Luminous Landscape Forum
The Art of Photography => User Critiques => Topic started by: Jeremy Roussak on July 07, 2009, 02:47:59 pm
-
I still think I don't understand bokeh!
Here are the two shots which started the commentary, and a few others. All were taken with the 100-400L on a 5D2. Was I using a bad copy? Can there be a copy (as opposed to a design) which has bad bokeh, though pretty sharp?
All are processed in LR, using the "landscape" capture sharpening preset, a bit of clarity and some fiddling with exposure, blacks and fill light. They're then exported as 80% jpegs with standard sharpening for screen.
It certainly looks as if the bokeh was worse when against the light. Perhaps it's better when stopped down a bit.
400mm, f5.6
[attachment=15191:400mm__f5.6.jpg]
400mm, f6.3
[attachment=15192:400mm__f6.3.jpg]
350mm, f7.1
[attachment=15188:350mm__f7.1.jpg] [attachment=15187:350mm__f7.1_2.jpg]
Any comments?
Jeremy
-
I'm not the official expert, but in simple layman's terms, here's what I see:
In #1, the background has a lot of reddish coloration that blends well with the coat of the deer, so the deer doesn't pop out. In #2, you don't have that problem, since the background is not that color. In #3, there's a tendency toward the problem, but not serious at that point. Although #4 has the same colorations as #3, because of the way the deer is positioned, it also doesn't stand out as well as in #3.
I've seen posts here describing how to alter backgrounds with layer processing that could help with this.
-
A quick once around the net shows a few things to checkout ...
How is it IS on vs IS off? I think Thom Hogan has mentioned bokeh issues with IS. (Or in his case VR. Not sure of the details. I just remember the comment.) Someone claims it is a problem with cheap UV filters. (Not quite sure how but ...) The brighter the background the more distracting the bokeh will be.
google "canon 100-400 bokeh" for more.
-
I still think I don't understand bokeh!
Here are the two shots which started the commentary, and a few others. All were taken with the 100-400L on a 5D2. Was I using a bad copy? Can there be a copy (as opposed to a design) which has bad bokeh, though pretty sharp?
All are processed in LR, using the "landscape" capture sharpening preset, a bit of clarity and some fiddling with exposure, blacks and fill light. They're then exported as 80% jpegs with standard sharpening for screen.
It certainly looks as if the bokeh was worse when against the light. Perhaps it's better when stopped down a bit.
Any comments?
Jeremy
Jeremy,
I'm not sure anyone understands bokeh. First of all, some lenses simply render good bokeh. Other's don't. I've never noticed that shooting against the light made a big difference, but I haven't really tried any controlled experiments. One thing that supposedly makes a difference is the shape of the diaphragm blades. Rounded ones often produce better bokeh. In many, perhaps most cases, shooting wide open will give you the best bokeh because the diaphragm blades are retracted. But I don't know of any guarantee of good bokeh other than simply having a lens that naturally produces good bokeh. I know that both my Nikkor 50mm f/1.4 and my Nikkor 24-70 f/2.8 give me good bokeh. My 24-120 f/3.5-5.6 mid range zoom usually gives me boheh that I don't like. I've seen reviews of lenses where two different reviewers gave completely different evaluations of the bokeh produced by the same lens, so maybe the quality of bokeh is a subjective thing. What we really need to answer the question is a bokeh expert -- if there is such a thing.
-
Jeremy,
I'm not sure anyone understands bokeh. First of all, some lenses simply render good bokeh. Other's don't. I've never noticed that shooting against the light made a big difference, but I haven't really tried any controlled experiments. One thing that supposedly makes a difference is the shape of the diaphragm blades. Rounded ones often produce better bokeh. In many, perhaps most cases, shooting wide open will give you the best bokeh because the diaphragm blades are retracted. But I don't know of any guarantee of good bokeh other than simply having a lens that naturally produces good bokeh. I know that both my Nikkor 50mm f/1.4 and my Nikkor 24-70 f/2.8 give me good bokeh. My 24-120 f/3.5-5.6 mid range zoom usually gives me boheh that I don't like. I've seen reviews of lenses where two different reviewers gave completely different evaluations of the bokeh produced by the same lens, so maybe the quality of bokeh is a subjective thing. What we really need to answer the question is a bokeh expert -- if there is such a thing.
That would be Mike Johnston over at the Online Photographer.
-
That would be Mike Johnston over at the Online Photographer.
I checked out the bokeh articles by Mike Johnston over on the Online Photographer. I found them interesting and informative. Thanks for the suggestion.
-
I checked out the bokeh articles by Mike Johnston over on the Online Photographer. I found them interesting and informative. Thanks for the suggestion.
That site is a great resource. He has also written articles for this site. Glad you enjoyed it.
-
One thing that supposedly makes a difference is the shape of the diaphragm blades. Rounded ones often produce better bokeh.
