This video has serious errors and should be taken down. It's really an embarrassment to the channel.

Tony & Chelsea Northrup thank you for revealing this matter. I love this topic it really helps me in choosing lens.. I want to buy your stunning photography, please tell me how can get the coupon code. Thank you.

The explanation is totally wrong and misleading.

@@xiaodisunzhou8496 I agree! People should look at this explanation instead: kzfaq.info/get/bejne/kLuSiJmW3bWylIk.html

I’m thinking of getting your books now. This video was well explained. If now only people could listen what was actually said and what was not. This comment section is like comedy. Really, really sad comedy.

The amount of light hitting the sensor is 4 times less as well. You would need an F/1.4 lens to compensate for the total light and DoF.

That's right, it's why full frame is always better at low light and less noise for this reason.

With plenty of light in a scene the FF will have less noise. With low light in a scene and the same amount of light on both sensors they will have the same amount of noise. I have never seen proof against this fact.

Are you saying a 5D and an 80D with same aperture/shutter and iso setting will exhibit the exact same noise in an image?

oo0Spyder0oo Considering that the 5D is so old the 5D will be marginally better. This test is unfair to the 80D because it's receiving 2 times less light.

It's cause people are in denial about this it seems.

or because if you study at least a little bit of the physics you would get that already, but hey, let's just watch youtube videos

Jagh Haringenamn Yes, it is confusing because the aperture does not change at all, the Northrups are just offering A WAY OF COMPARING depth-of field results on different sized sensors. THE APERTURE IS WHATEVER IT SAYS ON THE LENS...that's how much light it will transmit, so there is no need to worry about your exposure settings. If you think about it, the light level is the same at the centre of the image cast by the lens as it is at the edges of the image. Even if it was a tiny little sensor from a cellphone, the light level would be the same.

Iacuzzi is your degree from the university of snobby KZfaq commenters? 🤣

because this is simply not true. it would only be true if the crop sensor would be the exact same sensor technology as the full frame sensor. this guy's theory entirely ignores this. it is so easy to disprove. a sony a7r ii has two times better noise performance than a canon 5D ii yet both are full frame sensors. so according to this guy a f2 sony lens becomes f4 when you mount it to a canon 5D 2 body? nope. Equivalencies like this are totally useless because different companies use different technology for their sensors signal processing, this is why the newest m43 camera's sensor has better noise performance than canon aps-c cameras. We have to accept that different sensors will behave different way and use the depth of field equivalency only because that is useful, the exposure equivalency is useless. And another thing: if there is plenty of light, the noise difference will be unnoticable anyway- in daylight the crop sensor's noise level will be below the noticeable limit just like the larger sensors noise level. I can see why he uses it but it simply doesn't hold in all situations so he shouldn't spread it as a fact.

You shouldn't. FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

You've been mislead again, actually: FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

@@thetwinkleturnip godamn. such a big response to a comment I made 4 years ago! LOL. appreciate you clearing it up :)

@@riyazes My take on this: Crop factor will NOT change your f stop. You will not get more or less light. Your bokeh will be the same. You can think of it as just cropping down a full frame sensor in photoshop The 'equivalent focal length' is just a way for us to imagine how different focal lengths will look across sensors.

@@riyazes Get mind blown!!

@@fatboyslimz2554 haha. 4 years later still mind blown LOL

Fredrik Lindberg actually not only dof is affected, but also the view angle and the background compression. so the fuji f 1.2 behaves like a full frame f.1.8 in all except exposure, which remains f1.2 ... disclaimer I have. a Fuji mirror less system so I'm not biased!!

Nicola D'Elia Exactly, viewangle + background compression changes according to focallength. This means we agree since I wrote Tony could have stated that only the bokeh and effective focallength is different on crop sensor cameras, ergo your fuji f1.2 behaves lika a full frame f1.8 when comparing the images (with the exception of exposure which as you said remains f1.2). I never claimed only dof is affected..:)

Thanks!

Correct about the bokeh, DoF and aesthetics of the image. However, Tony you are mistaken about the light transmission decreasing on a crop factor. In Physics, a lens that allows for a certain amount of light over a given area gives the SAME theoretical values over the whole surface and does NOT change if you decide to only use the central portion of the same sensor instead. You need to run the same test using the same sensor and the same lens, once in Full Frame and once in Crop mode (as I did with my Sony A7RII, for example). Then open the two images in Photoshop and sample the very same area using the Eyedropped/Color Picker (it helps using an object with homogeneous color, like a colored sheet of paper, sampling the same point, say, near a corner). If both photos (FF and crop) are taken with the same shutter speed and f number and ISO then you will find that the exposure values (RGB, Luminance, etc) are pretty much the same, which will show you that the light hitting that very portion of the sensor stays exactly the same. In fact, you get to keep the same shutter speed and the same exposure, differently from what you'd get if you decided to use a Tele-extender instead.

