Full Frame Versus Micro 4/3: How Much Does Sensor Size Affect Depth of Field?

If you have one camera with a much larger sensor size than another camera, does it have a considerable impact on depth of field in your images? It's put to the test here.

These days, you have three main sensor sizes in most pro or semi-pro cameras on the market: full-frame, APS-C, and micro 4/3. In terms of sensor size, a full-frame sensor is quadruple the size of a micro 4/3 sensor. So, does that mean it has a commensurate difference in images, particularly depth of field? In this test, The School of Photography's Marc Newton endeavors to answer that question by taking a series of shots at varying f-stops with both cameras and putting the results side by side for you to compare. For the record, he uses a Canon 5D Mark II full-frame camera and an Olympus OM-D E-M1 Mark II micro 4/3 camera.

It's always good to see different types of cameras (and lenses) put together like this, so give the video a look, and let me know your thoughts in the comments below.

Log in or register to post comments

28 Comments

There's no need for this kind of test. It's physics.

How Much? It doubles it at the same _equivalent_ focal length and same aperture. 6 minutes of your life saved :)

Curtis Noir's picture

Haha! Thank you for saving my time

Robert Nurse's picture

It looks like Micro 4/3 has greater DoF.

Matt Williams's picture

uh oh, here we go again

Allow me to save you six minutes: sensor size does not directly affect DOF, but indirectly does so by requiring a shorter focal length as your sensor gets smaller to achieve the same frame (aka Angle of view).

Three things affect DOF: *real* focal length (not 35mm equivalent), aperture, and distance from subject.

Dan Seefeldt's picture

One of these days there will be an fstoppers article on what a camera is and the concept of capturing a still image of a moment in time.

Kenneth Muhlestein's picture

So close. Only TWO things directly affect DOF. Focus distance and size of entrance pupil (aka aperture) Focal length indirectly affects the aperture. An f-stop is the ratio of the focal length to aperture size. You're welcome and good night.

Matt Williams's picture

Except that isn't accurate if you're concerned with keeping the same angle of view. Which is what I said, and pretty much the only thing that matters to people.

If you keep the same working distance and entrance pupil size but change your focal length, you completely change your FOV. Therefore, if achieving the same AOV, focal length affects DOF.

You're welcome and good night.

Kenneth Muhlestein's picture

Clearly were talking about two different things. What i said is 100% accurate. But let me help you out. 100mm f4 on full frame = 50mm f2 on micro 4/3. (2x crop factor) You'll get the exact same FOV. AOV, and DOF. Of course when you use a smaller sensor you'll need to compensate with working distance or focal length. I'm not an idiot. Sensor size, focal length, and f-stop all INDIRECTLY affect DOF. Only focus distance and size of entrance pupil DIRECTLY affect DOF. If you can't understand this, don't worry, you're not alone. 95% of photographers don't understand this. They keep perpetuating incorrect info. Your second paragraph demonstrates your lack of understanding on the matter. Changing the focal length affects the entrance pupil.

Matt Williams's picture

And what if you don't change sensor size? Let's assume focus distance and size of entrance pupil are all that matter.

You're shooting FF. 50mm f/2 = 25mm pupil. Now change to a 100mm f/4 (still 25mm pupil). You have an entirely different AOV now, since you're at the same distance.

Kenneth Muhlestein's picture

God help you! If you aren't changing the sensor size, why the fuck are you changing the focal length? You Will end up with an entirely different composition. (And composition is the most important thing in photography) The whole point of doing the conversions is to get the same composition and look and depth of field. Are you so desperate to be right that you make yourself look like an idiot?!
Oh, and to answer your question, yes, 100mm f4 and 50mm f2 will produce the EXACT same depth of field at the same focal distance, but the 100mm will obviously see less cause it's zoomed in, hence the composition has changed and rendering this argument pointless.

Matt Williams's picture

You literally just admitted exactly what I've been saying. A 100mm f4 and 50mm f2 will produce the same DOF at the same distance, but with a different AOV.

THE ENTIRE TIME I HAVE BEEN TALKING ABOUT KEEPING THE SAME COMPOSITION. I ended my last comment with "You have an entirely different AOV now."

EVERY SINGLE COMMENT I MADE MENTIONED THAT

Therefore, TO KEEP THE SAME COMPOSITION, entrance pupil and distance are NOT the only direct factors.

Re-read what I have been saying and apply your own logic to it. You cannot argue that those are the only two direct factors (after I've been talking about retaining the same FOV) and then argue with me when I give you an example that proves that false.

