Crop factors carry with them a significantly more impactful footprint than a lot of people fully realize. Understanding them and how they relate to different aspects of photography goes far beyond simply saying that cropped sensors cover a smaller portion of a field of view, and thus images look as if they’re shot at the corresponding full frame-relative focal length. There are other considerations to be included such as the effect of ISO and aperture, and bokeh.
[REWIND: Full Frame vs Crop Frame Sensors]
There isn’t a day that rises where a debate on the benefits of full frame cameras over APS-C or micro four thirds doesn’t come about, however, usually what’s being touted are the quality differences in an image regarding dynamic range, noise, etc. You can find that information abound, so we won’t discuss that here now.
The video below does just about the best explanation, with relative visuals, explaining the differences in what you see, and how to handle each variance, that I’ve seen. Northrup is extremely clear, descriptive, and controlled in his display, so the points hit home, and they stick. I’m sending this to every single person who asks me about the differences to be had and noted from sensor sizes. The information is invaluable when it comes to purchasing new lenses, or cameras even. It’s also necessary to understand so when discussing different lenses and cameras, the proper comparisons can be made, and so you won’t sound like a fool.
Watch it, understand it, and be better off for it.
There’s a lot more to be learned from Tony, who has a brilliant and continual slew of informative videos like this one. See more from his YouTube channel.
30 Comments
Leave a Reply
You must be logged in to post a comment.
Problem is that Tony Northrup isn’t always right. Think of it like this – someone takes a photo with a camera and lens setup for the prevailing light conditions to say f8 at 1/60th sec at 200 ISO.He then decides to change the lens, there’s no need to alter any of those settings to obtain the same EV. Now he decides to go back to the original lens, but on a different body with a different sizes sensor – again no need to change the settings for the same lighting conditions. The same goes for when he uses the 2nd lens on the 2nd body. It’s that uniformity of apertures on different lenses and shutter speeds on different camera bodies and ISO on different sensor sizes that make the EV the same for any combination for the same lighting conditions. Of course if the lens/body/sensor size changes were made for a different depth of field, or a different amount of noise/circle of confusion etc, etc. then that’s different. But a combination of the same aperture, shutter speed and, ISO will all render the same EV for the prevailing light at the time. I have put this to the test with a Pentax full frame camera (K1 ii) an APS-C Pentax K70 and 4 Pentax lenses – F1.7 50mm, DA f2.4 35mm, DA f3.5 18-270mm and FA f2.8 24-70mm lens. At the same aperture, shutter speed and ISO for all lenses set up as close as possible to the same focal length, focused on the same subject in the same lighting conditions, the EV was exactly the same for each and every shot. I stand to be corrected with any evidence you have to the contrary.
Thank you very much for those very clear explanations. However, although I understand very well the logic of dividing aperture and multiplying focal length when passing from full frame to APSC, I don’t really see the logic of dividing ISO as well.
Can you explain please?
I just found and saw Tony’s video clip as I was searching the internet on this topic. Tony makes a valid point for which I created the proof. However, I came to the same conclusion but I used a different approach which I want to share with you. Unfortunately, I have to use a picture for the maths as I don’t think this site can handle mathjax to display equations.
This isn’t true same focal with same aperture will give same image project from your lens to the sensor!! so same depth of field!! but the crop sensor is only taken a part of that image and filling your picture with it. So same focal length different picture but same dept of field. He Is changing the focal length to get the same picture of course the depth of field will change . Think about It ;-)
Great article
This was beyond helpful! If I had only known about this a year ago! haha I currently use a camera with a cropped censor but will be making a upgrade to full frame sometime next year, so until then this info will be very useful. Thanks
What i gathered from Tony video, is that the ISO on crop cameras are boosted to at higher number, and labeled as the correct number for exposure purposes.
The comparison to water that instructors often use, can be used here as well. If a smaller sensor has the same number of pixels as a larger one, each pixel has to be smaller. Because of that each sensor pixel, is hit with less photons in the same amount of time than on a larger sensor. Because of that a smaller sensor is noisier, and Tony did in fact show that if you take ISO and aperture crop into account, the noise level is the same for crop and full frame.
Manby I’m dense, but it makes total sense to me. Our crop cameras won’t suddenly take worse (or better) pictures because of Tonys video, so no need to get defensive about it.
“A crop sensor is noisier because the pixels are closer together” – maybe, but on a full frame or medium format, the pixels are right next to each other as well – i don’t see any gaps in my photos ;)
My guess is a crop is noisier because each pixel is smaller, and thus gather less light…. and needs to be boosted, so we are able to get the right exposure.