​@andybertozzi Duh, of course exposure is the same, but the total amount of light across the whole sensor is what matters for photography. When you crop to 1.5x, you've multiplied the area of every speck of noise in your image by 1.5^2, so the cropped image is a little more than twice as noisy. This is exactly what happened if you stop down from f1.8 to f2.8 (actually more like f3.1) on a full frame camera. Hence why a cropped image looks exactly 'equivalent' to a full frame camera stopped down by 1 and 1/3rd of a stop.

Unfortunately, he's not correct and it is just tragic how many people have been mislead by this - FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

Yes I’m replying to a half-decade old comment, but I think this is the most important thing to consider. It’s not so much that the aperture is different on a crop sensor (it’s not), but the background compression compared to full frame will give different results that mimic different aperture. I think.

The Enthusiasm Project why does MFT have an iso of 2 stops lower compared to FF?

@C T the depth of field for lenses is the range of what’s in focus so it directly affects bokeh And this vid shows that 100mm on a 2X crop sensor doesn’t produce the same DOF as 200mm on full frame So what ART and BRI said was half right, if u put the same lens on a FF and APS-C camera, and cropped the FF to match the framing of the APS-C, you’d get the same amount of bokeh

FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

This is like saying “no one has ever died from falling, it’s the impact that kills them.” It’s technically correct but ignores the intrinsic link between the two things.

@@jomsies so if i want a ff image of a 23 f1.4 on a aps-c sensor.... im doing 23mm by f2.8? i suck at math but get the concept but im just lost with all the reading here

The formula for the depth of field is D = s / (1 ± N × 0.03 / c × (1000 × s - f) / f²) s: the distance (meters) N: the aperture value c: the crop factor f: the focal length (mm) The crop factor is there.

he used 200mm on the full frame and 100mm on the micro four thirds because that would make them about equal focal lengths, the 4/3 being 2x crop factor. So now that they are equal in focal length, he's explaining that due to the 200mm on the full frame, the aperture for that focal length gives more blur than the 100mm does, which is right. I get more blur/better bokeh from a 100mm EF lens on my full frame than the same on my APS-C body. Yes it's not exactly applying crop to the aperture, just showing that to achieve the same blur requires different aperture settings depending on the sensor in use.

Tom, I think you made a little mistake when you wrote: "Technically you could get more bokeh on a MFT lens because you would have to move the camera closer to the subject to get the same relative frame as the full frame camera. Decreasing object distance to camera decreases depth of field = more bokeh." Actually, when using the same focal length lens on two different sensor size cameras, to maintain the same relative subject size, you have to move twice the distance AWAY from your subject with the MFT camera vs. a Full Frame - since the field of view of the MFT appears to multiply the size of subject by a factor of two compared to a Full Frame camera. Thus, the need to double the distance between the subject and MFT camera also increases the DOF over the Full Frame (when trying to maintain the same field of view). Therefore, with any given lens, the MFT camera has less DOF and bokeh than a Full Frame camera when keeping the same relative subject size (field of view).

Randy Tesch Good point! Thanks for explaining that to me.

No problem, Tom! We are all just trying to learn new things and help each other. rstesch.smugmug.com

Tom Hartvigsen thank you 😊

Carsten Linke Thanks, Carten!

AminTheMystic No. It may look like it did, but it did not. Two images taken with the same lens at the same focal length an FF sensor and 2/3 crop sensor will have the same perspective although the smaller sensor will only show part of the image of the larger sensor. On the other hand if on the FF sensor you change the focal length so that the area of view matches that of the the one taken with the smaller one you also change the perspective. So yes, the subject on the FF sensor at 200mm will be of the same size as on the image taken with the sensor of a crop factor of 2, - that's the reason why people think the crop factor modifies the focal length - but the size of the other elements on the photo relative to same sized subject will be different.

AminTheMystic the difference is cropping a 100mm lens at 2.0 crop factor is not the same as 200mm FF at the same apreture. a 200mm lens will have a image that is twice as big, so will the bokeh at same apreture and same distance. cropping does not change depth of field but zoom does. the video is a bit confusing about the aperture times crop factor reasoning. you need to set the object at further distance on a crop sensor if using the same lens taking the shot meaning you are increasing depth of field as if you are using a lens with a smaller aperture. the matter of fact is the appreture does not need to be multiplyed to the crop factor because your settings will be the same to take the shot. its only the depth of field thats changing.

I think you're overthinking this... The concept of crop factor exists only to make it easier for us to compare the results we get with different sized sensors. As I mentioned in the video, crop factor doesn't impact camera settings; it impacts results.

