Summary on how to get more bokeh: 1) Have long focal length (80mm>50mm>35mm), step closer to the subject (closer>further), and have a wider aperture (f1.4 > f1.8 > f2 etc) 2) Keeping everything the same, the full-frame camera and crop sensor will have the same level of bokeh. The only difference is that the crop sensor's image will be more "zoomed in". AKA it crops in. 3) If you want to have the same level of "zoomness", you'll have to a) step back a bit with the crop sensor b) use a shorter focal length (wider) lens to get the same field of view. However, stepping back or using a wider lens will decrease bokeh as said in #1. 4) To solve this issue, you'll need to use a wider aperture lens. Conclusion: sensor size doesn't create more bokeh. The focal length, distance from the subject, and aperture size determine the depth of field. Crop sensors only zoom in (or crop) photos.
Or uhhh, just use a lens meant for the sensor size of the camera you have. Micro 4/3 sensor? Buy a Micro 4/3 lense and you will have no crop or "zoomness".
Pretty stupid statement. Sensor size has nothing to do with DOF. Larger sensor just need larger aperture or longer focal length. Sensor size itself doesn't affect DOF in any way.
This video is not wrong but the way information is presented may cause miss information. The sensor don't produce shallow depth of field, the distance to the subject and focal length does. The thing is that the crop and aspect of the sensor dictate the chose of both (distance and Lens selection).
This is exactly the way I would put it. Mi phone camera's tiny sensor is not at fault of the lack of bokeh; we have to blame the 4 mm lens that the sensor needs in order to have a decent field of view.
0:03 as you have said, "It does. But probably not in the way that you think." So it does. Of course its not the sensor size itself that affects the DOF, but the focal length adjustment that we need to do to achieve the same image coverage. And when we make the composition the same, the one with the bigger sensor would produce the shallower DOF.
That’s because not only you need to convert the focal length to match but also the F stop which I feel he missed in this video, if you do both you will get exactly the same image. A FF camera with a 100mm lens at F 2.8 to match that you would need a 50mm at F 1.4 and to make the matter worst to match the exposure you also have to convert the iso to but now things get more confusing so I will leave it there 🤣
@@phqnomenon yes there’s! You’ve to square the crop factor. If you’ve a GH5 the crop factor is 2x crop so it’s easy to figure out, let’s say you set your ISO to 400 on the GH5 to match that to a full frame you’ve to multiply 400 x 4 (1,600) if you match the 1-Focal length 2- Aperture 3-Iso you will get the same exact result the problem is camera companies are not making enough fast lenses for micro 4/3 cameras to match FF cameras
But it's not just about Knowing the answers is it? It's about understanding? When you understand, you create your own answers? Memory can rob us of Facts we have swallowed, but knowledge and understanding stays with us!
The best explanation on 'depth of field' I've ever come across. Good scientific approach; you identified all relevant variables (aperture, focal length, length to subject and sensor size) and showed what happens if you change one of them.
"Smaller sensors produce a shallower depth of field, because they have a higher pixel density." ? ?? ??? ????? ??????? Why would you want to further confuse people on this, WHY ?????????
Yeah that sentence itself is incorrect, which is ironic since the title of the video says that sensor size doesn't affect depth of field. As he explained though, higher pixel density does make depth of field shallower. I think he was trying to say that "smaller sensors often have higher pixel density, which produces a shallower depth of field" and just got the ordering mixed a bit.
Yes, the D850 will have shallower depth of field (not by a whole lot though). More pixels let's you distinguish more colours in a photo, and more pixels also let's you better distinguish how in focus different parts of the photo are. The same lens/aperture on a D5 as a D850 has the same interaction with light, but digital cameras have to record that infinite amount of light information onto a fixed resolution sensor. So on a D5 you will lose more information than on a D850, which makes the area of "acceptable focus" larger. This effectively makes the depth of field deeper.
If you keep the lens in the same place, use the lumix with a speed booster. (Say a .71) that cuts out 29% of the field of view, right? So that would mean you’d need to move the camera back less. Correct? Also the light could be converging faster with a speed booster shallowing the depth of field additionally? (Gh5 shooter over here)
This is helpful, and I can see why people get confused. So if I got this right. If you have FF and MFT, with same camera to subject distance, same lens, and same aperture, both will produce the same level of DOF or blurriness/bokeh in background. But the field of view will be different with the MFT being a much tighter shot. To match the cameras for FOV, where the composition matches, you would have to change either the camera to subject distance or the lens. If you change the lens, it will match the FOV, but because it is a different lens, it will have a different DOF. But that has nothing to do with the sensor, but the lens. And if you use the same lens, but change the camera to subject distance, you get the same FOV, but there is some difference in DOF because you changed the distance. Thanks for doing this. It is easier to remember when seeing a real life example.
The sensor itself, or in the old days, the film size, does not change the Depth of Field, 3 things do; distance to subject, ƒ-stop and focal length. The "normal" angle of view of a 35mm camera is 50mm, on medium format 6x6cm it is 80 mm and on a 4" x5" sheet film view camera it is 150mm. This means that the fact of the matter is for the same distance and ƒ stop a larger film/sensor size will provide shallower depth of field for the same angle of view. This is an optical fact that cannot be changed.
It’s great to see an explanation of DoF that’s based on optics science (physics) rather than uninformed opinion and passion for a particular sensor format. Bravo!
This is great information for people to have and helps dispel notions of things like "equivalent aperture" which only serves to confuse people. There is one point that was left out, though. You showed how the same lens at the same distance on FF can be cropped for the same result as APS/MFT. However, what is not shown is that because FF has a larger field of view, you can get physically closer to a subject to take a similarly framed shot without cropping. Take a portrait head shot for instance. 85mm FF, you frame in around the head and shoulders, and get a shallow depth of field. To get this same framing in APS or MFT, you only have two options: back up, or use a wider lens. Both of these options, as you demonstrated, will increase the depth of field, and therefore decrease "bokeh". You can always crop FF down to equivalent APS/MFT view, but there's no way to magically uncrop APS/MFT.
Slyest Fox That’s exactly what’s wrong with this video. How photographers actually do their job vs. some theoretical reasoning that tries to “disprove” the actual nature of things. Who would ever want to crop in on a 16mp MFT sensor anyway?
Yes but by cropping down to the APS-C size you will lose a little less then half the megapixels. A D850 at 45mp will be a 19mp image cropped to a D500 image uncropped is 20.9.
From what I've researched (and Tony Northrup also confirmed this) a MFT 50mm f/1.4 would be equivalent to a FF 100mm f/2.8. If you test this scenario, you should get nearly identical DOF.
Good and straightforward explanation! But in the process you have shown that practically speaking...sensor size matters. Because it is the relationship between the sensor size and the desired content which results in the image. So even though sensor size does not at all affect depth of field, the sensor size will dictate lens focal length-which will immensely affect depth of field! Same reason that medium and large format cameras, with relatively slow, but very long lenses, still can get great subject isolation and good bokeh. By debunking the myth, you actually have proved the point, that sensor size matters!
!!!!!!! WRONG INFORMATION !!!!!!!! Mate! You're wrong about Sensor Sizes! and YES, Larger Sensors Produce Shallower Depth of Field! For this experiment, They need equal situations without post-production things like CROPPING! to achieve the SAME RESOLUTIONS! So you should apply the "SAME FIELD OF VIEW" for both cameras! So, we have 2 options: 1- 100mm on full-frame and 50mm on 3/4 micro four thirds (same distance) 2- 100mm on both camera (different distance) now you will see, Larger Sensors Produce Shallower Depth of Field! In this video, You cropped the photo of the full-frame, and this is wrong! because you reduce the resolution of the full-frame photo!
I already knew all of this but Lee elaborated it perfectly and easy to understand. All you need from this video is the summary at the end. FF sensor doesn't create shallower DoF but the fact that you need a longer lens or moving closer to the subject to get the same point of view does.
Nate Kong This video substitutes theory for the reality of the real world. *Can* you get equally shallow depth of field with a small sensor? Sure. But given the gear that’s available, their prices, etc., good luck matching the DoF from a full frame 50mm f/1.2 on a MFT system. “Can it be done?!” Probably. But how about we talk about the world we actually live in? 🙈
Then can you please explain how if I take a photo with my M4/3 Panasonic with a 25mm f0.95 vs my Fuji GFX with the 63mm f2.8 , the GFX has SUBSTANTIALLY more shallow DOF and focus falloff. I have tried everything under the sun to do this comparison and it is simply impossible to create that medium format look with a smaller sensor.
Because you are using two different focal lengths. His point is that the same focal length from the same distance using the same aperture will produce the same DOF regardless of the size of the sensor. If you vary any of the three factors to adjust the field of view because of the sensor size you will effect the DOF.
This quote from dpreview sums up your videos thus far: Peter Cockerell "The compression you get using a long lens isn't a result of the lens, so much as the distance between your subject, your background, and the camera." God but these pedantic distinctions are tiresome. It may not be a result of the lens, but its observability is. If I take a shot of layered, misty mountain peaks in the far distance with a 400mm + 1.4x teleconverter, the "compression" (for want of a better term) is very obvious. If I take the same shot with a 16mm wide angle, the area covered by the original photo will occupy about 0.1% of the wide-angle shot, and will be totally lost in the rest of the image. No, the 640mm didn't create the compression, it just enables you to actually see it. I think most people understand this, implicitly, even if they don't necessarily articulate it. People who make videos like this are much more annoying in my opinion than people who say "lens compression". It's not even new or original. The same tired "explanations" come round time and time again. PS, thanks Peter. It is the same as folks that argue that sensor size doesn't affect DOF, yes yes, we all know the tech behind this but the point is there is a reason folks go full frame and medium, large format , and aren't shooting some cheap nikon coolpix camera, one of them is the sensor is bigger, and one of those reasons is it affords shallower dof in that particular package. PS, yes this is a total triggering video ;-)
Yup. Nice video. A 16mm f/1.4 on a crop sensor is a 16mm lens, it's just cropped to a 24mm FOV. 16mm won't give you the same DOF as 24mm. I don't know why this is so hard for people to understand.
question: at 4:41 the guy have printed shirt that say "nikon d850 tamron 70-200 f22" at 4:42 says "nikon d850 tamron 70-200 f2.8" how did you change the shirt without move the guy and where do i buy that shirts
Basically, having a larger sensor in and of itself does not mean a photo will have greater bokeh. However, due to having a larger sensor, the photographer must either use a longer focal length or physically step closer to the subject, which creates the shallower depth of field. I'ts not the sensor itself, it is the action required to match the same frame as APS-C or MFT sensors. This video wonderfully explains the issue, in my opinion. Great video!
Kudos Lee! You’re the first person I’ve seen on RUclips that has nailed this! Seems like everyone else has only skimmed the surface, probably because they themselves don’t have a solid handle on it. Tony Northrop came close, but he kinda glossed over the WHY. You’ve illustrated what Ken Wheeler drones on about repeatedly, yet never offers any comprehensive illustration. On another topic; Fstoppers has some of the only instructional content that I have gladly forked over money for. I found Monte Isom’s - Making Real Money course invaluable!
I've watched several explanations of depth of field on YT and yours has been one of the best, especially on Circle of Confusion,, thanks for your post I look forward to seeing more of your videos..........
