Meadow Curves II

At 13.5 megapixels, the Kodak DCS PRO SLR/n is at a distinctdisadvantage in terms of resolution compared to the 21 megapixel Canon5 D II (with which I shot almost this same slice of land a week ago). Resolution aside, a subjective assessment of image quality isdifficult. Lighting has changed. The sky has changed. The crop isdifferent, and the positions from which the shots were made are notquite the same. Even so, I have to say that I get the sense that thecolors and range of the Kodak are more agreeable to me than the colorsand range of the Canon--at least in this case, but also in many othercases. (The Canon version, "Meadow Curves," is in this folder.) At alittle over $500, the Kodak (which I bought on eBay just a few weeksago) certainly manages to do okay in good light, when time is not afactor. There are other times when I want to throw it in the nearestlake. Still, most of the time I am pleased with the results from theKodak (which uses Nikon lenses), and that is what finally matters. Inmy opinion, it is a shame that Kodak got out of the DSLR business. Comments welcome.

--Lannie

At 21.1 megapixels, the Canon 5D II has 1.56 times the number of megapixels of the Kodak DCS Pro SLR/n at 13.5 megapixels.  It does NOT have 1.56 times the resolution of the Kodak, however.  Rather, the resolution is a linear function (line pairs per millimeter), and so the actual resolution of the Canon compared to the Kodak is the square root of 1.56, which is about 1.25.  That is, the Kodak has about 80% of the resolution of the Canon, even though it has less than 2/3 the number of megapixels (which are spread over a surface area whose size must be expressed as a square function--thus the square root of that square function is used to restore linearity.

In any case, the resolution difference is not so vast as to leave the Kodak in the dust.  Here is a slice from the middle of the Kodak shot.

--Lannie

The first full-frame Kodak was the 14n, which I shot with Nikon lenses between 2004 and 2006.  Among its many problems was digital noise, which Kodak clubbed to death with a noise reduction algorithm that made cedar trees (among other things) into featureless patches totally devoid of detail--the program interpreted certain types of detail as digital noise and smeared it out.  ("Smeared" is the only word that I can find, although the effect was politely referred to as "water-coloring.")

Fortunately, among the improvements of the Kodak DCS Pro SLR/n was a new sensor and a different approach to noise reduction that did not cause this particular effect.

The results can be seen clearly in this attachment, which shows cedars looking pretty much the way cedar trees should look.

The damage had already been done to the Kodak's reputation, however, and, combined with other very real problems, the result was that Kodak soon gave up on DSLRs.

--Lannie

Here's one more slice from just left of center.

Not too shabby for 2004 digital technology. . . .

--Lannie

THIS SLICE OF THE CEDARS IS FROM THE CANON SHOT IN THIS FOLDER FOR COMPARISON.

Actually, the longer I look at the two slices of the same cedars, the more I like the natural-looking Canon colors.  It is true that the sky was different; the neighboring deciduous trees had had another day to grow in by the time of the Kodak shot; and there was a stronger wind blowing on the day of the Canon shot (April 6)--even so, taking into account all the possible differences in situations and conditions, the Canon shot is rather hard to beat.  That should not be too surprising, but I am still surprised at just how much more the Canon shows--and how it shows it.

--Lannie

And the angle was slightly different, complicating the comparisons. . . .

--Lannie

. . . not to mention the different lenses--AND I FORGOT TO MENTION THAT THE KODAK SHOT WAS MADE AT f/16, WHEREAS THE CANON SHOT WAS AT f/8!  (Neither was optimal, in other words, but especially not the Kodak shot.)

Thus all of the above is NOT a scientific analysis, but I didn't go there to do a scientific comparison--but now I wish that I had made comparisons ON THE SAME DAY OF SHOOTING, etc.  Oh, well, trying to compare shots from two systems made a week apart under different weather conditions, slightly different angles, etc.--what can one expect?

