Friday, March 17, 2006

ISO 1600 Comparison EOS 5D versus FujiFilm S3 Pro

A quick,simple,straightforward ISO 1600 test between the Canon EOS 5D and the FujiFilm S3 Pro is done by dPreview forum poster Bernie Ess from Germany. His simple comparison shots are found here:

He has JUST put the post up,and as I write this, there are no replies to the original post, but I think it's imminently clear: the EOS 5D delivers significantly more REAL detail, with less blotchy noise in the EOS file,easily bettering the S3 Pro when both cameras are set to ISO 1600. Feel free to stop in and see what Bernie has found with the EOS 5D and the S3 Pro.

Bernie has been very enamored of the 5D for a few months now,and he's closer and closer to buying one for himself. hell, I MYSELF would LOVE to buy a 30D so I could have its wonderful features and capabilities. Personally, I think it's clear that the EOS 5D is changing the face of D-SLR photography by throwing one heck of a monkey wrench into the works of all the other D-SLR manufacturers. With the 5D, Canon has produced one of the MOST-desired cameras in quite some time. The advantages of a larger capture area are several,and I find it VERY,very funny that a small but significant minority of web posters and bloggers are trying to "explain away" the benefits of the various larger-sensored D-SLR's on the market. Since Nikon,and Pentax,and Olympus do not offer full frame D-SLR models, there are some brand zealots who spend time and energy trying to defend the "APS-C" or Nikon's so-called Dx format sensor cameras and Oly's 4/3 sensor models. Basically, there are a number of sour grapes types whose favorite camera brand offers nothing even approaching full-frame,so these fellows write web posts and essays expounding upon topics like, Why full frame is unneeded" or "Why full frame suffers from corner issues that DX is free of". It's really rather pitiful.
If one does not understand a lot about capture size or "film format size" and how that relates to depth of field and focal length and perspective and camera-to-subject distance, lens angle of view, and a host of other nuts-and-bolts photographic principles and practices, one is not fully fit to talk about the issue of DX versus Full Frame capture. Suffice it to say, at the world wide web level, there are a mere handful of people who are technically well-grounded enough to make cogent arguments about the differences between APS-C and Full Frame capture cameras in actual use. Most people simply "do not understand enough" to fully comprehend the differences between capturing to a smaller format as opposed to capturing to a larger format. Just as many people do not have a full grasp of the concepts of fill-flash and flash-as-mainlight,and are on unsure technical footing in the flash arena,so are many people misinformed about the issues behind sensor size and lens focal length and depth of field issues.

Suffice it to say, small-sensor cameras (APS-C aka Dx,and 4/3) have their own problems,and bring with them some very significant penalties in terms of creative use of shallow depth of field with the lenses we have available to shoot with, and the full frame cameras bring with them some very real advantages (and some disadvantages.The arguments AGAINST full frame d-slr capture are largely specious arguments. A fair-minded indivudual cannot condemn full frame capture as unneeded or as unworthy of consideration. The Canon EOS 5D is a landmark camera. Maybe some day, Nikon will see the wisdom of ofering its customer base choices instead of trying to bet the entire farm on the tiny Dx-sized sensors. It's not that the DX format is "bad" or anything like that, but Canon does offer three differerent sensor sizes, with 1.3x, 1.6x, and 1.0x FOV models to choose from. WHY might Canon offer THREE different sensor sizes in its camera lineup? Choice is a good thing. People who argue against the need for indivudual choice in sensor size and in camera types are not being intellectually honest when they construct their "Dx is King, FF is bad" arguments.
**********Well, it's been 12 hours, and the post now has 58 total threads. It's instructive to wade through and to see who has said what.
A few points, as I see them: the Canon,as Bernie said, is "brighter" and this he,and I,and Robert Whiteman attribute to the Fuji's ISO 1600 rating being an over-statement of the degree of actual sensitivity. I know what the term "brighter" means, and the honest fact is that the Canon is yielding a "brighter" image. Just look at the images. Canon is "brighter". The exposure settings are very similar,and as long as the dawn lighting wasn't becoming rapidly brighter as the 1-minute shooting period elapsed, I think the Canon's brighter image would lead _me_ to conclude that the stated ISO of the Canon is good, at 1600, and that the Fuji camera's stated ISO is somewhat under 1600 ISO. I expect that at 1600 ISO the S3 would do better with +.5 EV comp dialed in, but that would cost a person shutter speed.For those who really NEED ISO 1600 or higher, shutter speed is usually the most pressing reason to go to elevated ISO's.
Last point: The difference in the amount of detail resolved by the EOS 5D with a Tamron 28-75 Di lens and a Fuji S3 Pro with a 50mm Nikkor lens is evident and it favors the Canon camera. With equal framing areas, the EOS 5D is applying more photosites to cover the SAME FIELD AREA of rooftop and building.Put another way, the Canon saturates the scene's entire area with MORE PIXELS worth of information. The first two test images cover the same FIELD AREA, but with the Canon, there are 13.8 million photosites depicting the scene; with the S3 pro, there are between 6.1 million and probably 9 million photosites depicting the scene; how many pixels the S3 is using is impossible for me to determine,but since the scene has no inordinately wide DR issues I do not think the S3's sensor would deploy the highlight-sensitive pixels in this dawn scene. The six megapixel D-SLR's do not have the same ultra-high line resolution counts as the 11,and 12, and 13 megapixel cameras have. The EOS 5D offers a lower-noise,higher-resolution image than the S3 Pro does. The fine,fine image quality the EOS 5D is capable of obvious to me in the more stringent professional studio type photos I've seen done with the 5D, but even in this quick and dirty dawn genre scene,the 5D proves its mettle with a lower-noise, brighter,and more-detailed shot than the other alternative.
News Flash: Since the introduction of the EOS 20D, Canon has been making remarkable progress toward boosting image quality while keeping prices fair, with a lot of choices in cameras from beginner's models to mid-point to top-end. The EOS 5D is the first three thousand dollar-class full frame digital ever offered for sale. And it is capable of capturing VERY high quality files under a wide range of shooting situations. The body has good AF, adequate speed, adequate buffer, a decent viewfinder,and offers creative, shallow depth of field photography to be done with lenses people actually own.

1 comment:

Impartial said...

You wrote:
Choice is a good thing. people who argue against the need or the rationale behind different choices are not being intellectually completely honest when they construct their "Dx is King, FF is bad" arguments.

"not being intellectually completely honest?" You mean they're STUPID! Great time for you to pull a punch. How uncharacteristic!

You need to post comments. It would make your blog much more entertaining for those of us who read your rants from time to time! Oh, and don't do drive-by tirades on people's pbase sites. That's not nice. Get a DPReview handle and post there or send a nice e-mail instead!