The real inforamtion is that at same f-stop and same FOV a bigger sensor produces a shallower depth of field. This is missleading because you don´t use the equalient focal-length, you use the same...
just on the spot! I am sorry to see that misinformation is presented in the video. Also apart from the sensor size the FOV or Focal Length has nothing to with the depth of field as long as the reproduction size is the same. What he calls as depth of field with a longer focal lens is just the compression effect, the reason why many portrait photographers prefer to work with lenses as long or longer than 85 mm.
If you doing something that will make FOV same, it is practical not technical (physic). This is example to make this issue more clear: I am standing in front of model with my Linhof 4x5 camera and 150mm lens, aperture wide open lets say at f5.6. I cut the 4x5inch negative to make 36x24mm negative. From a full body shot now I got a head shot. The depth of field between two formats is same, because it is from the same negative. That's what we call "cropped sensor". Sorry for my English
@@maspriyayi No it is not, the negative you cut has less depth of field as when you blow it to same size errors show more. The formula for depth of field is: D = s / (1 ± N × c × (1000 × s - f) / f²) where plus gives the near limit and minus the far limit. s is the distance, N aperture value, f focal length and c circle of confusion which depends on the negative/sensor size. For full frame it is 0.03 mm. Replace 1000 with 304.8 if you use feet instead of meters for s.
No sir, what he is saying in right... there is also a lot of confusion among some people that a, say 50mm f1.8 for full frame camera has a different f stop on an APSC camera, which is also wrong... using additional lenses, such as speedboosters, is different... to sum up a a given lens of a given aperture is same on any format...say again a 135mm f 2.8 is always that, a 135mm f 2.8...
sundar iyengar except for the fact that is provides a different angle of view depending on the sensor size. A 135 f2.8 will have different dof depending on its distance from the subject. To get the same shot as ff on apsc, you'd need to back up. The light intensity remains the same, since f stop is calculated by focal length and aperture size. Apsc back up so much when compared to ff, that a ff at 135 f2.8 will have the same dof as 135 f4 on apsc given the same angle of view (same composition).
Firestarter Agreed... That's why I mentioned about image magnification... To illustrate, if a photograph is taken on a 50mm lens at say, f2.8 and the same photograph taken with a 100mm lens at f2.8, if the mage of 50mm lens is cropped to that of a photograph taken with a 100mm lens at f2.8, both the photograph will have the same depth of field... May be, the out of focus area (bokhe) May be different, depending on the lens construction and other aspects and considerations... The RUclips video is nice and good.
Focal length doesn't matter in the way you think it does, the aperture is more important. 200mm f/2.8 means an aperture f/2.8 of ~71.43mm while 20mm f/2.8 equates to an aperture diametre of ~7.14mm. This is EXACTLY why the crop factor applies to both the focal length AND the aperture. 50mm f/2 with a crop of 2x has the same field of view AND DoF as a 100mm f/4 FF lens. The distance to your subject matters a lot since it goes into the equation squared. DoF ≈ 2u²Nc / f² = 2u²c / fD N = f / D u: distance to subject in mm N: f-stop/-number f: focal length in mm D: diameter of the aperture/entry pupil in mm c: circle of confusion (acceptable circle of confusion, that is perceived as a single point) c is actually often very arbitrarily picked but it should probably be around 1.5x to 2x the size of a photo site. The f-stop is not absolute, it is the ratio between focal length and the diametre of the so called entry pupil (or easier: aperture). N = f / D with N being the f-stop, f the focal length and D the diametre of the entry pupil. Of course, f and D are in mm. This also shows that the f-stop is unitless because mm / mm cancels each other out and you'll end up with a unitless factor. This is why you cannot just apply the crop factor to the focal length. People fail to apply the crop factor to both the focal length AND the f-number/-stop.
That is correct! The sensor literally 'crops' which is why APS-C sensors were (and still are) often called 'crop' sensors. Because the smaller sensor crops the image circle that is produce by lens, more so than a 'full frame' (35mm) sensor.
Not always. The lens designed exclusively for apsc and mft cameras are smaller to begin with. The focal lengths could be converted to FF equivalent but there's no enlargement of the photos.
@@tamakakashii7865 focal length is just a lenght, a measure. It doesn't matter if it's for a phone camera or a hubble telescope sensor. A 50mm lens designed for MFT is still a 50mm lens and would be equivalent to a 100mm FF. I've seen some fixed lens cameras advertising their focal length range with the 35mm equivalent length with an *asterisk that points to a foot note saying so, but they still has to specify that, because the real measurement is not that. And I don't think I've ever seen a MFT or a APS-C interchangeable lens being advertised like that, but I'm not too big on lenses models so if you could give me an example
@@TwskiTV you are correct on the focal length being a standard measurement. I was replying to the person above saying that image will be enlarged when viewed on computer, which is not when the crop happens.
Thank you for posting the video. I am so sick of hearing people claiming that larger sensors having a shallower DoF. The thing, though, is when you try to explain it to them, as you do in the video, some people still would refuse to understand it. Note that I say "refuse to understand" but not "unable to understand".
But at the same field of view the FF is shallower that's what we are trying to tell you guys but all you here in that sentence is : FF Shallower 🤦🏻♂️ God we're saying if you get the same field of view with both a crop and a FF then the FF depth of field will be shallower. But if you take both 200mm at 2.8 at the same distance then you get the same depth of field BUT you get less field of view on the crop because the crop sensor is more zoomed in god damn is it that hard ?
Is it the lens (aka focal length) or is the aperture? Fullframe vs m43 conversion in the example at 7:18. where you keep the same aperture. With the 50mm 2.8 on GH5 you have actually a 100mm f5.6 compared to fullframe and that's why you have a greater depth of field. So it's not because you use a shorter lens but a smaller aperture (5.6 vs 2.8). I wonder if Tony Northrup agrees on that. I just saw UCreations pointed to the same:)
...all fine but why is there a difference in the background distance in @7:12 I thought from the previous video that lens compression does not exist. So Nikon FF with 100mm and Gh5 with a 50mm are at the same spot, same distance to the objects but yet it looks like the background is closer with nikons 100mm than 50mm with gh5. What am i missing?
Not sure if this will ever get answered, but does altering the distance between the last glass element and the sensor (Flange Focal Length) alter the DOF, or more specifically the bokeh? Another way of putting this if the sensor size is the same and lenses are of the same aperture and focal length, and you compare a mirrorless camera vs a traditional DSLR, are the blurry bits the same? Thanks, and excellent content.
I shoot w M43 so no bias here, but full frame cameras are better for shallowed depth of field because perspective matters a lot. Being able to get a less cropped image of the same depth of field is essential because we can’t all just shoot with super long focal lengths.
I find one problem with your demonstration: you are using a full frame lens for both cameras. What happens if you use the same focal length, same distance and same f-stop but a full frame lens for the full frame camera and a different lens for the cropped sensor? Will there be a difference in the depth of field? I would love to see a video about this comparison.
Yup, DoF is similar to compression/distortion, a larger sensor doesnt affect it, but adjusting to get the same field-of-view does... which people often do when you have a bigger sensor.
@@jeepjoseph9036 Unless you are explicitly looking to create with a cropped sensor the same kind of photo you would get with a Full, (and why would you?), there is zero reason to calculate for full-frame - just learn how to take the best photo's possible with the camera you have and run with it. I started with 35mm film, back when I was in High School in the 80's, and only recently finally made the jump full-time to a DSLR. Currently I'm using a Nikon D5600, and while I was running the math in my head when I first started out, I quickly learned I was wasting my time - as I naturally adapted to using a crop sensor body I stopped trying to capture a scene I was looking at the way I would have with my 35mm film body, and instead began framing the best shots I could with the Nikon, for the Nikon. I'm sure I will eventually upgrade to a full-frame, if only because my wife is always bugging me to give her the 5600, but I'm in no rush at all, I'm totally content with the results I've been getting with a crop sensor, and don't see any reason to rush out and spend more $$$.
@@TwskiTV yes they are worthless But when you are pushing yourself & your camera to the limit the advantage of having a full frame can really help you.
I do have a question on the statement at 3:35, where he backed up the camera and says "If we zoom in to create the same field of view, or crop it afterwards". So no doubts that backing up will create a less shallow DoF and cropping to get a similar field of view will not influence that. However if "zooming in" meant to increase the focal length while being further apart, wouldn't that change the situations? If I take a DOF calculator, and say the 1st shoot was at 70mm f/2.8, 150cm distance and full-frame, I get about 3cm DoF. If I increase the distance x2, so 300cm, and compensate for that with a 140 mm focal length, I also get 3cm DOF, and the effective blur is even larger to the magnification of the longer lens.
@@LunarLightLtd1 You believe FF produces shallower DoF? Just curious on your arguments as to why. I still suspect that at a practical level, FF does produce a bit more shallow DoF.
@@barrycohen311 There are only two things that affect depth of field directly: focus distance and aperture. A full frame camera will force you to either get closer (shortening focus distance) or zoom in (increasing aperture diameter at the same f-stop) to achieve a similar composition as a smaller sensor camera. So, yes, a full frame sensor INDIRECTLY affects depth of field. This isn't heresay or speculation, this is optics and math.
This concept is finally starting to take roots! But be prepared for hordes of folks who aren't prepared to take a comprehensive approach to understanding the subject!
Ya, the early guys like Tony Northrup and that Canadian guy who's still popular would go on and on about not being able to get as shallow of DOF with cropped sensors, and it not gathering as much light as a full frame, and using the crop factor to adjust for exposure even!! All sorts of nonsense. A sensor for the most part is just a piece of film... it doesn't know nor care what lens you have on it, or what settings you have set. It literally just captures the light that hits it, no matter what it's size. The size doesn't change anything, it's just cropped.
@@urwholefamilydied I would like to disagree. Larger sensors have larger individual pixel sizes, when having the same resolution as a smaller sensor. And the light gathering ability is proportional to the surface area of each pixel. A larger pixel will let more light in. So you would get much better low light performance. But yes, DOF is a property of the lens and not the sensor. But indirectly, composition starts from the FOV, for which you would need a longer focal length lens for a larger sensor to get the same FOV as a smaller sensor with a smaller lens. This will affect the DOF.
@@Wildridefilms yes... it's why larger format cameras are known for being able to achieve much shallower DOF. And why really small sensors or film are known for not getting shallow DOF.
I have a question: Would a 50mm 1.4 produce more or the same bokeh than a 100mm 2.8 with a similar framing on a person. I'm interested because I want to aquire a 50mm 1.4 for my MFT camera.
Or more accurately, you can either use a different lens (e.g. M43 requires 2-times wider in terms of the stated aperture and 1/2 the focal length lens to be analogous in terms of FOV, light level, etc.of a FF camera), or you may be able to change the distance and exposure settings to get comparable DOF. However, it is not really accurate to state that the sensor doesn't affect the DOF, since you must compensate for it via a different lens or a different shooting solution and approach. Using your logic, you could theoretically state that megapixels doesn't dictate output resolution, since you might be able to shoot at a different distance and/or combine multiple shots to get the same final output resolution. I know what you are trying to state, but I assume that you may be asserting this to either (1) make people who have a smaller sensor understand that they have a possible work-around in some instances, or (2) to generate a lot of views and comments by people like me who wish to correct/clarify your statements. However, related to your secondary point, if (1and the "if" is critical), you have (1) a similar or greater pixel count on your small sensor camera and (2) you are not filling your frame with the subject and (3) you can justify cropping your image, and (4) there is a comparable shooting solution possible given your hardware and shooting conditions, that you can obtain a higher effective resolution with the smaller sensor camera. Wildlife photogs leverage this fact, which allows them to use smaller lenses (e.g. a 100-400mm on a M43 as an "equivalent" to a 200-800mm compared to a full frame sensor camera. That his is obviously not counting for the light lost if you have the same aperture lenses, since an f/2.8 FF is analogous to a f/1.4 on a M43 camera.
