Saturday, July 13, 2019

The Crop Factor Myth

Bees in a Wallflower Series 1-2
Almost from the first day that crop factor sensors were introduced (sensors smaller than the 35mm standard 36mm x 24mm) people have tried to claim that they were somehow special, that their diminutive size granted them almost magical qualities. I've seen everything from "crop factor sensors give you more depth of field" to "the exposure changes cause smaller sensors gather less light". Both of those statements are so incredibly false it's not even funny, and the explanation is really simple: The crop that a smaller than full frame sensor creates is no different than cropping in post. No matter how you do it cropping an image won't change anything other than how large the subject appears in the frame. To wrap your head around why let's try a simple thought experiment...

Canon makes two full frame camera with a 50MP sensor, the 5DS and the 5DS R. If you crop an image in post, taken with either camera, to a 1.6x crop then you're left with roughly an 18MP image. Cropping that image in post won't change the exposure, or magnification, obviously. But what would happen if you taped off an area around the sensor so now it has a 1.6x crop size? Not much other than the subject would be larger in the frame, the same effect you'd get if you cropped the full frame image in post. The magnification won't change because cropping an image, either by using a smaller sensor or cropping in post, won't reveal more detail in the subject. Cropping simply creates an enlargement. The exposure won't change either, since the same amount of light is striking each photo cell. There would be no difference in masking off the sensor, or cropping in post. No magic...

When I read that the physical size of a sensor alone changes the exposure of a scene it really puzzled me as to why someone would make that claim. But if you take a look at what they're saying they treat the entire sensor as a single light capturing element, and that's just not how digital sensors work. Each pixel on a digital sensor is a single light capturing devise, and a pixel's ability to capture light depends on its size -the bigger the pixel the more light it can "absorb" over a given time span. So bigger pixels are more light sensitive than smaller pixels, that much is true. But if the pixel size is the same two sensors will be equally light sensitive irregardless of the overall size of the sensor, as long as the underlying electronics are the same. Put the same lens on both a full frame camera and a crop factor camera, point them at the same scene, and the intensity of the light that's striking the pixels on both sensors will be exactly the same. If there is any difference in the exposure it will be due to differences in pixel size, or in how the camera's electronics amplify the signals produced by the pixels. Don't believe me? Go back to Canon's 50MP full frame sensor and mask off an area around it so that it's now a 1.6x crop factor sensor and nothing changes other than how large the subject will look in the frame, because it's cropped. The same amount of light is striking each pixel, so there won't be any change in exposure just because you've masked part of it off. Keep in mind that the size of the image circle projected by the lens does not change no matter what camera you put it on...

Is there an advantage to shooting macro with a crop factor sensor? Yes, and one of them is the ability to make a subject look larger in the frame at lower magnifications than when shooting the same scene with a full frame sensor. Shooting at a lower magnification will give you more depth of field, and I think that was the start of the myth concerning crop factor sensors and depth. Crop factor sensors don't give you more depth of field, being able to shoot at lower magnifications does since depth is simply a function of magnification and Fstop. But then again you could always take a shot with a full frame sensor and simply crop the image down to a 1.6x size.

If you have anything to add, or want to call me out for this post, just leave me a comment. I had to turn post moderation on because I was getting too many invitations to watch someone diddle themselves. So when you post a comment and it doesn't immediately show up don't worry, I will approve it as soon as I get the notification email even if your comment is negative. I've never filtered out a negative comment, just do me a favor and be civilized. Until next time, happy shooting!

Footnote: Another term making the rounds is "print magnification" and people are using it to describe how a subject looks larger in print when taken with a crop factor camera. But it's just enlarging the subject by cropping the image -and again it's no different than cropping a full frame image in post. I hate the term "print magnification" because it implies that the subject has been magnified, and it has not...

5 comments:

MonG9 said...

Hello John.
Of course, I totally agree with you, the "depth of field" feeling is depending of the size of the pixels and not of the size of the sensor. I own a 20 Mpx FF (5D MkII) and a 20 Mpx APS-C (7D MkII), and using this last one gives the possibility to obtain a better image quality at the same equivalent magnification factor (taking the crop factor into account). Magnification 3 on the MPE-65 becomes 4.8 due to the crop factor. Due to the limits of the MPE-65, using magnification 5 with the 5D MkII gives a lower quality picture than using magnification 3 with the 7D MkII. The pixels density of the 7DMkII is equivalent to the one of the 5DSr.
Thanks for your article.
Raymond

Dalantech said...

"Magnification 3 on the MPE-65 becomes 4.8 due to the crop factor."

Nope, it's still just 3x. Cropping an image, in post or with a smaller than full frame sensor, creates an enlargement of the subject and not an increase in magnification. Increasing the magnification would result in an increase in detail (ignoring diffraction). Cropping an image, no matter how you crop, will never reveal more detail...

MonG9 said...

You're absolutely right. That is the reason why I appreciate your article, it eliminates confusion. It is just that I always practice in function of the result I will have on the screen (or on paper), but definitely you are right, John, thanks.

Steveball said...

Hi John
I love your photos and your blog is really useful. I think two things you haven't mentioned are pixel density and the fact that lenses often have better image quality toward the centre of the image. I would also be very interested in your thoughts on how working distance fits into the discussion on sensor size and lens focal length in practical terms.

Stephen Ball

Dalantech said...

Hey Steven,
Pixel density impacts diffraction and dynamic range the most, with smaller pixels being more diffraction prone and typically having less dynamic range. Sensor manufacturers have gotten really good at tackling the dynamic range problem, so it's not as much of an issue as it was a few years ago. Diffraction can actually be amplified by motion, and it won't look like traditional motion blur. You'll just notice a loss in detail and blame diffraction for it, irregardless of the size of the pixels. Lens sharpness, or a lack of it, is another aspect that can increase diffraction effects. But as for using the center of a lens I haven't noticed much of difference, certainly not enough to justify centering the subject and cropping every shot in post. I think it's pretty clear that I think too much emphasis has been placed on absolute image sharpness, to the point where the macro discipline is full of razor sharp "snap shots". If an image is poorly composed and poorly lit then getting every pixel as sharp as possible (focus stacking, using the center of a lens, using the optimum aperture for the lens, etc.) won't save it. If a photo doesn't look good edge to edge then John Q. Public isn't going to look at it twice...

As for working distance, sensor size, and focal length: Photography is an art form, but too many people have tried to turn macro photography into a science project and as a result the discipline gets very little respect outside of the niche communities that most people are posting their images to. There's no formula for a good photo, or for creativity. So getting bogged down in discussions about technical aspects that the average viewer doesn't care about are pretty pointless in my opinion. Taking a photo of an insect is easy. Taking a photo of an insect that is appreciated by people who are normally freaked out by them is hard. It's the later that concerns me the most, and it's where I've taken my photography. So the only thing I care about is "Will the gear I'm using, and how I'm using it, allow me to take the shot that I want to take?". If the answer to that question is "Yes" then everything else is kinda irrelevant...