Kenneth Muhlestein's picture

I'm so sorry you have little dick syndrome! So desperate to be right. That you fail epically.
You said "focal length directly affects depth of field." I said "wrong, focal length affects entrance pupil which affects depth of field" hence only two variables.
You said, "to get the same field of view you have to change the focal length which changes the depth of field" "i agreed, but not for the same reasons you think. By Changing the Focal length, you are changing the entrance pupil size. I gave an appropriate crop factor calculation"
This is where you failed miserably: you said, "what if you don't change the sensor size?" Well then there would be no reason to change lens or do a crop factor calculation then, would there? By changing lens you would change the field of view and not have the same image making your point irrelevant.

Let's end this pointless debate. The fact that you know that 100mm f4 = 25mm entrance pupil tells me you agree with me 100%. Focal length affects the entrance pupil. By that logic, focal length is an INDIRECT variable that affects depth of field. Entrance pupil, focus distance. That's it. Goodnight.

Matt Williams's picture

Always cool when people resort to unoriginal insults when they have failed miserably to understand the discussion.

Kenneth Muhlestein's picture

I didn't misunderstand. Just correcting a common misconception.

Don't worry Matt, your point was clear from the get go. Everyone reading knows what you were saying and agrees.

Could someone please tell me if I am understanding right, as I am struggling a bit.

A full frame sensor with a full frame lens will have a certain depth of field, eg subject in focus with background blur.

At same distance, if I take that same lens and put it on a camera with a smaller sensor, it will have exactly the same bluriness" but I will be seeing a picture within a picture compared to the original image.

If I then put a different lens on the small sensor where the sensor corners actually hit the circle of the lens, I will see exactly the same as the original image with the big lens ?

If I had moved forwards with the little sensor with the big lens on it, I can match the image size of the original....but what happens to the bluriness of the subject ?

No if you change to crop factor camera a normal becomes a tele lens. You will need to step back to get same crop. Since you step back you loose some of the blur.

Yes, if you use the same lens on a full frame (FF) camera and a Micro 4/3rd (MFT) camera, they will have the same depth of field — only the smaller sensor will show, as you say, what is essentially a crop of the bigger sensor.

If you use a lens on the MFT with half the focal length of the FF (e.g. 50mm FF vs. 25mm MFT) and the same aperture, you will get the same picture, except for the depth of field (DOF), which would be deeper on the MFT.

If you use a lens on the MFT with half the focal length of the FF but open the aperture four stops (i.e. divide it by two) compared to the FF (e.g. 5.6 on FF and 2.8 on MFT), the pictures would look the same, same field of view (FOV), same elements, same aspect, same DOF. The same picture, essentially. But, this also means that an MFT lens would have to be four stops faster than the equivalent FF lens to be able to achieve the same DOF fully open.

Moving the camera to achieve an equivalent picture with the same DOF does not work, alas, as moving changes the relative aspects of elements in the picture. As the MFT camera shows less than the FF with the same lens, you would have to move back to get the missing elements into the picture. This would make closer elements in the picture look smaller relative to the distant elements and the DOF would increase as well.

I am not sure that this makes matters any clearer, though I hope so.

Thank you very much Mr. T for taking the time and trouble with that, which helped much. (Also also thank you Bjarne :) ).

Rk K's picture

Why are you going on and on about this?! The difference is two stops. At the same aperture and fov. That's it. It works for dof as well as noise.

John Sammonds's picture

Sorry get it correct the focal length is the same on all lenses its the field of view that changes, so you shot at 50mm full frame and 25mm micro 4/3 this is not the same. So to do this test correctly both should be at 50mm.

So for the same field of view (aka same composition which is what really matters) you will have a shallower depth of field. You are not going to slap on a 50mm on a 8"x10" large format camera. More likely a 300mm and f11 to get the same depth of field as a 50mm f2.8 FF

By changing variables other than just sensor size is hard to come with a solid conclusion.

Petr Svitil's picture

Not really, it shows that sensor size is not the only variable. When the Fuji GFX medium format series came out, the lenses available were pretty slow. So despite being larger than FF, the camera was unable to produce images with as shallow DoF as a FF.

Yeah, but the point is how much "sensor size" affect depth of field.

Yes for portrait shooting the full frame is superior. Much better depth of field but just as important is the improved dynamic range. With my crop camera I never shoot without lighting augmentation. With full frame I do. It lights shadows in the eye sockets much better. I shoot my crop for sports and Landscape can I use my full frame for portraits

Nick Rains's picture

No-one mentioned sensor resolution and viewing size. Both of these affect apparent DoF. My old Canon D60 has more apparent DoF than my current Leica SL. Both full frame.The correct term is 'apparent' DoF - only one point is actually in focus and the rest is a bit of an illusion.

If someone had included the specification of a normal observer viewing an A4 print at 25cm from your nose then you'd be a lot closer to the truth. Almost all DoF marks on lenses are based on this specification, and only for film resolutions too. Change anything and these are not accurate anymore.

DoF is a very slippery concept given that it's mostly an optical illusion, and it's not surprising that it's hard to explain to people.