The Bokeh Thread

[kyledroid]

Some may find this topic silly, while others may think it important. Truth be told, the jury's still out for me; however, let's start a discussion on the importance of Bokeh. What it means to a filmmaker, and how to achieve it. Most of you DIY'ers probably already know what Bokeh is, but for those who don't here's a recap:

Definiton: Bokeh (from the Japanese boke ぼけ, "blur"[1]) is a photographic term describing the subjective aesthetic qualities of out-of-focus areas in an image produced by a camera lens. (sited from http://en.wikipedia.org/wiki/Bokeh)

While doing research on this, I found some things interesting. For example, a good bokeh is usually produced from the camera lens. However, when dealing with dof adapters, ground glass must factor in as well. But the question is: When dealing with Bokeh, how much of it rests on the GG you use and how much does it rest on the camera lens? Thats why i started this thread. To shed light on what exactly causes bad bokeh. Is it just the GG, or it could it be more. Btw, here are two examples (of Bokeh) that clearly illustrate it's importance.


Example 1: Good Bokeh http://www.luminous-landscape.com/images18/bokehinpics.jpg

Example 2: Bad bokeh http://www.kenrockwell.com/nikon/18200/images/KEN_0193-bokeh.jpg

Of course, a matter of opinion is subjective all the way. All you readers out there might find the second picture more stimulating thant the first. Thank you for reading. And MAY THE BOKEH BE WITH YOU. http://www.dvxuser.com/V6/images/icons/icon10.gif

[-D-]

Okay, so, you've said the second one has a bad bokeh: why?

[egproductions]

-D- read the caption of the picture that will answer your question (its not a a bad example of bokeh, its a "bad example of bad bokeh" the two bads negate themsleves making it good bokeh)
In all seriousness I dont find the second picture to have bad bokeh per say but rather the wrong choice of background to throw out of focus. but not something to do with the lens doing a bad job becasue of the shape of the iris or anything like that.

[MattinSTL]

I don't know where you'd put this thread... I think most DIY'ers won't latch onto this unless it's about DOF adapters... Okay so maybe this thread does belong here.

I'm glad you started this thread... I was just talking about this with a friend last night as we were discussing the value in one lens over another. I'm a bokeh freak man... I love it.

Reflective lights or highlights in the BG should be uniform disks of light... no hot spots or rims... basically a good bokeh background should look like a blurry picasso.

Look in the second shot... at the light disks near the top center... they have outer rings and center bullseyes... things like that draw your eye away from the intended subject.

[tigeba]

This is shot with a Canon 85mm 1.8. I consider it decent, but not fantastic bokeh. You can see that it is very smooth and not splotchy like the "bad" example from above. I would say this is proably typical of most 85mm lenses as they are generally designed to produce good bokeh, sometimes at the expense of sharpness and distortion (e.g. Canon 85mm 1.2L)

http://static.flickr.com/39/87614846_3e2ba2fdd8.jpg

[MattinSTL]

Yeah that's good bokeh... I'd like to see more shots from you. Also let's see some bokeh where you can tell the environment also by the BG... like a city scene or something. I know that's a matter of preferrence and you take all kinds of shots, but I get my bokeh buzz when I can tell what the BG was and it looks like a painting of reality.

[tigeba]

I will throw some more stuff up this evening. I have spent the last few weeks loading grillions of photos into Apple Aperture. I think I can find some shots similar to what you are talking about.

I will go ahead and throw these in, since the whole reason I uploaded them in the first place to discuss bokeh and dof differences with various lenses.

The previous photo was taken at f1.8

The flower below is f2

http://static.flickr.com/36/87614845_83083f98f0.jpg

The statue is f4

http://static.flickr.com/43/87614844_7890338b1c.jpg

[kyledroid]

Giving examples of bad Bokeh is subjective, because its a matter of opinion. I found the second one to have bad bokeh for the reason that the out of focus areas seem to have some sharpness which distracted my eye from focusing on the branch. But like i said in my first post, and reiterated here, it's all a matter of opinion. Also, id love to see tests done with these dof adapters to see how much bokeh relys on the lens and how much it relys on the ground glass.

[TimurCivan]

its the lens. if its optics are good, it evenly disperses light into a smooth looking out fof focus area. if its a shady lens you get thoes HAloish looking out of focus areas.

GGmaterial shouldnt matter too much. since its just like trandslucent film surface, and the DVX would be reading that image. the biggest worry for GG materials is the dreded hot spot in the middle. if its not diffusing the light enough you get that hot spot inthe middle.

[ASG]

GG shouldn't matter too much? Nothing could be further from the truth Timur. The amount of diffusion of the GG is ABSOLUTELY CRITICAL to the bokeh you get on your adapter.

With the hotspot, if you add more diffusion you also get a worse hotspot. too little diffusion the hotspot is due to aerial image, then high diffusion it's down to lens falloff. hotspot is easy to fix so IMO it's not the main concern, for me it's the bokeh.

About "shady" lenses - Good lenses can also produce bad bokeh. I spent a lot of money on a Nikkor 50mm f1.2 and it produces a bokeh with a bright edge on film, technically bad bokeh. But when I use it on my adapter the bokeh is somewhat softer than on film, and it looks good to me.

To summarise, the influece of the lens bokeh properties is insignificant next to the influence the GG has on the bokeh your adapter produces. Sure once you have a good GG you'll notice the difference in bokeh between various lenses, but if your gg produces a soft hazy bokeh the you won't be able to see the bokeh properties of the lens.

