| Thoughts | Greate advance for photography in this decade by Contax N1,
But does not understand and ignores or denies it! Part of the reason for this is that apparently the sales organization of Contax/Kyocera also did not comprehend and could not tell anyone. These breakthrough features for photography for the new Contax N1 are the four focus points positioned to aid the "Rule of Thirds" and the "Focus Bracketing with Custom Function #8".
The first feature is being ignored in the June test report by PopPhoto and the Custom Function #8 is dismissed as being inconsequential. The editors are writing on page 132: Promising in theory, but questionable in practice, the N1's unique autofocus bracketing feature makes three pictures, around your subject. ----- (With Custom Function #8, you can double the shift in focus.) To prove the usefulness of the feature, we set up multiple test situations --alas, to little or no avail. -- Our guess? This feature is a handholding gesture for users who require absolute sharpness ---unnecessary in our estimation, because you would be hard pressed to find a soft subject among our dozens of tests ---(Their problem is the test setup, it's equivalent to testing the "first automobile" by putting six horses in front of it to find out how well it will perform as a Stagecoach!)
For the first time in the history of photography a tool for "Very Wide DoF" is available with a 35 mm SLR camera, to use handheld with fast shutter speeds.
The real application of Focus Bracketing with Custom Function #8 is a breakthrough for "Depth-of-Field". What the Custom Function #8 can do easily, combined with simple digital photo editing, is the "Scheimflug effect", which now requires a special "Tilt-Shift lens" with an average cost of $2169. For this, and good sample pictures, see the article in PopPhoto August 2000, page 96, "Do the Tilt and Shift". It is the "Tilt" feature, which can be done by software and function #8. (And, incidentally, the "Shift" feature is already being done for years by nearly all photo editing programs, even the cheapest, e.g. my $40 PictureIt99, but you will find no mentioning of this in the above article.) Given a photo opportunity, which "ask for" a Tilt-Shift lens, the Custom Function #8 can do the same faster and more conveniently. No high-cost T-S lens is needed, but more importantly, compared to T-S, the Function #8 can do metering with TTL-AE and Autofocus, and Zooms also are ok. Basically, you can do snapshots.
Are there any compromises? Not really, if you don't count the cost of the two additional frames of 35mm film. (For digital cameras it's for free.) What "Contax" forgot to do was provide the simple software-plug-in for Photoshop, which would fully automate the editing process. Now the job has to be done manually, but even that is rather simple, even for my Picture It-99. The three negatives are combined into one, the bottom 33% of the close-up with 33% of the middle-distance picture and then the top 33% of the far picture. (If you have one of the "Panorama program-plug-ins" you can automate the "stitching" by rotating the frames 90 degrees.)
A program specifically design for this task could actually do a better job than my manual procedure by comparing the three pictures for best "sharpness", the stitching part would only combine the best parts of each into the final print.
"Universal" Very Wide DoF?
What about scenes which cannot use the "Scheimflug effect", such as a landscape-photo taken through a very near window frame? Can Custom Function #8 of the Contax N1, combined with sophisticated software, handle photo opportunities like this? If yes, we would have a "Universal" Very wide DoF feature.
In June of 2001, when the report in PopPhoto was published, I believed this to be a simple program, but it turned out not to be so. The problem is that "objects" in the near-picture, when these obscure parts of the far-scene, will leave a "fuzzy halo" when the camera focuses on the far-scene. (Diffraction) The size of this "halo" is 1/2 the diameter of the "circle of confusion" around these objects. (I was made aware of this effect by "Andromeda Software", a company designing Photoshop plug-ins.) However, this deteriorating effect only works from near-to-far, not the other way around, and I am now convinced that there is an acceptable solution to the problem.
Again, this time in October 2001, I found a "manual procedure" to quickly handle this problem, but only for simple cases such as the mentioned landscape through a near window. The case however is "general" (universal) and so is the "procedure", and computers and programs don't get tired to do the same thing hundreds of times as I do. (For example, take the editing job for a fall-picture, "masking" all leaves on trees with a certain color, so that their "brightness" can be increased. I could do that manually on twenty leaves before I would give up, but my Picture It 99, will do thousands of these leaves in a few minutes and never complain.)