Tony Northrup Maybe I am to clever by half :) - still I think your video may be misleading. In fact from most of the comments I read I got the impression it was. The crop factor is the numerical factor the focal length of a lens on a crop body must be multiplied to determine what focal length lens on a full frame body would render the main subject to be the same size as on the image taken by the crop sensor - nothing more. What you say about the equivalent apertures that give the same DOF on both type of bodies is a good rule of thumb though.

Andreas Falco the keyword is that the sensor size changes only the field of view and nothing more. The crop factor helps to describe the field of view by adapting the focal length so that the field of view matches to the field of view what you would get on a FF body.

FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

Read sdp.io/crop if you want to understand this topic.

Tony & Chelsea Northrup can you argue?

I’m wondering if you actually watched this video, are you suggesting the examples shown are purposely edited to be misleading? They literally show the obviously different amounts of blur when compensating for field of view only (100 m4/3 vs 200 FF), and to open the aperture in the m4/3 to achieve the same field of view AND shallow depth of field from the same distance. This was very clearly shown and is absolutely correct. If you do what you’re saying with 100 mm 1.8 lenses on 2 different format cameras, yes, the depth of field would be equivalent, but the field of view will be very different (first example in the video I believe). If you do what most people wrongly say to do, “just back up,” you’re changing the framing and *angle of view* entirely, plus depth of field gets deeper the further you move away from the subject. The foreground and background will have a different relationship to each other, bringing compression into the equation. Whether that’s important in your use case or not is up to you, but the info presented in the video is 100% correct.

no. you can't move closer to your subject if the subject doesn't fit in the frame. the larger the sensor, the closer you can move at any given focal length, so subject distance depends on crop factor too. hence Tony is right.

+Oskar Thunberg No... with a Fullframe camera you would have to move closer to your subject to achieve similar framing thus giving you shallower DOF. If you left the cameras at the same spot, and simply cropped the full-frame... the shots would be EXACTLY the same! You are wrong and tony is misinformed.

+Charles J Gartner I totally agree and I believe Tony do to. I think you misunderstood the clip

+Oskar Thunberg Probably... or maybe tony is oversimplifying something that has many factors: Distance, Focal length, aperture, etc. My only point is, a lens always behaves the same way, cropping the sensor does only that... crop the exact same image. It's like putting masking tape on a full frame camera to crop the sensor, you're not going to get a different image, just a cropped one. The lens wont suddenly be confused and behave any differently.

+Charles J Gartner simply put: it's no secret bigger sensors give the possibility to shoot images with shallower depth of field. to get an idea of this effect, it's a good idea to apply crop factor to aperture. this does not mean the lens does anything differently because the sensor is bigger, it only means: what you can DO with the lens is effected by sensor size.

Practically, if the only thing you care is background blur, he is right. But the luminosity of the lens actually means another thing, the quantity of light that comes into the lens. A micro4/3 f1.8 lens is TWICE faster than a FF f3.7, but produce the same blur. This means that from the point of view of lens quality and specs, this video is misleading.

iDX1701 Thanks!

Don't forget that an F4 will need more time with the shutter opened to get the same exposure as your 2.8 lens (if you have one on your APS). Alternatively, you can shoot with the same shutter speed, but with higher ISO (more noise - if that compensates the benefits of higher resolution on the full frame, it's up to you). Or you can by a 2.8 zoom for your full frame, in order not to go higher at the ISO, but you are spending a lot of money. Unless you need professional quality photos, you may not need a full frame camera... Look, a Pentax K-50 (pentraprism, 100% viewfinder - where can you find that on Canon/Nikon for the same price?) with a Sigma 17-50 2.8 will do a very good job. A Sony A6000 would also be great, if only that had a decent lens! (Like a 2.8 standard zoom, not that... "power zoom" that sucks.)

Correct

Right, as I mention, crop factor (whether for focal length or aperture) isn't a factor when calculating your camera settings. The concept of crop factor exists only to compare the images that a lens is capable of creating. If you didn't apply the crop factor to the aperture, you'd get the same shutter speed and ISO, but the noise in the image would be much higher. If you do apply the crop factor to the aperture, you'd be using a lower ISO on the smaller sensor, providing noise levels more similar to that of the larger sensor.

No you don't, as with a smaller sensor you need to use lower ISO in order to prevent noise.

@@okaro6595 and so when you lower the ISO (because the sensor is smaller on a crop) to get less noise, the image is dimmer. That is why a 1.8 on a crop is only as bright as a 2.8 on a FF, when keeping the ISO to the same low noise level.

Its all about F number. Even on full frame high F number will result in DoF rather then Bokeh.

