I always assumed that the misconception about medium format DOF was due to people not understanding the changing focal lengths/stops between the various formats. One sure edge though to MF is the details and colors that can be resolved with those massive sensors. Great video!
The misconception isn't the math. It's that you require faster more expensive lenses the smaller the format is, to the point that they don't even exist. So with decent affordable lenses on a FF or Medium, you can accomplish things that are practically impossible in reality on a smaller format. To say that it is a myth because of a math formula, is really missing the point. If you want MORE depth of field, and don't need a fast lense or a handheld shot, no problem. But if you want a fast lense and a narrow depth of field, there IS a large advantage in larger formats. Equivilences be damned. Show me the wide variety of quality affordable sub f/1 lenses in the DX format. and we can discuss. Databyter
Conversely though FF has way more investment and engineering behind its products so usually you can't find a MF equivalent shallow DOF like you can in FF
The old real medium format was 6x6 cm (56 x 56 mm) and equipped with a standard lens of 80mm 2.8 it was very hard to get the same look on FF because you would need to use 35mm lens (and crop it to the square) and have aperture of 1.0 that generally does not exist for such a lens. Also, if you have FF with 35 1.4 and you want to get the same look in apsc you would need 23mm 0.9 that generally also doesn't exist. The thing is that medium format is now usually 44 x 33 and there are not many fast lenses, so in many cases FF have smaller DoF now. But if you use film 6x6 it's very difficult to replicate using FF.
We’ll said. And the old medium formats were generally still very sharp in the middle of the frame when the aperture was wide open. I think this ads to the “look” medium format is known for.
You are exactly right. This is why they were exclusively talking about the modern digital medium format size and arguing that the "magical look" simply isn't apparent in digital sensors
Agreed. This would have been a much better comparison if it was between a more proper medium format digital 53.4x40mm sensor, albeit modern ones are of course incredibly expensive
Much comes down to how easy a given focal length can be built to a given big aperture and then what field of view it has in a given format. An 85mm f1.4 is a real sweet spot on full-frame that can be bought cheaply to get a shallow depth of field that an APSC camera will need usually a very expensive lens to reach - a 56 f0.95 or so. Similarly, the Pentax 67 with its standard 105 f2.4 lens. You can make a 105 f2.5 cheaply. But a 50mm f1.1 or so on full-frame is not a cheap lens - at least until recently. So it's what people typically use that creates much of the impression of the differences. Pretty much anyone with a Pentax 67 will have that 105 f2.4 which will totally isolate a full body shot from the background. But such images are rare on 35mm and basically non-existent on APSC - even though they're achievable with all formats. What's really strange is the way you reviewers often don't change your language to reflect this - talking about a 56 f1.2 as a 'bokeh beast' or whatever on APSC while not giving the same enthusiasm to an 85mm f1.8 on full frame.
Totally agree and points very well made. I have used all 3 types of sensor extensively and now settled on APS-C Fujifilm and Hasselblad MF [Had GFX cameras, don’t care for them much]. Unless you’re printing huge there really isn’t enough difference to justify almost anyone moving to MF. Where there is a BIG difference is when I’m shooting interiors, which makes up the majority of my work. When shooting a room at 16mm on APS-C or 30mm on MF for the same FOV the difference in scene compression/distortion is huge and very noticeable. Even noticeable over 24mm on FF, and that’s why I shoot MF. If I wasn’t doing high end interiors APS-C has so many advantages and why I still have an extensive set of APS-C kit. 99% of people don’t need anything else, in fact shooting MF is generally an absolute pain - huge files, huge heavy cameras/lenses, mostly need to shoot tripod (yes IBIS is fine, but MF lenses don’t have fast apertures so you’re generally shooting with slower shutter speeds, often beyond the usefulness of IBIS) - not to mention the cost!
This is exactly the comment I've been scrolling for. The "magic" of MF isn't the shallow depth of field. It's the FOV, compression, and lack of distortion.
I don't know if this means anything, but when I got my GFX, I was headed out to a real estate shoot. I was itching to test out my new camera, so I got it out and shot some side by side with my FF Nikon. I was immediately impressed how geometrically "correct" the image was in the viewfinder! Granted, I was shooting at 35, not very wide, but still, I saw a very different image than what I usually saw with my other gear. (FF and APS-C). Is that what your talking about as far as less distortion is concerned?
@@problemat1que The lens can definitely make a difference, a good lens produces less distortion for sure, but you can’t compare shooting with a 16mm to a 30mm lens (24mm in full-frame terms) and that’s the difference between shooting APS-C and MF to get the same file of view - the compression is totally different, it’s simple physics. Shoot a room with both side by side and it’s very clear.
You totally missed the most important aspect in that discussion: Lens availability. In theory there is always an equivalent focal length and aperture. In practice however, only for mentioning one of many examples, you have no equivalent lens you can chose to use with either APS-C or MF to get the same look as an e.g. FF 85 1.2 (probably even f/1.4) .
And even then, doesn't the actual focal plane get smaller? Like even matching a FF 50mm 2.8 with a 35mm APS-C at 1.4 would look different because of the depth of the focal-plane itself. The bokeh/blur would be similar... but I'm pretty sure the FF would have a persons whole head in focus while the APS-C would only have the nose/eyes in focus leading to quite different look. People always give technical reasons why there is no "Large-format look" but there certainly is... it's easy to tell a movie was shot in medium format (65mm film-back). It looks WAY different than traditional s35 (APS-C).
for me personally, it's about the look i am getting at a wider focal length with shallow DOF. APS-C to FF, because of the CROP (or the lack of), I can stand closer to the subject (i shoot mainly portrait) and get a shallower DOF because of camera-subject-background distance than I can with an APS-C system. The same is true Medium/Large format. So, like most of you guys said, it's not the technical aspect of the camera systems we are aftering, it's the perceived look that we can achieve with one system compared to the other.
I generally prefer more depth of field than less and I like that I get it with a faster aperature on my m4/3. Of course with full frame you can raise the iso with minimal noise penalty so it really comes down to know your partner and learn to dance with her. Oh and if you want really shallow depth of field with apsc or m 4/3's then your partners going to be expensive. (Fast glass)
It's also about the background magnification that comes with using those longer focal lengths. It will appear more out of focus because it's magnified (apparently often confused with having a shallower dof), but to Chris' point I'm not sure if it makes much of a difference on 44x33 vs. like 6x8 where a 300mm lens has the fov of a ~135mm on 35mm full frame.
More articles like this please! Let's have Live models, good/bad lighting, more analysis of results, blind tests, and throw micro 4/3 into the mix too :)
The lens speed arms race is pretty much pointless. There are so many other factors that apply in actual real worl applications. The range of uses for f0.9 is pretty slim.
@@sulev111 hilariously saw this play out at a friend's wedding, two photographers showed up with Canon SLRs with one 50mm F1.2 each (and no other lenses). Kept shooting with frowny faces without flash way into the night under those old-timey dim light bulbs and delivered grainy, out of focus results, converted to Black and White to hide the Canon color noise. I would have been so mad if I was the client.