Crop Factor do not effect aperture, nobody get cheated.
http://www.bobatkins.com/photography/tutorials/crop_sensor_cameras_and_lenses.html
This is comparing apples with oranges. You have managed to confuse yourself because you are approaching this question from the very answer you are seeking. I am suprised you think that aperture needs to be multiplied. Here is a simple test for you (you can do this with your existing images). Keep everything the same (aperture, ISO, focal length, shutter speed) and just crop your full frame image to match the crop sensor frame. Are you suprised a little? Yes the image is exactly the same. This is because you haven’t compressed your background as you did when you used a 200mm and compared it to to your 100mm. This is why the lens has a real aperture value not an arbitrary value equivalent to an arbitrary format that is termed as ‘standard’ (for whom is it standard by the the way? Most people holding a camera these days have never seen or felt 35mm film). As for the comment about light per square meter and light gathered by sensor.
Physics 101:
The amplitude of a light wave is related to its intensity. Intensity is the absolute measure of a light wave’s power density. Brightness is the relative intensity as perceived by the average human eye.
You can tell from the above statement that as light falls unobstructed towards a surface (e.g. sensor) the intensity is the same no matter how big the surface. This is why settings don’t change in the same light situation for different sensors and it is also why the same amount of light will pass through a specific aperture no matter which body you use. The absolute measure of the aperture is: how much light it lets pass through the lens. Therefore the aperture written on your lens is the real aperture.
I don’t think he’s just talking about the amount of light (intensity), but maintaining the same amount of light when rendering the same DoF on a smaller sensor.
Doesn’t DoF differ because of the variety of angles your aperture settings allow the light to fall on your sensor?
I think Tony did a great job of explaining all this! I’m actually quite surprised that other people still get confused, or argue the wrong way, about all this.
This is why I’m excited for cameras like the Sony A6000, and lenses like the Rokinon 12mm f/2.0. It can give me roughly the same image quality, focal length, and DOF as a full-frame camera with an 18mm f/2.8 lens, however it does so for FAR less cash and in an incredibly small system size. This particular setup, for example, would allow me to run night timelapses using multiple cameras instead of just one big fat honkin full-frame camera and a 16-35 2.8 Heck I could probably cram THREE A6000’s and 12mm f/2’s into the same bag space as that full-frame setup!
=Matt=
Right, Matthew. What a lot of people seem to miss is that a perceived disadvantage of smaller sensors – more depth of field (DOF) – can actually be a benefit (unless you really want extremely shallow depth of field most of the time).
With a smaller sensor, you gain DOF. You don’t have to stop down as much as with a larger sensor for the same DOF. So you can use a faster shutter speed or lower ISO compared to larger sensors (thus counteracting the noise disadvantage). Not to mention that lens and camera could even be designed smaller/lighter (if it’s not a “full frame” lens used on a “crop” sensor). Oh, and smaller/lighter lenses, allow for more effective image stabilization because the stabilizer has to shift around and hold in place less weight.
Lots of advantages, if you think about it, and the reason why I migrated from Canon 5D Mark II to Olympus OM-D E-M5 and now E-M1. I use my 42.5 mm f/1.2 lens without ever stopping down for an entire portrait shoot, even in low light, whereas I’d have to stop down to f/3.5 for comparable depth-of-field and adequate sharpness/resolution on my 5D Mark II, where I’d probably also have to raise ISO by two stops to compensate.
Of course, the stabilized lens also lets me shoot hand-held at slower shutter speeds than I ever could with the equivalent unstabilized EOS lens, getting more real light onto the sensor instead of having to artificially boost brightness through raising ISO. All that with less weight that would slow me down and fatigue me.
And even if I had to raise ISO, the newer technology of the OM-D cameras compared to the old Mark II would mean at least the same or even better image quality at the same ISO levels – something I tested thoroughly before making the switch.
That’s just my own situation, but it would be a real pity for people to rule out certain possibilities just because they are mislead to think “full frame” is the holy grail and everything smaller is inferior or something, especially if they haven’t even tried it themselves.
While there is a difference, I’d be careful in terms of confusing people with more numbers or pushing a format. The DOF is also largely determined by positioning of subject vs camera & background.
