Busting the FF vs Crop DoF Bokeh Myth

Started Jul 27, 2009 | Discussions
wittgenstein Regular Member • Posts: 250
Is there any myth?

If you take a FF shot and make a crop of it, you'll get a crop camera shot.

A crop camera is only a FF camera taking only central pixels from the lens, so really DOF from it is the same.

Jan Madsen Senior Member • Posts: 2,398
Re: Experiment ignores FOV (Field Of View)

therickman wrote:

If you took a 4" X 6" printed photo from a full frame 5D, and with scissors cut the edges off to make the photo 2.5" X 3.75", that doesn't change the depth of field. This is the same thing as using a 1.6x crop sensor camera. All the crop sensor is doing is producing the central 63% of a full frame sensor.

Yes, but that would result in a rather mediocre APS shot. In order to get the same quality as the FF shot you need to calculate with a smaller circle of confusion, and then the DOF is not the same. You can't omit CoC from your statements.

-- hide signature --

  • Jan

Jan Madsen Senior Member • Posts: 2,398
Couldn't have stated it better (nt)
-- hide signature --

  • Jan

Jan Madsen Senior Member • Posts: 2,398
Re: You are adding to the myth

therickman wrote:

dherzstein wrote:

I don't think that the OP understands DOF.
--
-Dave
http://pixseal.com

I do. DoF is much different than FoV - which everybody here seems to be confusing.

  • and bokeh is different from DoF. You mix them up. And forget CoC.

Admitted, DoF is a complex issue...
--

  • Jan

Tannin Senior Member • Posts: 1,397
Re: Busting the FF vs Crop DoF Bokeh Myth

You are trying to do a fair comparison and cropping half of one picture out? Fair dinkum, you really need to think a bit harder before you start shooting test shots. These are completely meaningless. Go back and reshoot with equivalent fields of vie,w, then start a fresh thread. If you are lucky, people will have forgotten this one by then, and may even take it seriously.

But speaking of seriously, what on earth were you trying to demonstrate? That 1.6 centre crops taken from:

  • the same lens

  • at the same distance

  • from the same subject

  • at the same aperture

produce the same shot?

Duh. It would be a very strange world indeed if they were any different.

Nothing more to see in this thread. Move along please.

 Tannin's gear list:Tannin's gear list
Canon G9 X II Canon EOS 5D Mark II Canon EOS-1D Mark IV Canon EOS 7D Mark II Canon EOS 5D Mark IV +10 more
ovrebekk
ovrebekk Veteran Member • Posts: 4,701
You need to move closer with an FF camera

You need to move closer with an FF camera to get the same FOV with the same lens. That is why it is common to say that you get less DOF with FF.

As you say, keeping everything the same DOF is equal, but you seldom do this in practice. Say you want to fill an image with a face using an 85mm lens the FF camera will give you less DOF.

Kuivaamo Senior Member • Posts: 2,248
Preposterous? Oh dear.

therickman wrote:

Resizing an image doesn't change the depth-of-field. That's preposterous.

You seriously need to read up on DoF and CoC before you engage in these tests and debates. Resizing an image absolutely does change the depth of field.

As for the rest of the discussion going on in this thread, people are mixing two issues: 1) why your test is flawed in trying to prove that the DoF is the same for same lens, aperture and distance; 2) why your test is flawed at trying to prove that you generally get a narrower DoF when using a FF camera as opposed to a crop camera.

For #2, just compare your 50D/50mm/f2/5' shot against your 5D/85mm/f2/5' shot. It becomes clear that for the same FoV and perspective, FF does produce a narrower DoF at the same f-ratio (50mm x 1.6 = 80mm, so not exact but close enough).

For #1, you just need to familiarise yourself with what DoF and CoC mean and how they are derived.

EDIT: here are the images you've taken that show what I mean in #2:

GoldenSpark Senior Member • Posts: 1,037
Re: Experiment ignores FOV (Field Of View)

DOFMaster's calculator does't show the difference you claim.

It varies the circle of confusion between different cameras to account for different pixel pitch and hence differences in the acceptable sharpness figure.

For the same lens at the same distance and same f stop the image created at the image plane is the same. A crop camera will just crop it out more than FF.

There is no myth. It's a lack of understanding to do with field of view, as has been already explained.

To get the same field of view between crop and FF you have to change the focal length or subject distance, which will affect the DOF and bokeh.

dopravopat
dopravopat Senior Member • Posts: 1,180
Technical question

Assume that I want capture exactly the same scene with the 5D and 50D, with the same FOV and DOF, thus distance from subject shall be identical, focal length of the lens and aperture will be each time other to get the same look (same DOF, same bokeh).

Here is a simple Depth of Field Equivalents caluclator:

http://www.cambridgeincolour.com/tutorials/digital-camera-sensor-size.htm

For 50 mm lens on the 5D, focused to a subject in 5 meters distance and closed down to f4, I would need a 31 mm lens on the 50D, focused also to 5 meters and closed down to f2.5 to get the same FOV and DOF. Now the question. Would it give exactly the same bokeh, assuming the shape and number of aperture blades will be the same?

