This is basically a re-post of another thread on DPReview.
I use canon DSLRs and have some nice L lenses. I also have a Sigma SD14 (14 MP with a 1.7X crop factor) that I converted to use EF lenses. This puts me in the somewhat unusual position of being able to compare the IQ of the two systems, using the same lens on both. In this 50D vs SD14 comparison the Canon 200mm f/1.8 L was used, stopped down to f/4.
Dave In La wrote:
need to correct address for 50D .CR2....I got my download, but others may not correct it themself
That's odd, that address gets truncated for some reason. Thanks for pointing that out. The files can also be downloaded from here: http://cid-709471d4dadefb7b.skydrive.live.com/browse.aspx/shared
for interest, I will post crops from the RAW of both.
These are straight conversions with no adjustments made, not sure exactly what
to look for in sensor comparison.....so post away and I'll learn something
Thank you for posting these. I've always been interested in how a Foveon file (particularly from an SD14 or DP2) would compare to various others but I've never seen anyone able to do it with the same lens.
There's an obvious difference in resolution and white balance/exposure. Looks like the Sigma can make a very nice but small jpg file to me. If Sigma could get their sensor resolution up to that of the competition, they'd have a viable camera IMO.
Which is a better representation as seen by your eye.
Well... I'm a little colorblind and I have an amazing ability to overlook color casts unless two differing images are displayed side by side. With that said, and reading that the SD14 tends toward a greenish cast in it's default JPGs, I think the 50D is probably the more accurate of the two. I would assume color cast discrepancies could be largely corrected in the raw conversion process, but I can't say for sure. With my color perception problems I tend to be a minimalist when it comes to color adjustments.
As has been discussed numerous times on the Sigma board in the past at DPreview there is a slightly different interpretion of blues, greens and reds with the Foveon sensor. This shows in arky's image quite well. I would say a majoity of users of Sigma cameras like this coloration, "the Foveon look", and that's fine, it's a personal preference. I believe it to be a slightly warmer version of those colors. I use my DP1 for landscapes and I usually have to get the colors back to the way they actually appear, which is my preference. I live at a high elevation and have deep blue skies quite often but my DP-1 images give me more of a light azure blue sky.
I like the DP1 quite a bit. Great small carry landscape camera for walking when not wanting to carry big guns. It's images are quite sharp for a small camera. I have no problem with the off hues that I encounter with the Foveon sensor, it's a fast easy fix. They not only appear in ACR interpretations but Sigma's SPP as well, in fact more so. Maybe I'm just too use to "the Canon look"
I honestly hope Sigma, which just bought Foveon, will continue with improvements to the DP-X line of cameras and will have a smallish zoom some day. The Foveon sensor has potential to give CMOS sensors a good run for the money and I hope Sigma will continue with the development of the technology.
The Sigma image appears exposed more and has a higher green/blue component, perhaps even a bit of a WB shift.
Whatever the reason, the Canon image looks washed out in comparison.
The Sigma looks a little underexposed and too greenish whereas the Canon is a bit overexposed and magenta. Not sensor matters and both can be fixed with a couple clicks prior to RAW conversion.
I think it would be a good idea to set the white balance for each in Kelvin to prevent any color differences. I would also manually expose with identical settings. I have a feeling a fair bit of the difference is owing to the above.
Trevor W wrote:
I think it would be a good idea to set the white balance for each in Kelvin to prevent any color differences. I would also manually expose with identical settings. I have a feeling a fair bit of the difference is owing to the above.
You are probably right about the white balance difference.
The settings were identical though, AWB, ISO 100, aperture priority, f/4, both cameras metered the scene at 1/1000 sec with no exposure compensation. Any differences in exposure must be due to ISO sensitivity differences, or perhaps tone curves.
arky wrote:
You are probably right about the white balance difference.
The settings were identical though, AWB, ISO 100, aperture priority, f/4, both cameras metered the scene at 1/1000 sec with no exposure compensation. Any differences in exposure must be due to ISO sensitivity differences, or perhaps tone curves.
and in my RAW conversion, I made no adjustments from default on either
