When Will The Bokeh Craze End

[smoke665]

There is an interesting thing about background blur, and that is that the larger the aperture in millimeters the higher the level of background blurring. This is why lenses such as the 300 millimeter F / 2.8 gives such a high degree of background blurring at F 2.8 in comparison to say a 50 mm lens even when it is shot wide open at say F / 1.4.

Partially true. As in all things in life, there's always a "but". LOL

Background blur vs focal length, absolute aperture and relative aperture From the referenced article: assuming an "absolute aperture" the blur circle is disproportionate to the ratios of the focal lengths. For example a 400 mm lens would blur 12 times more then the same aperture on a 50 mm, but there's the distance from the subject required to get the same framing. The background blur circle (all else being equal) becomes greater the closer you are to a subject, which would give the 50mm lens the edge. Where the background blur becomes larger on longer focal lengths is when you get into "relative apertures". Lastly format has a bearing on the blur circle as the smaller the format the less background blur.

As a comparison to the above link I refrenced, here are some examples I've shot. In this first one 35mm @f/3.2, I don't remember the distance but I'm assuming it was close to the minimum focus distance on the lens, which is 12"

Going up to 135mm @ f/5.6, also shot at the minimum focus distance for the lens at approximately 20". Note the nice Bokeh balls.

Finally a 300mm @ f/6.7, also shot at the minimum focus distance for the lens, at just over 5'.

Three different lenses, three different apertures, and three different focal lengths, all distance adjusted to account for the minimum focus difference. The only difference really is the quality of the blur, dependent on the ability of the lens. The first two were moderately priced Pentax, and the last was a used $75 Sigma 70-300mm. Also proving you don't need to shoot at f/1.2 and add fake blur post to get a nice background.

 

[Tropicalmemories]

Couldn't we just let people do what they like for themselves? I've seen many applications of fake blur that were done beautifully.

Certainly. For me, there is something rewarding about doing something naturally but then if I held true to that I would shoot film.

Film? Bah! New fangled 19th Century chemistry and witchcraft. Almost as bad as glass plates.

If you want natural, use a camera obscura.

Or oil paint and an opium and absinthe addict artiste.

There's nothing natural about silver oxide or silicon chips, so just enjoy whatever photographic style takes your fancy.

And let's be honest, beyond the wonderful characters on this Forum, and a few long suffering close relatives, who really shows any genuine interest in the images we create?

"If I am not for myself, who will be for me?"

 

[JBPhotog]

And let's be honest, beyond the wonderful characters on this Forum, and a few long suffering close relatives, who really shows any genuine interest in the images we create?

My clients! LOL

 

[thereyougo!]

I think your question is impossible to answer. The latest generation of sophisticated cell phones has introduced remarkably good fake bokeh effects, and it might serve to prolong the trend. It's really impossible to say.

I think the use of fake bokeh in phones might actually serve to kill it off because of snobbery from some. Personally I'm nt a fan of bokehlicious shots that have no particular shot other than to demonstrate how someone has used their money to fuel their GAS.

 

[Braineack]

As good as computography is getting; the fake stuff will never look as good as real expensive glass.

It's more about "background blur" and how it renders everything -- including the transition into the blurred background.

 

[Derrel]

If you read the article carefully the author uses a disingenuous term. What I wrote is actually technically correct.I wrote The Wider the aperture in millimeters. He tries to equalize what would be f/1.4 on the 50 millimeter lens is about roughly let's say f4 on a 300 2.8. What I wrote is actually technically correct. I wrote The Wider the aperture in millimeters. He tries to equalize this by saying a picture with for example a 50 mm lens and aperture that is 25 mm across on a 300 millimeter lens would not be F2, as it would be on a 50 mm lens. Ergo what I wrote is actually technically correct. The author of the above article disingenuously calls this the same absolute aperture .. what I I was referring to was actual width of the aperture measured in millimeters.Take a look at what f/4 on the 200mm f/2 VR lens looks like and then shoot the same aperture value (f/4) on a 50 mm length lens. Yes indeed there is always a"but: and in this case the" butt" is the author of the above article that you linked to who uses a very disingenuous method of referring to apertures: F 2.8 on a 300 mm telephoto is a big hole. It's the same in Reverse with diffraction: on a 180mm relatively "normal lens " using 4 by 5 inch sheet film a small aperture such as f/64 is actually a pretty good-sized hole and diffraction at f/64 is not a real problem when using a 180 mm lenes on 4x5 by inch film, and in fact the light that goes through a physically-wider aperture does not cause a loss of sharpness due to diffraction, which is one reason underline the name of the Group of f64. On a small-format camera such as a 35 mm film camera, a 50 millimeter lens has long been considered normal. If we were to use a small aperture such as F/16 on our 35 mm film camera we get quite a bit of loss of sharpness due to to diffraction. using a much larger lens with a larger piece of film, apertures are correspondingly larger and small aperture values such as f/16 on a view camera with 4×5 inch sheet film and a 180mm normal lens causes virtually no loss of sharpness due to to diffraction. On a smaller camera with a much smaller piece of film or sensor, small aperture values like f/16 cause a serioud loss of sharpness as the light rays are forced through what is,physically, a narrow opening.