How fast can Chris say a whole bunch of numbers? I feel as if I'm being sold a used car. The real depth-of-field myth that needs to be busted is the notion that less DOF is always better.
The myth that needs to be busted is the notion that there is right and wrong. Different things are used for different artistic expressions. There is no serious photographer who seriously thinks "less DOF is always better", so you're arguing a strawman. People use short DOF when it's the feeling they want, and their viewers like. Not because of some unwritten principle.
@@onegrapefruitlover I mean yes there definitely are people in all areas of arts that do a specific things purely for the principle. But it's not really a "myth" that needs to be argued against IMO. Those photographers will hopefully find different ways to make interesting photos eventually.
What your saying Chris makes a lot of sense. I'm a Fujifilm shooter and for me to achieve a Full frame DOF equivalent to 1.2f (which is readily available and can be found vintage) in crop-c that would be a 0.8f which is not only difficult for fuji to create in AF but gonna be bloody expensive. Bottom line, with the technology we have today the best and most affordable way to get the highest DOF is to use a full frame camera because there are so many fast FF lenses in the market what can beat the DOF produced by a 1.0f FF lens?
I always thought of medium format having better light falloff, not focus. The dynamic range of the shadows to light is so much more gradual in medium format. I actually was teaching someone about this recently and they were able to, only knowing this, correctly identify blindly, full frame shots(from a Sony) vs Fuji medium format shots. because sharpness comes from the lens, colour comes from the sensor, and depth of field comes from the aperture and distance from the subject to the background. But the light gradient comes from the sensor size.(ofc this is being a tad reductive but you get the idea) it's about the only consistent thing I've noticed whenever people do sensor size comparisons.
If someone could tell the formats apart without checking details it must have been due to the different processing of the files, be that in the raw converter profiles, or manufacturer jpeg engine.
Just curious, is there a way to copy or get close to that medium format gradation in full frame or APS-C cameras? Adobe profiles in LR are relatively contrasty compared to a linear profile that you get with Capture One. There are free 3rd party linear profiles for LR (I have a site for that) and it allows the image to look flatter and smoother in terms of gradation of tones. One reason I ask is because that smooth gradation of tones might not make a large difference when an image is on Instagram (both because of compression and size) so I wonder how much of a difference it REALLY makes in the end.
@@vernonsza there is, you could do a stitch of several different photos taken from the same scene. you could do a pano stitch, or multi-bracketing. basically all you are doing is capturing multiple shots at various exposures, since the smaller sensors have less dynamic range than the medium format. there are actually many videos about getting the "medium/large format look" on smaller sensor cameras. and all of them revolve around this principle. in the end I think the difference is only significant if you are looking for it specifically. and that I think is the greater point that Chris wanted to make in this video.
Agreed. But I don't think these 'small' medium format cameras like the gfx have much (if any) of an advantage here. I'm sure one can pick the difference when you know what to look for, but there is not really a discernable look like one sees on the larger sensors. I follow the architectural photography world a bit and one can always immediately recognise work shot on these expensive systems. I also read an interview with someone (arch photog) who switched from Canon to the Fuji gfx for this purpose (adapting TS-E lenses) but switched back after some months because there was no perceivable benefits evident in the light fall-off which a lot of folks are chasing.
As a M4/3 shooter, my only problem is mostly not the depth of field, but the low light performance, and there is no alternative for something like a 50mm f1.8, or a 24mm f2.0 so in that case, much cheaper to invest into a full frame camera, insted of a fast m4/3 lense
Here is a simple math trick: Divide the focal length by the f-number. What you get is the size of the entrance pupil, the thing that gathers the light. You will see that it's the same for all compared lenses. The rest is parallax. The thing with your thumb jumping left and right if you open/close your eyes.
I’m so glad you made this video! I’m so sick of hearing people saying “Love that Medium Format Look”. There is no Medium Format look. A 50mm f1.4 is the exact same regardless of which sensor size it’s used on. The only difference is that one can stand closer to the subject with larger sensors vs smaller sensors which allows the larger sensor to get shallower depth of field with the same field of view. But Fuji glass is too slow. Their f4 zooms are only f3.2 equivalent. They can’t even replicate the shallow depth of field as a FF f2.8. Not to mention, they are more expensive and have substantially less zoom range. Even Fuji’s primes are too slow with most being a f2 (f1.6 equiv) and one at f1.7 (1.4 equiv). Sony and Canon are already releasing f1.2 primes. I do believe the medium format sensor does deliver better ISO performance and more color depth, but one is going get a better “Medium Format Look” from a Full Frame system now days over Fuji’s offering.
Great video Chris. I think APS-c comes off very well in that comparison. I was thinking about getting a full frame camera but now I'll stay withy my Fuji X-T30. I wonder how micro 4/3rds would have compared along side of those larger sensors.
Other than what other people pointed I am also interested in the compression a longer lens will give you. I love thé 85mm field of view but i want to get more background compression. The only way to do it is to go on larger sensor and use a longer lens. Same applies for crop sensors. 56mm can give you 85mm but not in the compression of an 85mm lens.
Compression is only effected by spacial relationships between camera, subject and background. You can not take a photo that is the same distance from your subject with the same field of view and have any more or less compression. A 50mm on full frame that is 20 feet away from the subject will have the exact same compression as a 25mm on m43 or any other equivalent in any other format.
A very comprehensive "One time for all" segment about the sensor/lens conversion and final look. It will spare some photographers a lot of time. Thank you !
Great episode! How about color accuracy and the gradation from light to dark areas? I am considering what my next camera should be. I’m choosing between a Fuji GFX 50s II, a Canon or Sony Full frame camera, and the new Fuji X series camera coming out in 2022. I currently have the XT3 and I love it. But, I’m tempted and just want to make an educated and thoroughly thought out decision.
Despite knowing that FF is the sweet spot technically, I currently use Fuji because I enjoy their colour rendition and user experience. If I were you I'd wait to see what Fuji comes out with, see how X-T5 performs.
Stay with Fuji man, their new 1.4 lenses is all you will ever need. The new 33 1.4 is just a hell of a lens. Plus colors in jpeg on Fuji are super nice. Plus plus look of the cameras. There's no match at Sony with X-Pro 3 or X100V
I've been wanting to switch to a camera with 16-bit color. (GFX 100, any Hasselblad)... A few months ago, I switched to DxO Photolab 5 Elite and it rendered my X-T{2, 3, 4) photos better than even Capture One. This made me think I don't need a new camera any time soon, as Photolab is bringing life to old photos. Also, I am looking forward to seeing what those new X Series sensors can do for me.