I think that is saying more about Mr. Northrop than his understanding of the subject. FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

Unfortunately it's wrong. FOV radius at any given working distance is not linear, while DOF with regard to working working distance is. Anyone who has spent any time using a camera should intuitively realize this. What this means in practice isfilm plane This means that any compensation for depth of field will not necessarily double the projection of magnification, so moving in closer won't necessarily compensate depth of field for crop factor. This disparity naturally will decrease with focal length as FOV approaches zero degrees, and is why between 100mm and 200mm the difference in the two images is negligible. However, the same is not true of wider lenses as the FOV approaches 180 degrees (as it is applied to rectilinear lenses).

If you were both using the same aperture, shutter speed, and ISO, they should have the same level of brightness. The crop factor won’t affect your exposure. It affects the depth of field, angle of view, and image quality. If you were seeing different exposures with exactly the same settings in the exposure triangle, it’s probably because of manufacturers playing games with their ISO values to make their images seem cleaner at a given setting. I know from experience that Fuji does that, and some others do too.

Because it is not true. The video is misleading. The real thing that changes is the crop factor. Luminosity of the lenses stays the same in all the systems. At parity of lens, the picture of micro4/3 are basically the same (except for noise) than FF ones, but 2x cropped.

Drew Davis Thanks, and you're welcome!

shaolin95 Can you explain how this is not accurate? I am not a photo pro by any means, so I am asking honestly.

shaolin95 tkarlmann You have missed the point. Most of the arguments against Tony's video have to do with the way people understand these topics (aperature/Fstop, focal length, 'crop sensor', equivalency, light gatherning etc). The point is that most non-FF cameras/lenses will be marketed in terms of FF focal lengths - take the new Panasonic LX100, for example. It is a '24-75mm' lens according to marketing material, but it's clear this is FF equivalent. It is also claimed to be a '1.7-2.8' lens. So would a picture taken by the LX100 @24MM look identical to a D810 w/24MM 1.7 lens, everything else being equal? Absolutely not. But it is marketed so that you would think you're getting a 1.7-2.8 equivalent lens. To me, this is like selling a product in the US as costing "200". Everyone here would assume you mean it's 200 dollars, but in reality, the manufacturer is saying it's 200 British pounds; very deceptive. Either be consistent with your scheme for specifications, ie, a ~12-37MM lens, F1.7-2.8 or say it's a ~24-75 F3.4-5.6 (FF eq). I believe this is what Tony was explaining, and giving examples of what to expect as far as photo quality and DoF. All the other stuff is beside the point, and moot (open to discussion).

shaolin95 Yes, you have explained this incorrectly over and over. f-stop of an optical system is the ratio of the lens's focal length to the diameter of the entrance pupil, it is a simple equation. So when you use crop sensor and make that FF spec'd 100mm into an effective 160mm, the ratio also changes because the entrance pupil is the same size but the focal length has increased. I'm not going to do the math but let's say it makes an 100mm f/2.8 into an 160mm f/4. Let's put it another way and this will blow your mind since you seem to think this is about light transmission not a ratio of focal length to pupil size. If they designed a 160mm f/4 made for the 1.6x crop factor sensor size it would transmit the same amount of light per area as the 100mm f/2.8 on a full frame (or that same lens on a the 1.6x crop sensor).

Drew D the problem is that you are trying to measure 'blurriness' with raw numbers. You have to remember just like 'field of view' is a relative term, the 'blurriness' is also a relative term. When camera manufactures write F1.7 that is a totally accurate physical measurement and you can't say its 'like' F3.4. F-number is the ratio between focal length and the diameter of the aperture and it will always remain constant for a particular lens regardless of the sensor format you use it on! Technically it has nothing to do with 'blurriness'. Blurriness is just a bi-product that we happen to appreciate. Like guitar amp distortion in rock/metal music.

That looks like the image stabilization of the lens, and the reason why you should disable it when you're on a tripod.

Oliver O'Neill No, it was not the lens IS system, I don't know why the the camera was shaky, but the jiggle effect you see in the video was made by applying stabilization in post.

***** weird, I've gotten that same effect with IS. had to disable it for it to behave. Maybe it was just confused.

IS should always be turned off when using a tripod.

Oliver O'Neill And now I saw that Oliver already mentioned this…lol. KZfaq hiding comments… ;)

kratuna0 Thanks!

