Over at the not-always-reliable dpreview, Canon's Chuck Westfall apparently sent an email stating that in JPEG, the 5Dii is quite a bit better in high ISO than the 1DsIII, but in RAW there is only a slight advantage to the 5Dii. It seems the small gaps in the 5Dii's microlenses are smaller compared to the original 5D, but not necessarily when compared to the 1Dsiii, and the big difference between the cams lies in Digic IV processing. I think the 5Dii being slightly better than the A900 in high ISO, with the A900 slightly being better in low ISO is what we're looking at, and it seems either camera will produce stunning images, so we should all be happy.
Here is the email from Chuck:
" Hello, Amin:
> You ask a good question for which there's no easy answer. Bottom line is, the noise differential between RAW images from the 1Ds Mark III and the 5D Mark II will be affected to a great extent by the user's choice of RAW conversion software. Once that's been established, there will also be some variation in image quality based on the way the image is edited in that particular RAW converter.
> Let me put it to you this way: Using Canon's DPP 3.5.1 raw conversion software with all image quality settings equalized, noise levels will be almost the same for both cameras, with a slight advantage to the 5D Mark II because of the improvements in its CMOS sensor compared to the 1Ds Mark III. YMMV with other converters.
> On the other hand, there is a noticeable improvement in the quality of in-camera JPEG images for the 5D Mark II versus the 1Ds Mark III at any comparable ISO setting when other in-camera image quality settings are equalized. As you suggest, this is due, in large part, to the difference in image processing algorithms between the two cameras, which in turn is affected by the difference in image processors.
> Best Regards,
douglasf13 wrote:
I think the 5Dii being slightly better than the A900 in high ISO, with the A900 slightly being better in low ISO is what we're looking at, and it seems either camera will produce stunning images, so we should all be happy.
Actually, I think that remains to be seen.
From everything I've seen so far, if the a900 is better at low ISO ("low" here meaning only at 100 ISO) it is very slight -- and these differences might be quite subjective. I would further point out that this is from 5D MkII jpegs and a900 raw -- we have yet to see 5D MkII raw files, which should prove to be even better than the jpegs. At high ISO's, and in this case I would start "high" at 400 ISO, the differences are not slight to my eyes, and the differences increase with higher ISO's. Mind you, we have not seen a good head to head comparison here with images of the same subject and same situation and settings, so this is clearly a provisional preliminary assessment that is subject to change. But to my eyes, and I've downloaded a number of a900 raw files to examine, the differences at high ISO's are not slight. We will have to disagree there, at least for the time being -- but as long as we are all happy. ;-) I hope, however, to be pleasantly surprised by a competent comparison once someone so inclined can get their hands on both cameras.