Andrew -- you are 100% correct. I have shot since the 60's and worked with digital since the 90's as a professional who shoots for various magazines, and newspapers. In consulting for an independent photographic technology review company, I find a lot of similar confused explanations like this one from F-Stoppers. In contrast to FS, your explanation is accurate and complete, and I think you should be working for FS... or, at the very least, doing their accuracy checks for them!
Thank you Ms Munceli. I am glad to see that many on this site caught F-Stoppers poorly structured analysis and misleading conclusions. However, they did manage to confuse and derail many others. I see that they mislead some into thinking that they can put a 50mm f/2.8 on a M43 and get comparable results to a 100mm f/2.8 on a full frame, just by changing distances and cropping.
I'm sorry for being late and possibly wrong, but: I don't think there is any "lost light" due to crop factor, you can test this by using the same lens on full frame and apsc at the same aperture, or your phone and a equivalent focal length lens on whatever DSLR (or whatever) if you only have one big camera. Just apply the crop factor to each aperture (nearly always expressed as 35mm equivalent) and keep the ISO the same for each camera, you'll find that the actual shutter speed is the same instead of the expected due to crop factor. I think the actual reason for poorer image quality in APSC is pixel size and greater interference between those pixels.
You are not wrong, and that is the basis for applying crop factor and related adjustments to get parity and other compensated effects. However, as far lost light, it depends on, first, what you mean specifically by "lost light" and second, whether you employ an aperture related speed booster or not. If you mean lost photons when you say lost light, then, unless you employ a speed booster you will not capture the same total number of photons with the smaller sensor-- all other things being equal (and by equal, I do not mean compensated for by crop factor or related adjustments, which are not always possible without artificial light sources or other physical/external means). With a speed booster of the right factor, all of those photons you would have gotten with the larger sensor would be captured with the smaller sensor, resulting in a higher effective photon density via the same total number of photons-- hence the "speed booster" effect. Without the booster, they are effective not captured at those settings-- and thereby "lost". Often such considerations, such as the ones surrounding this video's assertions and viewers' pushback are due to the ambiguities of language. For example, FStoppers is not correct in asserting that sensor size does not have a relationship to depth of field. However, that is not what they meant or even explicitly said (it was strongly implied though, and supporting arguments bolstered that interpretation). If you can compensate for the size limitations of a given sensor by aperture, distance or other such compensation mechanism, then sensor size does not HAVE TO affect depth of field. However, since you cannot always compensate for that (otherwise nobody would ever need a DSLR, as you could shoot everything in all conditions with a cell phone); and as such, it is both commonly accepted and practically correct to assert that increased sensor size effects or allows for a shallower depth of field.
Great video. On 7:51 (what I'm interested in compare) the background seems to be closer to the object. You said to get the same dof on m43 mounts but you back up .... the camera or the subject? I think you back up the subject, cause if not, your experiment show me that finally there is a collateral effect on m43, and would be the apparent distance of background to subject. Could you clear my doubt? Thank you
I did notice that there was one aspect overlooked: actual aperture size. Changing the f/stop or changing the focal length while retaining the same f/stop both change the actual aperture size and hence DoF. Would have been interesting to have that also listed underneath the other data on the overlay.
@@stephaneoliveira2495 The f-stop is focal length by the aperture diameter. If we maintain the same distance to subject (in order not to change perspective) then the blur circle is proportional only to the aperture diameter. In other words, the amount of blur is proportional to the ratio of focal length and f-stop. That's this video in a nutshell. Moving the camera only introduced an extra factor and unnecessary confusion.
Hey there, do you guys educate/teach people how to use a micro 4/3 camera to take professional looking pictures (e.g., food, smooth waterfalls, night/star gazing pics, fast movement pics) and videos (mainly for near distance vlogging and increased bokeh + sharpness)?
You don't always need shallow depth of field to make the image look good. It's more of a matter of taste. On the other hand if you choose to use full frame, you ALWAYS have to carry the big camera body and the big lenses. If you choose to use "mirrorless" full frame, the lenses will be even bigger(and more expensive). Just look at those huge mounts.
Why do you feel the need to make your choice appear good? It has nothing to do with the Video. It's as if I was trying to make points on why you should shoot Nikon instead of Canon. FF Vs APS-C is a valid discussion with pros and cons on both sides, and everyone decides differently, but frankly this video is about a physical explanation of Dof. Besides, to answer your argument: On a FF body you don't need to take a FF lens. Modern bodies can shoot crop with an APS-C lens. You have the choice if you want to take light or heavy lenses with you. On APS-C bodies you can only shoot crop, so no choice there. People always make it out like you ALWAYS HAVE to put FF lenses on a FF body. There are still many valid points in favour of APS-C bodies, but the one you named is none of these.
@@jamskof "shallow wins every time,....." == it's a matter of taste. Anyway, neither of them are selling very well, It's losing the market to the cell phone cameras. Do you know any cell phone which has the shallow depth of field with big aperture?
Really good explanation of how aperture, focal length, and distance affect DoF, but the sensor stuff could really confuse people. The larger sensor gives you a larger field of view at a given focal length and distance from subject. So you are able to achieve shallower depth of field for a given composition with the larger sensor.
Larger sensors do not produce shallower depth of fields however all of these factors affect how the depth of fields are achieved in comparison to smaller sensor: 1) Larger sensor size lead to photographer moving closer to the subject to get the same field of view (yes more versatile but one important factor lager sensor you didn't mentioned gather more light and hence affect Exposure Triangle too) 2) Longer focal length at closer distance due to point 1) further generates shallower depth of field effect 3) Lower F stop lens further exaggerate the shallower depth of field effect In short, all of these 3 factors (distance/focal length/aperture) in place, you get the creamiest bokeh or shallowest depth of field. The opposite is correct too. In photography, all 3 work in complement of each other at all times so this video title should be the "Depth of Field Triangle" similar to that of the "Exposure Triangle" with a highlight that "large sensors do not directly produce shallower depth of field". I get where you are coming from but when viewers see the title, some will disagree.
Come on, facts are so last century. People need to FEEL like they’re being told the truth, or scared into listening to a veiled marketing pitch. The biggest shysters in media are the ones who rely on false provocation (aka “triggering” everyone) over reputation and skill. Looks like fStoppers are circling the drain. Shame.
to my understanding the shallower DOF from dslrs compared to m43 comes from different distance from lens to sensor. or is this oversimplified? when you use a metabones adaptor on a m43 you kinda simulate the distance that a dslr has, right?
I don't think so. Focal length is measured from the optical center of the lens to the sensor. If that changes then I'd have to say the focal length chances and then DOF would be effected.
No it's like this: when you scale up the sensor to make a comparison you also scale up the lens, for the same geometry. So a m43 with a 25mm/1.4 lens becomes full frame 50/1.4. It's the same 'geometry' it's the same type of lens design (1.4), it's just scaled up. They both produce the same FOV the same 'picture', the lens will be in the same class of difficulty the same design just a different size. But.. the opening, the max entrance pupil of the bigger lens will be 35mm whilst the smaller one is 17. That means the smaller lens will produce half the size of the bokeh of the larger one. It's all about the size of the entrance pupil that's the basic, primary source of bokeh. With a very small 'bug eye' camera all the world would be in focus. It would all be 'at infinity'. The bigger the camera the smaller the world gets and the more things are out of focus.
Higher pixel density doesn't haven't anything to do with depth of field. 100mm on FF is 100mm in FF. F2.8 on FF is f2.8 on FF. 100mm FF on m43 is 200mm. Here is actually what is happening. Excuse the T approximation for exemplification. F2.8 on FF and m43 isn't the same in terms of DOF. If you're using a FF lens in an m43 body You are using the center of the lens, so because you've changed you DOF you are now using a FF 100mm f2.8 as a 200mm f5.6 on m43, but only in terms of DOF. You assumed t stop value is still f2.8 in terms of the amount of light the lens is capable of allowing in. In other words, your t stop value stats static but you DOF aperture value changes accordingly. This has absolutely nothing whatsoever AT ALL with pixel density. That is absolute nonsense. Accurately put, if for example you were using an apsc sensor that had 24mp, you would have a 36mp equivalent sensor on FF based solely on resolution and ppi. But again this depends on the sensor. Sensors has different pixel pitch within their own sensor format size. They would behave accordingly. In other words you would have higher resolution cropping an image with higher pixel density than without, that is a no brainer. But that has nothing to do with DOF. This video has a lot of misinformation from a professional manufacturing POV. Quite simply put, you can't compare sensors this way. Shooting at f2 on a FF sensor will always have a shallower DOF on a targeted object than shooting at f2 on an apsc or m43, simply because of crop factor. Yoire using lens compression and none optimized lens examples to make a confusing case that is just not scientifically accurate. You're also excluding t stop valuation. A video like this, while enjoyable debate for professionals, is actually not a helpful video for amateurs or beginners to come across because it gives them misleading information that they can't disseminate, let alone grasp. Distortion from compression isn't bokeh. An accurate comparison would have been something like 50mm FF at f2 would have almost the same DOF as 35mm apsc at f1.4. Then too would have to address ISO for exposure, or shutter speed.
Final question: 50:2 FF vs 35:1.4 APSC, seems like they have same sizes of aperture(25mm in diameter), so does this mean same amount of photons will hit each pixel in theory(ingore lens quality) if these two bodies have same MP count?
They aren't wrong. The sensor size has no effect on your Depth of Field as it is a physical property of the lens. The idea that a 100mm f2.8 becomes a 200mm f5.6 on a MFT is factually wrong. It remains a 100mm f2.8 except you are closer. So you back it up to a 200mm focal range distance to get a similar image. However, the aperture is still 2.8 the only thing that has changed is the distance. This will give the look of the 5.6 except it isn't. It is just as bright, and unsharp as the f2.8. It's not overly confusing for beginners and is important information for them to have. The DOF is entirely a lens related effect. NOT a sensor based effect. Understanding this and understanding how one creates a shallow depth of field is important for all photographers to understand. Knowing this, you will be able to create shallow depth of field on a MFT camera, for some reason, people seem to believe that is impossible, and you will be able to create deep depth of field on a large sensor camera.
Great explanation and illustration. I'm subscribed. It's difficult to explain things to people 'cause they won't listen. The best thing about this video is it demonstrates the truth. Yeah, some won't believe it anyway like when Galileo said the Earth goes around the Sun despite solid reasoning. This video is both the solid reasoning and good illustration. I liked the drawing showing why it actually the way it is. *A superb job. Commends a lot of respect! You really nailed it! You are probably a superb pro photographer of the highest caliber* Most photographers can't even get things in focus, or avoid motion blur and shoot in manual while spreading their misconceptions, wrong ideas etc.
nice video. really well explained. I think that the thing is with cropped sensors there is a lower treshold of depth of field, so at some point you simply cant get more. full frame and maybe medium format allow you to go for those extremely shallow depth of field. your nikon with a f1.4 lens produces unmatched depth of field and I love it
@Fstoppers All you say is correct, but in the last part you missed to explain how you can get the exact same shot on a smaller sensor. If you keep the composition of your 100mm F2.8 shot, you can get the same DOF on M43 by matching the actual physical aperture diameter of both the 50mm and the 100mm lens. Your 100mm F2.8 lens has an aperture diameter of 35.7mm. To get the same aperture diameter on the GH5 with a 50mm lens, you would need an F1.4 aperture.