The depth of field depends also on the sensor size but in the opposite way people think. Smaller sensors have shorter depth of field with the same focal length because the image will be magnified more. The formula does the depth of field is: D = s / (1 ± N × CoC / c × (1000 × s - f) / f²) s: the distance (meters) N: the aperture value c: the crop factor f: the focal length (mm) CoC: The circle of confusion on full frame, typically 0.03 (mm). The + gives the near limit, the - the far limit (if negative then infinity)

Yeah some pukka property there. Keep up your acquisitions ;-)

Adanna Egbe 😂😂

pay no attention to this. it's rubbish. Basically what's going on is this: Lenses focus light into discreet "packets" called the "Circles of Confusion", or "CoF". When you look at OOF portion of an image, these circles are sufficiently large that they read as being "blurry". That is the reason why you get "bokeh balls" through things like leaves. So when we're talking about sharp focus, what we really are saying is that the CoF is sufficiently small enough to read as being "sharp". This is also why larger apertures give less FOV, because the CoF is larger. But the focus point is always a 2D plane, there is no "DOF" only a range of space where the CoF is going to be small enough to be "sharp". Ok. So when we change the format size the CoF is relatively larger on smaller sensors, resulting in a more blurry background. But this also goes for the whole of the image. In fact, if we compensate the magnification by changing our working distance, the bokeh will be the same, though the DOF will increase. Likewise we can decrease the aperture (say from f/2.8 to f/5.6) and the bokeh will be the same. This latter point is what he was getting at, but it's a pretty weird way of talking about it. Bokeh should be seen as a function of magnification and working distance, and not a function of aperture. But *absolutely* the aperture does not change with sensor size, only the relative magnification.

@@squirming_squirrels if you don't understand, just get a full camera and it will be better :P that's the conclusion from the video :P

Scottie Nguyen uhm. I’ve shot everything from 110 to 4x5. This video is nonsense.

You haven’t adjusted your camera position when you changed your cameras to allow for the same FOV hence F stops remain the same for each lens with the similar background blur

Just our producer's random screen grab, nothing personal :)

I figured that. I was just joking. Great video Tony.

***** I'm your first subscriber on your KZfaq channel.

Compression is only a factor of equivalent focal length... In other words, multiply the crop factor and the compression is exactly the same.

@@TonyAndChelsea Compression does not exists. Or more precisely, it is an impression which is related to the field of view, the latter is the only relevant factor.

When you keep the subject size the same (and why wouldn't you), depth of field definitely changes.

It changes because it is a 45mm lens, not a 90mm lens. It has the DOF of a 45mm lens, obviously. I wasn't doubting you, you are my favourite photography channel because you know your stuff and give excellent advice. I just had the impression by some comments that people had misunderstood what you were saying. I was merely saying it in a different way. Keep up the excellent work! I love your (and Chelsea's) videos. ☺

+Tony & Chelsea Northrup I didn't really explain things well before, but this is what I have issue with: when you compared the full frame and a 200mm lens against the Micro Four Thirds camera with a 100mm lens - you made a statement along the lines of, "we didn't get the same shallow depth of field out of the Micro Four Thirds camera we expected because we weren't applying the crop factor to the aperture". The reality is the bodies and sensor sizes are totally irrelevant. We didn't get the same depth of field because we compared a 200mm lens to a 100mm lens at the same distance and at the same aperture. A 200mm lens has a shallower depth of field than a 100mm lens at the same distance and same aperture settings, period . Whichever camera body/sensor size is hanging off the back of the lens has no effect on how that lens performs in terms of DOF.

Exactly the truth. The depth of field will only be affected by "focal length", "aperture" and "distance". With the same aperture and distance, a 200mm lens does have a shallower depth of field than a 100mm lens. Sometimes people say FF camera can get shallower depth of field than apsc/M43 camera, that's just because they have came closer to the object when using a FF camera (with using the same lens). (Sorry for my poor English, as I am not a native English speaker =P)

You get equivalent of 90 mm f/3.6. You cannot say 45 mm depth of field as 45 mm on FF is entirely different. Also check these that I got from DoFmaster. 45 mm f/1.8, focus at 3 m. Micro 4/3 DoF 0.23 m. Same on FF DoF 0.47 m. People, you have all the world's knowledge at a few clicks. Use it. It is not like the 70s when one had to do to library to learn things.

Maybe it's time to stop doing research.

Jessica Jay , see my comment, the explanation is not correct. Otherwise, all lens manufacturers become liars.

+Charles Sianturi Yeah, you'd get similar results if you used f/18 on a full-frame camera. That's why those low-light photos look so bad :).

+Charles Sianturi it also is why iphone images look so surprisingly good sometimes... if you wanted huge depth of field on the fullframe, your back at a high iso while the phone is still wide open with a low (but noisy) iso. so a big part of the iso advantage of the bigger sensor only works if your willing to shoot with tiny depth of field. and well. the max quality is much higher, as iso 100 on the phone might compare to say 1600 on the fullframe and you cannot go lower

The bokeh will be bad, it is still the same amount of light coming in. It is kind of like taking a super wide angle picture and cropping it.