@@calokid nice to hear that software can you give me some examples of fujifilm shots of your in lowlight especially the new bodies before when I used the fujifilm camera especially the fujifilm X-T2 before I saw noise grain at 3200 and my limit to that camera is 3200 ISO and if I need to take a shot even with noise is 6400 ISO. But right now I don't know the noise performance of fujifilm x100V, XT-4 and newer models and I want also the result you edit from the software you use because I'm Capture One user from the time I use Fujifilm thanks in advance 😁📸👍
@@junichinomura4810 The low-light advantage of some of the newer X-Series cameras really has to do with IBIS. I've shot with the X-T2 and loved the sensor. Every X-Series camera I've used starts to get noticeable noise around 3200, so no real change. On an X-T4, however, I can shoot handheld as low as 1/15th of a second, others, with a steadier hand can shoot as low as 1/8th of a second. The advantage of IBIS photography would only apply to photography in which tripods do not apply. You could try Photolab yourself, and we could also decide on experiment criteria for exploring how low light looks on X-T4 or we could process one of your RAW files in different software to see the differences.
i think this is where the misconception comes from. when talking about equivalency, most people are aware that 35 mm in APS-C will have the same look as around 52 mm in FF, but they don't realize that the aperture would need to be adjusted as well. and they just end up comparing 52 mm in f/2 in FF having more bokeh vs 35mm in f/2 in APS-C (while in fact based on the exact equivalency, they should have done 52 mm in f/2.8 -- closest number to f/3 since there is no f/3). the media has not helped either by always mentioning the equal zoom for FF in sentences like, "16 mm lens, or equal to 24 mm in FF", but then not mentioning the equivalent aperture. and another reason for the misconception is just based on lens availability. say i have a 50 mm f/1.2 lens in FF and i love that look. to get that same look in APS-C, i would need to get a 35 mm f/0.8 and well, good luck finding a lens like that. because of that, people then start saying that FF is better than APS-C since frankly, it can produce images that APS-C just can't. (or maybe it can if that lens is available, but im sure it would be very expensive).
Very fine explanation of the technical part of this mith. Just two (big) points let out of the discussion. The bigger the format, the bigger the lenses the lesser the aberrations, distortions, and much sharper because the properties and qualities of the glass become less relevant than on small format lenses. Also the bokeh - even if you correctly explained that the d.o.f would be (roughly) the same - would be much creamier on bigger formats while the in focus part will be sharper and with less aberrations well known on very fast lenses. Last thing: you correctly used (digital) medium format (33x44mm) in the test, which is by the way a "crop" format and quite small compared to much bigger film sizes like 6x7 (56x68mm).
The 100mp Fuji cameras, have the same pixel density as the A7R IV and the 26mp Fuji cameras. Meaning the precision of manufacturing for the optics has to be about the same to have the same level of aberrations and such on a per pixel level.
Theoretically it doesn't matter, as the image exists in a point, not an area. But practically there is a difference in how longer focal length go out of focus "faster". The really difference will be when you stop down and want deep depth of field. Because diffraction limit is more severe on smaller focal lengths, apertures and sensors. Althaus why people do f/64 on large format and get Soo much better results than f/18 on FF. "Medium format" as in 6x7 or 6x9 has a negative that's close to the interpupillary distance. So it looks and feels closer to your human perception. And practically. The fastest lens you can get on MFT is the Voigtländer Super Nokton 29mm f/0.8 - you can't get faster and wider even with focal reducers. Therefore going with a bigger sensor gives you some results you can not reach otherwise.
The interesting thing comes in that to get equivalent depth of field we end up with different apertures, so you can shoot medium format in daytime at a desired depth of field and be less likely to blow out highlights as the aperture will be more closed down to achieve it than apsc. Also, at night, we can gather more light on apsc without being restricted by razor thin depth of field that we would have on a larger format at the same aperture.
At night, you would just shoot a full frame camera with a narrower aperture and higher ISO if you don't want so thin depth of field. The better sensor performance would mean it's no sacrifice. But you do have the *option* of a faster aperture equivalent, letting you get better low-light noise performance or faster shutter speeds. There's no way to spin APS-C as an advantage in low light.
Depth of field is only really influenced by focus distance and aperture diameter in mm. Exposure is influenced by the focal ratio, otherwise known as F-stops. Interestingly, focal length directly impacts exposure, requiring a larger aperture to maintain the same F-stop. In the case of larger formats, longer focal lengths are used to achieve the same field of view as smaller formats. These longer focal lengths reduces exposure, so to maintain exposure, you use a larger diameter aperture. This brings exposure back to parity, and reduce depth of field. I’d probably consider both APS-C and FF to be kind of the sweet spots for photographers. Sensor advancements allow shooting at darker F-stops at night, while the availability of very fast lenses for both systems allows for extensive flexibility.
For me medium format is less about DOF and more about tonal gradations between light and shadows. Maybe that’s a myth too, or perhaps the MF photographers whose work I admire are just really good at lighting.
I guess if you can afford (and want to buy) a medium format camera, you're more likely to be good at photography and have more experience in that craft.
It’s very true. This “test” was designed to achieve a desired result. Yes, I can make an xt4 image “look like” a GFX image, but the quality of that image, resolution, ability to edit, shadow and highlight depth, will be significant different. It’s all the same on Instagram. Much different when you’re printing a large bridal portrait.
There is indeed a difference if you take for example a X1D and a full frame lens like a 50mm 0.95 with large enough coverage for those medium-ish format digital sensors.
Hi Chris. You are talking about background blur which may vary but depth of field remains equal at any given aperture regardless of sensor size. I love this channel and have been following you guys since camera store
One thing I struggle with on Micro Four Thirds is ISO, as you would expect with a smaller sensor. My f2.8 lenses are more like f5.6 on full frame in terms of depth of field. However, since sensor technology is constantly changing and improving, I wonder how much further you can push the ISO on, say, a Sony a7 IV vs. my OM-D E-M1 Mark II and get similar noise levels. I usually think of two stops as a rule of thumb, but is that really the case? On my camera, I consider anything above ISO 6400 for emergencies only, and I wouldn't want to do much cropping above 1600. Unfortunately, I find myself reaching 3200 and 6400 more than I would like to when I'm shooting wildlife due to the fast shutter speeds.
I don't use M43 but use(d) FF, APSC, and 1". Crop factor works as you'd expect, and FF will look like the ISO is 1/4 of M43. That said, M43 has really, really good stabilization, so the difference will mostly appear with moving subjects.
My issue with M43 (I sold all my cameras and lenses (Lumix G9, etc.) was ISO, DR, and color noise, but mostly pixel density. The size of the G9 M43 sensor and the density of its pixels gives M43 a more digital look (less organic). This is also seen in some newer APC-S and FF cameras. Even though some look for pixel perfection, for me it was almost resembling iPhone quality. Raws were like artificially sharpened and I could easily tell the size of sensor by the look of the images.
@@borispradel1037 but can u imagine 10 years down the road when gh9 comes what can happen? Iso performance will be atleast twice or thrice better. And computational photography technology is expected to come to m43 before apsc due to smaller size.
@@josh_boak yup 12800 ISO is the usable for lowlight scenario in emergency use also for colored photographs of any fullframe camera bodies right now 👍😁📸
Interesting video, Chris however, though I know your "focus" was on depth of field equivalence, I think you glossed over the major advantage of medium format when compared to the smaller formats...the fact that when you print very large the MF image requires less enlargement hence more sharpness and detail are retained in the final print. And just FYI, I used to develop large format b/w sheet film, and a contact print from an 8x10 negative is the sharpest, most detailed image one can get, since there is Zero enlargement. Thanks.