He complicates a bit by introducing physics... I'd explain like this - equal focal will give you equal DoF at the same aperture. Take a photo on a FF and crop it by 1.5 (or so) and you'll get the same photo as if you shoot with that same lens on an APS-C body. The smaller field of view on an APS-C is what makes you go further away from the subject or take lower focal length to get larger part of the scene - both of these will increase your DoF... but obviously, neither of the two is the sensor :-)
That answer was an emotional one. Unfortunatly most photographer, including lot calling themself "pro" just don't have any clue about optics. Optics is a science, it's a branch of physics and one that has been well studied since more several millenium. It define precies terms such as focal length and depth of field. And the way all this works is all backed up with centuries of scientifical experiments and studies. Those are now established fact ! Ho yes i know the way the real worl works doesn't exactly fit the way most of you photographers want to see it, but that doesn't change what it is. I know that hurt your feelings, but the real world doesn't care about your feelings, it's been there far befor you born and will still be there long after you're dead. So you think explaining how things realy work is "misleading" ? hahem , well yes you are right, this could lead people to question manufacture's marketing BS. That's probably no good. So her's the deal, that guy put up a video backed up by both centuries of physics science and show exemples that demonstrate that they actualy work that way. So you call it BS ? ok then go on, do the same work, give us some proof of what you pretend, back it up with citations of scientific work that goes on your way... Unles that, what you say is just trolling. Just my 2cts...
When you crop the image of the full-frame camera (6:31), you actually get the smaller sensor size, and that's why DOF is the same. If you made the same image, you could prove that larger sensors DO produce shallower DOF.
"To keep this simple we're going to ignore the variables that are the key to the actual subject of discussion when the subject comes up." (Slightly paraphrased.)
good point, but in not all scenarios you can be able to control the distance of the subject. i'm using both crop and fulframe but i want to do a shot with shallower depth of field, i always use my fullframe.
So you are saying for two cameras with differing sensor sizes and lenses of the same focal length the depth of field will be the same, but the field of view will differ. Whereas for the same two cameras with lenses of equivalent focal lengths, the camera with the larger sensor will have the shallower depth of field. In other words, yes the larger sensor does produce a shallower depth of field (when comparing lenses of equivalent focal lengths).
"lenses of equivalent focal lengths" is a really bad way of saying "Two different lenses" Two different lenses will create two different depths of field. 1 lens will have the same depth of field regardless of what is placed behind it.
1) The image that comes out of the lens is always the same (2.8 on FF is 2.8 on Crop as well) 2) Background blur has 3 factors: Aperture, Focus Distance, Focal Length 3) To get the same picture with a crop camera like FF you have to back up a little bit due to the 1.6 crop ----> The focus distance gets further away ----> Less Background Blur
isn't the aperture multiplied in crop sensor so in this case, 2.8 times 2x crop giving us f 5.6 in the m43, so to get the same aperture don't u have to use f1.4 for m43 and f2.8 for full-frame to get the same results??
Starting at 6:11, you bust the pervasive myths about applying crop factor to f-number and you demonstrate how you make changes in optical systems and framing on crop sensors that account for the changes in depth of field relative to full-frame. Thank you.
I genuinely found this very interesting! I’ve always believed that full frame or even medium format cameras produced shallower DOF, though as explained in the video, it’s actually the choice of lens producing the DOF, as the crop factor alters with the different sensor sizes. Great work!
When comparing photos taken by different camera sensors with different lenses, you also need to multiply the aperture by the crop factor, not only the Sensors and/or Focal lenght.
Exactly. This video is just a wrong point of view. Cause and causality is reversed and it just leads to more confusion. If somebody would like to understand this phenomenon I recommended this video: ruclips.net/video/hi_CkZ0sGAw/видео.html . Much better examples and clearer explanation.
what's the equivalent of 1.8 in full frame to crop? is there any crop lenses available ? none, so yes, full frame as for now, has shallower depth of field.
I think what he's saying is that you can get the same depth of field with an APS-C sensor compared to the full frame sensor. It's not the sensor but the distance to the subject and focall length. FOV changes based on sensor so you have to move with the FF to get the same FOV as the APS-C
eric00214 oh I got that, but what he eventually admitted is that due to practical and typical shooting distances, FF has the effect of giving a shallower depth of field, which is something he should have just stated in the beginning..
Thanks... this is a much better explanation than the one from Tony Northrup who keeps telling Camera manufacturers to change the specifications of the lens for APS-C cameras.
While scientifically speaking, the sensor does not impact depth of field since the lens spits out the same image behind it regardless of what is behind it. But practically speaking, the larger the sensor is, the easier it is to get bokeh. For example When you see a full body portrait with a completely blurred background (that is a natural lens blur. What would it take to get that same effect on a camera with a 1/2.3 inch sensor? Most of the common terms used, are not based on the pure physics of the light, lens and sensor, instead is it based on the end result and what is practical. What is sought after is getting a specific depth of field while getting a specific framing. It is all about being able to quickly communicate what equipment you want and what end result you want.
scientifically speaking, depth of field isn't just what the lens puts out. theoretically, there is precisely one distance exactly in focus, and it is infinitely thin. ignoring format, the depth of field for any lens is *zero*. you have to factor circle of confusion into to get a range of acceptable focus that appears sharp in your final print, because circle of confusion is the largest size an out of focus circle can be while still appearing as a point. circle of confusion relies on enlargement and viewing acuity/distance. for a given print size, the larger sensor will be enlarged less, so it has a larger circle of confusion. that means things can be further off that precise point of focus while still appearing acceptably sharp. that means that for a given lens, aperture, and distance, the larger format will have *more* depth of field. not less. it's just that to get the same angle of view, the photographer with a larger sensor will use a longer lens, which will have a larger physical aperture for a given f/stop...
I was stating it more from the perspective of if you were to take a lens and hold it in front of a piece of paper, you will see a circular image projected on the paper. Since a lens is passive, making that paper larger or smaller, will not change the image circle. When you use a smaller sensor, you are effectively physically cropping that image circle. Because of this, if you keep the lens at the same exact position and only swap the body, you will get a more smaller FOV, but the depth of field will be exactly the same. The change in DOF comes from when you adjust your framing (mainly be moving back a bit when you go to a crop sensor, or move to a wider lens.
Razor2048 on a crop sensor, you're also enlarging more for a given reproduction size. that happens to matter. trying to talk about DOF without COC is a bit odd, since COC is the thing that defines what's in focus. COC changes with format size.
The depth of field changes noticeably if you from 135 to medium format or even more, large format. But aps-c and full frame sensors don't have big enough of a difference.
Lee the iconoclast. Lens compression doesn’t exist. DOF isn’t related to sensor size. What’s next? Strobes don’t actually make a scene brighter. Sensors don’t actually record images. I can’t wait.
I have only recently begun paying attention to FStoppers, but it seems that Lee is becoming increasingly pedantic. These clickbait videos are more concerned with arguing that the terminology is not absolutely precise than the real world scenarios that photographers face that lead to these concerns.
Totally agree with you... If you look at a DOF chart; the larger sensor (needing less enlarging to attain a specific size product) has, if shot from the same distance and with the same focal length, a wider DOF than the smaller sensor. However, as you mentioned, photographers will usually back up or use a longer lens with a larger sensor which does produce de facto, a more shallow DOF.
Really great explanations for the most part! If you want practical knowledge as a photographer on how to reason about DOF on different cameras, there are better videos for that. But if you're interested in the actual science behind DOF, this was a really informative video. If you feel that this video made things even more confusing for photographers, just keep in mind the video doesn't preclude using rules like "multiply by crop factor". Those are useful rules. If in addition you want to know more of the science, then keep an open mind and learn from the video.
Summary on how to get more bokeh:
1) Have long focal length (80mm>50mm>35mm), step closer to the subject (closer>further), and have a wider aperture (f1.4 > f1.8 > f2 etc)
2) Keeping everything the same, the full-frame camera and crop sensor will have the same level of bokeh. The only difference is that the crop sensor's image will be more "zoomed in". AKA it crops in.
3) If you want to have the same level of "zoomness", you'll have to a) step back a bit with the crop sensor b) use a shorter focal length (wider) lens to get the same field of view. However, stepping back or using a wider lens will decrease bokeh as said in #1.
4) To solve this issue, you'll need to use a wider aperture lens.
Conclusion: sensor size doesn't create more bokeh. The focal length, distance from the subject, and aperture size determine the depth of field. Crop sensors only zoom in (or crop) photos.
Very informative video and comment, thank you for that
your a g
Or uhhh, just use a lens meant for the sensor size of the camera you have. Micro 4/3 sensor? Buy a Micro 4/3 lense and you will have no crop or "zoomness".
Amazing how many ppl disliked the video. Some people are just resistant to knowledge. Your explanation was really good.
So no, but yes
So so, but actually its confusing
Exactly
No, but no. Moving closer or using a longer lens does.
so no, but no, but still, you will get a shallower depth of field with a full frame camera ;)
yeahhh a metal head
The correct statement is " Larger Sensors produce a shallower DOF at the same field of view"
and that would be add more focal length for full frame by 1.6x, so yeah the key is focal length
Best summary ever.
Pretty stupid statement. Sensor size has nothing to do with DOF. Larger sensor just need larger aperture or longer focal length. Sensor size itself doesn't affect DOF in any way.
@@Alex-fk3ni He said "at the same field of view", which is correct, and a good summary of how this practically plays out in the field
@@Alex-fk3ni you cant use a camera without a lens bigbrain
This video is not wrong but the way information is presented may cause miss information. The sensor don't produce shallow depth of field, the distance to the subject and focal length does. The thing is that the crop and aspect of the sensor dictate the chose of both (distance and Lens selection).
This is exactly the way I would put it.
Mi phone camera's tiny sensor is not at fault of the lack of bokeh; we have to blame the 4 mm lens that the sensor needs in order to have a decent field of view.
well said
I could not have said it shorter or better.
Brainstorming Films amen
You lost me at "miss information." So, if information gets married, will she be "Mrs. Information?"
0:03 as you have said, "It does. But probably not in the way that you think." So it does.
Of course its not the sensor size itself that affects the DOF, but the focal length adjustment that we need to do to achieve the same image coverage. And when we make the composition the same, the one with the bigger sensor would produce the shallower DOF.
Truee
That’s because not only you need to convert the focal length to match but also the F stop which I feel he missed in this video, if you do both you will get exactly the same image. A FF camera with a 100mm lens at F 2.8 to match that you would need a 50mm at F 1.4 and to make the matter worst to match the exposure you also have to convert the iso to but now things get more confusing so I will leave it there 🤣
What MrKockabilly said.
@@CabralCreates Do you know how to match the iso? is there some formula for it?
@@phqnomenon yes there’s! You’ve to square the crop factor. If you’ve a GH5 the crop factor is 2x crop so it’s easy to figure out, let’s say you set your ISO to 400 on the GH5 to match that to a full frame you’ve to multiply 400 x 4 (1,600) if you match the 1-Focal length 2- Aperture 3-Iso you will get the same exact result the problem is camera companies are not making enough fast lenses for micro 4/3 cameras to match FF cameras
Ignore everything he says skip to 6:00
Thats all you need to know.
