I love your sophisticated rebuttal! So, all lenses are IDENTICAL in the way they render, according to you. Right?
Oh Derrel, you know full well that's not what I mean.
Of course not all lenses are identical, and certainly there is science involved in optical design. Come on, man. You KNOW that's not what I am saying here.
But nothing in that article reflects anything objective, just a bunch of delta charts with no actual data to back it up with in an attempt to make a wholly opinion-based analysis seem more analytic. Starting with the notion that a lens "captures reality". That assertion is *such* a load of crap, not the least of which is that "reality" cannot be measured objectively.
What optical design elements reflect 'reality' anyway? What metric are we using to define 'reality'?
Reality is *inherently* unscientific. You can say that "this lens design does that under this condition" but you cannot say "this lens design is more realistic" or "natural", at least not from a scientific perspective.
Sean, I spoke about ONE web article, and some rather mathematically and scientifically complex topics discussed by The A.P.. Not JUST one brief web article, but two things, of which the videos are i heavily science oriented, and offer side-by-side comparisons of primes versus zooms.
But, honestly, it was unclear exactly what you meant, with your five-word dismissal. So please let me explain. There is plenty of science, and also some art in lens design. The assertion in the article by Yannick was simple: that newer lenses with huge element counts are over-corrected to the point of flat, sterile images that lack a sense of dimension in 3-Dimensional subjects. And that my TPF compadre, is easily observed. Look at a Helios 44-2's image, you know
the old lens with the terriffically swirly bokeh; look at a Nikkor AF-S G 58mm f/1.4 and its photos (the NEW, $1600, G-series lens designed with high amounts of field curvature; and then look at the image at 50mm from a Tamron 24-70 Aspherical VC lens at 50mm zoom position. All three lenses have vastly different image character. Or, look at the example of the 50mm AF-D Nikkor versus the 13-element Sigma 50mm f/1.4 ART lens...different. Obviously "different" interpretations of reality, from an old lens with terrible/terrific
swirly bokeh; a new Nikon lens designed
to look 3-Dimensional, and one designed for
test chart scores to "beat Canon" and "beat Nikon" in the normal lens war that so many people want to get in on.
And so on,and so on, and so on. All one needs to do is to LOOK at images made by
modern optics, those new-fangled, uber-corrected lenses, and then compare the pictures shot with older lens designs with 5,6,7,8, or 9 elements. The very-newest, HIGH-performance lenses shoot test charts quite well. And many give harsh, ugly bokeh because every last bit of spherical aberration has been corrected out, and in general the primary design emphasis with, as he calls them
modern optics, is ultimate test chart resolution, and apparent sharpness or the sense of visual acuity.
Not three-dimensional rendering, but test chart "scores" for selling in the age of the internet.
This post began with an OP lamenting selling off an old lens, the 300/4 non-AF-S Nikkor prime, and replacing it with one of these newer
modern optics. Your position is pretty well summed up when you wrote about your feelings regarding lens drawing and lens renedering differences, quote "
*such* a load of crap, not the least of which is that "reality" cannot be measured objectively."
Okay. That's your opinion.You'e free to believe that we cannot measure the way lenses show reality. (But we can, by measuring many factors like aberrations, distortions, light fall-off,and so on).
But if you cannot instantly look at a
Petzval lens shot, or a
Helios 44-2 shot, and recognize how those lenses differ from say, the flat, sterile, PERFECT image from a 60mm AF-S G Micro-Nikkor, then you're obviously not using the words "measure" or "reality" in any common use of those words. It's DEAD-EASY to measure chromatic aberration, spherical abberation, coma correction or lack of, contrast, lens resolution, and vignetting. That's why your flippant, 5-word comment made so little sense to me.
The author's premise is that old lens designs and modern optics make different types of images. You call that idea a load of crap? And claim that we can NOT measure how lenses compare to reality? Maybe I missed something in your 5-word reply. If so, I apologize, I really do. But Yannick's premise makes a lot of sense to me. Compare two lenses: Nikkor 35mm f/1.4 Ai-S and Sigma 35mm f/1.4 ART. Wayyyyyyyyyyyy different renderings of "reality". Wayyyy different. One design is from 1969, the other is less than half a decade of age.
Sorry, but I think the article
The Trouble With Modern Optics carries a heavy dose of truth. Different lenses "capture reality" to use your words, in very different manners. The common words are
lens drawing style, or
lens rendering. The author has noted something that is plain to see
: modern optics have for the most part, been engineered with emphasis on uber-high resolving power, and their perfected images are causing less of a sense of depth to 3-D objects. Pretty simple...strip out spherical aberration to make the test chart scores soar, buuuuuuut in doing so, make the slight out of focus differences and the background blurring look nervous,edgy, weird, and unappealing. I think the intent of the article about modern optics is to call attention to a shift that's becoming very widespread. TEST CHART scores for some makers (Sigma, Zeiss) at the expense of other makers who are still running with older designs.