How similar would the two images look like, considering DOF and bokeh?

Do you have the Sigma 30 f1.4 to try such a test? I am aware that the Sigma 30 1.4 renders a different bokeh than the Canon 50 1.4, but I would like to see it anyway, with the apertures 4 (FF) vs 2.5 (1.6 crop). If you have the lenses and time, I would appreciate you do such test.
--
Erik
http://jalbum.net/users/eziak

 dopravopat's gear list:dopravopat's gear list
Canon EOS 30D Canon EOS 60D Canon EF-S 18-135mm F3.5-5.6 IS Sigma 8-16mm F4.5-5.6 DC HSM Sigma 18-35mm F1.8 DC HSM Art +5 more
Kuivaamo Senior Member • Posts: 2,248
Re: Experiment ignores FOV (Field Of View)

GoldenSpark wrote:

DOFMaster's calculator does't show the difference you claim.

It varies the circle of confusion between different cameras to account for different pixel pitch and hence differences in the acceptable sharpness figure.

Not quite; the CoC given by Dofmaster for e.g. the 20D, 40D & 50D is the same. Likewise, 5D & 5D Mk2 have the same CoC. This is because pixel density does not affect DoF; CoC is only affected by the output size and viewing distance (given a certain level of eyesight). The "default" output size is 8x10" while the viewing distance is 25 cm for a person with good vision.

The reason CoC is different for different formats is that a larger format needs to be enlarged less to achieve an 8x10" (or any given size) print, and thus a larger CoC criterion is permissible.

The reason why the OP's test appears to disagree with Dofmaster's calculations is that he essentially resized the images to different output sizes when he made the crop comparisons. What he's showing us is that if you take a 1.6x crop from a FF image, you get the same image as you would with a 1.6x crop camera with the same lens, aperture and shooting distance. This should not be a surprise to anyone, and thus his actual results are trivial.

Kuivaamo Senior Member • Posts: 2,248
Re: Technical question

dopravopat wrote:

How similar would the two images look like, considering DOF and bokeh?

Bokeh depends on the lens, so the answer to your question is "it depends". Even if both lenses have circular aperture blades, the bokeh can be markedly different.

Kuivaamo Senior Member • Posts: 2,248
Correction.

Kuivaamo wrote:

2) why your test is flawed at trying to prove that you generally get a narrower DoF when using a FF camera as opposed to a crop camera.

This should have read: "why your test is flawed in trying to prove that you generally don't get a narrower DoF when using a FF camera as opposed to a crop camera."

Anonimo Contributing Member • Posts: 583
Ironic (but true) answer

Rick, you just demonstrated that a given lens, set at a given aperture and focused at a given distance will always form the same image, regardless of which camera it is mounted on. Congratulations!!!

Now you are ready to move to the next level! Follow the link below and discover what lens equivalence is about.

http://www.josephjamesphotography.com/equivalence/

Those who are more advanced can go straight to the original thread about DoF and check the math.

http://forums.dpreview.com/forums/read.asp?forum=1032&message=32498836

Once Rick is done with the article I suggest him to set an experiment to bust the lens equivalence myth. He shall reproduce two practical situations:

a) Portrait shoot
5D with an EF 85mm f/1,8 ($570) at f/2 and ISO-400 versus
50D with an EF 50mm f/1,2L ($1.600) wide open at ISO-160

b) Landscape shoot
5D with an EF 24mm f/2,8 ($490) at f/4,5 and ISO-400 versus
50D with an EF 14mm f/2,8L ($2300) wide-open at ISO-160

I might help to arrange lenses for the 5D since Rick gets the lenses for the 50D in advance.

P.S.: I hope that Rick hasn't bought the 50D to get an edge when shooting tele. Otherwise he might be now upset to know that all he needed to do was to crop the pictures taken by the 5D. Oh God, I shouldn't have busted this myth!!

Don't take it personally The true message is hiden behind the irony.

Regards, Anonimo

Spencer Regular Member • Posts: 365
Re: Busting the FF vs Crop DoF Bokeh Myth

Perhaps you've seen many posts saying DOF is shallower in FF and interpreted this as the poster not realizing it is FOV which is the real difference between FF vs Crop.

But the fact is that to get the same FOV between a Crop and a FF you need to be closer to your subject or use a longer lens on the FF. Both cases the FF will produce a shallower DOF.

This is how we shoot in real world situations. That is why calculators take FOV into consideration. Photographers take FOV into consideration.

I understand what you were trying to prove in your test but that argument really is moot since the whole idea of DOF in FF vs Crop is based on Field of View. It's not a myth when you apply this to real world photography.

 Spencer's gear list:Spencer's gear list
Canon EOS 5D Canon EOS 5D Mark II Canon EOS 20D Canon EF 15mm f/2.8 Fisheye Canon EF 50mm F1.8 II +7 more
Gazer Contributing Member • Posts: 675
I forget where is...

the indoor/outdoor sensor on my camera? Or maybe it is on the lens? You must have some truly magical equipment where the bokeh/DoF changes whether it is indoor or outdoor.
--
Steve
http://www.pbase.com/mastels

 Gazer's gear list:Gazer's gear list
Canon EOS-1D Mark III Canon EOS 5D Mark III Canon EOS 6D Canon EOS-1D X Mark II
OP therickman Senior Member • Posts: 1,305
Re: I agree with Cheburashka

cobra_genial wrote:

you have to make the comparison with the same FOV...

