the 45mm doesn't provide shallow enough DoF... [image + shooting notes]

[SLOtographer]

Let me just say that I appreciate the work presented and the info behind the work. Thumbs up and keep it coming!

 

[CharlesB58]

noirdesir wrote:
Whenever I hear somebody saying (in a forum of smaller sensored cameras), 'shallow DOF is overrated', it almost always sounds like a defensive argument to me. rkeller's image was a nice example of a good image with just the right background blur but the post still seemed partly motivated by wanting to proof a point that I think doesn't really need proving.

I've been involved in photography for over 30 years, at various levels of professional engagement. My first taste of a 35mm slr involved being entranced by how nice a portrait I took looked due to the OOF background from shooting a 50mm lens at f1.8. I know when to use shallow DoF for particular effects and when not to. It depends on what I want the photo to achieve, both visually and narratively. I get paid to have that understanding. I learned the basics of photography when shooting just about any lens wide open was something you did only as a last resort. Most lenses 30 years ago simply weren't sharp enough, or suffered noticeable vignetting, if shot wide open.

I have no need to defend my choice of gear. My clients like my photographs.

I use 4/3 for various reasons. DoF seldom is an issue for me because my style of portraiture is primarily environmental. When necessary I position my subject in front of a non-distracting background and/or choose a focal length that will increase the blur of the background as needed.

In my experience, too many people want that nice shallow DoF because they think it will make up for choosing a poor background or the wrong focal length to subject distance,(many don't even think about such things anyway) or simply because they think that portraits must have shallow DoF to "look professional". I've seen people say such things outright on these forums. That's what I mean by "depending on gear": people want the gear to make up for poor personal technique.

(An example of this narrow way of thinking is threads in which people said they saw no purpose in the 75mm f1.8 lens. In their thinking, it's too long for portraits and not long enough for sports or birding. For a portrait and concert photographer such as myself, it's almost an ideal lens.)

My point isn't about whether shallow or deep DoF is superior to the other. It's about people trying to sound authoritative on the subject who really only think within certain confines. For someone whose personal style depends primarily on shallow DoF for portraits, I'd be the first to recommend a larger format for that purpose. The Canon 135 f2 on a 5D2 makes for a great portrait combo, for a particular style of portraiture. I know someone who takes almost all his senior portraits and posed wedding photos with that set up, and I truly admire his work. But we're talking about only one style of portraiture, not the only style.

For someone whose style is more along the lines of environmental portraits such as those done by Arnold Newman or Annie Liebovitz, m4/3 works just fine as we see in the OP's example. So does FF, when it comes to purely technical considerations.

I agree the point might not necessarily need proving as the OP presented it. On the other hand, a lot of novices peruse this forum. To read some threads, you would think that m4/3 is a total bust when it comes to any subject in which shallower DoF is desired. So I think perhaps the OP can be excused for showing how proper understanding of the gear combined with solid technique can produce admirable results.

 

[Moti]

Bob Meyer wrote:

ryan2007 wrote:

Three things effect depth of field. Focal length, aperture and camera to subject distance.

Focal length doesn't affect DOF except as it affects camera to subject distance. Take two shots from the same positions, one with a 100mm f2.8, the other with a 50mm f/1.4, then crop and enlarge the second image to the same size as the first. DOF will be identical

That is only half of the truth because based on your twisted assumption, you can now go one step forward and prove that aperture also doesn't effect DOF because f1.4 and f2.8 in your example give the same DOF...

Nice try,

Moti

--

http://www.pixpix.be

http://www.musicalpix.com (under construction)

 

[Anders W]

Bob Meyer wrote:

ryan2007 wrote:

Three things effect depth of field. Focal length, aperture and camera to subject distance.

Focal length doesn't affect DOF except as it affects camera to subject distance. Take two shots from the same positions, one with a 100mm f2.8, the other with a 50mm f/1.4, then crop and enlarge the second image to the same size as the first. DOF will be identical.

