it produces is the same size as a 3.4 MP camera would make.

It has it's own RAW file and won't shoot jpegs.

Most importantly,. it can't use Canon lenses....

Look up the "Foveon" sensor,. and it'll explain the 3.4 Vs. 10 MP thing.

In a nutshell,. the SD9 is the best looking 3.4 MP camera in the world.. but it does not top the 10D. :)

[Belmondo]

Local camera store just took one of the Sigmas in on trade ---- for a 10D (no foolin'). It does take the nicest 3mp pictures you're likely to see, but there are a lot of deficiencies with the camera (see Steve' DigiCam review on it). A lot of potential, very little of which has yet been realized.

Save your money.

[ron chappel]

The camera does actually make a 10MP image without interpolation .I recall reading that it is in actuallity WAAAY better than a 3Mp camera would be but isn't quite what a 10Mp camera should be.

In the end it makes images VERY similar to a 10D/D100/SII Pro with only minor differences-make your buying decisions based on lens availablitity,resale value,ease of use,etc

[defordphoto]

It most certainly does not create a 10mp file. No way. No how. Maximum file size is 2268 W 1512. That ain't even close to 10 true megapixels.

The Foveon chip might be the thing of the future, but it's got some big problems and produces some great 3mp pictures, but that's as good as it gets.

Stand away from this camera. Turn to your left. See the Canon cameras? Walk straight ahead. Purchase camera. Enjoy it for many years to come.

[iwatkins]

I've used both the Sigma SD9 (for a couple of months) and played with the new SD10 for a week.

Sure, the SD10 fixes a lot of the problems (e.g. two different types of batteries on the SD9, doh!!) but it still doesn't feel "right". Colour rendition is very very good though. At times a little better than the 10D dare I say.

However, the whole thing does have problems. The SD9 would sometimes produce very strange colours in certain lighting conditions, especially shooting into the sun and the SD10 has improved on this but still not perfect.

The SD9 produces a 3.4 MP image.

I will sy that the Sigma software is excellent, much better than Canon's offering. However, I can process RAW from the Canon 10D in PS CS directly so don't need Canon software.

I have a Canon 10D, and for now I'm staying with it thank you very much :)

Cheers

Ian

[slin100]

If you want some entertainment, go check out the discussions on rec.photo.digital (news://rec.photo.digital). There you will find posts by someone who apparently stands alone in making outrageous claims and spreading technical misinformation about the superiority of the Sigma SD9. One of the claims is that it is a 10.3 MP camera, which could not be further from the truth.

The Foveon sensor is quite an amazing piece of technology. But Sigma's use of the sensor is fatally marred by the absence of an anti-alias filter. The images produced by the SD9, at first glance, appear amazingly sharp. Closer inspection will reveal severe aliasing artifacts which manifest themselves as moire, jaggedness, and other aberrations.

As others have mentioned, the SD9 also exhibits odd color. A common observation is that it renders skin tones with too much yellow. Personally, I find this a bit surprising as one would expect a Foveon sensor to exhibit higher color accuracy. It would seem that the sensor's color response is not as ideal as the technology would lead one to believe.

[iwatkins]

Yep, on closer inspection of images from the SD9 (and to a lesser extent on the SD10) it may actually be a problem with the auto white balance of the images.

I've had some strange ones. :)

Anyway, both the Sigma cameras are back with their rightful owners now so I can now give all my time back to getting some shooting done with the 10D.

Cheers

Ian

[Longwatcher]

Just a technical note as I understand the theory of the foveon sensor and light physics (which means I might not have it right).

Since the foveon sensor uses the priciple that different wavelengths of light pass through silicon to different depths, the issue will be that under certain light conditions the light will actually penetrate further into the silicon, or at least more (or less) light as a percentage will pass deeper into the silicon detector.
Thus in certain light conditions the foveon sensor would need to correct for the more photons hitting that region of the pixel and this would be very difficult to do for every occasion, because how does the algorithm now it is the light itself and not the color causing the problem.

The above will probably only make sense to a few, but it was something I had not thought about the problems with using a Foveon sensor that may make less acceptable until those problems can be resolved, which may require a secondary Bayer sensor on the camera to correct for certain light effects.

Just a strange thought.