Bokeh

[Sandy Vongries]

[Nick D.]

So, then what's the point of buying a lens with VR/IS/VC, etc? :mad::confused::(

VR is a weapon of last resort, when light is low and you have to use slow shutter speed, can't use tripod, etc. Looks like everybody become obsessed with VR or IBIS, in reality they just introducing another moving element in lens, camera or both. VR is trying to keeps subject in focus by moving lens element or sensor, but projection of out of focus parts of image moving a bit differently, creating halos:( Nothing beats good old heavy tripod.

And they can charge you some extra for new feature, always good for manufacturing company:)

Edited February 26, 2019 by Nick D.

[Ed_Ingold]

Image stabilization (IS) tries to keep the subject steady, by sensing small, random motions due to camera shake. Continuous focus (AF-C) tries to keep the subject in focus.

 

IS can work against you when using a tripod, because it tends to hunt in the near absence of camera shake. This can be observed slow drift in the viewfinder, or as doubling, at the pixel level or even macro level. Modern lenses with IS seem to avoid these problems, but may cause an annoying lag in response when you pan or tilt, a rubbery effect. Used wisely, IS is a tremendous asset, nearly eliminating the need for a tripod.

 

Continuous focus is annoying when the subject moves out of range of the focus spot, causing the lens to hunt, or you attempt to use the focus-and-recompose technique. Some cameras use pre-focus (without pressing the shutter release) to speed response when you're ready to take the shot (optional in a Sony A7xxx), which is a huge drain on the battery as you walk about.

[steve_gallimore1]

I think something that is often overlooked in these discussions is that the background and the subject/background distance contribute significantly to the appearance of the final image.

 

I've been wondering about the odd star-shaped iris of the Industar-61 I posted earlier in this thread. I think the designer's intention was to produce a less distracting bokeh than the 'bubbles' typical of most lenses at the time. The 'stars' have less discernable edges than circles or hexagons at normal viewing distances.

 

Hexagons and pentagons are the most distracting and 'artifical' looking, in my opinion.

[Ed_Ingold]

To be clear, IS compensates for motion of the camera, not the subject. These motions are sensed by accelerometers for pitch, yaw, roll and translation.

[Ricochetrider]

I sometimes use the blurred background to help frame the shot...

 

[davidrosen]

Thank you Apple for making bokeh a verb, as in, “You bokehed my child in the background!”

[Ricochetrider]

I think something that is often overlooked in these discussions is that the background and the subject/background distance contribute significantly to the appearance of the final image.

 

 

I think so too, see my shot above where I framed the shot around the background. I don't always do this but I feel the bokeh'd background can sometimes be significant in setting the scene and/or tone of the composition. This shot, BTW was taken with my Hasselblad 500CM and Zeiss 80mm f2.8 lens.

[steve_gallimore1]

More Industar-61 strangeness, I do enjoy playing with this lens.

 

First, the aperture, is there any other lens that has this? (I know of the Volna-9 and a Mir)

f2.8 (wide open) f4

 

f5.6 f8

 

f11 f16

 

Sorry, it's not easy to juggle a lens in one hand, the camera with a macro lens in the other, try to line everything up and focus on the aperture.

 

Now the results of a little test this afternoon, which netted me a couple of nice portraits of my long-suffering partner.

 

f2.8

 

 

f4

 

f5.6

 

f8

 

 

f11 would have mostly put the background in focus, so I stopped there, but it would have been hexagonal.

 

 

There you have it, I give you, the Industar-61 LZ, choose your bokeh!

 

 

Strangely, most of the time, I don't really give it much thought.

[Ed_Ingold]

Lenses which produce a "soap bubble" effect for OOF point sources, bright rim, dim center, tend to produce granular, busy bokeh. Placement of the diaphragm relative to one of the nodes of a compound lens affects the shape of it's image in OOF areas. The Industar lens above is a classic example of failure on both counts.

[ben_hutcherson]

Hexagons and pentagons are the most distracting and 'artifical' looking, in my opinion.

 

Here's a lens that at least at one time was quite popular and seems to have well-liked bokeh wide open, but is somewhat divisive when stopped down.

 

A lot of my use of this particular lens is at or near infinity, and I don't have a scan at hand of an image that really shows an OOF background. I've never been too bothered by it, but then when I'm using it at closer distances I also tend to have it wide open or nearly so.