For any given sharpness criterion (circle of confusion), one can say either that DoF is a function of a) aperture, focal length and subject distance or b) of aperture and subject magnification (where subject magnification is of course in turn a function of focal length and subject distance). Only the first is generally true. The second is a good approximation under some conditions but not a universal truth. See here, point 3:

http://toothwalker.org/optics/misconceptions.html

 

[Paul De Bra]

If you have the room to step back further to frame the subject in the same way with a longer lens the difference between camera to subject versus camera to background becomes smaller and this (partly) undoes the effect of using an equally fast but longer lens.

With such an intended picture the only two real parameters that influence the background blur are aperture (but in the image this is maxed out) and sensor size (which is fixed given the camera). Changing the lens for another f/1.8 lens isn't going to help a lot. (It may help some as the formulas are not as simple as my simple argument here about the distance issues.)

 

[Anders W]

Paul De Bra wrote:

If you have the room to step back further to frame the subject in the same way with a longer lens the difference between camera to subject versus camera to background becomes smaller and this (partly) undoes the effect of using an equally fast but longer lens.

Stepping back so as to get the same subject framing with the 75/1.8 as with the 45/1.8 yields approximately the same DoF. However, the perceived background blur will be greater with the 75/1.8 than with the 45/1.8 due to the fact that it will magnify the background more. See here, under the heading "Background blur" for a good illustration of the effect of focal length with subject magnification as well as DoF held constant:

http://toothwalker.org/optics/dof.html

In the case exemplified in this thread, the increase in perceived blur when switching from the 45 to the 75 would be rather pronounced since most of the background is rather far away from the subject. The effect becomes minimal only when the background is very close to the subject. But that case is rather uninteresting inasmuch as it wouldn't allow any blur to speak of in the first place.

With such an intended picture the only two real parameters that influence the background blur are aperture (but in the image this is maxed out) and sensor size (which is fixed given the camera). Changing the lens for another f/1.8 lens isn't going to help a lot. (It may help some as the formulas are not as simple as my simple argument here about the distance issues.)

 

[Moti]

You really hit the nail on it's head.

Cheers

Moti

 

[KLO82]

I liked your image.

Two thoughts on DoF, not completely related to your post:

1. I have seen many people compare the ability of making shallow DoF at different apertures by comparing depth of field numbers, but these numbers does not tell the whole story. For example, consider the 85mm on a FF camera focused at 5 ft. 85mm @ f2.8 will give you 2 inch DoF, whereas @ f 4 DoF is 2.9 inch, not much of a difference, right? But the thing is, the portions of the photo which are OOF both @ f2.8 and f4 will look very different at these apertures, which these DoF numbers do not tell. For example, see this link: http://www.the-digital-picture.com/Reviews/Canon-EF-85mm-f-1.2-L-II-USM-Lens-Review.aspx , see the mouse over photo that shows same scene at different apertures. You will see that the difference in photos at 2.8 and f4 is quite large, even though difference in DoF is not much.

2. I was just thinking, will it be possible one day, that softwares will be able to simulate blur at different larger apertures? For example, I have taken a photo @ f5.6. From the focal length and shooting distance inormation, the software will be able to show exact same DoF and blur at f4 , 2.8 etc. It will not be able to increase DoF though, for example shooting at f2.8, but trying to show f22 later on the software.

 

[Paul De Bra]

Indeed: even though the background isn't really blurred more it looks as if it is because it gets magnified when using a longer lens. Whether a more or less magnified background is desired really depends on taste and on the actual situation. So it is a matter of image composition, not of depth of focus, that determines which lens is best for the situation.

 

[Bob Meyer]

ryan2007 wrote:

Bob Meyer wrote:

ryan2007 wrote:

Three things effect depth of field. Focal length, aperture and camera to subject distance.

Focal length doesn't affect DOF except as it affects camera to subject distance. Take two shots from the same positions, one with a 100mm f2.8, the other with a 50mm f/1.4, then crop and enlarge the second image to the same size as the first. DOF will be identical.

Nope,

What do you do if your 40 feet from the subject. You have a 50 1.4 and a 100 2.8 which is still too short and you can't crop the image to "fake it" or whatever. You need the lens to do the work.

If you need to fill the frame by cropping is not good unless you want to loose that "pixel" information.

I can take a 300 mm 5.6 lens to your 50 1.4 and not have to crop anything at the working distance.