Thanks lol
But it's not just about Knowing the answers is it? It's about understanding? When you understand, you create your own answers? Memory can rob us of Facts we have swallowed, but knowledge and understanding stays with us!
i AGREE, THE FIRST 6 MIN WERE TOO ALL OVER THE PLACE. (stupid caps lock lol)
Thanks mate
lol thanks
The best explanation on 'depth of field' I've ever come across. Good scientific approach; you identified all relevant variables (aperture, focal length, length to subject and sensor size) and showed what happens if you change one of them.
"Smaller sensors produce a shallower depth of field, because they have a higher pixel density."
?
??
???
?????
???????
Why would you want to further confuse people on this, WHY ?????????
Yes, hes wrong when he said that. :)
Yeah that sentence itself is incorrect, which is ironic since the title of the video says that sensor size doesn't affect depth of field.
As he explained though, higher pixel density does make depth of field shallower. I think he was trying to say that "smaller sensors often have higher pixel density, which produces a shallower depth of field" and just got the ordering mixed a bit.
Do you really believe that if you shoot a D3 and a D850 the D850 is gonna have a bigger bokeh???
Yes, the D850 will have shallower depth of field (not by a whole lot though). More pixels let's you distinguish more colours in a photo, and more pixels also let's you better distinguish how in focus different parts of the photo are.
The same lens/aperture on a D5 as a D850 has the same interaction with light, but digital cameras have to record that infinite amount of light information onto a fixed resolution sensor. So on a D5 you will lose more information than on a D850, which makes the area of "acceptable focus" larger. This effectively makes the depth of field deeper.
Ray J Wow!
How do you know these things... I am impressed!
If you keep the lens in the same place, use the lumix with a speed booster. (Say a .71) that cuts out 29% of the field of view, right? So that would mean you’d need to move the camera back less. Correct? Also the light could be converging faster with a speed booster shallowing the depth of field additionally? (Gh5 shooter over here)
This is helpful, and I can see why people get confused. So if I got this right.
If you have FF and MFT, with same camera to subject distance, same lens, and same aperture, both will produce the same level of DOF or blurriness/bokeh in background. But the field of view will be different with the MFT being a much tighter shot.
To match the cameras for FOV, where the composition matches, you would have to change either the camera to subject distance or the lens.
If you change the lens, it will match the FOV, but because it is a different lens, it will have a different DOF. But that has nothing to do with the sensor, but the lens.
And if you use the same lens, but change the camera to subject distance, you get the same FOV, but there is some difference in DOF because you changed the distance.
Thanks for doing this. It is easier to remember when seeing a real life example.
agreed
The sensor itself, or in the old days, the film size, does not change the Depth of Field, 3 things do; distance to subject, ƒ-stop and focal length. The "normal" angle of view of a 35mm camera is 50mm, on medium format 6x6cm it is 80 mm and on a 4" x5" sheet film view camera it is 150mm. This means that the fact of the matter is for the same distance and ƒ stop a larger film/sensor size will provide shallower depth of field for the same angle of view. This is an optical fact that cannot be changed.
It’s great to see an explanation of DoF that’s based on optics science (physics) rather than uninformed opinion and passion for a particular sensor format. Bravo!
This is great information for people to have and helps dispel notions of things like "equivalent aperture" which only serves to confuse people.
There is one point that was left out, though. You showed how the same lens at the same distance on FF can be cropped for the same result as APS/MFT. However, what is not shown is that because FF has a larger field of view, you can get physically closer to a subject to take a similarly framed shot without cropping. Take a portrait head shot for instance. 85mm FF, you frame in around the head and shoulders, and get a shallow depth of field. To get this same framing in APS or MFT, you only have two options: back up, or use a wider lens. Both of these options, as you demonstrated, will increase the depth of field, and therefore decrease "bokeh".
You can always crop FF down to equivalent APS/MFT view, but there's no way to magically uncrop APS/MFT.
Slyest Fox That’s exactly what’s wrong with this video. How photographers actually do their job vs. some theoretical reasoning that tries to “disprove” the actual nature of things. Who would ever want to crop in on a 16mp MFT sensor anyway?
Yes but by cropping down to the APS-C size you will lose a little less then half the megapixels. A D850 at 45mp will be a 19mp image cropped to a D500 image uncropped is 20.9.
From what I've researched (and Tony Northrup also confirmed this) a MFT 50mm f/1.4 would be equivalent to a FF 100mm f/2.8. If you test this scenario, you should get nearly identical DOF.
This has been one of the most important lessons I've learned about photography. Thanks for explaining this in such a straightforward way.
Good and straightforward explanation! But in the process you have shown that practically speaking...sensor size matters. Because it is the relationship between the sensor size and the desired content which results in the image. So even though sensor size does not at all affect depth of field, the sensor size will dictate lens focal length-which will immensely affect depth of field! Same reason that medium and large format cameras, with relatively slow, but very long lenses, still can get great subject isolation and good bokeh. By debunking the myth, you actually have proved the point, that sensor size matters!
You saw there's no response to the Facts..it's all some useless, fake debate, trying to keep the ship from sinking...
I'm convinced the guy makes inflammatory controversial videos to get a high amount of views. It's pretty cringey.
@@zr_1234 i call it stupidity...
!!!!!!! WRONG INFORMATION !!!!!!!!
Mate! You're wrong about Sensor Sizes! and YES, Larger Sensors Produce Shallower Depth of Field!
For this experiment, They need equal situations without post-production things like CROPPING! to achieve the SAME RESOLUTIONS!
So you should apply the "SAME FIELD OF VIEW" for both cameras! So, we have 2 options:
1- 100mm on full-frame and 50mm on 3/4 micro four thirds (same distance)
2- 100mm on both camera (different distance)
now you will see, Larger Sensors Produce Shallower Depth of Field!
In this video, You cropped the photo of the full-frame, and this is wrong! because you reduce the resolution of the full-frame photo!
I already knew all of this but Lee elaborated it perfectly and easy to understand. All you need from this video is the summary at the end. FF sensor doesn't create shallower DoF but the fact that you need a longer lens or moving closer to the subject to get the same point of view does.
Nate Kong This video substitutes theory for the reality of the real world. *Can* you get equally shallow depth of field with a small sensor? Sure. But given the gear that’s available, their prices, etc., good luck matching the DoF from a full frame 50mm f/1.2 on a MFT system. “Can it be done?!” Probably. But how about we talk about the world we actually live in? 🙈
Then can you please explain how if I take a photo with my M4/3 Panasonic with a 25mm f0.95 vs my Fuji GFX with the 63mm f2.8 , the GFX has SUBSTANTIALLY more shallow DOF and focus falloff. I have tried everything under the sun to do this comparison and it is simply impossible to create that medium format look with a smaller sensor.
Because you are using two different focal lengths. His point is that the same focal length from the same distance using the same aperture will produce the same DOF regardless of the size of the sensor. If you vary any of the three factors to adjust the field of view because of the sensor size you will effect the DOF.
This quote from dpreview sums up your videos thus far:
Peter Cockerell
"The compression you get using a long lens isn't a result of the lens, so much as the distance between your subject, your background, and the camera."
God but these pedantic distinctions are tiresome. It may not be a result of the lens, but its observability is. If I take a shot of layered, misty mountain peaks in the far distance with a 400mm + 1.4x teleconverter, the "compression" (for want of a better term) is very obvious. If I take the same shot with a 16mm wide angle, the area covered by the original photo will occupy about 0.1% of the wide-angle shot, and will be totally lost in the rest of the image.
No, the 640mm didn't create the compression, it just enables you to actually see it. I think most people understand this, implicitly, even if they don't necessarily articulate it. People who make videos like this are much more annoying in my opinion than people who say "lens compression". It's not even new or original. The same tired "explanations" come round time and time again.
PS, thanks Peter.
It is the same as folks that argue that sensor size doesn't affect DOF, yes yes, we all know the tech behind this but the point is there is a reason folks go full frame and medium, large format , and aren't shooting some cheap nikon coolpix camera, one of them is the sensor is bigger, and one of those reasons is it affords shallower dof in that particular package.
PS, yes this is a total triggering video ;-)
Yup. Nice video. A 16mm f/1.4 on a crop sensor is a 16mm lens, it's just cropped to a 24mm FOV. 16mm won't give you the same DOF as 24mm. I don't know why this is so hard for people to understand.
question: at 4:41 the guy have printed shirt that say "nikon d850 tamron 70-200 f22" at 4:42 says "nikon d850 tamron 70-200 f2.8" how did you change the shirt without move the guy and where do i buy that shirts
saul gonzalez Should this be funny?
no
you must be fun at parties
Is this is a serious question? 🤔😂
(If so, it's just text overlay)
Basically, having a larger sensor in and of itself does not mean a photo will have greater bokeh. However, due to having a larger sensor, the photographer must either use a longer focal length or physically step closer to the subject, which creates the shallower depth of field. I'ts not the sensor itself, it is the action required to match the same frame as APS-C or MFT sensors. This video wonderfully explains the issue, in my opinion. Great video!
Well explained... One of the most clearly explained... Northrups video seems more complicated
That's because he explains it correctly. Northrups videos are specific enough not to be misunderstood.
Kudos Lee! You’re the first person I’ve seen on RUclips that has nailed this! Seems like everyone else has only skimmed the surface, probably because they themselves don’t have a solid handle on it. Tony Northrop came close, but he kinda glossed over the WHY. You’ve illustrated what Ken Wheeler drones on about repeatedly, yet never offers any comprehensive illustration. On another topic; Fstoppers has some of the only instructional content that I have gladly forked over money for. I found Monte Isom’s - Making Real Money course invaluable!
Thank you for not clowning yourself into long useless goofy intros, and actually going direct to the point. Subscribed!
I've watched several explanations of depth of field on YT and yours has been one of the best, especially on Circle of Confusion,, thanks for your post I look forward to seeing more of your videos..........
The real inforamtion is that at same f-stop and same FOV a bigger sensor produces a shallower depth of field. This is missleading because you don´t use the equalient focal-length, you use the same...
just on the spot! I am sorry to see that misinformation is presented in the video. Also apart from the sensor size the FOV or Focal Length has nothing to with the depth of field as long as the reproduction size is the same. What he calls as depth of field with a longer focal lens is just the compression effect, the reason why many portrait photographers prefer to work with lenses as long or longer than 85 mm.
If you doing something that will make FOV same, it is practical not technical (physic).
This is example to make this issue more clear: I am standing in front of model with my Linhof 4x5 camera and 150mm lens, aperture wide open lets say at f5.6.
I cut the 4x5inch negative to make 36x24mm negative. From a full body shot now I got a head shot.
The depth of field between two formats is same, because it is from the same negative. That's what we call "cropped sensor".
Sorry for my English
@@maspriyayi No it is not, the negative you cut has less depth of field as when you blow it to same size errors show more.
The formula for depth of field is: D = s / (1 ± N × c × (1000 × s - f) / f²) where plus gives the near limit and minus the far limit. s is the distance, N aperture value, f focal length and c circle of confusion which depends on the negative/sensor size. For full frame it is 0.03 mm. Replace 1000 with 304.8 if you use feet instead of meters for s.
At 7:27 minute, what if you use a 50mm f/1.4?
