Does 3-d pop exist? Does this image have it?

Apparently people differ strongly on which images have 3D pop. Or what 3D Pop even means. It seems. I browsed a lot of images on Flickr, searching for 3D, Pop, Depth keywords. Hm, yeah there were some but rather rare. In the end I found out a Flickr group completely devoted to 3D Pop:

3D Pop | Flickr

Can we agree that this group contains lots of images with a strong sense of depth (ok, illusion of depth) and that this is a desirable property? Some look even 3D to me, but yeah, that's me.

Lots of fast primes from lots of brands, not many zoom lenses that I could see. Apparently there is lots of CaNikon too, it's not like vintage/Voigt/Leica primes are all over the place and modern designs can't do it.

Hey, did I ever see Sony? Not sure...

(Oh and IF you browse that group, make sure to hit "L" for screenfilling image for a better impression)
An additional (personal) note from me.

I don't find the shallow depth of field close up images (like the lizard) convincing examples of 3D-pop. I think we have learned to expect shallow depth of field in close up and this lessens the effect.

The exact same type of shot with the subject at a middle distance is much more effective. I think this is the cause of the nostalgia for medium format film full length portraits with the famous Pentax 67 105mm f/2.4 and the like. The depth of field and rate of fall off for middle distance subjects with these cameras and lenses surprises us. We don't expect to see that kind of DoF fall off in subjects at that kind of distance. I doubt the effect is anything specifically to do with the design of that famous lens, it just has the right specs for those middle distance full length portraits to exhibit that surprise-value level of DoF fall-off. I note also that the effect diminishes where there are competing subjects in the sharp zone. It seems to work best with a centralised sharp subject with space around it set against a gently falling away Dof without extreme blur.

I imagine the Bokerama/brenizer images enjoy a similar basis. The effect may be akin to the surprise/ambiguity we see in the "toy model effect" from the misuse of tilt lenses. We are not expecting to see that precise DoF at those subject distances and it stimulates the 3D effect (or the miniature effect).

--
2024: Awarded Royal Photographic Society LRPS Distinction
Photo of the day: https://whisperingcat.co.uk/wp/photo-of-the-day/
Website: https://www.whisperingcat.co.uk/wp/
DPReview gallery: https://www.dpreview.com/galleries/0286305481
Flickr: http://www.flickr.com/photos/davidmillier/ (very old!)
 
Last edited:
Apparently people differ strongly on which images have 3D pop. Or what 3D Pop even means. It seems. I browsed a lot of images on Flickr, searching for 3D, Pop, Depth keywords. Hm, yeah there were some but rather rare. In the end I found out a Flickr group completely devoted to 3D Pop:

3D Pop | Flickr

Can we agree that this group contains lots of images with a strong sense of depth (ok, illusion of depth) and that this is a desirable property? Some look even 3D to me, but yeah, that's me.

Lots of fast primes from lots of brands, not many zoom lenses that I could see. Apparently there is lots of CaNikon too, it's not like vintage/Voigt/Leica primes are all over the place and modern designs can't do it.

Hey, did I ever see Sony? Not sure...

(Oh and IF you browse that group, make sure to hit "L" for screenfilling image for a better impression)
There are lots of images with strong subject separation (or what was called "pop" in this thread) in that group and a handful with "3D pop" (either completely or in part):

(mostly 3D pop for me, blur not as strong due to the perspective and look of the lens)
Agreed, although there aren't really "surroundings", something for the eyes to grasp, but there's still depth (in reality as well as in the image).
(mostly 3D pop for me, also some additional use of blur and perspective elements/lines)
Yes, although the subject isn't clearly defined. It would work much better with something with clear borders, a bit shiny, red. Also, if the wood wouldn't be this saturated, the effect would probably stronger.
(only in part, works well because of the small area in focus, the line to it and the busy blur, which creates the illusion of more DOF than there is)

(only in part, relies heavily on blur, so mostly "pop")
This is 3D Pop to me, but it could be better. There's a very good transition from sharness to blur on the ground. Also there are different objects visible at different distances from the sharp parts which create some form of layers. Still, it would work much better, if the subject itself were completely in focus (which would only work if it were smaller, obviously - without changing the blur on the ground). Also the strong colors on the ground work against emphasizing the subject against its surroundings.
(only in part, also relies on blur - might be 50/50 for me)
I agree, this is more in the category of pop. The transition to the blur is too sudden. The only observable transition is on the dog itself (which makes it a mixed case, alright). The background is recognizable, but only barely so. Further back, no depth can be discerned.
As mentioned before there are of course lots of nice examples which rely purely on blur or a combination of blur + additional factors like colors, lines etc. It's generally pretty rare to see images with pure 3D pop and I don't think this thread showed many effective examples as well.

If you really draw a clear distinction between the categories 3D pop and pop, I think there are not many shots left which fall into the former... and of those the vast majority will likely rely on a strong use of color (like it has been explained here as well) or differences between bright and dark areas. And only after that lens choice becomes a defining factor... So in a way it's not surprising why there are no definitive answers.

For me personally it doesn't make sense chasing after that at all. I think the most effective shots are often a combination of different elements, including things like blur, elements of perspective etc.