@Ziggi Mon ehm. You can . Because you can use your fullframe or large Format or so and Exposé longer or use flash, keeping ISO low

@Ziggi Mon no You are just wrong.

Lam Tran I suggest watching all three of my crop factor videos for detailed information.

This comment helps so much.

She’s gorge

@@arunca190 He's quite handsome too, but a bit smug looking 🙊💍

Tony was voted the most sexy nerd a few years ago :-)

+Arson Yeah same here. Sure I only have my 18-55 (f/3.5-f/5.6) and I thought "Eh, I can live with 28.8-88" but now that it turns out it's also f/5.6-8.9 it's a bit cruddy.

+Arson Don't feel robbed. The aperture written on your lens is correct. Tony's just saying that you get less background blur with a smaller sensor camera than you'd have gotten on a larger sensor camera. He just seems to think that redefining what the word "aperture" means is a useful way to put it (it's not).

+RuzzyShuya It isn't f5.6-8.9, it's f3.5-5.6. You don't get such shallow depth of field as you would on a camera with a larger sensor, that's all. It's just as capable in low light as an f3.5 lens on any other camera. Your smaller sensor might have a bit more noise at high ISO than a camera with a bigger sensor would, but so what? Does it take great photos? Course it does.

+Graham Ashton "just as capable in low light" =/= "more noise at high ISO". That's the whole point. If your camera has more noise at higher ISO, it's less capable in low light.

+Eric Vogler Yes, the camera is. The lens isn't. In other words, an f3.5 lens is an f3.5 lens, and it's the f-stop that dictates how good that lens is in low light. The bit that Tony repeatedly gets wrong is that he states that the lens isn't as good because the sensor is less good at high ISO. It might not sound like an important detail, but *because* the f-stop *defines* how good a lens is in low light Tony's confusing the issue, trying to redefine what the f-stop means.

If you want to get technical, the two primary factors in depth of field are aperture and magnification on the sensor. Larger sensors have higher magnifications when the subject appears the same size as the picture. That's why DoF is shallower for larger sensors given the same f/stop number. Some people factor in pixel size because they define the DoF as being the area where each pixel appears sharp... but since we rarely view images on a pixel-by-pixel basis that's a fairly abstract method.

Tony Northrup DOF is smaller because circle of confusion is higher. That's it. And circle of confusion depends on sensor size, i.e the ratio between coc is equal to crop factor.

^^^ THIS THIS THIS THIS! Aperture changes the CoC, not the DOF. DOF isn't even really a "thing". It's just where the CoC is sufficiently small enough to render as sharp. So when the CoC is larger relative to the sensor then the DOF appears reduced.

Jezus...relax man.

Idiot....the f notation is flawed from beginning as a measurement of lens,its the same only as far as exposure goes in different format, but artistic effect is completely different in different format for the same aperture

The f-stop of an optical system is the ratio of the lens's focal length to the diameter of the entrance pupil, it is a simple equation. The entire world should be using this equation which is independant of camera format. So when you use crop sensor and make that FF spec'd 100mm into an effective 160mm, the ratio also changes because the entrance pupil is the same size but the focal length has increased. For exposure, your camera uses a light sensor to determine the amount of light striking the sensor, not f-stop. The camera from there can determine relative f-stop for things like aperture required for Av, exposure compensation or exposure bracketing. There is no free lunch.

+ 1 tkarlmann. Sick of people confusing this. Especially "pro's"

Robert Barta Following your logic then the center of the 100mm lens gets darker cropped, instead of brighter, when we crop away the vignette to a 160mm FOV. Cropping away vignette makes a resultant luminance gradient darker? REALLY?

+TheRcEngineer I'm not sure what you mean; at no point in the video do I discuss megapixels impacting exposure... they don't. If you want more info on this topic, visit sdp.io/crop.

+Tony Northrup Thank you so much for answering me, maybe I got you wrong. In short I understood the change in aperure naturaly changes the exposure length. I am confused now, will check your link. As I understood you said the crop factor does not only alter the focus length but also the aperture. After the first shock (considering the kit lens is probably f10) I thought this does not make sense. Because this would also imply full frame sensors with 40 MP now would laso have double the aperture comparing to 20 MP sensors. AsI wrote I will check the link, hope I misunderstood you. Thank you for your great videos, learned a lot from them.

+TheRcEngineer I think you're confusing sensor size and pixel density. Imagine a full frame image like the ones in the video, now draw a box over the top of that image the size of a (for instance) APS-C sensor, so the box is smaller than the full frame image. All of that image, and therefore that light outside of the cropped sensor is effectively not there if it never hits the sensor. It doesn't matter if your full frame sensor had 10MP and your cropped sensor has 100MP, the full frame sensor gets the whole image projected onto it where the cropped one loses some of the light that is projected beyond it's edges.

