04-16-2018 06:38 PM - edited 04-16-2018 10:12 PM
It seems to me that the quality of images, when discussing cameras and/or lenses, is often distilled into numbers. This is perfectly fine and I endorse the use of actual, objective measurements when discussing technology.
However, in my case, my photos are almost never printed these days but viewed on a screen, where ultimate resolution is no longer a real limiting factor. For me, perspective, field of view, colour rendition, contrast, etc are more important in achieving the result I'm looking for.
This was brought home to me when I recently said I preferred the images from my 24-70 f/2.8L to those taken with my 24-105 f/4L. I said the 'image quality' wasn't good enough. That was wrong. In every measurable way, the image quality from the 24-105 was more than adequate for my purposes. So what did I really mean? I truly do prefer those taken with the 24-70 but, because I'm not a tachno-nerd and don't enjoy getting into pixel-level analysis, I found it hard to describe what it is exactly that I don't like so much about the 24-105.
If I put my brain in gear and bought new glasses, perhaps I could nail it, but in the process of doing the detailed analysis required, I would probably lose the will to live. So, in the absence of such, I will simply say I like the 'style' or 'characteristics' of one over the other and, to my eye at least, it is very noticeable.
Maybe it's something similar to the way in which some people love the look of Lomo or vignettes, and others love monochrome. It's just a subjective preference and isn't easily explained by numbers.
04-16-2018 08:30 PM
You’d probably enjoy some of Roger Cicala’s blogs at LensRentals (he’s the owner). He once did a blog on lenses that aren’t “technically” scoring the best... and yet he just likes the images from those lenses much more.
There is a LOT to a lens besides “sharpness” (which is what everyone seems to talk about). Ironically, Roger points out that there is no such thing as “sharpness” (he uses the term “accutance” and then points out that there are many factors that go into that).
Full disclosure: I’m a science geek who also happens to love photography. But you’ll catch my science-geekiness coming through in my explanations (and it wont be subtle). So while I appreciate color rendition & contrast... I don’t get too hung up on such things because (a) there is no color ... that’s an illusion of the brain and can *easily* be tweaked in software, and (b) every image can benefit from some post-production tweaking for color & contrast.
I confess admiring some of the images I’ve seen shot with the Petzval lens (it probably has hideously awful MTF scores, but these optical flaws result in a “swirly” bokeh can be attractive). On the other hand, if every shot I took looked like that, the over-use of the effect would wear thin.
I tend to like lenses that can produce a shallow depth of field and a creamy blurred background. The EF 50mm f/1.8 II (with it’s 5-blade aperture that wasn’t well-rounded) created a distinctly “not smooth” blur... that had a jittery look to it (not attractive) even though in sharp areas it was actually very good. (The new 7-blad aperture of the “STM” version of the 50mm f/1.8 is much much better and it turns out Canon didn’t actually change the optics... at all. It’s the same lenses. They changed the aperture mechanism and the focus mechanism and improved the building quality but left the optics alone).
There are some lenses that don’t control chromatic aberration (color-fringing) well and that’s generally never an attractive look. But sometimes chromatic aberration can be fixed in software (by separating the red, green, and blue channels... re-scaling them to “register” correctly, and then re-merging them). Some computer software has lens profiles that does this automatically.
I often deliberately increase vignetting on images to call attention to a subject. I try to avoid the vignetting being too obviously noticeable. All lenses at extremely low focal ratios SHOULD have a vignetting pattern (that’s part of the physics of light) but vignetting is EXTREMELY EASY to correct in software. So I never credit or fault a lens for this.
04-17-2018 10:30 AM
All camera gear should be what the owner wants. What the rest of the world says is moot. If it is doing what you want it to, that is number one.
And that is precisely why I test and play with so many lenses. What comes out of DXO or any of the so-called labs is meaningless for what happens in the actual worldly use of lenses.
I guess some people read graphs better than they can look at photos.
09/26/2024: New firmware updates are available.
EOS R5 Mark II - Version 1.0.1
EOS R6 Mark II - Version 1.5.0
07/01/2024: New firmware updates are available.
04/16/2024: New firmware updates are available.
RF100-300mm F2.8 L IS USM - Version 1.0.6
RF400mm F2.8 L IS USM - Version 1.0.6
RF600mm F4 L IS USM - Version 1.0.6
RF800mm F5.6 L IS USM - Version 1.0.4
RF1200mm F8 L IS USM - Version 1.0.4
Canon U.S.A Inc. All Rights Reserved. Reproduction in whole or part without permission is prohibited.