So let's look at a few actual pictures. These all come from the same sharing site and several are made with the same lens, but there's no need to read anything into that.
A digression first
Before I start, though, I just wanted to comment on one thing in passing: in looking for a deliberately soft lens that doesn't go for the analytical/clinical/forensic look, I happened across one of those very cheap third-party manual-focus lenses (I sometimes jokingly call them "fourth-party" lenses—you know, sorta like "off-off-Broadway"—but nobody gets the joke). It only costs around a hundred bucks, and it looks like it might be quite nice for intentionally less-than-sharp portraits.
So then what does the first review say? "This lens is total crap! It's soft at every aperture!"
...Which seems unfair. He's criticizing the fact that the $100 lens he bought is not the same as a $500 lens. Whose fault is that though? His, if you ask me. If you insist on a lens that renders like a $477 Nikkor, then save your money for a bleepin' $477 Nikkor, ya broke-ass, no-cash-having jamoke. Leave the soft $99 lenses to those who are looking for soft lenses.
There, now I feel better.
Picture 1
Okay, now on to the pictures. When you open this first picture, just look at it sitting a normal distance from your screen, without being critical. Pretty standard, right? And looks okay. At a glance, you know what's going on: sharp, in-focus blooms in front, fading into progressively more blur in the distance. So now take a look.
But zoom in by clicking to enlarge it two times, and we find to our astonishment that virtually nothing in the picture is sharp, save a few leaves on the left-hand side of the frame. None of the blooms are sharp, not even the one at the lower right which seemed like the prime candidate at first glance.
But remember 3R prints, the once-ubiquitous standard so-called "drugstore prints" that were 3.5x5 inches in size? This would look fine at that size. Those 3R prints were optimized to mask slop and imprecision of many kinds. It was a system designed to routinely make silk purses out of sow's ears.
Picture 2
Or take this one. Now, in my humble opinion at least, this portrait (which gets the important things right, the expressions and what it implies of the relationships—and it's a nice enough composition, too) has a few more pressing technical problems, specifically a tonal scale that's way too short and what looks to me like a virtual green filter (you can see it in the gums, which should be a much lighter tone that doesn't draw attention, and that characteristic mottling of the skin tones. Light skin often has slight reddish mottling which is nearly invisible, or rather inoffensive, to our eyes, but it's highlighted in an unpleasant way when you make the reddishness darker with a green filter). But again, if you look at it uncritically at the original screen size, it looks fine. But, enlarged, the boy (middle figure) is clearly less sharp than the other two and the face on the left is less sharp than the face on the right.
When I was a custom printmaker (I always said "custom printer" but that was before a "printer" was a black plastic box made by Epson), I would often call clients to inform them that a certain negative would only go to a certain size. My clients were mostly photographers, and they would almost always assume that I was talking about technical parameters. They might reply something like "I've enlarged Tri-X negs from the Hassie to 16x20 before and they looked fine." Sometimes I was indeed talking about technical parameters, such as the size at which a slight touch of camera shake became too obvious. But I was usually using my judgement about the size at which that specific image would most effective. For example, if I were printing this picture, I might suggest a smaller size as the max, to avoid drawing attention to the disparity of sharpness between the three faces. Printing it overlarge would just draw the viewer's attention to the differences in sharpness, and those differences are absolutely immaterial to the picture.
I might even cheat a bit and actually soften the sharpness of the face on the right. Have to be careful with that, though, because that would never be something you want to call attention to.
Picture 3
It's actually not easy for non-specialists to detect very small amounts of blur, whether it's out-of-D-o-F blur or camera shake. I'm more sensitive to it than the average bear*, but I can't claim to be infallible. Sometimes you almost intuit its presence depending on how large you're seeing the image. On this third and final picture, there are three women dressed in blue T-shirts front and center. Look at the middle woman first when you open the link.
She looks fine, right? Comma, at that size. But, again, after you enlarge this and look critically, you can see more easily that she's actually not at the plane of best focus. The woman on the left, with the tiara and the portrait of the photographer in her purple sunglasses, is the most sharp. The smiling woman on the right is now clearly out of the depth of field, even though she's in the D-o-F at the smaller size.
I would probably have simply placed the focus on the middle of the three women and stopped down one more stop. That can sometimes be a bit risky, of course, and I'm betting you know why: because if you misjudge and get the woman closest to the lens—the woman with the tiara—a little too soft, you draw too much attention to that. (As a general rule, subjects closest to camera position need to be sharp, and it can be distracting if they aren't.) Hence the extra stop of increased D-o-F, as a safety margin...assuming I'd have the presence of mind to adjust in the heat of the moment, which I probably wouldn't. But those are the things that experience teaches you to judge.