+TheRcEngineer The "change" in aperture with the crop factor does not change the lightlevels/brightness/exposure, it just change "the image" (field of view/background blur). To be able to compare one must compare not only field of view (crop factor for mm) but also background/depth of field (crop factor for f-stop).

dude, there is nothing deal with the MP. we are talking about the aperture and DOF

Right on the money, thank you!!!

This!

dofsimulator.net/en/

Thank. You. Someone with sense.

You said: "But if you were to keep the same focal length and just backup away from the subject to maintain the same framing on the subject the bokeh or background blur is exactly the same.". That's actually not correct....if you move farther away and refocus the lens, the depth-of-field increases and the background will be less blurred. For proof of this fact, just look at the DOF scales form your lens maker (or simply observe the DOF scale on your lens if it has one). If you want to keep similar bokeh you would need to open the aperture, if you have the range to open it and are not already shooting wide open. Even doing that the character of the bokeh will be different. Tony's crop factor multiplier is valid, but just an approximation BTW. Those who know optics design know why....can't explain it here simply.

ye its a common sense but people just make another theory to get famous.

Glad to hear it! Thanks!

well said.

What you are talking about are T-stops used in cinema. F-stops are the size of the aperture compared with the focal length. In other words, a 50mm f2 lens has an aperture of 25mm, a 16mm f4 lens has an aperture of 4mm and a 200mm f4 lens has an aperture of 50mm. The reason the number series is an odd one (f1, f1.4, f2, f2.8, f4, f5.6 f8 etc) is that each number allows twice the light through and each aperture diameter is about 1.41 times larger than the previous one. But set the physics aside... Tony gives you an easy to youse rule of thumb to evaluate the elements of a pictures in reference to the sensor size in a real world application, which is a great tool to have.

Glad to help!

Chaz Spellman Right, he'll have a lower ISO, but he'll be gathering the same total amount of light, and thus the images will have about the same noise, despite the Fuji's lower ISO. Well, Canon's not the best example here, because their full frame bodies are so old, but a Nikon full-frame body would compare perfectly. I say repeatedly that crop factor has no impact on camera settings. Crop factor impacts every part of your final image quality, and that's the point of this video. If you want to know more, watch Crop Factor Part 2 and Part 3. And yes, medium format cameras have a crop factor < 1, and you should apply it to both the focal length and the f/stop when making comparison.

I've watched them. I think the confusion is the Title alone. If this is a tutorial on how and why to adjust for Depth of Field then that's what it is, but there has been some miscommunication lost in translation about the part where saying all shooting data remains the same, but here are some other factors to consider, that seems to contradict, though it may very well not.

Chaz Spellman It's depth-of-field, total light gathered, and total image noise--the three factors that the f/stop influences that also impact final image quality.

Tony Northrup "Total light gathered" by the lens, "total light transmitted by the lens" does not change just because you put a different camera on the back.

Tony Northrup Yeh, except that the pixel pitch on the 50MP Canon 5Ds is the same as a 24MP Crop sensor's pixel pitch. So, what's the loss, what's the gain? If the per-pixel IQ is better on the 5Ds than on a 24MP (budget) crop sensor the problem isn't light gathered, cropping or aperture, is it? (I won't even get into the advantage of Sigma's Foveon sensors over the inherently lossy Bayer arrays of Canikon land). The differential in IQ comes from the grade of technology being used, not the "crop factor," & as Foveon sensors demonstrate, not even the fine acutance of a sensing array. As I said before, your salient point is fine: A smaller system's lenses need a larger aperture to achieve the same shallow bokeh. You're right, Tony, but then this is old hat to large format photographers whose wide angle lenses are ~ 130mm. The telephoto effect explains this differential instantly: The longer the lens, the shallower the DOF & the more diffuse the bokeh of a given F#. But every other anecdote you use to support that statement is arguable, to say the very least. Of all the anecdotal thinking I find on the internet, the worst is the fuzzy-headed nonsense that crop factors ='s teleconversion (it doesn't). Pixel-pitch differentials & pixel-level acutance can afford a relative gain, but cropping DOES NOT, never has, never will. But the only way a 1.5x crop factor affords any teleconversion or "extra reach" is if the lens is also sharp enough to resolve to the finer resolution or pixel pitch of a 24MP crop sensor versus a 24MP full frame. The REAL issue here is soft glass being sold top dollar. The crop-factor myth is hiding the sins of soft glass being touted as first-rate, when only 80+ lp/mm is adequate these days. The advent of the 24MP, 3.5u micron pixel pitch, 90 lp/mm sensor is exposing lenses to torturous uses, revealing that the lens manufacturers could bear more scrutiny for the lenses they're pushing. The only thing that's covering soft glass right now is the inherent lossage of Bayer sensors, with the 30% RGGB penalty. Instead we need to look forward to better camera tech... we've already crossed the digital-vs-film threshold for fine acutance (B&W T-Max reliably resolves ~ 85 lp/mm (not the theoretical 120 lp/mm), color film at least half that). In a future scenario, all new cameras will sport Foveon sensors (not Bayer) with true megapixel equivalence in the output & NO interpolation, no color-conversion loss, so invariably sharp. That, and the minimum standard for any lens would be 75 lp/mm, the median ~ 85 lp/mm. But with 100+ MP Bayer sensors in the labs, we'll keep fussing over (less and less) moire, aliasing & color shifts for a time, until at some point the Foveon patent will run out & then we'll all be in high cotton... :P