Most important factor in depth of field is final image size relative to viewing distance. Look at the same print from further away and the depth of field will be greater. Hence popularity of incredibly shallow depth of field when images are only viewed really small on Instagram. Same depth of field does not work for image to be viewed on phone and one to be viewed as 20" print (unless you view it from far away). So strangly this justifies fast, heavy, expensive glass just to post on Instagram. Having had fast lenses in the past I looked through my lightroom catalogue and discovered I more often ruined shots wide open than not so now shoot with an f4 zoom in FF and it is great. I guess on crop I'd need an f2.8. Generally fast glass is a fetish as it is no longer needed to compensate for slow films.
@@de-graftasafo-adjei4646 Edit of what I posted on that video: DPReview definitely triggered you didn't they! Depth of field is actually ZERO for all lenses and can only defined in practical terms by what an acceptable size for the circles of confusion on the sensor/film plane is (en.wikipedia.org/wiki/Circle_of_confusion) and that depends on the angle subtended at the eye of the *final image*. Literally all you have to do is look at a image with shallow depth of field on your monitor at a normal viewing distance and then look at it from the other side of the room and you will see that the depth of field appears to increase. Take a portrait where the ears are slightly out of focus as an example. From the other side of the room they will look in focus. In that case there is nothing to do with the camera at all :) I shoot 8x10 (and whole plate) and contact print them and they look great but I know if I were to enlarge them the depth of field would be way too shallow - unless the prints were viewed from far away (Father Ted reference). So does the 8x10 have shallower depth of field than my APS-C with an f/2 lens on it? Well it all depends on the print size! (also subject distance which is a biggie). It's so funny seeing people get wound up about this. Naughty boys at DPreview provoking everyone. I'm just stoking the flames.
In my experience, going from canon 5d III(fullframe) to fuji xt3(aps-c) I just do not have the same tools at hand regarding field of view and depth of field. My concern is mostly getting a wide angle with great separation. That is not as simple(if impossible) with my fuji as it was with canon. For example saying that fuji's 16-55mm f2.8 is same as canon's 24-70mm f2.8 is just rubbish. 16mm just does not produce the same separation as 24mm. + 16mm gets a lot bendier(barreling).
The analogy I like to use is gears on a bicycle. You certainly can pedal up a steep incline in a high gear, but the amount of work required to do that would be more than most people would like. It’s a rough analogy, granted. So, while you certainly can match looks amongst these different sensor sizes, there’s clearly a “comfort zone” for each format. We saw this in the early days of prosumer digital cinema cameras when customers were seeking out ultrawide lenses for their first generation blackmagic cameras. Not disagreeing at all with the comparisons presented here… but, as a thought experiment, imagine if micro-4/3 had been included in this roundup.
the real difference can be found in colour rendition. medium format sensors capture a 16bit image that reduces banding resulting in more lifelike colour/contrast transitions.
The logic of the test is strange in my opinion: "let we adjust the aperture for the medium format camera so it has the same depth of field as an APS-C camera. Now you see, it has the same depth of field!" )) Basically the medium format makes it easier to make shallower depth of field lenses. GF 110mm f/2.0 is equivalent in terms of depth of field to a 55mm f/1.0 on an APS-C camera. And the latter lens will be definitely inferior in terms of quality wide open. And GF 80mm f/1.7 simply doesn't have any equivalent in the XF world. Let alone zoom lenses. And this difference is even more pronounced if you compare larger film formats to a "full frame". Basic f/3.5 lens for 6x9 will produce the same depth of field as f/0.75 on an APS-C camera.
Whenever I hear arguments like this video, the diminishing returns you suffer trying to match the “medium format look” with anything but is never addressed, especially in the comment section. This is the first I’ve seen addressing just that.
@@kapapa3764 yes we do. And in fairness Fuji GFX (and any other relatively affordable "medium format") is not so much of a "medium format" if compared to FF.
Actually - there is a difference (although, most people will not care too much - but for my shooting, it is important to get a certain look). It's about the depth of field you can get at a specific distance with a specific focal length. That's what can make medium format images look "special" or maybe the real word is "different". Everything else is correct for the general case, it's just that there are situations where it is not possible to frame a subject at a specific distance and get the same fall off. For example, it can be very difficult to get a wide angle full bodied portrait with shallow depth of field on an APS-C camera, without either long zooms or exotic lenses - but even then, it's not the same look because you don't get the same distortion characteristics. The thing is - a 24mm f1.4 IS a 24m f1.4 no matter what the sensor size. It doesn't change. But obviously the framing will be different at the same distance. So that's where the FOV and DOF equivalency comes in - you need to change the distance to the subject to get the same framing, and therefore DOF necessarily looks like it is different when on different sensor sizes.
But the equivalency is the key. I can very well have 1.2f on 35mm for full frame and I won't be able to buy equivalent lens (0.75f) for ASP-C, hence bigger sensor => shallower DOF possible.
First of all, another awesome video, even though I am like 1 year late to it...BUt what you are confirming is hat in order to maintain the same DoF in smaller sensors you need larger apertures! So I would argue that definitely it is easier to get shallower DoF with the MF than with the FF and with the FF it's easier than with the ASPC...Might be just semantics, but using the same lens with the three sensors, while the angle of view or apparent focal distance changes, aperture does not. F2 is F2 in any sensor. Now, the larger the sensor the most difficult/expensive to build large apperture lenses. An APS-C F1.6 lens is cheaper than an F3.5 MF lens. And yet there would be no gain in terms of DoF. In orther to have a gain you would need a F2.8 or F2...So, looking at this way, yes, nothing magical or special about MF.
Yes thank you this is why micro four thirds with 25mm f/1.4 works for me! Gives the same photos as full frame 50mm f/2. Unless you need lots of megapickles, which I do not. I gotta say, the reason MFT died is sensor development stalled and why weren't they putting f/1.2 apertures on all their lenses from day one? Even the Voigtlander lenses with f/0.95 are not too narrow depth of field, they're totally usable wide open on MFT.
The 6 year old sensor on the EM1m2 already outperfomed, and still does, the best FF body by more than 1 stop of DR shooting at quivalent ISO settings from ISO200 and above. This M43 sensor punches way above its weight.
I feel like this video kinda proved the opposite. It's a situation where a standard medium format lens gets you this baseline depth of field but meanwhile on the APSC you have to blow a bunch of money on their very high end super wide aperture lenses to try and replicate the DoF benefits of a larger sensor. I'm not a "gear matters" kind of person with gear acquisition syndrome or anything but when you have to make the other cameras work a Lil harder to replicate something that the other one does naturally, the natural one seems the winner?
Yeah but what about the cost of the medium format bodies? Doesn't really seem like you are saving a lot of money by going with that option, if any at all.