(Oh, and that "portrait" of the photographer in the sunglasses? I'm betting it's the other photographer. It seems to me that the three ladies are looking at someone taking their picture, and it's not the guy who took this picture. Because of the angle of their heads, I'm betting we're seeing the other photographer in the sunglasses, the one who they're looking at. Just my interpretation.)
How to evaluate your own pictures
So how can you start to get a sense for the optimum sizes of your own pictures? There's actually a pretty easy way, although it takes a fair amount of experience. Assuming your eyesight works for the distances involved and your room is big enough, just fill your editing monitor with the image, and stand up and back away from the monitor! As you move farther and farther away, notice how the image subtly changes and how the level of visible detail gradually diminishes. If you study what you see and learn to "feel into" the image and how it works, you'll start to get a sense of how much each image can take of enlargement, and the way different degrees of revealed detail "work" for that particular picture.
As mentioned above, though, I'd encourage you not to look at mere technical parameters, but rather to look at how the specific image is functioning in terms of content and meaning. All of what we've been discussing is a judgement call, a balancing act, based on what in the picture you want to call attention to, and what would be a distraction that would detract from the picture if it is spotlighted too much. This is influenced by technical parameters but is not a technical issue at root. It depends on each specific picture, and how you want that picture to work.
Mike
(Thanks to Junpei, Jim, and Glen, the photographers of these three pictures)
Ed. note: The second link was missing when I posted this. Gah. Sincere apologies for the inconvenience.
*That's actually a Yogi Bear reference, and if you even recognize that name you're showing your age as badly as I am. :-) Five life points if you know who the cartoon bear was named for.
ADDENDUM: Here's some truly trivial trivia for ya. Do you know why Yogi Bear had a collar?
The answer, says Wikipedia, is because it allowed animators to redraw only his head when he was seen speaking, and not have to redraw his entire body. This was a cost-cutting measure by Hanna-Barbera, the studio that produced the cartoon. It reduced the number of frames needed for a cartoon from ~14,000 to ~2,000.
Will you help sustain "The Online Photographer"?
Also sponsored by...
Original contents copyright 2021 by Michael C. Johnston and/or the bylined author. All Rights Reserved. Links in this post may be to our affiliates; sales through affiliate links may benefit this site.
(To see all the comments, click on the "Comments" link below.)
Featured Comments from:
Gavin Paterson: "It’s Yogi Berra, baseball player, but I do have an advantage, as I worked for the Australian subsidiary of Hanna Barbera in the '80s, when we still produced animated series for USA networks, such as the Flintstones, Scooby Doo, Yogi Bear, and the Jetsons—now that’s showing my age!"
V.I. Voltz: "Yogi Bear was named after Yogi 'You can observe a lot by just watching' Berra. My kids watch YouTube; we had to make do with cartoons, which, at first, were in black-and-white."
Rob de Loe: "You can do as many of these posts as you like...."
hugh crawford: "I got the bear reference, but you missed the opportunity to call out your Boo Boo. Did you know there is a Daws Butler stream on Spotify?"
Mike replies: I never knew his name before! Thanks for that.
w98abee: "The Yogi Bear reference was like deja vu all over again."
Guy Couture: "Thanks, Mike, for this clear explanation. I think it's the first time I see it explained this way. No doubt you're a good teacher."
Mike replies: Gracias Guy.
Jim Kofron: "Photography is 90% mental, the other half is camera gear...."
Steve Renwick: "The bear's namesake might have said, 'Most pictures are not as sharp as they look.'"
Lawrence Huggins: "Why would you bother to devote any of your time to writing about three such unexceptional photos?"
Mike replies: Because they are...
Cary Talbot: "Nice examples to illustrate your point, Mike. Reminds me of something I read in a photography book by Scott Kelby. He was stressing the importance of zooming in on your image when 'chimping' it on the back LCD to make sure you nailed the focus because 'everything looks tack sharp on a 2.5-inch screen.'"
Mike replies: Yes. And, in like wise, just as with the old 3R prints the small size of a smartphone-as-display masks a multitude of sins.
John Robison: "Ah yes, 3R prints. I well remember an ad by Keystone Camera for a cheap 126 camera, comparing results from it to a Hasselblad. Of course, in a magazine reproduction at a 3R print size, there didn’t appear much difference.
"When working at a camera store, more than once customers would bring in a strip of 35mm negatives with the sample 3R print and ask for a 8x10 enlargement. Grabbing a 10x magnifier would often reveal that the neg was just not up to that degree of enlargement. It was surprisingly hard to explain this to customers. Some would insist that we send in their order anyway. Once a year, at inventory time we always had a box of enlargements refused by customers that were written off and discarded.
Mike replies: They never want to take your word for it, do they?