I can't believe people like you exist! Take an old 55mm zenit lens that "probably" means it's for 35 mm film and compare it to the kit lens. Then look back at what you wrote here! You working for Canon by the way? :) Or maybe you applied, but they rejected you for LACK OF KNOWLEDGE

Matthew Ferguson lol, you really don't get it pal. They do have different optics, but the focal lengths are the same. These lenses can be made cheaper and smaller to fit with the smaller sensor. If you would put these on a full frame you would get the same image as you would get with a EF lens (of the same focal length) only with heavy vignetting as the lens simply doesn't need to take these extra bits on the side into account.

Matthew Ferguson Ok what is this? Trying to prove you are a complete idiot? TEST IT YOURSELF BEFORE YOU TALK NONSENSE, because reading articles and listening to SMARTER people is obviously not helping.

Matthew Ferguson yes that's it, The full frame sensor 'sees' more of the image that is projected through the lens.This vignetting, though, occurs because that part of the image gets cropped off by the cropped sensor anyways, so there is no need to optimize that part of the image. And you can't fit an EF-s lens on an EF camera because of the mount, but there are EFs type lenses by Tamron and Sigma that come with an EF mounth. (like the Tokina 11-16mm) these are also lenses made especially for EFs cameras and so give heavy vignetting on EF cameras. Basically every lens projects the same exact image on a sensor though in case of a smaller sensor the sides simply get cut off. If you were to crop a picture shot on 50mm lens on a Full frame camera to 62% (in Photoshop) you get the exact same picture shot on a 50mm lens with an apsc sensor. And if this would involve an EFs lens you would be cropping off what would also be the parts that vignette on the full-frame.

Matthew Ferguson mount a 55mm EF lens to an APSC camera and you will get the same image as a 55mm EF-S lens mounted to the same camera. Focal length is not defined by field of view, otherwise, everything would have to be based on the largest of large format and work its way down, making 35mm fairly close to the bottom rung. Also, EF-s lenses cant be mounted to a FF canon camera, but DX and DT lenses from Nikon and Sony respectively can be mounted to their respective full frame cameras. Focal length describes magnification on the sensor, and regardless of crop or no crop, the magnification of a 50mm lens will remain the same on that sensor regardless if it an apsc lens, 35mm lens, medium format lens, etc. The projection area from the rear of the lens will change based on its design, but that is not part of the definition of focal length.

Tony's salient point: That on smaller system cameras, to get the same bokeh, you need faster glass, up to 2 stops faster. Look at it this way: Large format 'togs already know this, that their 150mm lenses w/ the same FOV as one of our 50mm lenses, they get much shallower DOF at the same F# as us... But Duh! That's the telephoto effect! So an Olympus Micro 4/3rds lens might want a 28/0.9 to a Canikon 35/1.4 lens to get the same DOF per FOV. I admit that Tony kinda fogged it up w/ some other statements either best omitted or rephrased (esp. the bits about less light hitting the sensor & the mystical crop-teleconversion effect )...

Well, let's get specific. Most people who have an issue with this video misunderstand the fact that aperture defines light gathered per square inch, not the total light gathered by the sensor. Why don't you read through some of the other comments for this video and see if I've already addressed your concerns. If it's something different, let me know and we'll discuss it like engineers. I'm open to correcting any mistakes, but so far, every discussion I've had with viewers (that didn't end in a rage quit) has led to them acknowledging that the facts of the video are correct.

Ehm, no. The actual aperture doesn't change. But as soon as you say that APS-C lens X compares to FF lens Y you have to apply the crop factor to the aperture as well. The reason for this is that the ctual f-number is calculated with N=f/D with f being the focal length and D being the diameter of the entrance pupil. If you want to compare a lenses field of view you might say that it has the same FoV as lens Y on a FF sensor, but you don't change the actual diameter of the pupil. If you apply the crop factor to f it automatically applies to N as well.