@@alexbustamante6532 yeah i thought about that too, but that also depends on the body you're getting. Fuji GFX 50R retails 4.5k USD, almost always on sale for 3.5 or even 3k. My A7RIV retails 3.5 as well, so it's effectively the same cost. Only time cost really becomes a factor i feel is if you're genuinely debating an APSC vs Full Frame vs Medium Format, but depending on the body, your price range between FF and MF is negligible
The 56mm is $1000, the Sony $1600, the 110mm is $2200. In APS-C you can go up to 50 F1 at $1500 (not coincidentally similar to the Sony) or down to 50 F2 at $450, making it a very flexible system for both size and price.
@@problemat1que yeah if we're using the price argument obviously APSC is far superior at being cost effective but the video isn't as much focused on the price end, moreso focused on the final image. Obviously price matters, it's just a secondary subject here
This is only applicable in the context of shooting on native lenses to each system. The beauty of mirrorless is having the ability to adapt. 44x33 is actually a great compromise because in many cases you can still get full (or close to full) sensor coverage with FF lenses. Pairing digital medium format with those larger apertures gives an insane amount of compression. The biggest advantage is compression at wider focal lengths. This also doesn’t take into account natural dynamic range and detail retention in post processing.
compression comes from camera distance, not the lens. the lens limits fov, as does the sensor, so moving to medium format gives you more fov within the same focal length, but doesn't do anything to perspective.
Is there a difference, yes, super noticeable, nope, replicatable, not really. lemme explain: * Lets assume parallel rays come in and exit the rear lens at an angle to meet in the center of the sensor... (most diagrams are drawn like this) * Now, notice the rays form a cone if you will, like this: > * Whats important is that flange distance sometimes in medium format, of the distance from the back lens to the folcal plane is longer, then the cone angle is smaller. * Ok, when something goes out of focus with a longer cone (focus in front or behind the focal plane) there is now a little circle on the sensor. * the size of this circle also changes with the angle of that cone, hence the flange distance. * if you have a long flange distance than something far out of focus doesnt seam to be that out of focus.. * when the flange distance is short, that cone angle is big and things can jump out of focus faster and even more out of focus the further things are. This was a tough one for me when I went to medium format, I couldnt get super out of focus fall off, the out of focus was so gradual I kept going back to my full frame for some shots. Dont misunderstand me, the medium format had more out of focus, and thinner focus plane where I was shooting beauty at f/16-f/22. Overall what I saw the major difference was thinner focal plane and smoother transitions in the focal range
I feel like this video misses the point. No one is concerned about how to make every image look the same across these cameras. People want to know how medium format CAN look different. Take a 50mm f1.4 lens and try it on every one of these cameras with the same subject and compare them. Do the same with a 135mm lens, and with an ultra wide lens and compare the differences. There is a difference in how each sensor size affects the image, but you will never see that if you are only focused on how to make them look identical. Also it is important to show the larger sensor medium format cameras, regardless of how expensive they are, since we are specifically looking at how sensor size affects the images produced.
A true comparison would be to put a Hasselblad V mount lens on all 3 cameras and take a shot at same aperture, same focal length on all 3. Then tell me there is no difference. Nobody shoots medium format to emulate APS-C! Depth of field is just scratching the surface of what a large sensor is all about. Larger sensor = better low light at any given resolution vs same resolution on smaller sensor. There's a reason 100 megapixel has come to a GFX camera before a Sony A7R. Larger sensor = wider compatibility with more lenses. You can crop to GFX to FF and use adapters to vintage glass designed for SLRs or cinema cameras, as well as using medium format lenses from Fuji, Hasselblad, and so on. To see the full potential of a medium format lens on FF you need a focal reducer and that in itself has compromises, won't work with mirrorless medium format lenses. To say the video is a gross oversimplification of photography and why people choose medium format or larger sensors than APS-C is an understatement. Why try to reduce everything down to the most simple spec imaginable and use different apertures, different focal lengths to achieve that? What does it achieve? Nothing!
and that will be very idiotic, people buy full frame/medium format for larger pixels, better low light, tonal gradations, bla, bla, bla, not for your mathematical deep of field calculations, what have no relevance whatsoever...what next you will put a phone sensor and demonstrate how good it is ? Damn that was so shallow...............
So basically, focal length and aperture stays the same on all format. the only thing that changes is the size of image circle and sensor size. is that right ?
Depth of field is determined by the f stops on your lenses, not the format of the camera. The lower the f stop on your camera, the lower the depth of field. The higher the f stop, the higher the depth of field.
When we say “DOF” what we really mean is background blur. “DOF” is about what is in focus and background blur is about what is out of focus. How much blur an out of focus area will have is dependent on the size of the aperture opening. On a 110mm lens at f/2 the aperture opening is 50mm where as on the 56 1.2 the aperture opening is 46mm. That’s almost a 10% difference. This is why when comparing size of the bokeh balls on the indoor comparison the MF took the cake over the APS-C. The further away the background is from the subject the more this will become apparent because the lens with the lager aperture opening will have a larger maximum blur spot potential. When comparing to the full frame tricky part is, yes, the MF isn’t that much larger than FF so the advantage isn’t all that noticeable
Larger apertures often have more aberrations visible, which can affect the perception of quality. The greater resolution also allows you to better discern what is and isn't in clear focus, giving the illusion of "more depth of field". There are some differences at larger scales but not because of the format but rather the lenses. Matching the DoF of say a f/5.6 lens on 8x10 would require a full frame f/0.7 lens, and that doesn't really exist.
Something I noticed pretty straight away when I started with a Nikon D3300 several years back, some companies are much better at supporting the APSC format than others. Fuji ofcourse knocks it out of the park, while I think Nikon and Canon string you along with some decent APSC lenses here and there, hoping you will 'upgrade' to full frame. If you do your research and want to commit to a smaller/lighter kit, it's very much worthwhile researching what APSC-specific lenses said company offers first. Thankfully many third parties fill the APSC-starved gaps of Nikon and Canon on the DSLR side. Things are rolling out a little more slowly with the RF and Z mount. Here's hoping we see more quality 3rd party autofocus lenses make their way to those mounts in the coming years.
@@Bayonet1809 It‘s defined beeing between 4,5cm x 6cm and up to 6cm x 9cm and thats far away of 3,3cm x 4,4cm. This is a non existing film format, so it can’t be Medium Format what is defined by film.
Hey DP, I know you’ve already covered the iPhone 13 as a standalone video, but what I think is lacking is a iPhone vs entry level camera review. There’s many iPhone vs $5000 dslr videos but none are comparing the same price points and would be very interesting to see how they would compare, either be a old dslr or a new dslr with a decent lens?
I think the term “magical” is overused. There’s a bigger difference between APSC and FF than FF and Medium Format. You don’t need FF but it’s definitely better to me. But to each their own.