 

I'll also mention that I have three other lenses with the same basic aperture-perhaps the 250mm f/5.6 would be the one to really look at...

 

[steve_gallimore1]

Lenses which produce a "soap bubble" effect for OOF point sources, bright rim, dim center, tend to produce granular, busy bokeh. Placement of the diaphragm relative to one of the nodes of a compound lens affects the shape of it's image in OOF areas. The Industar lens above is a classic example of failure on both counts.

 

Oh, I agree, it's very much a 'special effects' lens, I wouldn't normally use it for portraits.

 

I did rather provoke it though, shooting at 3pm with the sun through the leaves.

 

Out of curiosity, I think I'm going to try the same shot today with a selection of different lenses.

[Uhooru]

Many nice examples here, but no need to worry about it anymore. It seems Apple will provide some "bokeh" feature on an iPhone camera...:)

but back to the thread. . .

[steve_gallimore1]

I was a bit late and the light wasn't the same, so the results are less extreme, I think.

 

I've tried to crop the same region in each photo, the pine tree branch is standing in for my wife.

 

In no order.

 

Konica Hexanon 40mm

f1.8

 

f2.8

 

 

Macro Hexanon 50mm

f3.5

 

 

Jupiter-8 53mm

f2

 

f2.8

 

 

Mir-1 37mm

f2.8

 

 

Helios-44 58mm

f2

 

f2.8

 

f4

 

 

The Helios (CZ Biotar) and Jupiter (CZ Sonnar) are certainly the most dramatic of the bunch, the Konica 40mm performed better than expected, I was expecting hard-edged hexagons, the Mir blew-out the top of the image.

 

Interesting how the bokeh of the Helios continues to blur the background even at f4, I wouldn't have thought that the 8mm difference vs the Macro Hexanon would make that much of a difference.

[glen_h]

Thank you Apple for making bokeh a verb, as in, “You bokehed my child in the background!”

 

There is a quote, supposedly attributed to Noam Chomsky: "Any noun can be verbed."

 

(Note that the spell check is red underlining verbed.)

[glen_h]

Now that we have discussion for bokeh from actual lenses, there is now fake bokeh

from the new iPhone. (And, I hear, some older phones.)

 

So we don't need real bokeh anymore?

[steve_gallimore1]

Now that we have discussion for bokeh from actual lenses, there is now fake bokeh

from the new iPhone. (And, I hear, some older phones.)

 

So we don't need real bokeh anymore?

 

Can we have some more phoney bokeh please? I'm actually curious to see it.

 

 

Sorry, the pun was too good to resist.

 

 

I'm an engineer and the technical aspects of photography, the mechanics, optics, chemistry have always interested me, but I think one of the things that has really hooked me is that there is always that little spark of magic in the optics and the chemistry, that 1% pure luck that creates the character, differentiates one formula from another and makes art out of science.

 

Otherwise we'd all be using the same lenses, film and developer.

 

 

So how does the computer-generated stuff stack up? I've written a few small programs here and there, so I know there can be art in algorithims too.

[The Shadow]

It’s interesting that bokeh generated by an iPhone would be called “fake.” Does that mean bokeh generated by a lens would be “real?” If so, then what’s generated by squinting my eyes or crying? I think the starting point may be to recognize the significance of the artifice of photography and art, at which point the supposed difference between what’s real and what’s not starts to ... um ... blur! :rolleyes:

[glen_h]

 

There’s always something new under the sun.

 

 

Reminds me years ago, there was a (partial) solar eclipse, and we didn't have anything to view it with.

 

Then I noticed that the shadows under a tree, through the leaves moving around, were all crescent shaped.

 

That is, shadow bokeh!

 

We normally don't notice that shadows of things like leaves are circular,

as we also mostly don't notice circular bokeh, but do notice when it isn't circular.

[DavidTriplett]

I think the starting point may be to recognize the significance of the artifice of photography and art, at which point the supposed difference between what’s real and what’s not starts to ... um ... blur!

Which takes us back to the Nikkor DC lenses... Is their bokeh "fake" because it can be controlled and manipulated? I don't think anyone would suggest so. Hence, what if we manipulate using electronic, rather than optical means? Seems to be a circular argument. I think the only issue that matters is, in the end, the final image.