That's a myth pushed mostly by lens manufacturers. As a matter of fact, the aperture's shape has only minimal effect on a lens' bokeh. Other factors are more significant by one or two orders of magnitude. The single most important factor is the characteristics of the lens' residual spheric aberrations.
In many, perhaps most cases, shooting wide open will give you the best bokeh ...
Just the contrary is true: With most lenses, at full aperture bokeh is at their worst.
I've seen reviews of lenses where two different reviewers gave completely different evaluations of the bokeh produced by the same lens, so maybe the quality of bokeh is a subjective thing.
Most people fail to understand that bokeh, in the first place, is a property of the image, not the lens. What we call "lens bokeh" for brevity should really be, "the way the lens' out-of-focus rendering contributes to the image's bokeh." An image's bokeh depends on many factors, among them (in no particular order) depth-of-field, perspective, background distance, background texture, background brightness, background contrast, and lens properties. So the lens' out-of-focus rendering is just one out of many factors. That's why it is so hard to judge the lens' out-of-focus rendering from an image's bokeh. It is next to impossible to compare two lenses' out-of-focus renderings from images taken at different times under different lighting conditions of different subjects. When one image looks better than the other, it still does not necessarily mean the lens used had better out-of-focus rendering.
-- Olaf
-
Interesting what you say Olaf.
About the starting images, I must confess I smile about that bokeh discussion a bit Maybe its not the best look, but I have decided, for my images, the general composition, etc., is much more important than the look of the bokeh. I will start care about bokeh when I consistently make so good images that the bokeh becomes the limiting factor, that is, when all other factors from my side are well done.
Christian
-
Interesting what you say Olaf.
About the starting images, I must confess I smile about that bokeh discussion a bit Maybe its not the best look, but I have decided, for my images, the general composition, etc., is much more important than the look of the bokeh. I will start care about bokeh when I consistently make so good images that the bokeh becomes the limiting factor, that is, when all other factors from my side are well done.
Christian
Bravo, I like your approach.
Peter
-
That's a myth pushed mostly by lens manufacturers. As a matter of fact, the aperture's shape has only minimal effect on a lens' bokeh. Other factors are more significant by one or two orders of magnitude. The single most important factor is the characteristics of the lens' residual spheric aberrations.
Just the contrary is true: With most lenses, at full aperture bokeh is at their worst.
Most people fail to understand that bokeh, in the first place, is a property of the image, not the lens. What we call "lens bokeh" for brevity should really be, "the way the lens' out-of-focus rendering contributes to the image's bokeh." An image's bokeh depends on many factors, among them (in no particular order) depth-of-field, perspective, background distance, background texture, background brightness, background contrast, and lens properties. So the lens' out-of-focus rendering is just one out of many factors. That's why it is so hard to judge the lens' out-of-focus rendering from an image's bokeh. It is next to impossible to compare two lenses' out-of-focus renderings from images taken at different times under different lighting conditions of different subjects. When one image looks better than the other, it still does not necessarily mean the lens used had better out-of-focus rendering.
-- Olaf
Olaf, Thanks for the very worthwhile summary. I guess it boils down to what I said earlier: "I don't know of any guarantee of good bokeh other than simply having a lens that naturally produces good bokeh."
Interesting what you say Olaf.
About the starting images, I must confess I smile about that bokeh discussion a bit Maybe its not the best look, but I have decided, for my images, the general composition, etc., is much more important than the look of the bokeh. I will start care about bokeh when I consistently make so good images that the bokeh becomes the limiting factor, that is, when all other factors from my side are well done.
Christian, Hear, hear! That's the most sensible statement I've seen about bokeh.
-
I guess it boils down to what I said earlier: "I don't know of any guarantee of good bokeh other than simply having a lens that naturally produces good bokeh."
Definitely not.
-- Olaf
-
Interesting what you say Olaf.
About the starting images, I must confess I smile about that bokeh discussion a bit Maybe its not the best look, but I have decided, for my images, the general composition, etc., is much more important than the look of the bokeh. I will start care about bokeh when I consistently make so good images that the bokeh becomes the limiting factor, that is, when all other factors from my side are well done.
Christian
You might consider learning how your lenses behave.
-
You might consider learning how your lenses behave.
Ah cmon dont assume I dont try to learn about lenses, or am not interested in bokeh-discussions only because I currently disregard bokeh-quality for what I do. Is it really neccessary to add, that I find the discussion in itself perfectly legitimate? I thought this goes without saying
-
Definitely not.
-- Olaf
Olaf, Fair enough, as you pointed out most of the quality of the bokeh comes from the nature of the scene itself, but, given a static setup, same scene, same focal length, same aperture, etc., one lens will render bokeh that's different from what another lens will render. I'll bite. How do you determine whether or not a lens will give you good bokeh other than to try it in various situations?
-
Double Post!