Come on, now you are just triggering people...
media.giphy.com/media/ui1hpJSyBDWlG/giphy.gif
Called click bait... RUclips it
No sir, what he is saying in right... there is also a lot of confusion among some people that a, say 50mm f1.8 for full frame camera has a different f stop on an APSC camera, which is also wrong... using additional lenses, such as speedboosters, is different... to sum up a a given lens of a given aperture is same on any format...say again a 135mm f 2.8 is always that, a 135mm f 2.8...
sundar iyengar except for the fact that is provides a different angle of view depending on the sensor size.
A 135 f2.8 will have different dof depending on its distance from the subject. To get the same shot as ff on apsc, you'd need to back up. The light intensity remains the same, since f stop is calculated by focal length and aperture size.
Apsc back up so much when compared to ff, that a ff at 135 f2.8 will have the same dof as 135 f4 on apsc given the same angle of view (same composition).
Firestarter Agreed... That's why I mentioned about image magnification... To illustrate, if a photograph is taken on a 50mm lens at say, f2.8 and the same photograph taken with a 100mm lens at f2.8, if the mage of 50mm lens is cropped to that of a photograph taken with a 100mm lens at f2.8, both the photograph will have the same depth of field... May be, the out of focus area (bokhe) May be different, depending on the lens construction and other aspects and considerations... The RUclips video is nice and good.
Focal length doesn't matter in the way you think it does, the aperture is more important. 200mm f/2.8 means an aperture f/2.8 of ~71.43mm while 20mm f/2.8 equates to an aperture diametre of ~7.14mm. This is EXACTLY why the crop factor applies to both the focal length AND the aperture. 50mm f/2 with a crop of 2x has the same field of view AND DoF as a 100mm f/4 FF lens.
The distance to your subject matters a lot since it goes into the equation squared.
DoF ≈ 2u²Nc / f² = 2u²c / fD
N = f / D
u: distance to subject in mm
N: f-stop/-number
f: focal length in mm
D: diameter of the aperture/entry pupil in mm
c: circle of confusion (acceptable circle of confusion, that is perceived as a single point)
c is actually often very arbitrarily picked but it should probably be around 1.5x to 2x the size of a photo site.
The f-stop is not absolute, it is the ratio between focal length and the diametre of the so called entry pupil (or easier: aperture). N = f / D with N being the f-stop, f the focal length and D the diametre of the entry pupil. Of course, f and D are in mm. This also shows that the f-stop is unitless because mm / mm cancels each other out and you'll end up with a unitless factor. This is why you cannot just apply the crop factor to the focal length.
People fail to apply the crop factor to both the focal length AND the f-number/-stop.
So the 2x crop of a MFT camera doesn’t actually convert the 50mm to a 100mm, it crops to only show a 100mm FOV.
That is correct! The sensor literally 'crops' which is why APS-C sensors were (and still are) often called 'crop' sensors. Because the smaller sensor crops the image circle that is produce by lens, more so than a 'full frame' (35mm) sensor.
Otherwise cellphones would make multiple thousand dollars telephoto lenses useless
Not always. The lens designed exclusively for apsc and mft cameras are smaller to begin with. The focal lengths could be converted to FF equivalent but there's no enlargement of the photos.
@@tamakakashii7865 focal length is just a lenght, a measure. It doesn't matter if it's for a phone camera or a hubble telescope sensor. A 50mm lens designed for MFT is still a 50mm lens and would be equivalent to a 100mm FF. I've seen some fixed lens cameras advertising their focal length range with the 35mm equivalent length with an *asterisk that points to a foot note saying so, but they still has to specify that, because the real measurement is not that. And I don't think I've ever seen a MFT or a APS-C interchangeable lens being advertised like that, but I'm not too big on lenses models so if you could give me an example
@@TwskiTV you are correct on the focal length being a standard measurement. I was replying to the person above saying that image will be enlarged when viewed on computer, which is not when the crop happens.
Thank you for posting the video. I am so sick of hearing people claiming that larger sensors having a shallower DoF. The thing, though, is when you try to explain it to them, as you do in the video, some people still would refuse to understand it. Note that I say "refuse to understand" but not "unable to understand".
But at the same field of view the FF is shallower that's what we are trying to tell you guys but all you here in that sentence is : FF Shallower 🤦🏻♂️ God we're saying if you get the same field of view with both a crop and a FF then the FF depth of field will be shallower. But if you take both 200mm at 2.8 at the same distance then you get the same depth of field BUT you get less field of view on the crop because the crop sensor is more zoomed in god damn is it that hard ?
7:55
Is it the lens (aka focal length) or is the aperture? Fullframe vs m43 conversion in the example at 7:18. where you keep the same aperture. With the 50mm 2.8 on GH5 you have actually a 100mm f5.6 compared to fullframe and that's why you have a greater depth of field. So it's not because you use a shorter lens but a smaller aperture (5.6 vs 2.8). I wonder if Tony Northrup agrees on that. I just saw UCreations pointed to the same:)
...all fine but why is there a difference in the background distance in @7:12 I thought from the previous video that lens compression does not exist. So Nikon FF with 100mm and Gh5 with a 50mm are at the same spot, same distance to the objects but yet it looks like the background is closer with nikons 100mm than 50mm with gh5. What am i missing?
Distance from the subject is the key.
Not sure if this will ever get answered, but does altering the distance between the last glass element and the sensor (Flange Focal Length) alter the DOF, or more specifically the bokeh? Another way of putting this if the sensor size is the same and lenses are of the same aperture and focal length, and you compare a mirrorless camera vs a traditional DSLR, are the blurry bits the same?
Thanks, and excellent content.
I shoot w M43 so no bias here, but full frame cameras are better for shallowed depth of field because perspective matters a lot. Being able to get a less cropped image of the same depth of field is essential because we can’t all just shoot with super long focal lengths.
Thanks for not only telling, but also showing the mechanics.
Great VIDEO!
Thank you for helping me understand this. I never got an answer that really made sense to me.
I find one problem with your demonstration: you are using a full frame lens for both cameras. What happens if you use the same focal length, same distance and same f-stop but a full frame lens for the full frame camera and a different lens for the cropped sensor? Will there be a difference in the depth of field?
I would love to see a video about this comparison.
Yup, DoF is similar to compression/distortion, a larger sensor doesnt affect it, but adjusting to get the same field-of-view does... which people often do when you have a bigger sensor.
Which is the part people seem to conveniently forget when they try to defend their "NUH UH, IT DOES CHANGE IT." position.
then how can you get the FF 85mm f1.4 dof and perspective on crop ? 55mm f0.9 ? is there any lense out there ? fullframe is still lid
I just bought a full frame to get rid of those calculations with aparture, focul length, iso & depth of field
Little hint...
Those calculations mean nothing.
@@looneyburgmusic what do you mean?
@@jeepjoseph9036 Unless you are explicitly looking to create with a cropped sensor the same kind of photo you would get with a Full, (and why would you?), there is zero reason to calculate for full-frame - just learn how to take the best photo's possible with the camera you have and run with it.
I started with 35mm film, back when I was in High School in the 80's, and only recently finally made the jump full-time to a DSLR. Currently I'm using a Nikon D5600, and while I was running the math in my head when I first started out, I quickly learned I was wasting my time - as I naturally adapted to using a crop sensor body I stopped trying to capture a scene I was looking at the way I would have with my 35mm film body, and instead began framing the best shots I could with the Nikon, for the Nikon. I'm sure I will eventually upgrade to a full-frame, if only because my wife is always bugging me to give her the 5600, but I'm in no rush at all, I'm totally content with the results I've been getting with a crop sensor, and don't see any reason to rush out and spend more $$$.
Those calculations are the stupidest thing ever
@@TwskiTV yes they are worthless
But when you are pushing yourself & your camera to the limit the advantage of having a full frame can really help you.
I do have a question on the statement at 3:35, where he backed up the camera and says "If we zoom in to create the same field of view, or crop it afterwards". So no doubts that backing up will create a less shallow DoF and cropping to get a similar field of view will not influence that.
However if "zooming in" meant to increase the focal length while being further apart, wouldn't that change the situations?
If I take a DOF calculator, and say the 1st shoot was at 70mm f/2.8, 150cm distance and full-frame, I get about 3cm DoF.
If I increase the distance x2, so 300cm, and compensate for that with a 140 mm focal length, I also get 3cm DOF, and the effective blur is even larger to the magnification of the longer lens.
Best explanation I have heard on this controversial and often misunderstood subject.
Except it's wrong.
@@LunarLightLtd1 You believe FF produces shallower DoF? Just curious on your arguments as to why. I still suspect that at a practical level, FF does produce a bit more shallow DoF.
@@barrycohen311 There are only two things that affect depth of field directly: focus distance and aperture. A full frame camera will force you to either get closer (shortening focus distance) or zoom in (increasing aperture diameter at the same f-stop) to achieve a similar composition as a smaller sensor camera. So, yes, a full frame sensor INDIRECTLY affects depth of field. This isn't heresay or speculation, this is optics and math.
@@LunarLightLtd1 Fair enough. There is a lot to consider with this subject.
@@barrycohen311 check out this video, it'll change your mind. If not, there's no hope. ruclips.net/video/1bzHn2cKwLI/видео.html
I don't want to zoom in 200mm on someone's face, so do I back off a little or do I decrease the focal length to have the dof as shallow as possible?
This concept is finally starting to take roots! But be prepared for hordes of folks who aren't prepared to take a comprehensive approach to understanding the subject!
Ya, the early guys like Tony Northrup and that Canadian guy who's still popular would go on and on about not being able to get as shallow of DOF with cropped sensors, and it not gathering as much light as a full frame, and using the crop factor to adjust for exposure even!! All sorts of nonsense. A sensor for the most part is just a piece of film... it doesn't know nor care what lens you have on it, or what settings you have set. It literally just captures the light that hits it, no matter what it's size. The size doesn't change anything, it's just cropped.
@@urwholefamilydied I would like to disagree. Larger sensors have larger individual pixel sizes, when having the same resolution as a smaller sensor. And the light gathering ability is proportional to the surface area of each pixel. A larger pixel will let more light in. So you would get much better low light performance.
But yes, DOF is a property of the lens and not the sensor. But indirectly, composition starts from the FOV, for which you would need a longer focal length lens for a larger sensor to get the same FOV as a smaller sensor with a smaller lens. This will affect the DOF.
@@Wildridefilms yes... it's why larger format cameras are known for being able to achieve much shallower DOF. And why really small sensors or film are known for not getting shallow DOF.
I have a question:
Would a 50mm 1.4 produce more or the same bokeh than a 100mm 2.8 with a similar framing on a person.
I'm interested because I want to aquire a 50mm 1.4 for my MFT camera.
Or more accurately, you can either use a different lens (e.g. M43 requires 2-times wider in terms of the stated aperture and 1/2 the focal length lens to be analogous in terms of FOV, light level, etc.of a FF camera), or you may be able to change the distance and exposure settings to get comparable DOF. However, it is not really accurate to state that the sensor doesn't affect the DOF, since you must compensate for it via a different lens or a different shooting solution and approach. Using your logic, you could theoretically state that megapixels doesn't dictate output resolution, since you might be able to shoot at a different distance and/or combine multiple shots to get the same final output resolution.
I know what you are trying to state, but I assume that you may be asserting this to either (1) make people who have a smaller sensor understand that they have a possible work-around in some instances, or (2) to generate a lot of views and comments by people like me who wish to correct/clarify your statements.