The author you linked us to is quite clever in talking about aperture values in terms of absolute width and his mathematical graphing is correct as far as it goes, but in the real world we actually look at relative apertures, relative to the focal length of the lens that is in use and relative to the actual in- camera exposure settings. As working photographers we refer to apertures by their F value, and not by their actual width as measured with a ruler For example if I take a 300mm f / 2.8 lens and shoot it at f/2.8 that is a much wider physical aperture than F2.8 is on a 50 or a 24 mm lens, Ergo at f/4 a 300 mm gives me more background blur than does a 50mm shot at at 4. The guy is basically full of b******* and is demonstrating his technical knowledge of photographic arcana, but he is doing a huge disservice to people who are trying to understand what real lenses do in the real world. For example if you want to shoot Sports in poor light and want a defocused background you'll get a more-blurred background with a 300 millimeter lens set at F 2.8 then you will with a shorter focal length if the same f-stop value. I know what the h*** I'm talking about, and the guy above is basically full of b******* and just making a very arcane and non-useful point. I have never seen a photographer measure the absolute aperture width and say, " oh, I would like to shoot at an aperture that is 25 mm in widt." That's not how we do it.
Trust me,I spent $7,000 on a pair of lenses that would completely blow out my backgrounds. There is a reason for the existence of the 400 millimeter F / 2.8 lens, And the 300 mm F 2.8 lens, and the 200mm f/2 lens. Let us look at the 200 mm f/2..... at f/2 the physical width of the aperture is roughly one half of the length of the lens, or 100 mm across. With a 50 mm lens f/2 is half of the length of the lens, or roughly 25 mm in width. With the 200mn f/2 VR Nikkor your backgrounds are blown out as shown by the above two sports photos that I took.Had I used the same relative aperture, f /2 and a shorter lens, the backgrounds would have had a much greater degree of recognizability; as we go to longer lenses, the same aperture, as an f/ value or put and other way "in relative terms", is greater. What I mean to say is that on a 200 millimeter lens F2 is a huge hole, 100mm across.... on a 50 mm lens f/2 is a hole that is a mere 25 mm across. Based upon calculations in the article that you referred us to, I believe that we would have roughly 20 times more defocus when shooting at f/2 with the 200 mm lens then we would have with a 35 millimeter lens and shooting at the same F value , f/ 2.0 Sometimes you have to pay very close attention to what is being written to understand an issue or a point of practical photography . What I was saying is that that as focal lengths get longer,the amount of background blur increases substantially as focal length gets longer because as I have here pointed out, with longer lenses fractional aperture values ( f / values) become hugely greater in area. For example take a look at the huge hole in the 200mmf2 at f/2, and look at the modest width of the aperture value of f/2 in a 50 mm lens.

Your appeal to Authority and quick web search goes against what hundreds of thousands of experienced Shooters understand, and that is why I so strenuously object to the way the author of the article you linked us to uses some really disingenuous ways of characterizing this issue. At f/4 let's say, a longer lens produces greater background blurring than does a shorter lens. Shoot your 35 mm prime lens at at f/4 and backgrounds are not all that blurry. Shoot your 300mm telephoto lens at at f/4 and backgrounds are fairly blurry. A quick web search can turn up all sorts of support for almost any position, but 45 years in photography has shown that the physically wider the aperture the more the background is blurred. An f/4 300 mm telephoto produces a pretty good degree of background blur and the hole width is how big? 1/4 the width of the length of the lens. With a 50 mm lens f/4 is roughly 12.5 mm across, and backgrounds are not that blurred. Your author is attempting to equalize f/values which are fractional values which take into account the length of the lens in use.... I have yet to encounter a light meter that measures light according to the measured width of the aperture opening itself, but instead we use a fractional value or an F stop reading.