-
Ah cmon dont assume I dont try to learn about lenses, or am not interested in bokeh-discussions only because I currently disregard bokeh-quality for what I do. Is it really neccessary to add, that I find the discussion in itself perfectly legitimate? I thought this goes without saying
If you know one lens is going to render the OOF areas better than another why would you ignore that?
-
Definitely not!
Fair enough, as you pointed out most of the quality of the bokeh comes from the nature of the scene itself ...
Exactly.
... but, given a static setup, same scene, same focal length, same aperture, etc., one lens will render bokeh that's different from what another lens will render.
Yes---that difference is what we call 'bokeh' and actually mean 'lens bokeh'.
How do you determine whether or not a lens will give you good bokeh other than to try it in various situations?
That's the point: various situations. To learn about the quality of a lens' bokeh you have to try it in a bunch of different situations---at different apertures, with low-contrast and high-contrast backgrounds, at various distances, and so on. Many people try judging the bokeh of their lens after just one or two shots. And the comparison of the bokehs of several lenses is another can of worms.
-- Olaf
-
That's the point: various situations. To learn about the quality of a lens' bokeh you have to try it in a bunch of different situations---at different apertures, with low-contrast and high-contrast backgrounds, at various distances, and so on. Many people try judging the bokeh of their lens after just one or two shots. And the comparison of the bokehs of several lenses is another can of worms.
Okay, we've gone around the flagpole twice, but it seems to me we're back to what I said earlier: "I don't know of any guarantee of good bokeh other than simply having a lens that naturally produces good bokeh."
In any case, Olaf, you're the first person I've run across who had some actual information on bokeh. Thanks.
-
Sorry, another double post. I'm having some problems with my web connection.
-
Christian, Hear, hear! That's the most sensible statement I've seen about bokeh.
Overlooked your answer - Well I came to this out of my practical experience, the bokeh in my images always looked decent, it was just never so important for me, or so bad like in the images here.
And I must say Olaf has a strong point here. Although I was aware that my lenses produced very different bokeh for all sort of settings, scenery and light, I never really thought about it so much to actually draw the conclusion he wrote. (And probably could not have said it so good...) For example, I was always pleased that for its price, the 18-55 Kit from Canon has a very good bokeh in Macro situations. But NOW I would not be surprised if someone pointed out that pretty much ALL macro images have good bokeh in most situations. So for Kikashis case that would mean, the lens is not worse or better then any other sample, instead the particular shooting situation was just bad. I think thats the point Olaf was trying to make.
Christian
//edit: Overlooked, not oversaw
-
Overlooked your answer - Well I came to this out of my practical experience, the bokeh in my images always looked decent, it was just never so important for me, or so bad like in the images here.
And I must say Olaf has a strong point here. Although I was aware that my lenses produced very different bokeh for all sort of settings, scenery and light, I never really thought about it so much to actually draw the conclusion he wrote. (And probably could not have said it so good...) For example, I was always pleased that for its price, the 18-55 Kit from Canon has a very good bokeh in Macro situations. But NOW I would not be surprised if someone pointed out that pretty much ALL macro images have good bokeh in most situations. So for Kikashis case that would mean, the lens is not worse or better then any other sample, instead the particular shooting situation was just bad. I think thats the point Olaf was trying to make.
Christian
//edit: Overlooked, not oversaw
This makes me curious. We can get all kinds of books on photographic principles, editing programs and techniques, even developing techniques for film work. But other than specific (and sometimes expensive) books written for a specific camera like you see in Barnes & Noble etc., are there texts that cover this sort of thing in detail, with a comprehensive list of camera bodies and lenses? I'm guessing it would be a great seller, since every camera and lens combination seems to incorporate a lot of black magic (or it just accidentally works in most cases).
-
This makes me curious. We can get all kinds of books on photographic principles, editing programs and techniques, even developing techniques for film work. But other than specific (and sometimes expensive) books written for a specific camera like you see in Barnes & Noble etc., are there texts that cover this sort of thing in detail, with a comprehensive list of camera bodies and lenses? I'm guessing it would be a great seller, since every camera and lens combination seems to incorporate a lot of black magic (or it just accidentally works in most cases).
Im afraid if the bokeh in a certain situation is horrible, you have very limited possibility to change that, so such a book would become superfluous. Furthermore, people working in film or in a studio situation know these situations from experience anyway. You learn that, you dont read about it. From a physics or engineering standpoint however it might be interesting, but its not neccessary for shooting. Maybe lens engineers or camera mens who use a lot of different lenses could probably sum this up.
If you where really curious about this you could grab 2-4 different lenses of the same lenght, a tripod, and with some lens changing make comparisations in the field in order to get a feeling for it.
-
http://theonlinephotographer.typepad.com/f...ehrankings5.pdf (http://theonlinephotographer.typepad.com/files/bokehrankings5.pdf)
-
Thanks, Dark. I'd say that that article is dispositive. The bottom line seems to be that good bokeh is in the eye of the beholder.