I will say that the Kodak Pro SLR/n with the Nikon 24mm f/2.8 did a respectable job--and maybe more than respectable.  The Kodak-Nikon combo was simply outgunned by the 5D II and an L series zoom.  No surprises there. . . .

The surprise might be in how well the Kodak-Nikon combo from 2004 holds up against the Canon 5D II with the famed EF 24-70mm f/2,8L USM lens.

--Lannie

SUMMING UP:  The Kodak has its own color patina that gives it a certain charm, in my opinion.  Whether that is worth the trouble of operating it is another question.  Its limitations are obvious enough.

It is a quirky camera, frustrating to use at times.  Using it requires some patience, but, as I have hoped to show, its photos do have "something" that is not to be found in the super hi-res Canon shots.  Whether that "something" is desirable will remain a matter of personal preference.

This particular copy of the Kodak and the Nikon 24mm f/2.8 lens together cost about $750 on eBay.

The Canon combo of the 5D II and the EF 24-70 f/2.8L cost over $3500 when new.  Used today, the combined cost would likely be something over $2500 on eBay.

So. . . the used Kodak combo today costs about the price of a fairly modest contemporary lens.  I think that it would be worth buying or keeping, if one had the means to do so. 

That is my personal judgment from doing this series of test shots in this folder.  It has been fun.

--Lannie

I think I'm partial to the Kodak color....

 

I like the colors, too, Diane, especially as seen in this photo:

http://photo.net/photodb/photo?photo_id=15525412&size=lg

--Lannie

Hi Lannie,

There are fewer lens than can deliver top notch performance on the 20+ mega pixel cameras.

This number becomes even smaller when even high populated sensors emerge, like the Nikon D800 series.

Kodak realized this fact, and the color response from those DCS cameras remains among the very best today.

Best Regards,  Mike

 

Thanks, Mike.  I always wait for you to weigh in.

Have you ordered the D800E yet?

--Lannie

it's just a really beautiful picture.  j

p.s.:  the kodak sharpens too much for me.  if you've a mind to, see if there's an option to reduce it.  the details look a little like the M8 renders without sharpening turned off.  i hate to be racist, but you know, the japanese engineers are just BETTER.

Jamie, I used USM at 150 during post processing on most of these--including the Canon shots, too.  I can't say that I like it that sharp, either.

I had the sharpening turned off when I shot the Kodak.

In any case, the over-sharpening is not the camera's fault, but mine.

--Lannie

MORE FROM THE CANON SHOT OF APRIL 6 (also in this folder). . .  This slice is from the area just left of the cedars near the center.

--Lannie

AND YET ANOTHER FROM THE CANON SHOT OF APRIL 6 (OF THE SAME SCENE AS POSTED). . .  These tree trunks can also be seen above about one-third of the way from the right side.

--Lannie

Here is the Canon file from which the last three in-line "slices" were made.

http://photo.net/photodb/photo?photo_id=15523152&size=lg

Two points are noteworthy:

(1) Much higher sharpening was used on all of the in-line photos above than was used in the posted photos at the top of this page and in the Canon shot on the link just given.  (Please do not ask why.  I did not think this through before starting.)

(2) The Canon shot just linked to was actually made a full week before the Kodak shot which appears here, not "one day" as I said in one post above.  The result of this one-week lag is that many branches filled out with leaves during that one-week interval.  This is especially obvious in the bush to the lower right of the cedars, which looked almost like bare branches in the Canon shot, but looks so full of leaves as to be almost a part of the cedar in the Kodak shot.  This difference can be clearly seen in the in-line photos of the cedar from both cameras.  The cedar(s) are barely left of center in the larger posted panoramic photos, in both cases.

This ends this very messy, non-scientific comparison--at least on my part where posting in-line photos is concerned.  I may have another comment or two to make in response to those made by others.