[Ed_Ingold]

I'm confident that some genius (genii) will produce software with convincing bokeh, much as software implements many audio effects in nearly perfect fidelity. Convolving reverb software traces sound waves in real auditoriums using a battery of microphones, then reproduces those patterns on each byte of the audio file. Closer to home (i.e., PNET), ray-tracing software not only facilitates lens design, but emulates the results in the form of MTF curves.

[Timo Hartikainen]

Pentax FA 50mm f1.4 @ f2.8

[ben_hutcherson]

Which takes us back to the Nikkor DC lenses... Is their bokeh "fake" because it can be controlled and manipulated? I don't think anyone would suggest so. Hence, what if we manipulate using electronic, rather than optical means? Seems to be a circular argument. I think the only issue that matters is, in the end, the final image.

 

I think the key difference is in the fact that the DC lenses DO achieve it optically by over or under correcting spherical abberation.

 

As I understand it, the iPhone basically takes two photos-one in focus and one out of focus-and then uses software to pick out the subject and overlay the "sharp" subject with OOF.

 

The effect works decently well when you have a simple image-a portrait of one or two people in front of a relatively distinct background.

 

it can get a bit more complicated when you have more going on in the image. As we all know, there really is no such thing as depth of field, just one plane of sharp focus and an amount of blur that's considered acceptably sharp. When the OOF areas are captured at the same time as the subject, you can see a smooth drop-off in sharpness both in front of and behind the subject, something that can't be replicated the way the iPhone does it.

 

In addition, I've seen examples where the algorithms didn't do a perfect job of identifying all the regions of the subject, and gave some strange artifacts(although I'm sure this can be and is continually improved). I've also seen places where subjects at very different distances from the camera were in focus, but the background was still blurry-something that really doesn't make sense to me.

 

I will admit to my own attempts to blur backgrounds in post before. My technique has been to select the subject(s) and the apply a Gaussian blur to the background. I will only attempt it when there's a VERY obvious and clear delineation between the two, and usually also when I unintentionally got something distracting in the background. My attempts have mostly just made a mess :) , although it's given a result I was happy with once or twice. I've even attempted something similar in the darkroom by cutting a mask for the subject(a LOT of work) and then exposing the background out of focus. Both doing it digitally and on an enlarger, I've usually ended gone to other cliché techniques that are a bit more repeatable, like adding some vignetting to draw attention to the subject and away from a distracting background.

 

I think that the "iPhone way" can and does give good results in very limited circumstances, but overall I'd rather just do it in the camera.

[glen_h]

Which takes us back to the Nikkor DC lenses... Is their bokeh "fake" because it can be controlled and manipulated? I don't think anyone would suggest so. Hence, what if we manipulate using electronic, rather than optical means? Seems to be a circular argument. I think the only issue that matters is, in the end, the final image.

 

It seems to me that you could make real fake bokeh from the images from Lytro:

 

Lytro - Wikipedia

 

which has enough information to figure out the distance to each object in the image.

 

I suppose with a stereo pair, you could also figure out distance.

[The Shadow]

I think that the "iPhone way" can and does give good results in very limited circumstances,

I also try to keep in mind that my work with iPhones and my digital work in general doesn’t necessarily have to be trying to do or come up with results that look like what more traditional cameras or film came up with. Just as digital has some possibilities that are unique compared to film, so iPhones have some possibilities that can be achieved precisely by NOT seeking similar results to other cameras or mediums. Experimenting with and exploring iPhone photography for its own unique qualities instead of or in addition to ways in which it emulates other more traditional means and looks has a lot of potential and can be exciting.

 

Early on in the history of photography, photographers assumed they had to emulate traditional painting in order to be accepted as artists or even as being of any significance. Pictorialism resulted from that mindset and has much to recommend itself and is still of great value. But photographers moved on and eventually found that photography was its own unique medium and didn’t have to live up to the aesthetic standards of the more traditional arts. That lesson can apply to iPhones. It’s not that we have to give up completely on traditional notions of what a photo or blur or bokeh look like, but we can also move on and embrace those qualities of iPhone photos that won’t resemble what we’re more used to but that may offer exciting possibilities for new aesthetics, new looks, and new paradigms.