On smaller sensor size cameras, you can easily compensate by buying faster lenses. I have the 1.2 nocticron and most people with FF cameras won’t have lenses that fast. Of course there are exceptions, but only very few people can handle shallower DoF, due to focusing&positioning requirements. I often see portraits where one of the eyes is out of focus, and I doubt it was done intentionally…
I do agree with this point, Tom- I do just fine with 1.5x crop sensors, for example, by simply minding my background positioning and subject distance. Want shallower DOF? Often times a step to the left or right will frame your subject against a background that is 100 feet away instead of 50 or 10 feet away, thus dramatically improving your shallow DOF look.
https://slrlounge.com/photo-of-the-day-a-kiss-at-sunset
(Shot on a 1.5x crop sensor with a 50-150mm f/2.8 crop-sensor lens. DOF / bokeh looks pretty nice to me!)
Tony did a very good job with this video. Only very few people who are teaching photography techniques pay attention to this non-trivial topic of crop factors.
My understanding is lens brightness is calculated independent of sensor size.
An M43 Lumix 12-35mm and FF Nikkor 24-70mm are both legitimate f/2.8 lenses.
Imagine two sky lights of differing diameters each shedding a circle of light on a dark floor. Neither circle of light is brighter or dimmer than the other.
Perceived depth of field can vary due to the so called “crop factor”, but not brightness.
I found this useful: http://www.dpreview.com/glossary/optical/focal-length.
Thank you. :)
I saw this video being shared in some forums with people trying to argue that an f2.8 on full frame becomes an f4.2 on crop sensors. This video is doing more harm than good for the novices who are trying to learn the exposure triangle. If you want to say the DOFs are different then fine, but don’t say its actually a different equivalent aperture because these people are thinking of exposure when they hear that.
f/2.8 on an APS-C (1.5x crop) creates similar images to f/4.2 on a full frame body, so that’s correct.
I do specifically say in the video that it doesn’t impact your exposure, and that crop factor exists only for the purpose of comparing the resulting images you’ll get with different lenses and sensor sizes.
You’re 100 percent right. And testimony for are the comments on this video, where people are expressing their willing to trade their smaller sensor cameras. tony’ video is much more confusing and misleading (not for me) than any manufacturers marketing slogans have ever been. What is his reason for doing this?
My understanding is lens brightness is calculated independent of sensor size.
An M43 Lumix 12-35mm and FF Nikkor 24-70mm are both legitimate f/2.8 lenses.
Imagine two sky lights of differing diameters each shedding a circle of light on a dark floor. Neither circle of light is brighter or dimmer than the other.
Perceived depth of field can vary due to the so called “crop factor”, but not brightness.
I found this useful: http://www.dpreview.com/glossary/optical/focal-length.
Thank you. :)
Great comments and thx for the kind words. I was thinking it was the lighting on second thought esp now that you mention it. For some reason ive been reading up on lens defects e.g. lens compression and was wondering if that played a role in side by side images at their equivalent ranges. From my understanding if the ranges are equivalent then the compression or flattening effect should be similar? or I might be totally off
Ray
Like the video – helpful and well explained. The only thing I’d disagree with is the call to camera manufacturers to publish crop factor differences in aperture. Think this would actually add more confusion as the aperture is correct when it comes to exposure and that is the most important thing.
Well, the focal length is correct when it comes to exposure, too. The camera settings will be the same for a given amount of light, with respect to shutter speed, ISO, and aperture, but because both ISO and aperture measure light *per square inch*, you’re getting less light with the same settings when using a smaller sensor… and to me, the total light gathered by the sensor (the single biggest factor in image quality) and depth of field are more important than the settings.
The fact is, ISO and aperture are metrics designed for the film era, and we’re stuck with them. Both are very misleading when you compare images taken by cameras with different sensor sizes, because they weren’t designed for that purpose. That’s why I made the video; I’d like people to know how to use these outdated metrics to accurately predict what a final picture will look like on different, modern cameras.
Anyway, I stand behind wanting camera manufacturers and websites like this one to either stop using crop factor, or apply it equally to both the focal length and the aperture. My only beef is when they apply the crop factor to one and not the other, because that doesn’t provide an accurate comparison of image qualities. That leads to people spending thousands expecting results they can’t get, and that disappoints people and hurts photography as an art.
I’ve been following both SLR Lounge and Tony Northrup’s YouTube channel for quite some time, and have always respected both of them. When Tony first published this video on his channel, I watched it but didn’t comment, but now that it’s featured here I feel compelled to respond.
While I applaud Tony’s attempt to clear up some misconceptions about the so-called “crop factor”, and find his visual comparisons a neat idea, I’m afraid that the conclusions being drawn by some viewers/readers are unfortunately quite misleading.
I didn’t comment on the video because I believed Tony knew better and it wasn’t his problem when some people misunderstood what he tried to demonstrate. But reading his comment here makes me wonder, so I’d like to clear up some misconceptions:
“because both ISO and aperture measure light *per square inch*, you’re getting less light with the same settings when using a smaller sensor… “
That’s simply not true. Nowhere do ISO or aperture measure anything by sensor size.