@@jamesburns679 There's nothing magic about it if you compare them with the same >equivalent< apertures, sure. But FF has more lenses with larger equivalent aperture lenses than APS-C, and that's where the 'magic' happens. There are tons of 1.4 lenses for FF, but an APS-C lens would need to be f0.95 to get the same DOF, with everything else equivalent. And the ironic thing then is that the APS-C 0.95 lens will most likely be more expensive than the FF 1.4 lens. Look for example at how expensive the Fujifilm 56mm 1.2 is, compared to equivalent FF 85mm 1.8 lenses. No APS-C system has anything equivalent to an FF 85 1.4. So, no-one claims that the FF magic is in the sensor itself. Rather, it's in the lens selection. FF really is the sweet spot between sensor size and lens selection with large equivalent apertures. And I'm saying that as a current Fuji user. I love their user experience and colour rendition, but there are times where I really miss my Sigma 35 1.4 ART, since Fuji doesn't have anything properly equivalent.
Thank you for trying to educate people on this topic! There's another "magical" thing: APS-C/MFT lenses that "have bright f/1.2-2.8 aperture" are magically lighter compared to FF lenses with such apertures. Well, they are just darker. I don't mean having darker lenses is bad, you just shouldn't pretend that physics and geometry don't exist :)
Also, folks say oh medium format has better low light sensitivity, but you have to take into account the fact that you can shoot at f/1.8 on APS-C easily, and that aperture would be ridiculous and unusable on medium format, so you can compensate for low light with the lens rather than increasing the ISO.
So many people and so many videos make this too difficult. The only factor affecting your image is magnification, that’s it. The only critical physics benefit of larger formats (technically) is angular resolution when tilting or swinging lens or film planes.
@@Frontigenics a "look" is a very subjective thing, but if you mean equal depth of field, It's not possible due to the magnification difference you'll experience using an APS-C sized sensor. You can't have the same field of view and same depth of field because the magnification will be different, and therefore either one of them will have to change.
Another example of marketing, as you basically say at the start. Most digital MF isn't anything like the size of MF, But the perception to the consumer, or rather to the enthusiast, is that MF represents a huge difference in size/quality like it did in the film days.
This is not medium format)))) not even 6x4,5))) Fuji is a crop of medium format)))) But all the magic of medium format is not a dop! All magic is about perspective!!! Perspective of 80mm but view of 50mm!! Or even 45mm!! This is a true magic!
I think MF would be good for large prints, where detail is part of the presentation. APS-C and M-43 have the advantage of having a large DOF, which comes in handy for some applications. When the smaller sensors can stack images with sensor shift, you can get some great results on static subjects that need the larger DOF advantage. My desire for MF is for the ability to photograph moving subjects with the ability to capture great detail for large prints (which is entirely a stylistic choice, not a technical one). Currently I have a Canon RP, and the advantage from moving to a larger sensor from M-43 is the reduction of noise for indoor shots, and 6 more megapixels. Plus, I love the JPG colors canon makes. I tend to work in more from a purest perspective, "cropping inside the camera" and making color adjustments "in the camera". I appreciate it's a slower process, but again that's a stylistic choice. There's a camera out there for everyones style and technical workflow, and they are all valid! Heck, even using a phone camera is valid! I use the mantra "you can make a technically perfect photograph, but if it's a boring image, it's just a boring image". Have skill in your technique, but remember that content is king. These types of videos are extremely helpful, because they tell us photographers the abilities and limitations of each camera. It helps us decide if the camera is going to aid our style or hinder it. Plus, the presenters are just entertaining to watch! Thanks DP Review, Jordan, and Chris!!
Side note, I think it would be entirely reasonable to evaluate a lens using "angle of view" and not the millimeters of the physical design (because there is variation on angle of view between lenses of the same millimeter). But, just like I'll never learn the metric system, I'm sure it would be like pulling teeth to try and get people to start using AOV.
I recently purchased a 45mm Pentax 6x7 lens and attached it to my Pentax 645z. Was told the 35mm equivalent of that lens would be 23mm but it looked exactly like a 45mm 645 lens I already had for my Pentax 645z. Finally I decided to place that 45mm 6x7 lens on my Pentax K-1 full frame and aside for the different frame sizes (pentax 645 has a 4x5 frame) it looked closer to a 55mm equivalent than a 23mm lens I had that was native to 35mm full frame. In other words I saw almost zero difference in terms of what was included in the frame when mounting the 45mm 6x7 lens to the 645 as the 45mm 645 lens mounted on the 645. What am I missing aside from the money I spent thinking this 45mm 6x7 would give me the same focal length of a 23mm lens native to full frame?
I think it's related only to modern digital 'medium' format. On 6x9 film difference is more obvious. But there is no sensor of that size. I don't even mention large format.
it's the same on film too, but there are other benefits of a MUCH larger image capturing medium than DoF. All MF film still has a tonal range advantage I think over digital...and still handles highlights in a much more pleasing manner than digital.
Medium Format is magical for enough reasons to be better than FF. It doesn’t mean I don’t like all the sensor sizes. I just know which one I like the most MF, period.
There's no difference... besides all the differences. I mean, everything technically is as pointed out, but put someone who doesn't know the technicals on the street with a different sensor size than they normally use and let them take photos the same way they always have. They'll tell you there darn well is a massive difference. I found this out the hard way when just starting to use a Minolta Autocord while in Santa Fe, photographing my family in front of the church. The most beautiful family portrait ever taken _with the guys in the second row obviously out of focus!_ In practical use, there's a difference.
Well the aperture is equivalent throughout the three , no crap the depth of field is the same/similar. If you use the same equivalent focal lengths with the same aperture for each lens , then youll see the difference. For instance all at ,F2 and equivalent focal length. I always looks at depth of field per a given aperture. If all lenses theoretically have a f1.8 with equivalent focal length then i think you would have a obvious winner.
So for the lenses tested there is little difference. Sure. What you did not test is something like the 80f1.7 on GF. This sort of lens and the mitakon 65 f1.4 allow environmental portraits that would be hard to replicate on 35 and basically a non starter on smaller formats.
At uni anyone using less than 6X7 was teased in the dark rooms, that's the real power. back in the film days there was a real quality gain for large hand prints, today I assume it's the same for large prints. As ever it also says "look at me, my camera is so big and costs so much so pay me more for work" which I assume is a big part of it today.
the title of the video ought to have been "There's nothing magical about DIGITAL medium format depth of field" I have a mamiya RB67, and as Chris pointed out, the old school film camera NEGATIVES really were that much bigger than the 35mm film cameras, especially 6x7 and upwards - so a noticeable difference in LOOK could easy be seen in the pictures taken with a 6x7 v/s one taken with a 35mm film camera. for digital medium format, yes, Chris is on point that there really isn't much of a difference compared to 35mm full frame digital - nonetheless, there is still that much of fine detail, color rendition, beautiful shadows and highlights that medium format STILL delivers "slightly" better than the 35mm full frame cameras. to each his own i guess, I use a 16mp Olympus EM1 now, seems very good for the kind of photography i do - but then there's those who can see the difference in digital medium format camera pictures or are required to use medium format cameras for one reason or the other. good video, as usual !
DPReview is trying to put a title to stir people emotion. My oversimplified take from the video is that "There's Nothing Magical About Different Formats of DOF - it's about equivalency". I'm surprised Chris and the team put out a video and the title like that.