However, related to your secondary point, if (1and the "if" is critical), you have (1) a similar or greater pixel count on your small sensor camera and (2) you are not filling your frame with the subject and (3) you can justify cropping your image, and (4) there is a comparable shooting solution possible given your hardware and shooting conditions, that you can obtain a higher effective resolution with the smaller sensor camera. Wildlife photogs leverage this fact, which allows them to use smaller lenses (e.g. a 100-400mm on a M43 as an "equivalent" to a 200-800mm compared to a full frame sensor camera. That his is obviously not counting for the light lost if you have the same aperture lenses, since an f/2.8 FF is analogous to a f/1.4 on a M43 camera.
Andrew -- you are 100% correct. I have shot since the 60's and worked with digital since the 90's as a professional who shoots for various magazines, and newspapers. In consulting for an independent photographic technology review company, I find a lot of similar confused explanations like this one from F-Stoppers. In contrast to FS, your explanation is accurate and complete, and I think you should be working for FS... or, at the very least, doing their accuracy checks for them!
Thank you Ms Munceli. I am glad to see that many on this site caught F-Stoppers poorly structured analysis and misleading conclusions. However, they did manage to confuse and derail many others. I see that they mislead some into thinking that they can put a 50mm f/2.8 on a M43 and get comparable results to a 100mm f/2.8 on a full frame, just by changing distances and cropping.
Andrew Denis thanks!
I'm sorry for being late and possibly wrong, but: I don't think there is any "lost light" due to crop factor, you can test this by using the same lens on full frame and apsc at the same aperture, or your phone and a equivalent focal length lens on whatever DSLR (or whatever) if you only have one big camera. Just apply the crop factor to each aperture (nearly always expressed as 35mm equivalent) and keep the ISO the same for each camera, you'll find that the actual shutter speed is the same instead of the expected due to crop factor. I think the actual reason for poorer image quality in APSC is pixel size and greater interference between those pixels.
You are not wrong, and that is the basis for applying crop factor and related adjustments to get parity and other compensated effects. However, as far lost light, it depends on, first, what you mean specifically by "lost light" and second, whether you employ an aperture related speed booster or not. If you mean lost photons when you say lost light, then, unless you employ a speed booster you will not capture the same total number of photons with the smaller sensor-- all other things being equal (and by equal, I do not mean compensated for by crop factor or related adjustments, which are not always possible without artificial light sources or other physical/external means). With a speed booster of the right factor, all of those photons you would have gotten with the larger sensor would be captured with the smaller sensor, resulting in a higher effective photon density via the same total number of photons-- hence the "speed booster" effect. Without the booster, they are effective not captured at those settings-- and thereby "lost".
Often such considerations, such as the ones surrounding this video's assertions and viewers' pushback are due to the ambiguities of language. For example, FStoppers is not correct in asserting that sensor size does not have a relationship to depth of field. However, that is not what they meant or even explicitly said (it was strongly implied though, and supporting arguments bolstered that interpretation). If you can compensate for the size limitations of a given sensor by aperture, distance or other such compensation mechanism, then sensor size does not HAVE TO affect depth of field. However, since you cannot always compensate for that (otherwise nobody would ever need a DSLR, as you could shoot everything in all conditions with a cell phone); and as such, it is both commonly accepted and practically correct to assert that increased sensor size effects or allows for a shallower depth of field.
Great video. On 7:51 (what I'm interested in compare) the background seems to be closer to the object. You said to get the same dof on m43 mounts but you back up .... the camera or the subject? I think you back up the subject, cause if not, your experiment show me that finally there is a collateral effect on m43, and would be the apparent distance of background to subject. Could you clear my doubt? Thank you
Finally!!! Thank you very much, guys, this makes so much sense now!
Got stuck in arguments trying to explain this to people and they dont get it. glad you made this video.
I did notice that there was one aspect overlooked: actual aperture size. Changing the f/stop or changing the focal length while retaining the same f/stop both change the actual aperture size and hence DoF. Would have been interesting to have that also listed underneath the other data on the overlay.
Interested in this too
@@stephaneoliveira2495 The f-stop is focal length by the aperture diameter. If we maintain the same distance to subject (in order not to change perspective) then the blur circle is proportional only to the aperture diameter. In other words, the amount of blur is proportional to the ratio of focal length and f-stop. That's this video in a nutshell. Moving the camera only introduced an extra factor and unnecessary confusion.
Hey there, do you guys educate/teach people how to use a micro 4/3 camera to take professional looking pictures (e.g., food, smooth waterfalls, night/star gazing pics, fast movement pics) and videos (mainly for near distance vlogging and increased bokeh + sharpness)?
You don't always need shallow depth of field to make the image look good.
It's more of a matter of taste.
On the other hand if you choose to use full frame, you ALWAYS have to carry the big camera body and the big lenses.
If you choose to use "mirrorless" full frame, the lenses will be even bigger(and more expensive).
Just look at those huge mounts.
Why do you feel the need to make your choice appear good? It has nothing to do with the Video. It's as if I was trying to make points on why you should shoot Nikon instead of Canon.
FF Vs APS-C is a valid discussion with pros and cons on both sides, and everyone decides differently, but frankly this video is about a physical explanation of Dof.
Besides, to answer your argument:
On a FF body you don't need to take a FF lens. Modern bodies can shoot crop with an APS-C lens. You have the choice if you want to take light or heavy lenses with you.
On APS-C bodies you can only shoot crop, so no choice there.
People always make it out like you ALWAYS HAVE to put FF lenses on a FF body.
There are still many valid points in favour of APS-C bodies, but the one you named is none of these.
yes you do. shallow wins every time, come on up to the big leagues croppy man.
@@VisaxLP
"FF Vs APS-C is a valid discussion with pros and cons on both sides...."
==
Depth of field isn't one of the them.
@@jamskof
"shallow wins every time,....."
==
it's a matter of taste.
Anyway, neither of them are selling very well,
It's losing the market to the cell phone cameras.
Do you know any cell phone which has the shallow depth of field with big aperture?
Really good explanation of how aperture, focal length, and distance affect DoF, but the sensor stuff could really confuse people. The larger sensor gives you a larger field of view at a given focal length and distance from subject. So you are able to achieve shallower depth of field for a given composition with the larger sensor.
Larger sensors do not produce shallower depth of fields however all of these factors affect how the depth of fields are achieved in comparison to smaller sensor:
1) Larger sensor size lead to photographer moving closer to the subject to get the same field of view (yes more versatile but one important factor lager sensor you didn't mentioned gather more light and hence affect Exposure Triangle too)
2) Longer focal length at closer distance due to point 1) further generates shallower depth of field effect
3) Lower F stop lens further exaggerate the shallower depth of field effect
In short, all of these 3 factors (distance/focal length/aperture) in place, you get the creamiest bokeh or shallowest depth of field. The opposite is correct too. In photography, all 3 work in complement of each other at all times so this video title should be the "Depth of Field Triangle" similar to that of the "Exposure Triangle" with a highlight that "large sensors do not directly produce shallower depth of field".
I get where you are coming from but when viewers see the title, some will disagree.
Come on, facts are so last century. People need to FEEL like they’re being told the truth, or scared into listening to a veiled marketing pitch. The biggest shysters in media are the ones who rely on false provocation (aka “triggering” everyone) over reputation and skill. Looks like fStoppers are circling the drain. Shame.
+++ best explanation here
Depth of field triangle! I like the way you explain :)
to my understanding the shallower DOF from dslrs compared to m43 comes from different distance from lens to sensor. or is this oversimplified? when you use a metabones adaptor on a m43 you kinda simulate the distance that a dslr has, right?
I don't think so. Focal length is measured from the optical center of the lens to the sensor. If that changes then I'd have to say the focal length chances and then DOF would be effected.
No it's like this: when you scale up the sensor to make a comparison you also scale up the lens, for the same geometry. So a m43 with a 25mm/1.4 lens becomes full frame 50/1.4. It's the same 'geometry' it's the same type of lens design (1.4), it's just scaled up. They both produce the same FOV the same 'picture', the lens will be in the same class of difficulty the same design just a different size. But.. the opening, the max entrance pupil of the bigger lens will be 35mm whilst the smaller one is 17. That means the smaller lens will produce half the size of the bokeh of the larger one. It's all about the size of the entrance pupil that's the basic, primary source of bokeh. With a very small 'bug eye' camera all the world would be in focus. It would all be 'at infinity'. The bigger the camera the smaller the world gets and the more things are out of focus.
Higher pixel density doesn't haven't anything to do with depth of field.
100mm on FF is 100mm in FF. F2.8 on FF is f2.8 on FF. 100mm FF on m43 is 200mm.
Here is actually what is happening. Excuse the T approximation for exemplification.
F2.8 on FF and m43 isn't the same in terms of DOF. If you're using a FF lens in an m43 body You are using the center of the lens, so because you've changed you DOF you are now using a FF 100mm f2.8 as a 200mm f5.6 on m43, but only in terms of DOF. You assumed t stop value is still f2.8 in terms of the amount of light the lens is capable of allowing in.
In other words, your t stop value stats static but you DOF aperture value changes accordingly.
This has absolutely nothing whatsoever AT ALL with pixel density. That is absolute nonsense.
Accurately put, if for example you were using an apsc sensor that had 24mp, you would have a 36mp equivalent sensor on FF based solely on resolution and ppi. But again this depends on the sensor. Sensors has different pixel pitch within their own sensor format size. They would behave accordingly. In other words you would have higher resolution cropping an image with higher pixel density than without, that is a no brainer. But that has nothing to do with DOF.
This video has a lot of misinformation from a professional manufacturing POV.
Quite simply put, you can't compare sensors this way. Shooting at f2 on a FF sensor will always have a shallower DOF on a targeted object than shooting at f2 on an apsc or m43, simply because of crop factor. Yoire using lens compression and none optimized lens examples to make a confusing case that is just not scientifically accurate. You're also excluding t stop valuation.
A video like this, while enjoyable debate for professionals, is actually not a helpful video for amateurs or beginners to come across because it gives them misleading information that they can't disseminate, let alone grasp.
Distortion from compression isn't bokeh.
An accurate comparison would have been something like 50mm FF at f2 would have almost the same DOF as 35mm apsc at f1.4.
Then too would have to address ISO for exposure, or shutter speed.
Final question: 50:2 FF vs 35:1.4 APSC, seems like they have same sizes of aperture(25mm in diameter), so does this mean same amount of photons will hit each pixel in theory(ingore lens quality) if these two bodies have same MP count?
They aren't wrong. The sensor size has no effect on your Depth of Field as it is a physical property of the lens. The idea that a 100mm f2.8 becomes a 200mm f5.6 on a MFT is factually wrong. It remains a 100mm f2.8 except you are closer. So you back it up to a 200mm focal range distance to get a similar image. However, the aperture is still 2.8 the only thing that has changed is the distance. This will give the look of the 5.6 except it isn't. It is just as bright, and unsharp as the f2.8.
It's not overly confusing for beginners and is important information for them to have. The DOF is entirely a lens related effect. NOT a sensor based effect. Understanding this and understanding how one creates a shallow depth of field is important for all photographers to understand. Knowing this, you will be able to create shallow depth of field on a MFT camera, for some reason, people seem to believe that is impossible, and you will be able to create deep depth of field on a large sensor camera.