“The fact is, ISO and aperture are metrics designed for the film era, and we’re stuck with them. Both are very misleading when you compare images taken by cameras with different sensor sizes, because they weren’t designed for that purpose.”
Aperture, like focal length, is wholly independent of the capturing medium (be that film or sensor). Those values don’t change, no matter what camera the lens is mounted on, that’s why they are such important specs.
What changes based on sensor size is field of view only, which makes it look like a different focal length if you compare it with the historically widespread 35mm film format’s well-known focal lengths. A 42.5 mm lens on a 2x “crop factor” camera has the same field of view as a 85 mm lens on a “full frame” camera. It’s still a 42.5 mm lens, though, and thus has more depth-of-field than the longer 85mm lens because of the smaller focal length. (Other differences, like bokeh, are either a result of the depth-of-field difference or because of different builts of lenses.)
What doesn’t change at all is the amount of light hitting each sensor pixel – that’s why exposure settings (shutter speed, aperture, and ISO sensitivity) are the same for a given amount of light, independent of sensor size. So the amount of light passing through a f/1.2 lens is the same (neglecting the difference between f-stops and T-stops since transmission is a lens property and naturally independent of sensor size), no matter if it’s mounted on a “crop” or “full frame” camera.
ISO, contrary to aperture and focal length, is based on the medium – film or sensor. Talking about digital photography, ISO generally speaking depends on sensor sensitivity, noise, and the appearance of the resulting image.
The reason why smaller sensors tend to be noisier is not because there’s less light per pixel, but because pixels are smaller (to get the same resolution/megapixels as a bigger sensor) and have a worse signal-to-noise-ratio because of that. Technology is steadily progressing, though, and current cameras (even with smaller pixels) tend to have less noise (thus higher ISO abilities) than older models.
I know I’ve been pretty technical with my explanations, but it’s important to understand these concepts if one doesn’t want to come to wrong conclusions. It would be fatal to draw the conclusion that only “full frame” sensors provide great image quality and that “crop factor” cameras (and their lenses) are advertised with misleading specs.
Remember, “full frame” is just a newer name for the old 35 mm (135) format, which is “small format” compared to “medium format” and “large format”. So “full frame” is also a “crop” size when compared to the larger formats, it’s just a matter of (an arbitrary) convention.
For further reading and better understanding, check these links:
https://en.wikipedia.org/wiki/Aperture
https://en.wikipedia.org/wiki/F-number#T-stop
https://en.wikipedia.org/wiki/Film_speed#ISO_12232
http://en.wikipedia.org/wiki/Depth_of_field
Finally, I hope I didn’t come off as too negative with this comment or bore you guys with technical stuff. In the end, it’s not the size (of your sensor or camera) that matters, but how you use it (to create your images). :)
Dead on, Stefan.
These comparisons need to be done with all of the same settings and taken at the same time, with maybe a trigger on both cameras.
the light could have changed a bit. All of the variables need to be locked in to prove that the difference is in the sensor sizes.
more like a science experiment, you know…
Certainly the light did change, and that aspect surely could’ve been controlled in say a studio. However, to illustrate the differences and how to measure equally the settings needed to be changed. For an example, to show what settings are required on a 4/3 shot to achieve a similar look in a full frame – the values had to shift. Cheers, Jason.
I’m looking at the comparison images, it looks like her face seems flatter in the Olympus photo vs the canon photo, or is it because the model is at a different angle when taking the photos?
Hi Ray, I think it could be partially due to a few things. One the light is different. In the video you’ll see the model stays in the same spot (close enough), and the lens remains constant. The Micro 4/3 shot is softer though – this could be due to the aperture being lower. Not entirely sure. Good question, I’ll go inspect the video again. Cheers man – and you’ve got some nice bridal shots!
Longer focal lengths do compress facial features, but if they’re same with the crop factor, the compression should be the same, too. But Chelsea’s human, and she moves and such, which could be a factor. Anyway, once you account for crop factor, facial compression doesn’t differ.
Kishore mentioned that the Olympus images seemed softer, and I didn’t notice that when looking at the original photos, nor is it that’s not really part of what I was trying to convey, but it would be a bit softer because the pixel density is so much higher on the micro four-thirds camera, and the pixel density probably exceeds the full frame lens’ resolution. That Canon 70-200 is outrageously sharp, but it’s designed for full frame cameras with a lower pixel density.
Oh, and thanks for sharing the video, Kishore!