If you shot all camera formats at f2 for example all the Depth of fields would be completely different, idk if this comparison works since you’re shooting all camera f stops at the same equivalence. I think Medium format would have a shallower DoF since it has a larger sensor, f 1.8 fullframe is still a f 1.8 on crop sensor, the only thing that changes is depth of field and focal length. The aperture is still as open no matter what format.
@Hardware Software if you shoot all formats at the same aperture. Your results will look different. A 1.8 fullframe is the same as a 1.8 crop sensor. The thing that changes is depth of field and the crop of the focal length. This test is incorrect because they’re compensating by putting the apertures to the equivalence of “depth of field” Which isn’t an honest test.
@Hardware Software the only thing that needs to be compared is equivalence to focal length and set at the same aperture. If i shoot a 56(85 equiv.)f1.2 on a crop sensor and shot 85 1.2 on a fullframe the depth of field would be vastly different, the fullframe would have better background blur 100% So if you shoot on a fujifilm MF 110mm which is a 87mm equivalent, lets hypothetically say it had an aperture of f1.2 The depth of field would be more immensely shallower then a FF Achieving the MF look ?
@@djstuc exactly, the whole point is show us the difference in DOF for each format. Set at the same aperture, with the same field of view. Im not sure, if this guy gets what we’re trying to say?
@Hardware Software bro lol If fujifilm made a 110mm f1.2 tomorrow Theres no way a fullframe manufacturer can make a lens that is 85mm f0.94 with autofocus. You know how expensive that would be to purchase? And absurd that would sound. lol The larger sensor will have better DOF, better dynamic range, better picture quality because the sensor is larger
@@djstuc honestly have a feeling that the GFX would look more creamy. I watched a ton of comparisons vids on crop sensor vs FF when i was deciding to buy a camera. And you can always tell which camera was which, off of depth of field. The fullframe always had the best DoF So if they compared MF to FF with similar apertures for sure the MF would smoke the FF
You can kinda figure out for yourself how that would work. This video literally proves that the math works out. So by that same principle, you can do the math yourself to get an understanding.
I always assumed that the misconception about medium format DOF was due to people not understanding the changing focal lengths/stops between the various formats. One sure edge though to MF is the details and colors that can be resolved with those massive sensors. Great video!
The misconception isn't the math. It's that you require faster more expensive lenses the smaller the format is, to the point that they don't even exist. So with decent affordable lenses on a FF or Medium, you can accomplish things that are practically impossible in reality on a smaller format. To say that it is a myth because of a math formula, is really missing the point. If you want MORE depth of field, and don't need a fast lense or a handheld shot, no problem. But if you want a fast lense and a narrow depth of field, there IS a large advantage in larger formats. Equivilences be damned. Show me the wide variety of quality affordable sub f/1 lenses in the DX format. and we can discuss. Databyter
@@DatabyterI agree that for shallow depth of field, MF and FF have an advantage over APSC and smaller, but I don't see an advantage of MF over FF.
Conversely though FF has way more investment and engineering behind its products so usually you can't find a MF equivalent shallow DOF like you can in FF
The old real medium format was 6x6 cm (56 x 56 mm) and equipped with a standard lens of 80mm 2.8 it was very hard to get the same look on FF because you would need to use 35mm lens (and crop it to the square) and have aperture of 1.0 that generally does not exist for such a lens.
Also, if you have FF with 35 1.4 and you want to get the same look in apsc you would need 23mm 0.9 that generally also doesn't exist.
The thing is that medium format is now usually 44 x 33 and there are not many fast lenses, so in many cases FF have smaller DoF now. But if you use film 6x6 it's very difficult to replicate using FF.
We’ll said. And the old medium formats were generally still very sharp in the middle of the frame when the aperture was wide open. I think this ads to the “look” medium format is known for.
You are exactly right. This is why they were exclusively talking about the modern digital medium format size and arguing that the "magical look" simply isn't apparent in digital sensors
Agreed. This would have been a much better comparison if it was between a more proper medium format digital 53.4x40mm sensor, albeit modern ones are of course incredibly expensive
Yes!!! My absolute favorite format was 6x7, and boy was that "look" special.
but now FF has 35mm/f0.95 and other f/0.95 lenses, so... no difference
Much comes down to how easy a given focal length can be built to a given big aperture and then what field of view it has in a given format. An 85mm f1.4 is a real sweet spot on full-frame that can be bought cheaply to get a shallow depth of field that an APSC camera will need usually a very expensive lens to reach - a 56 f0.95 or so. Similarly, the Pentax 67 with its standard 105 f2.4 lens. You can make a 105 f2.5 cheaply. But a 50mm f1.1 or so on full-frame is not a cheap lens - at least until recently.
So it's what people typically use that creates much of the impression of the differences. Pretty much anyone with a Pentax 67 will have that 105 f2.4 which will totally isolate a full body shot from the background. But such images are rare on 35mm and basically non-existent on APSC - even though they're achievable with all formats.
What's really strange is the way you reviewers often don't change your language to reflect this - talking about a 56 f1.2 as a 'bokeh beast' or whatever on APSC while not giving the same enthusiasm to an 85mm f1.8 on full frame.
Totally agree and points very well made. I have used all 3 types of sensor extensively and now settled on APS-C Fujifilm and Hasselblad MF [Had GFX cameras, don’t care for them much]. Unless you’re printing huge there really isn’t enough difference to justify almost anyone moving to MF. Where there is a BIG difference is when I’m shooting interiors, which makes up the majority of my work. When shooting a room at 16mm on APS-C or 30mm on MF for the same FOV the difference in scene compression/distortion is huge and very noticeable. Even noticeable over 24mm on FF, and that’s why I shoot MF. If I wasn’t doing high end interiors APS-C has so many advantages and why I still have an extensive set of APS-C kit. 99% of people don’t need anything else, in fact shooting MF is generally an absolute pain - huge files, huge heavy cameras/lenses, mostly need to shoot tripod (yes IBIS is fine, but MF lenses don’t have fast apertures so you’re generally shooting with slower shutter speeds, often beyond the usefulness of IBIS) - not to mention the cost!
That's more a function of the lens optical quality than the sensor size
This is exactly the comment I've been scrolling for. The "magic" of MF isn't the shallow depth of field. It's the FOV, compression, and lack of distortion.
I don't know if this means anything, but when I got my GFX, I was headed out to a real estate shoot. I was itching to test out my new camera, so I got it out and shot some side by side with my FF Nikon. I was immediately impressed how geometrically "correct" the image was in the viewfinder! Granted, I was shooting at 35, not very wide, but still, I saw a very different image than what I usually saw with my other gear. (FF and APS-C). Is that what your talking about as far as less distortion is concerned?
@@alejandroarredondo1997 Yes, exactly that!
@@problemat1que The lens can definitely make a difference, a good lens produces less distortion for sure, but you can’t compare shooting with a 16mm to a 30mm lens (24mm in full-frame terms) and that’s the difference between shooting APS-C and MF to get the same file of view - the compression is totally different, it’s simple physics. Shoot a room with both side by side and it’s very clear.
