Well for me D200 has clearly better specs than 20d. That is what I meant when I wrote that it is better camera than 20d. I have no idea about picture quality thou.
I've got the files, will try to post soon. Between having a ton of stuff to do at work, and NIkon Capture being practically unusable, it may be a while.
The D200 files already look better through NC than ACR, fwiw.
Hrow wrote:
Maybe it is just me but the IQ from the 20D seems much better than the D200.
Actually looking at the images again at home rather than on my work laptop screen, I'd agree the 20D's images do appear to be sharper. It seems to me that they are noiser though and I wonder whether that has something to do with the apparent difference in sharpness between the two cameras? Perhaps the NR in the D200 is hurting it's sharpness?
Hrow wrote:
PS to BenM... the Nikon 17-55 is a very highly rated lens that is about double the price of the 17-40. Logic would suggest that If anything it should perform better than the 17-40 but as we all know, that may not necessarily be true.
21farms wrote:
i don't have a way to post the RAW files but i'm happy to share them with whoever wants them. the NEF is over 15mb though and i'm not sure my email program allows me to send out such a large file. any thoughts here?
sam, thanks...i'm looking forward to seeing what you come up with as i have no experience with NC. i also sent you another set of files of a slightly different angle of the same general scene but taken at ISO100 using a tripod. the 20D file is slightly underexposed though.
BenM, i'm sure the 20D looks sharper at ISO800 because of there is more NR going on with the 200D that cannot be turned off. at ISO100 (which i just uploaded), i think you'll see a dramatic difference.
To my eye the nikon shots still have more pleasant colour even after setting the white point & the canon is sharper with more contrast. As you say, this could be all about the lenses... would be interesting to try a test with a nikon adapter on the canon!
Okay, here's the results using Nikon Capture and DPP. There's the "as shot" versions, and then version with better WB and no sharpening to minimize noise.
D200, Nikon Capture all "as shot" settings:
D200, Nikon Capture, 5600K WB, no noise reduction, no sharpening, contrast normal:
20D, DPP, as shot settings:
20D, DPP, 5600K WB, normal contrast, no sharpening:
It looks like they are failry similar at ISO100, but by ISO800 the 20D clearly is rendering much more detail with less noise. The D200 looks like it's got a heavy noise reduction filter going on that blots out fine detail (small words on signs, etc.) yet the D200 shots are still REALLY noisy in my opinion. Not sharp, not more detail, just noisy. The WB and color differences are only trivial -- they are easily fixed in or out of camera with some minor saturation/WB changes. I'm surprised that the D200 didn't do as well at high ISO. And the Nikon noise reduction... ick. It makes the tailgate look all mottled and splotchy. Though the 17-55 f2.8 should outperform the 17-40L if cost was the defining factor, it doesn't look like it does -- the 17-40L seems to exhibit much less CA and better sharpness.
It'd probably be better to use a light meter, and in addition to the same ISO and Focal length.
And it might be better to use primes instead, say an 85 1.8 vs 85 1.8 (if you have those).
*Set identical exposures
*Apertures(equal as well as stopped down [ie if you stop the nikon down to f4, then stop the canon down to f5.6])
*Set the same kelvin
*Turn off in camera sharpening
*Turn off in camera color balance
*Turn off in camera color adjustments
I am very unimpressed by the Nikon and never expected to see so much of a difference. At this point I am wondering if there is not a software problem going on as the noise levels are drastically higher on the D200, much more than is reasonable at this level of camera.
Also, I don't think the lens is having much impact unless the Nikkor has been dropped repeatedly and run over by a truck. I am not seeing that much of a differnce in sharpness but rather it seems to be the noise that is softening things up.
kalieaire wrote:
It'd probably be better to use a light meter, and in addition to the same ISO and Focal length.
And it might be better to use primes instead, say an 85 1.8 vs 85 1.8 (if you have those).
Why? The differences in character of noise is pretty obvious in any case.
kalieaire wrote:
*Set identical exposures
*Apertures(equal as well as stopped down [ie if you stop the nikon down to f4, then stop the canon down to f5.6])
*Set the same kelvin
*Turn off in camera sharpening
*Turn off in camera color balance
*Turn off in camera color adjustments
Then do the same in ACR.
Why would you use ACR? Then you're subject to ACR's handling of each file. I think it's better to assume that Nikon and Canon know how to get the best out of their files, which is why I used Nikon Capture and Canon. I know that that is the case with Canon files, not so sure about Nikon. I personally do not like ACR's processing, so I really don't care to see it's results here.
sam, thank you very much for doing the NC and DPP conversions. i have added them to my pbase gallery for easy reference.
kalieaire, i tried my best to limit the variables as much as i could given my time constraint but i also intend to play around some more over the holidays.
hrow, my 17-55/2.8DX is brand new. the ISO100 shots were with 85/1.8 primes from nikon and canon and both lenses are in mint condition. you can see a marked improvement in sharpness in the ISO100 shots but i'm not sure if that's due to the lens or the lower ISO...probably a combination of both.
Isn't the 17-55mm f2.8 supposed to be one of Nikons best Lenses? I see people rave about this one a lot. Actually I'm surprised to see the 17-40L getting the sharper shots - going by these, considerably sharper shots.
Norm, I fully expected that your 17-55 was in good/great shape and was only trying to point out that I didn't think we were looking at differences caused by lens quality. I am also not sure that we are seeing fundemental problems with the camera. I'm thinking software. Have you to tried to shoot high res JPEG? Would be interesting to take the RAW converter out of the mix and see what happens with the noise.
Hrow wrote:
Norm, I fully expected that your 17-55 was in good/great shape and was only trying to point out that I didn't think we were looking at differences caused by lens quality. I am also not sure that we are seeing fundemental problems with the camera. I'm thinking software. Have you to tried to shoot high res JPEG? Would be interesting to take the RAW converter out of the mix and see what happens with the noise.
But RAW is the format where you'll see the camera at its best. Shouldn't the output of Nikon's own RAW software produce the best results?
As for differences in shot settings, the two files that I was given of the sports bar for the ISO 800 test were shot at identical settings - ISO 800, 1/250th, f/8.
do you have opinion on whether there is always Some sharpening done on Canon cameras even if they are set to 0 and shot in Raw? I remember Askey saying something to that effect but cannot remember the context. I am also wondering if the D200 is Always doing some NR at higher ISO? Both look pretty good here I think, but the canon files here look sharper but so hard to determine the reason
Vole wrote:
Isn't the 17-55mm f2.8 supposed to be one of Nikons best Lenses? I see people rave about this one a lot. Actually I'm surprised to see the 17-40L getting the sharper shots - going by these, considerably sharper shots.
i'm surprised by this as well. this makes me not feel like the 17-40L is a great bargain!