A telephoto lens inherently has shallow depth of field and a wide angle lens has greater depth of field no matter the f-stop. The aperture enhances that effect.

Take a the equivalent 300 mm lens at 5.6 and the other at 2.8 from the same 20 foot distance. That 2.8 lens has shallower DOF and will blur the background more so.

This is photo 101 stuff.

 

[clengman]

Bob Meyer wrote:

ryan2007 wrote:

Bob Meyer wrote:

ryan2007 wrote:

Three things effect depth of field. Focal length, aperture and camera to subject distance.

Focal length doesn't affect DOF except as it affects camera to subject distance. Take two shots from the same positions, one with a 100mm f2.8, the other with a 50mm f/1.4, then crop and enlarge the second image to the same size as the first. DOF will be identical.

Nope,

What do you do if your 40 feet from the subject. You have a 50 1.4 and a 100 2.8 which is still too short and you can't crop the image to "fake it" or whatever. You need the lens to do the work.

If you need to fill the frame by cropping is not good unless you want to loose that "pixel" information.

I can take a 300 mm 5.6 lens to your 50 1.4 and not have to crop anything at the working distance.

A telephoto lens inherently has shallow depth of field and a wide angle lens has greater depth of field no matter the f-stop. The aperture enhances that effect.

Take a the equivalent 300 mm lens at 5.6 and the other at 2.8 from the same 20 foot distance. That 2.8 lens has shallower DOF and will blur the background more so.

This is photo 101 stuff.

 

[rkeller]

Anders W wrote:

Paul De Bra wrote:

If you have the room to step back further to frame the subject in the same way with a longer lens the difference between camera to subject versus camera to background becomes smaller and this (partly) undoes the effect of using an equally fast but longer lens.

Stepping back so as to get the same subject framing with the 75/1.8 as with the 45/1.8 yields approximately the same DoF. However, the perceived background blur will be greater with the 75/1.8 than with the 45/1.8 due to the fact that it will magnify the background more. See here, under the heading "Background blur" for a good illustration of the effect of focal length with subject magnification as well as DoF held constant:

http://toothwalker.org/optics/dof.html


In the case exemplified in this thread, the increase in perceived blur when switching from the 45 to the 75 would be rather pronounced since most of the background is rather far away from the subject. The effect becomes minimal only when the background is very close to the subject. But that case is rather uninteresting inasmuch as it wouldn't allow any blur to speak of in the first place.

With such an intended picture the only two real parameters that influence the background blur are aperture (but in the image this is maxed out) and sensor size (which is fixed given the camera). Changing the lens for another f/1.8 lens isn't going to help a lot. (It may help some as the formulas are not as simple as my simple argument here about the distance issues.)

Good point. Using EXIF data from the image:

• Focus distance: 2.655 m/8.71 ft
• Angle of view: 22.2 degrees and 1.04 m/3.41 ft (but I also cropped the image slightly). [Interesting question: does Olympus EXIF calculate the linear measure as a diagonal based on the focus distance and the AoV for the lens?]
• DOFMaster.com says DoF near limit = 2.57 m/8.42 ft; far limit = 2.75 m/9.02 ft; total = 0.18 m/0.60 ft

If I used the 75mm f/1.8 wide open with the same main subject framing, I would have been shooting with a focus distance of 75/45*2.655 m = 4.425 m/14.51 ft. DOFMaster says:

• near limit = 4.34 m/14.20 ft; far limit = 4.52 m/14.80 ft; total = 0.18 m/0.60 ft

And as discussed: the background elements would be comparatively larger in the frame, and while they would not have more absolute blur, they would appear to have more blur.

 

[Anders W]

Bob Meyer wrote:

ryan2007 wrote:

Bob Meyer wrote:

ryan2007 wrote:

Three things effect depth of field. Focal length, aperture and camera to subject distance.

Focal length doesn't affect DOF except as it affects camera to subject distance. Take two shots from the same positions, one with a 100mm f2.8, the other with a 50mm f/1.4, then crop and enlarge the second image to the same size as the first. DOF will be identical.

Nope,

What do you do if your 40 feet from the subject. You have a 50 1.4 and a 100 2.8 which is still too short and you can't crop the image to "fake it" or whatever. You need the lens to do the work.