Great explanation and illustration. I'm subscribed. It's difficult to explain things to people 'cause they won't listen. The best thing about this video is it demonstrates the truth. Yeah, some won't believe it anyway like when Galileo said the Earth goes around the Sun despite solid reasoning. This video is both the solid reasoning and good illustration. I liked the drawing showing why it actually the way it is. *A superb job. Commends a lot of respect! You really nailed it! You are probably a superb pro photographer of the highest caliber* Most photographers can't even get things in focus, or avoid motion blur and shoot in manual while spreading their misconceptions, wrong ideas etc.
7:12 Answer is there, just Multiply the Aperture BY 2 on the D850, and problem is solved.
yep, applying crop factor to both focal length and apperture! a small miss.
nice video. really well explained. I think that the thing is with cropped sensors there is a lower treshold of depth of field, so at some point you simply cant get more. full frame and maybe medium format allow you to go for those extremely shallow depth of field. your nikon with a f1.4 lens produces unmatched depth of field and I love it
Hallelujah, you got it spot on the money, I wish everyone would watch your video !
@Fstoppers
All you say is correct, but in the last part you missed to explain how you can get the exact same shot on a smaller sensor.
If you keep the composition of your 100mm F2.8 shot, you can get the same DOF on M43 by matching the actual physical aperture diameter of both the 50mm and the 100mm lens. Your 100mm F2.8 lens has an aperture diameter of 35.7mm. To get the same aperture diameter on the GH5 with a 50mm lens, you would need an F1.4 aperture.
He complicates a bit by introducing physics... I'd explain like this - equal focal will give you equal DoF at the same aperture. Take a photo on a FF and crop it by 1.5 (or so) and you'll get the same photo as if you shoot with that same lens on an APS-C body.
The smaller field of view on an APS-C is what makes you go further away from the subject or take lower focal length to get larger part of the scene - both of these will increase your DoF... but obviously, neither of the two is the sensor :-)
Finally a true no nonsens, no emotionaly driven video on tis subject, thank you verry much !!!
Just my 2cts...
That answer was an emotional one. Unfortunatly most photographer, including lot calling themself "pro" just don't have any clue about optics. Optics is a science, it's a branch of physics and one that has been well studied since more several millenium. It define precies terms such as focal length and depth of field. And the way all this works is all backed up with centuries of scientifical experiments and studies. Those are now established fact ! Ho yes i know the way the real worl works doesn't exactly fit the way most of you photographers want to see it, but that doesn't change what it is. I know that hurt your feelings, but the real world doesn't care about your feelings, it's been there far befor you born and will still be there long after you're dead.
So you think explaining how things realy work is "misleading" ? hahem , well yes you are right, this could lead people to question manufacture's marketing BS. That's probably no good.
So her's the deal, that guy put up a video backed up by both centuries of physics science and show exemples that demonstrate that they actualy work that way. So you call it BS ? ok then go on, do the same work, give us some proof of what you pretend, back it up with citations of scientific work that goes on your way...
Unles that, what you say is just trolling.
Just my 2cts...
When you crop the image of the full-frame camera (6:31), you actually get the smaller sensor size, and that's why DOF is the same. If you made the same image, you could prove that larger sensors DO produce shallower DOF.
If i use a 180mm macro 3.5 lens, wil this lens have more DOF instead of a 100mm macro at 3.5?
"To keep this simple we're going to ignore the variables that are the key to the actual subject of discussion when the subject comes up." (Slightly paraphrased.)
Is that why this was so damned confusing?
good point, but in not all scenarios you can be able to control the distance of the subject. i'm using both crop and fulframe but i want to do a shot with shallower depth of field, i always use my fullframe.
So you are saying for two cameras with differing sensor sizes and lenses of the same focal length the depth of field will be the same, but the field of view will differ. Whereas for the same two cameras with lenses of equivalent focal lengths, the camera with the larger sensor will have the shallower depth of field. In other words, yes the larger sensor does produce a shallower depth of field (when comparing lenses of equivalent focal lengths).
Duncan Moakes It's the lens, not the sensor.
"lenses of equivalent focal lengths" is a really bad way of saying "Two different lenses"
Two different lenses will create two different depths of field.
1 lens will have the same depth of field regardless of what is placed behind it.
So when I put a 50mm 1.8 on a crop body, am I getting the same results as if I used a 85mm f1.8? (80mm to be exact)
1) The image that comes out of the lens is always the same (2.8 on FF is 2.8 on Crop as well)
2) Background blur has 3 factors: Aperture, Focus Distance, Focal Length
3) To get the same picture with a crop camera like FF you have to back up a little bit due to the 1.6 crop ----> The focus distance gets further away ----> Less Background Blur
isn't the aperture multiplied in crop sensor so in this case, 2.8 times 2x crop giving us f 5.6 in the m43, so to get the same aperture don't u have to use f1.4 for m43 and f2.8 for full-frame to get the same results??
Starting at 6:11, you bust the pervasive myths about applying crop factor to f-number and you demonstrate how you make changes in optical systems and framing on crop sensors that account for the changes in depth of field relative to full-frame. Thank you.
What is that tripod head that you are using? And is it fluid?
I genuinely found this very interesting! I’ve always believed that full frame or even medium format cameras produced shallower DOF, though as explained in the video, it’s actually the choice of lens producing the DOF, as the crop factor alters with the different sensor sizes. Great work!
When you use DOF you can fall into a trap of confusion? What about Depth of Focus, DOF? They are NOT the same!
Can believe that people are still complaining about the video. It was amazing! Thanks!
When comparing photos taken by different camera sensors with different lenses, you also need to multiply the aperture by the crop factor, not only the Sensors and/or Focal lenght.
Exactly. This video is just a wrong point of view. Cause and causality is reversed and it just leads to more confusion. If somebody would like to understand this phenomenon I recommended this video: ruclips.net/video/hi_CkZ0sGAw/видео.html . Much better examples and clearer explanation.
what's the equivalent of 1.8 in full frame to crop? is there any crop lenses available ? none, so yes, full frame as for now, has shallower depth of field.
Longer lenses have shallower DoF because the physical aperture size generally gets larger as you zoom, not because they have a longer focal length.
Bepl h no. longer focal length don't *generally* have larger aperture.
Try the kit lens: 18mm / 3.5 = 5.1mm aperture growing to 55mm / 5.6 = 9.8mm aperture
Bepl h you don't divide focal length with aperture f-stop
How else do you work out the physical aperture size?
Bepl h f-stop *is* physical aperture size. t-stop is how much light coming into your sensor.
@6:58 are you sure you did not move the camera position for the two images?
So, all things being "equal".... the full frame gave you a shallower depth of field. WTH are you even doing ...
I think what he's saying is that you can get the same depth of field with an APS-C sensor compared to the full frame sensor. It's not the sensor but the distance to the subject and focall length. FOV changes based on sensor so you have to move with the FF to get the same FOV as the APS-C
eric00214 oh I got that, but what he eventually admitted is that due to practical and typical shooting distances, FF has the effect of giving a shallower depth of field, which is something he should have just stated in the beginning..
Sir
What is difference between a full frame sensor and a crop factor
Thanks... this is a much better explanation than the one from Tony Northrup who keeps telling Camera manufacturers to change the specifications of the lens for APS-C cameras.
So 35mm on apsc has a field of view as 50mm but a depth and perspective of a 35mm? Is this correct?
While scientifically speaking, the sensor does not impact depth of field since the lens spits out the same image behind it regardless of what is behind it. But practically speaking, the larger the sensor is, the easier it is to get bokeh. For example When you see a full body portrait with a completely blurred background (that is a natural lens blur. What would it take to get that same effect on a camera with a 1/2.3 inch sensor? Most of the common terms used, are not based on the pure physics of the light, lens and sensor, instead is it based on the end result and what is practical. What is sought after is getting a specific depth of field while getting a specific framing. It is all about being able to quickly communicate what equipment you want and what end result you want.
scientifically speaking, depth of field isn't just what the lens puts out. theoretically, there is precisely one distance exactly in focus, and it is infinitely thin. ignoring format, the depth of field for any lens is *zero*. you have to factor circle of confusion into to get a range of acceptable focus that appears sharp in your final print, because circle of confusion is the largest size an out of focus circle can be while still appearing as a point. circle of confusion relies on enlargement and viewing acuity/distance. for a given print size, the larger sensor will be enlarged less, so it has a larger circle of confusion. that means things can be further off that precise point of focus while still appearing acceptably sharp.
that means that for a given lens, aperture, and distance, the larger format will have *more* depth of field. not less.
it's just that to get the same angle of view, the photographer with a larger sensor will use a longer lens, which will have a larger physical aperture for a given f/stop...
I was stating it more from the perspective of if you were to take a lens and hold it in front of a piece of paper, you will see a circular image projected on the paper. Since a lens is passive, making that paper larger or smaller, will not change the image circle. When you use a smaller sensor, you are effectively physically cropping that image circle. Because of this, if you keep the lens at the same exact position and only swap the body, you will get a more smaller FOV, but the depth of field will be exactly the same. The change in DOF comes from when you adjust your framing (mainly be moving back a bit when you go to a crop sensor, or move to a wider lens.
Arachnophilia, spot on!
Razor2048 on a crop sensor, you're also enlarging more for a given reproduction size. that happens to matter.
trying to talk about DOF without COC is a bit odd, since COC is the thing that defines what's in focus. COC changes with format size.
The depth of field changes noticeably if you from 135 to medium format or even more, large format. But aps-c and full frame sensors don't have big enough of a difference.
Lee the iconoclast. Lens compression doesn’t exist. DOF isn’t related to sensor size. What’s next? Strobes don’t actually make a scene brighter. Sensors don’t actually record images. I can’t wait.
I have only recently begun paying attention to FStoppers, but it seems that Lee is becoming increasingly pedantic. These clickbait videos are more concerned with arguing that the terminology is not absolutely precise than the real world scenarios that photographers face that lead to these concerns.
Yeah, strobes don't do anything because you can just use a longer shutter speed without strobes!
Looking for a fast lens for the gh5! Any suggestions?
Schrodinger's cat loves photons :)
shouldnt you multiply the fstop with the crop factor aswell, because that could be why there is less blur in the background at 7:25 or something..
You are correct if only there would exist f1.4 m4/3 zoom lenses and f0.7 prime lenses.
Shallow depth of field is almost always bad.
The thing with most MFT lenses is you can shoot them wide open. Often not the case with ff lenses
All modern FF lenses perform or are expected to perform well wide open.
@@kiisseli1337 Often they need stopping down a little
Totally agree with you... If you look at a DOF chart; the larger sensor (needing less enlarging to attain a specific size product) has, if shot from the same distance and with the same focal length, a wider DOF than the smaller sensor.
However, as you mentioned, photographers will usually back up or use a longer lens with a larger sensor which does produce de facto, a more shallow DOF.
Thanks I'll get rid of my FF and use my phone from now until I can find an even smaller sensor. 😀
Really great explanations for the most part! If you want practical knowledge as a photographer on how to reason about DOF on different cameras, there are better videos for that. But if you're interested in the actual science behind DOF, this was a really informative video.
If you feel that this video made things even more confusing for photographers, just keep in mind the video doesn't preclude using rules like "multiply by crop factor". Those are useful rules. If in addition you want to know more of the science, then keep an open mind and learn from the video.
7:55 - thank me later...
If you are shooting a subject the FF forces you to move closer for the same head shot making the DOF shallow with the same focal length equivalent.