I did it on page two:
http://forums.dpreview.com/forums/readflat.asp?forum=1029&message=32518209

-- hide signature --

Insert obligatory quote here...

http://www.picsure.net

OP therickman Senior Member • Posts: 1,305
Re: Busting the FF vs Crop DoF Bokeh Myth

Tannin wrote:

You are trying to do a fair comparison and cropping half of one picture out? Fair dinkum, you really need to think a bit harder before you start shooting test shots. These are completely meaningless. Go back and reshoot with equivalent fields of vie,w, then start a fresh thread. If you are lucky, people will have forgotten this one by then, and may even take it seriously.

I didn't crop anything. The images in my OP are straight out of the camera, no crops. I displayed the setup pics... did you read it? Apparently not. Also, you didn't read my follow-up where I did a FoV comparison:
http://forums.dpreview.com/forums/readflat.asp?forum=1029&message=32518209

But speaking of seriously, what on earth were you trying to demonstrate? That 1.6 centre crops taken from:

  • the same lens

  • at the same distance

  • from the same subject

  • at the same aperture

produce the same shot?

I NEVER implied that. Did you read my OP? Or did you see everone else slamming me and decided to jump right in like a 5-year old not knowing the facts!?

Duh. It would be a very strange world indeed if they were any different.

Yes it would.

Nothing more to see in this thread. Move along please.

Please do.

-- hide signature --

Insert obligatory quote here...

http://www.picsure.net

Deanis Senior Member • Posts: 1,525
On a basic level....

FF only gives "better bokeh" as framing the same subject with the same lens at the same aperture as a crop-sensor means you have to step closer to the subject, thus blurring the background further than what was possible with the crop-sensor.
--
My Gallery:
http://flickr.com/photos/27982239@N02/

OP therickman Senior Member • Posts: 1,305
Re: Preposterous? Oh dear.

Kuivaamo wrote:

therickman wrote:

Resizing an image doesn't change the depth-of-field. That's preposterous.

You seriously need to read up on DoF and CoC before you engage in these tests and debates. Resizing an image absolutely does change the depth of field.

How so? Most people print 4X6. If I take this image...

and print it at 4X6, then enlarge it to 20X30, that's not going to make the "Photography" and "Lighthouse" books any clearer or blurrier in the picture. The books are staggered 1 foot and 2 feet behind the subject, and that's how it's been captured at that moment in time.

In the picture above, let's say the DoF is about 5 inches - 2 inches in front and 3 inches behind the Ansel Adams book. The DoF is the same whether it's printed as a 4X6, an 8X10, or 20X30 enlargement. The picture isn't changing, just getting bigger.

That's like saying if I take a picture of two men - one standing 10 feet behind the other, and I then enlarge that printed image, he's no longer 10 feet behind the other guy. That's preposterous. DoF is captured by the camera at that moment in that shot, and doesn't change.

-- hide signature --

Insert obligatory quote here...

http://www.picsure.net

OP therickman Senior Member • Posts: 1,305
Re: Ironic (but true) answer

Anonimo wrote:

Rick, you just demonstrated that a given lens, set at a given aperture and focused at a given distance will always form the same image, regardless of which camera it is mounted on. Congratulations!!!

Now you are ready to move to the next level! Follow the link below and discover what lens equivalence is about.

http://www.josephjamesphotography.com/equivalence/

Those who are more advanced can go straight to the original thread about DoF and check the math.

http://forums.dpreview.com/forums/read.asp?forum=1032&message=32498836

Once Rick is done with the article I suggest him to set an experiment to bust the lens equivalence myth. He shall reproduce two practical situations:

a) Portrait shoot
5D with an EF 85mm f/1,8 ($570) at f/2 and ISO-400 versus
50D with an EF 50mm f/1,2L ($1.600) wide open at ISO-160

b) Landscape shoot
5D with an EF 24mm f/2,8 ($490) at f/4,5 and ISO-400 versus
50D with an EF 14mm f/2,8L ($2300) wide-open at ISO-160

I might help to arrange lenses for the 5D since Rick gets the lenses for the 50D in advance.

P.S.: I hope that Rick hasn't bought the 50D to get an edge when shooting tele. Otherwise he might be now upset to know that all he needed to do was to crop the pictures taken by the 5D. Oh God, I shouldn't have busted this myth!!

Don't take it personally The true message is hiden behind the irony.

Thanks. But I'm not a moron. Had you read this follow-up post, you'd have realized that before you spouted your patronizing sarcasm:
http://forums.dpreview.com/forums/readflat.asp?forum=1029&message=32518209

-- hide signature --

Insert obligatory quote here...

http://www.picsure.net

Keyboard shortcuts:
FForum MMy threads