[TimurCivan]

hmmmm i want to test now........

i can see your point. but i am not sold on the idea that GG makes a huge difference in the appearence of bokeh. because if it changes bokeh that means it will significantly change the infocus parts too. Thsi just means somethign more to test in the 35adapter extravaganza.....

[Wayne Kinney]

The Bokeh is fixed to the 35mm lens, but the GG also effects how it 'renders' that bokeh, depending on its level of diffussion.

Francois Poitras from dvinfo posted this, which is a great example:


I did test a variety of GGs. Here are some results showing the difference between a less diffusive GG and a more diffusive GG in the exact same lighting conditions. The diffusers used in this case are engineered diffusers which have a 10° difference in angle FWHM.

Camcorder Panasonic PV-GS400
Vibrating adapter
Canon 50mm F1.4

Less diffusive GG, camcorder at F4 (http://www.ltfp.com/clips/DiffuserA_CamF4.jpg)

More diffusive GG, camcorder at F3.4 (http://www.ltfp.com/clips/DiffuserB_CamF3_4.jpg)

By the way, both these setups are unusable because of grain showing in good lighting.

You can see on the CamF3_4.jpg image, the highlights of the 3 balls are more disk shaped.

[ASG]

hmmmm i want to test now........

because if it changes bokeh that means it will significantly change the infocus parts too. Thsi just means somethign more to test in the 35adapter extravaganza.....

Generally I don't think so, the aerial image and the diffused image converge on in focus parts so the diffusion level doesn't affect those parts of the image much. What can happen with high diffusion sometimes is those in focus parts becomes softer, like if you use two diffusers face to face you get a fully diffused bokeh but the in focus stuff is never quite in focus, it's not nice to look at.

Of course we must defer to the man with three balls.

[MattinSTL]

I don't think bad bokeh is subjective at all... If you only show a person a single picture of a good subject, but with bad bokeh... and they have no photography experience then they may still think it's a good shot... but I think you'd be in the minority if you called that second shot above good bokeh.

With just a little understanding of photography it gets easy to spot good bokeh from bad... most people who have little to no experience evaluating photography can't explain why they prefer one shot over another (in an apples to apples bokeh showdown)... but they will always pick the shot with the better bokeh.

If DOF matters then bokeh matters... bokeh IS DOF... or at least the quality of it. When you get a jaw dropping shot everybody just sees the crisp part of the image and thinks you did an incredible job... but it's just like any aspect of production... how man times have you watched a movie or tv and because you now do video production you can spot all the techniques? I sit there and laugh at good lighting on people's faces in places where there would be no light... and things like that... all the time analyzing.

Well bokeh is the same thing... in the above scenario where everybody praises the great shot for the crispness of the subject... unless they know better they don't realize that the QUALITY of the BG, which is often times more then 50% of the visible frame... is the foundation of the quality of the subject.

Your own eyes do not see bokeh like a lens sees bokeh... but it's a lot closer to the good bokeh then the bad... I have terrible vision w/o glasses or contacts and I can speak from experience... the smoother and more progressive the DOF fall-off is... the more real the shot looks. Good bokeh doesn't sharpen random details in the BG or manipulate the reflected light.

Tigeba... yes I have an 85 1.8 on the way right now... for the very same reason.

Anybody else that wants to know, go to pbase.com and pick any 85 1.8 for sample photos... it won't take long to understand what good bokeh looks like.

[TimurCivan]

The Bokeh is fixed to the 35mm lens, but the GG also effects how it 'renders' that bokeh, depending on its level of diffussion.

Francois Poitras from dvinfo posted this, which is a great example:



You can see on the CamF3_4.jpg image, the highlights of the 3 balls are more disk shaped.


well i'l be damned it does make adiffrerence. ok, we shall test bokeh as well.

i found the sgpro had nice bokeh anyway. Waynes GG is magical.

[Dennis Wood]

My testing so far has involved using a 50mm lens attached to a Minolta SLR lens with shots taken on different speeds of film, compared to the same lens and framing on the Brevis/video camera. This is the only way to objectively evaluate the differences. It's also an interesting visual exploration of the apparent exposure range compression occuring with the adapter. The Brevis is a very good film match at f 2.8/ f 4 and up. I don't believe any adapter can exactly replicate film bokeh at f 1.2 without very large hits in terms of light loss and visible grain. If the adapter is unusable at F2.8 due to excessive grain showing up, it becomes very limiting. Further, extremely low DOF is not the norm in my observations of film footage. So, as has been discussed elsewhere, all of the adapters will have their unique characteristics. We're moving forward nicely with developement of a 2 diffuser system which I hope will address all of these potential choices with the Brevis implementation.

Your testing should run the full range from the lens aperture fully open to fully closed to evaluate changes in DOF, diffusion, and hokey (ha) bokeh. Taking a film shot with an f1.4 lens on a sunny day will require in excess of 1/1000s and application of ND filters to even take the shot. My trusty old SLR camera tops out at 1/1000s and choked on these shots.

[TimurCivan]

yea these tests are going to be kinda arduous.

thats fine it will be fun.

[kyledroid]

Bokeh quality isnt subjective, but opinions are. I think thats what we have to remember. If u like the bokeh yur adapter is giving off beit the lens and/or the gg, then great. If not, then this thread and other threads like it should help us dof enthusiasts out by discovering what exactly is causing good and bad bokeh