 

[Derrel]

I would encourage you to re-read what I actually wrote. Take a 300 mm lens and set it to f/2.8 and with even the same framing of an object at different distances go out and photograph a box that spans the width of the frame. You will see that with a 24 mm lens and at f/4 and a 50 millimeter lens at at f/4 and a 300 mm lens at at f/4, that despite the same-sized object across the width or height of the frame that in the 300 mm lens photo the background is very blurry, and in the 50 mm shot the lens produces a fairly recognizable yet still blurred background, and with the 24 millimeter lens you will have a background that is easily-recognizable despite all three photos having roughly the same "depth of field", your actual photographic proof will be revealed. As lenses grow longer the physical width of the aperture in f/value increases substantially. If we measure, in millimeters the actual width of the hole, we can jigger things around and produce a really cool graph as did your author. But we're not after graphs here , we are attempting to understand how background blur actually works in the real world, using f/stops and shutter speeds and actual lenses to make actual pictures.

 

[Derrel]

A really good read, including a particularly descriptive photo illustration comparing a 50 mm lens , and an 85 mm lens, and a 135 mm lens is to be found at the following article: Bokeh and Background Blur, by the author Bob Atkins. He shows three photos of a camera box. The box is the same size in all three photos, and the camera-to-subject distance has been changed to get the same-sized foreground object, which is a Canon 20D camera box. All three photos were taken at the same aperture value which was f/2.8.

With the 50 mm lens the physical width of the aperture was 17.8 millimeters across, with the 85 millimeter lens the aperture width was 30.4 millimeters across, and with the 135 mm lens the aperture value of f 2.8 was represented by a hole that was 48.2 mm across.

As I mentioned the foreground object is the same actual size in all three photos and what we see is a clearly-recognizable background from the 50 mm lens, and a less- recognizable background from the 85 mm lens, and a distinctly out of focus and blurry background from the 135 millimeter lens .No graphs, no technospeak, and no attempt to prove some arcane point.

 

[Derrel]

Go to the Bob Atkins article that I listed and look at 3 photos of the same scene and The Identical object in the foreground. There you will see conclusive proof of the principle that as we use lenses which are longer, those lenses produce a greater degree of background blur at the same f-stop value... as a fairly experienced photographer I know from years quote literally decades that a fast long length such as a 135 mm f/2 produces photos which have tremendously blurred backgrounds, and that short lenses such as 24mm and 28mm lenses create photos that have backgrounds which are fairly recognizable.

Three photos made at the same location, the same subject, with each of the photos containing The Identical foreground ground subject, rendered at exactly the same size in all three photos, with lenses of three different focal lengths, but all at the same f/stop value, and not three different photographs made with three different lenses at varying f stops. It's really quite amazing, but you can easily be mislead by someone who is throwing numbers and graphs at you fast and loose, and who is out to prove some point that has almost no relevance in the real world.

I would challenge you to begin measuring your actual,measured aperture width when you are shooting flash photos, and you willingly disregard the focal length of the lens in use. Imagine if we were to try to determine the correct flash exposure for a 12- millimeter wide aperture with lenses between 300 mm and 18 mm in length. That's why we use the f/stop system when we try to take actual photos.

 

[smoke665]

@Darrell I didn't disagree nor intentionally try to disrespect you in the first comment, merely tried to point out that there were other considerations besides focal length. I did disagree with the last long post on the basis of the aggressive attitude of the response, and not whether I question your knowledge. In any case it's late and I've no desire to debate anything this close to my bedtime. I have a busy day tomorrow, but I will attempt to review your comments and reference article with a clearer head.

 

[DanOstergren]

Couldn't we just let people do what they like for themselves? I've seen many applications of fake blur that were done beautifully.

Certainly. For me, there is something rewarding about doing something naturally but then if I held true to that I would shoot film.

I'm sure it's very similar to the very rewarding experience that I feel after finishing a great retouch on a portrait. Often even those who complain about other photographers using artificial blur techniques in a photograph will take their own photos into a post production software and apply "artificial" edits to their photos, and 9 times out of 10 they can't even do a decent job in doing so. The point I'm getting at is that most people edit their photos and apply artificial changes to them. We could all perhaps be less forceful of our own preferences on other people's work.