--Lannie

This is a poster child for the old adage "Your Camera doesn't matter".  It doesn't matter as much as the technique and the photographer.  And we're not talking Holga vs Hasselblad. (I'd argue that in that case your camera DOES matter)   One could argue all day the merits of a more recent camera's abilities vs an older model but I think that most viewers would be hard pressed to see the difference in normal sized prints. The sharpness here is apparent even in a web sized offering.  I like the framing effect of the shadow area at the bottom and there is just enough cloud detail in the sky to prevent it from being a boring mass.  Good exercise!  Best, LM.

Thanks, Len.  I learned something from trying this, inept as my comparisons might have been.  (I really must learn to do diptychs in Photoshop--along with another gazillion tricks that I never bothered to learn--so that I could show the excerpted "slices" side by side.)

I have come to the same conclusion as you regarding the image quality for most practical applications.  Of course the Canon has more resolution, and arguably more natural color, but above all it is more convenient to use--and the battery life is incredible compared to the Kodak.  (I got some usable batteries from China for the Kodak about a month ago--for peanuts, with free shipping to boot.)  The Canon 5D II is also a very good low-light camera, as the Kodak most definitely is not.

The Kodak is a pain, as was the 14n which I used between 2004-06, but I still like it and might try taking some shots that I can print.  I am still, however, learning its quirks, and thus still pulling my hair out at times in the field. The improvements over the 14n still make the DCS Pro SLR/n worth trying some more--and I hope to do precisely that.

Thanks for being willing to give so much attention to this "enterprise."  It has been fun--and it makes me want to take the old beater out some more and see what I can get.

Using Nikon lenses again (even if hardly state of the art) also makes me want to try the new D800E, but a change in my financial status would be required to make that a realistic possibility.  There is just something about Nikon and Nikon lenses that I never got from Canon, not during my FD epoch, and not during my EOS epoch. 

One last point is about a matter of some considerable surprise: a lot of people are still using the big full-frame Kodaks and buying and selling them on eBay, in spite of the inconveniences that go with operating them.

Again, thanks for stopping by and making such helpful and supportive comments.

--Lannie

Hi Lannie,

I have tested the prototype Nikon D 800 E  in the lab. It is an excellent camera.

As I expected, there are a few lens that will really match up well . In my collection, I have found some Leica, Hassleblad, Mamiya, Nikkor and Canon lens that will do fine with the Nikon D800 E camera.

I have my version on order, as I can not just keep the evaluation samples from my clients.

The camera is as important as the lens because unlike film cameras, where the media could be changed , the sensor is fixed. One needs to best match the characteristics of the lens to the sensor to obtain the full advantage of the system.

I own hundreds of lens, because no one company dominates in all areas of design and build quality.

Nikon does represent the widest range of specialized lens across the board, but Canon also has a few lens that step up to the plate. Many third party lens have done a reasonable job to introduce adequate choices for those on a budget. My Pentax, Olympus, Leica, Mamiya, Contax,Voigtlander, Carl Zeiss, etc. all offer particular designs that place them high up the list of being excellent lens.

The premium medium format lens are at the top of the list and will easily produce excellent imagery using the newest 36 mega pixel Nikon D 800 E.

Last point: In my lab records, every sensor used in today's DSLR cameras has been fully characterized. This means that I know how the sensor behaves using the electronics driving the sensor (in the camera) as well as how the sensor behaves outside the camera under different driving circuits. While all sensors share the same basic physics , their realized performance varies from manufacturer to manufacturer. The final images produced are also sensitive to the post processing algorithms used, and unfortunately, not too many people really know how to best process their images afterward.

Best Regards,  Mike

Mike, thank you for more comments about the significance of getting cameras with good sensors--and knowing your sensors.

I do have a few of the more common Hasselblad Zeiss lenses.  Maybe I can put them in my hope chest, along with my dreams of owning the D800 E. 

--Lannie

Thanks a lot for your careful  study.  Love it.  Best Regards.

Thank you, Wanghan.

--Lannie

It was me, Dude.  It was me.

Just look where it got him, picture on money and all.  best, j