I'll add some examples in another post.
I find this to be effective (also on the space shuttle, so no chromostereopsis there):
Space Shuttle (primeuser)

Also good, although subject too wide, not enough space around it. The rest is all very effective. Nice transitions around the subject:
Young woman sitting on metal chairs

I find this to be effective due to the transitions, this time as layers (Ice cream cup in front blurry), back also blurry and more strongly so as you go back:
Boy sitting in front of his ice cream cup

Pop, (no 3D): Not enough depth cues in the background. The young woman pops out of the blur:

Young woman, POP only

Quite effectuve due to the smoke, lowering contrast in the background, making the subject stand out more:

Boy with sparkler

Spiderman ticks many boxes on my list for effective 3D Pop:
very clear cut subject, perfecly sharp, colorful, red;
wall in front has some transition of sharpness from front (left) to back (right). Background is very blurry, but still recognizable.
Could be even better, if the background weren't so colorful and if there were more visible transition to the back.:
Spidey

This works well (even tough is leaves little space around):
Havanna (?)

This works well despite weirdly overworked colors and contrast (the very nice transition, wet and the shiny ground are helping, I think).

Boy on the path
...

--

TheOtherSideOfBokeh
 
Last edited:
I was going to write something even more blunt, but obviously that would just be showing bad manners. I assume the low element count theory comes from the following article, which I think proves nothing:

Low element count and 3D pop
For me, this article is a good example for the mass of BS on Internet:
Glass is a capacitor. ...
Show that the Author has no connection to facts, optics, physics, etc. Reading articles like this poisons the mind.
I feel the same, which is why I overreacted a bit when reading about the low element count.
--

Flickr
TheOtherSideOfBokeh
To me it's an intriguing theory, but we have to stay open for facts. Zeiss is famous for 3d and not low element, so there is a fact. Still I just viewed DPR first A7 review samples, and then they had an adapted lens gallery too, with old Pentax and Vivitar lenses. While I admit the regular gallery was not prime-only, the second gallery looks way better to me. Especially the Pentax SMC shots.

(I looked at A7 because superlight body and possibly fun for small primes)


Bas
 
I was going to write something even more blunt, but obviously that would just be showing bad manners. I assume the low element count theory comes from the following article, which I think proves nothing:

Low element count and 3D pop
For me, this article is a good example for the mass of BS on Internet:
Glass is a capacitor. ...
Show that the Author has no connection to facts, optics, physics, etc. Reading articles like this poisons the mind.
I feel the same, which is why I overreacted a bit when reading about the low element count.
--

TheOtherSideOfBokeh
To me it's an intriguing theory, but we have to stay open for facts.
How can a theory be intriguing if it is based on complete ignorance physics and optics?

Every usable theory must have a solid fundament.
 
I see the same human inability to recognize image properties every time I teach my "cameras as computing systems" course. Students have trouble correctly identifying the various types of artifacts caused by lenses, sensors, etc., even after being taught how to recognize each.
Do they have cameras that they can use to produce the various effects ?
They are supplied with cameras for the course. In the most recent offerings, they use (and reprogram) both a Canon PowerShot SX530 and an ESP32-CAM. They also are shown, and directly handle, about a dozen mirrorless, DSLR, and film SLR bodies and several dozen lenses with wildly varying characteristics.
I would guess that many of them would never have used an ILC.
Before the course, that is oddly true. Many haven't used a DSLR either.
I can imagine that if all they have is phones, teaching about depth of field (for instance) would be difficult. Of course, phones are even more computery than DSLRs.
DoF isn't the hard stuff. It's things like recognizing what (very extreme!) lens defect is seen in this photo:

218863dac53643d7ab53cea29ef5d9e7.jpg

BTW, that's an example of Axial (aka Longitudinal or "Bokeh") Chromatic Aberration from a Vivitar 200mm f/3.5 -- which suffers this problem worse than any of my 300+ other lenses. That said, the colors are boosted a bit to make them obvious when on a laser-printed hardcopy exam, but not boosted as much as you probably think.
Maybe a course called "How phones think" would be effective ? Then you can mention that there are also things called "cameras" that can take wide angle and tele photos, and show examples.
Concepts like wide and tele are pretty well understood. Although, just to be technical (which the course very much is), tele does not mean longer-than-coverage-diameter focal lengths; tele means a lens formula where focal length is longer than physical distance to the center of the lens. ;-)

The problem is that people in general are not good at viewing images critically. By the end of the course, they're OK at this stuff -- but it's a struggle getting folks past the "that's pretty" thinking.

Incidentally, the course I usually do now is a little different: Programmable Cameras and IoT. I still have them do some stuff with PowerShots, such as programming one to be a timer that starts/stops the count when it detects motion or to automatically do optimal-length HDR capture sequences and merge them in camera. In the IoT part, I had them build a little light spectrum analyzer and a movement-tracking wireless network camera, both using ESP32-CAM. Not very different from before, but now there's always a project that combines imaging and direct control of physical devices (servos, steppers, etc.).
 
3D pop is mostly about lighting, composition, and the lens used. It does not necessitate shallow DoF:

[ATTACH alt="Open "original" to appreciate the pop"]3634990[/ATTACH]
Open "original" to appreciate the pop

[ATTACH alt="Open "original" to appreciate the pop"]3634992[/ATTACH]
Open "original" to appreciate the pop
 

Attachments

  • 34a37d0ef00f4cf49224dd08672813ad.jpg
    34a37d0ef00f4cf49224dd08672813ad.jpg
    8.2 MB · Views: 0
  • 0c38cef29d13421693c66f74c45cba03.jpg
    0c38cef29d13421693c66f74c45cba03.jpg
    11.7 MB · Views: 0
Last edited:

Keyboard shortcuts

Back
Top