 

[Derrel]

Why do you think I spent $4,000 on the 200 mm F/2 and $2,900 on a used 300 mm F / 2.8? So that I could shoot newspaper sports photos with extremely blown-out, blurry backgrounds when using a Nikon D2x and a 1.5 x crop sensor. I'm sorry that you are trying to prove a point that is not in accordance with what I have been trying to say, and I applaud you for shedding some light on the discussion, even if it is from some mathematics geek who seems to want to try to prove a point that is basically pointless in actual photography. We pick an f/stop such as a 2.8 or 4 or 5.6.



If you go to the Bob Atkins article mentioned above by name, and look at the three photos all taken of the same object it's the same sized foreground object in the frame, and by necessity, the three photographs we're all made at three different camera to subject distances. Mr. Atkins used a 50 mm lens, and an 85 mm, and a 135mm lens. All three lenses were set to the f/value of 2.8. The actual aperture widths were 17.8 millimeters, 30.4 mm, and 48.2 millimeters.

The photos show the same-sized forground object, and three different degrees of background blurring. In the photo made with the 135 mm lens the background is most decidedly much blurrier than in the photos made with the 50 ml lens or with the 85 millimeter lens.

The 135mm focal length I have been shooting since 1980, and I'm pretty familiar with how a 135mm blows backgrounds out of focus. The 180 mm f / 2.8 I have been shooting since 1986, and it is a good lens length when you wish to blow backgrounds pretty well out of focus at f/2.8 or f/4 and to have a fairly-highly-defocused background at f/5.6.

have owned two 300 mm F 2.8 lenses since 1998, and in actual picture taking, the 300 f/2.8 is quite good at creating extremely- defocused backgrounds, and so is the Nikon 200mm F2 . When used at F / 2, the 200 VR creates a beautiful color wash when you shoot portraits, while at longer distances it is razor sharp and there is plenty of depth of field.

 

[JBPhotog]

Not to throw a wrench into this rather microscopic discussion but, diffraction is an issue with large format cameras. The optimum f-stop for 4x5 film is F22 to F32 as referenced by my Sinar P with the employment of the scheimpflug principle. The main reason F64 was the “go to” f-stop was due to the lack of technical accuracy in lens and film plane central pivoting to achieve perfect scheimpflug angles. Sinar recognized this limitation and developed asymmetrical tilts and swings of the lens and film standards including a depth of field dial. There was and is a difference in sharpness at F22 compared to F64 or F45 for many 4x5 lenses.

I owned at one time both a Sinar F and P and shot thousands of sheets of film for ad agencies over the years and they were a joy to use.

 

[smoke665]

@Derrel as mentioned earlier, I have a busy day and time is limited. I did have a chance to scan the article by Mr. Atkins. I want to go back and reread it more slowly but one thing I caught is the similarity of statements in both the article I referenced and the one by Mr. Atkins. Both talk about the effect of distance to the subject and the effect on background blur. From the Atkins article where he talks about the 50mm close to the suject "This also means that the background close to subject will be blurred more by the faster lens. In this case the 50mm lens at f1.4 gives slightly greater blurring for objects up to about 1m behind the subject in focus. However as you go further back, the lens with the largest physical aperture starts to show the most blur" . From the article I referenced "In other words, as the background gets further away (independently of the subject), "longer lens means more background blur" becomes more true." Clicking on the link to Mr Atkins calculator page he again confirmed " Based on this you can see why a 600/4 lens wide open (physical aperture = 150mm) will blur the far distant background far more than a wide open 50/1.4 lens (physical aperture = 35.7mm). In fact it will blur it 4.2x more. However if shooting the same subject at the same magnification (let's say 0.01x, which would by typical for a full body shot of a person), the depth of field of the 50/1.4 will be about 1/3 of that of the 600/4 and so will blur close in objects more."

I don't have the time right now but I will, as well as do some experimenting on my own, to compare the two articles. I suspect that on closer review, and taking the articles in their entirety rather then snippets, that both reach at least the same general conclusions. I did find this statement from Mr Atkins insightful "There's no really simple "rule of thumb" that will tell you whether relative aperture (f-stop) or absolute aperture (size of the aperture in mm) will be most important in determining the degree of blurring at a given distance behind (or in front of) a particular focus point. " I may be reading to much between the lines on this but I take this to mean that blanket statements on focal length and aperture do not apply to "all" circumstances.

 

[The Barbarian]

I'm not all that interested in bokeh, but when I found this old 50mm f2.0 SMC Pentax-M...

I have to admit, it has its uses.