If you need to fill the frame by cropping is not good unless you want to loose that "pixel" information.

I can take a 300 mm 5.6 lens to your 50 1.4 and not have to crop anything at the working distance.

A telephoto lens inherently has shallow depth of field and a wide angle lens has greater depth of field no matter the f-stop. The aperture enhances that effect.

Take a the equivalent 300 mm lens at 5.6 and the other at 2.8 from the same 20 foot distance. That 2.8 lens has shallower DOF and will blur the background more so.

This is photo 101 stuff.

 

[Anders W]

Anders W wrote:

Bob Meyer wrote:

ryan2007 wrote:

Bob Meyer wrote:

ryan2007 wrote:

Three things effect depth of field. Focal length, aperture and camera to subject distance.

Focal length doesn't affect DOF except as it affects camera to subject distance. Take two shots from the same positions, one with a 100mm f2.8, the other with a 50mm f/1.4, then crop and enlarge the second image to the same size as the first. DOF will be identical.

Nope,

What do you do if your 40 feet from the subject. You have a 50 1.4 and a 100 2.8 which is still too short and you can't crop the image to "fake it" or whatever. You need the lens to do the work.

If you need to fill the frame by cropping is not good unless you want to loose that "pixel" information.

I can take a 300 mm 5.6 lens to your 50 1.4 and not have to crop anything at the working distance.

A telephoto lens inherently has shallow depth of field and a wide angle lens has greater depth of field no matter the f-stop. The aperture enhances that effect.

Take a the equivalent 300 mm lens at 5.6 and the other at 2.8 from the same 20 foot distance. That 2.8 lens has shallower DOF and will blur the background more so.

This is photo 101 stuff.

 

[Tim F 101]

If your primary concern is depth of field, don't buy a micro four thirds camera. You know in advance that FF cameras do absurdly narrow DOF much better. Buy one of those. You can get a 5DII for a song these days. If you need something with a big sensor that fits in a purse, a Sony rep would love to talk with you about their RX1.

In similar news, my Honda Fit does an excellent job ferrying a surprising amount of construction supplies on almost no gas but sucks at the dragstrip.

 

[Brian D. Schneider]

Everdog wrote:

Why did you leave NIkon1 owners out too?

That's plain mean.

Are there any?

 

[bobn2]

Brian D. Schneider wrote:

Everdog wrote:

Why did you leave NIkon1 owners out too?

That's plain mean.

Are there any?

The idea that Nikon doesn't sell any 1 series is a myth held dear by some on this forum. It sells pretty well, on the basis that it has no real competitors for what it does.

 

[acahaya]

and the image. Very interesting thread until it drifted towards the unavoidable equivalence discussion (yawn)

The only thing i am not 100% happy with are the highlights behind the head, otherwise this is an interesting image with just enough DOF and a nice 3D effect. I also like the b&w, colors might be distracting from the main subject.

Sabine

 

[bobn2]

acahaya wrote:

and the image. Very interesting thread until it drifted towards the unavoidable equivalence discussion (yawn)

Don't worry, Illy, the Bustard and I are doing our best to change the subject.

 

[rkeller]

acahaya wrote:

and the image. Very interesting thread until it drifted towards the unavoidable equivalence discussion (yawn)

The only thing i am not 100% happy with are the highlights behind the head, otherwise this is an interesting image with just enough DOF and a nice 3D effect. I also like the b&w, colors might be distracting from the main subject.

Sabine

Thanks! I think any thread on this forum, given enough time, will devolve into the same mess!

Someone else mentioned the highlights also, and I agree that they are potentially distracting. I did shoot a few different variations (about 20 total). Turns out this was the one the client and his publisher picked. I like the B&W too, but it also turns out they are going with a color version. So there you go!

Book to be published on Yale University Press in August 2013: "The Electronic Silk Road: How the Web Binds the World Together in Commerce" http://yalebooks.co.uk/display.asp?K=9780300154597&st1=Silk+road

I'm waiting for someone to publish the following:"The Electronic Rogue's Gallery: How DPreview Forums __________" [fill in the blank]