I have a good explanation and description for the needed computer program. Such a program (Photoshop plug-in) would only use existing and proven elements, the novelty is in the basic concept. The easiest way to describe this is to go through the example I did manually, in a step by step fashion, with the actual photos of the landscape photographed through a very close-by window frame. (Unfortunately, even with strong JPEG encoding, the photos will take several Megabytes --- too much for an unsolicited e-mail attachment --- so please request it if you are interested.)
These are the two methods for extreme wide Depth-of-Field, provided by the Contax N1 and new software, which works even for a handheld "snap-shot". The first is very much equivalent to the method of the Tilt-Shift lens (the Scheimpflug effect), but with several other advantages.
The second method is "universal" for all photo subjects and was never available before. However, this universal method creates a very minor distortion in the final picture, which I believe is not detectable without very precise comparison-measurements at the actual scene at which the photo was taken. (If such measurements were actually taken they would give an "illusion" as if the three focus-bracketed frames were not taken at the same focal length. For a 50mm lens the first frame might shows 50mm, the second 51 and the third 52, a difference not detectable without very precise measurement.)
A Concept of Ansel Adams and the Future of Photography
The PC-programs I described for "Extended DoF" with Custom Function #8 of the Contax N1 might be new, but the "basic concept" is old as Ansel Adams. This great man clearly understood that the technical process of photography has three basic elements. The camera and lens, the film and its development, and the tools and processes for creating the print. (See his books, The Camera #1, The Negative #2, and The Print #3.) All three must "work together" and be matched as one complete system. The camera alone does not have to generate the finished picture, neither does the film or the CCD nor the print program. It is immaterial at what stage (#1, #2, #3) the different parts of a picture are perfected. The described DoF programs are only shifting part of the task from the lens to the editor in the computer. To create and perfect the photo, (Ansel's "Fine Print") the three tools and their processes must be inter-locked as an integral system.
The future revolution in photography has actually very little to do with the replacement of the chemical film with a CCD or CMOS target, but everything with combining "Lens and Camera" with the computational power of a digital processor. As example, take two important parameters of photographic lenses, linear distortion (barrel and pincushion) and chromatic aberrations, not getting red green and blue (RGB) at the same focal plane. (This is the reason for the high cost of APO Tele lenses). A tutorial in the Edmund Industrial Optics catalog states: --- Distortion is a geometric aberration. This means that information is not actually lost, [sharpness is not lost] but simply misplaced on the image plane. --- However, once the distortion has been accurately measured [e.g., by Popphoto's lens test], it can actually be removed from the image by using software - [e.g., a Photoshop plug-in by Andromeda Software]
The same is true for chromatic aberration; the colors are just misplaced, but this time not in the horizontal/vertical plane but in depth (focal plane). In some digital cameras there are actually three silicon chips, one for each color and these could be slightly re-focused to match a certain Tele lens. However, it is probably cheaper to just use the autofocus mechanism to do the very small adjustments needed. In a film camera, the film has three (or four) different chemical components, all mixed together for the different colors and the autofocus mechanism must be used to bracket three exposures, just as the Contax N1 does for DoF.
For this correction of distortion the digital-camera-computer can do the complete job, and for film cameras, concerning chromatic aberrations, all the PC-program needs to know is which scanned negative is for what color, but for the correction of linear distortion we need to tell the PC-computer what the "misplacement" is. A good place to put these "misplacement" measurements is inside the lens, the ROM of its AF computer. And for correction of linear distortion for film cameras a data-file of the lens is also needed for the photo-editing program.
Let me close with a visionary quote from Ansel Adams on digital photo editing from 1983, long before there was a Photoshop and 2-GigaHertz Pentium computers.
The scope of control with the electronic image has not been explored, but I feel confident astonishing developments await us in this area.
|