You totally missed the most important aspect in that discussion: Lens availability. In theory there is always an equivalent focal length and aperture. In practice however, only for mentioning one of many examples, you have no equivalent lens you can chose to use with either APS-C or MF to get the same look as an e.g. FF 85 1.2 (probably even f/1.4) .
And even then, doesn't the actual focal plane get smaller? Like even matching a FF 50mm 2.8 with a 35mm APS-C at 1.4 would look different because of the depth of the focal-plane itself. The bokeh/blur would be similar... but I'm pretty sure the FF would have a persons whole head in focus while the APS-C would only have the nose/eyes in focus leading to quite different look. People always give technical reasons why there is no "Large-format look" but there certainly is... it's easy to tell a movie was shot in medium format (65mm film-back). It looks WAY different than traditional s35 (APS-C).
I think the Magic is, you’ve got a 35mm look with the perspective of a 50mm lens. (Background Compression)
for me personally, it's about the look i am getting at a wider focal length with shallow DOF.
APS-C to FF, because of the CROP (or the lack of), I can stand closer to the subject (i shoot mainly portrait) and get a shallower DOF because of camera-subject-background distance than I can with an APS-C system.
The same is true Medium/Large format.
So, like most of you guys said, it's not the technical aspect of the camera systems we are aftering, it's the perceived look that we can achieve with one system compared to the other.
I generally prefer more depth of field than less and I like that I get it with a faster aperature on my m4/3. Of course with full frame you can raise the iso with minimal noise penalty so it really comes down to know your partner and learn to dance with her. Oh and if you want really shallow depth of field with apsc or m 4/3's then your partners going to be expensive. (Fast glass)
Sigma 30mm 1.4 for fuji is under $300 :)
It's also about the background magnification that comes with using those longer focal lengths. It will appear more out of focus because it's magnified (apparently often confused with having a shallower dof), but to Chris' point I'm not sure if it makes much of a difference on 44x33 vs. like 6x8 where a 300mm lens has the fov of a ~135mm on 35mm full frame.
It’s cool to show equivalence in different formats. A better comparison would be to use the best lens combination and push each format to its limits.
More articles like this please! Let's have Live models, good/bad lighting, more analysis of results, blind tests, and throw micro 4/3 into the mix too :)
Its all about the fastest lenses available for a system, not the size of the sensor (only).
The lens speed arms race is pretty much pointless. There are so many other factors that apply in actual real worl applications. The range of uses for f0.9 is pretty slim.
@@benjamindover4337 never underestimate the power of young moms doing photography as a sidehustle shooting a whole wedding wide open.
@@sulev111 hilariously saw this play out at a friend's wedding, two photographers showed up with Canon SLRs with one 50mm F1.2 each (and no other lenses). Kept shooting with frowny faces without flash way into the night under those old-timey dim light bulbs and delivered grainy, out of focus results, converted to Black and White to hide the Canon color noise. I would have been so mad if I was the client.
@@problemat1que lol. I would love to see these pictures for a good laugh.
I am telling this for more than a hundred times.
Given the same aperture, it's just the magnification of the image that plays a part here.
so sad that I just brought a second hand gfx50r
but it's colorperformance may look good compare with sony and nikon
still happy to have one
How fast can Chris say a whole bunch of numbers? I feel as if I'm being sold a used car.
The real depth-of-field myth that needs to be busted is the notion that less DOF is always better.
The myth that needs to be busted is the notion that there is right and wrong. Different things are used for different artistic expressions. There is no serious photographer who seriously thinks "less DOF is always better", so you're arguing a strawman. People use short DOF when it's the feeling they want, and their viewers like. Not because of some unwritten principle.
@@nikoolix There's definitely a lot of people out there aruging that shallower DOF is better. Not good or serious photographers, but a lot of people.
I couldn't agree with you more.
@@onegrapefruitlover I mean yes there definitely are people in all areas of arts that do a specific things purely for the principle. But it's not really a "myth" that needs to be argued against IMO. Those photographers will hopefully find different ways to make interesting photos eventually.
I guess this video is for people that are young enough to still be sharp...
Thank you! It finally does come down to size, weight, cost and personal preference. To each his own. Let's all just go out and shoot. :)
Dont agree, the test is flawed and here is why ua-cam.com/video/fHCtop_yfvY/v-deo.html
What your saying Chris makes a lot of sense. I'm a Fujifilm shooter and for me to achieve a Full frame DOF equivalent to 1.2f (which is readily available and can be found vintage) in crop-c that would be a 0.8f which is not only difficult for fuji to create in AF but gonna be bloody expensive. Bottom line, with the technology we have today the best and most affordable way to get the highest DOF is to use a full frame camera because there are so many fast FF lenses in the market what can beat the DOF produced by a 1.0f FF lens?
I always thought of medium format having better light falloff, not focus. The dynamic range of the shadows to light is so much more gradual in medium format. I actually was teaching someone about this recently and they were able to, only knowing this, correctly identify blindly, full frame shots(from a Sony) vs Fuji medium format shots. because sharpness comes from the lens, colour comes from the sensor, and depth of field comes from the aperture and distance from the subject to the background. But the light gradient comes from the sensor size.(ofc this is being a tad reductive but you get the idea) it's about the only consistent thing I've noticed whenever people do sensor size comparisons.
If someone could tell the formats apart without checking details it must have been due to the different processing of the files, be that in the raw converter profiles, or manufacturer jpeg engine.
Just curious, is there a way to copy or get close to that medium format gradation in full frame or APS-C cameras? Adobe profiles in LR are relatively contrasty compared to a linear profile that you get with Capture One. There are free 3rd party linear profiles for LR (I have a site for that) and it allows the image to look flatter and smoother in terms of gradation of tones. One reason I ask is because that smooth gradation of tones might not make a large difference when an image is on Instagram (both because of compression and size) so I wonder how much of a difference it REALLY makes in the end.
@@Bayonet1809 probably not, it was through a computer screen and the video was prints and insta posts.
@@vernonsza there is, you could do a stitch of several different photos taken from the same scene. you could do a pano stitch, or multi-bracketing. basically all you are doing is capturing multiple shots at various exposures, since the smaller sensors have less dynamic range than the medium format. there are actually many videos about getting the "medium/large format look" on smaller sensor cameras. and all of them revolve around this principle.
in the end I think the difference is only significant if you are looking for it specifically. and that I think is the greater point that Chris wanted to make in this video.
Agreed. But I don't think these 'small' medium format cameras like the gfx have much (if any) of an advantage here. I'm sure one can pick the difference when you know what to look for, but there is not really a discernable look like one sees on the larger sensors.
I follow the architectural photography world a bit and one can always immediately recognise work shot on these expensive systems. I also read an interview with someone (arch photog) who switched from Canon to the Fuji gfx for this purpose (adapting TS-E lenses) but switched back after some months because there was no perceivable benefits evident in the light fall-off which a lot of folks are chasing.
As a M4/3 shooter, my only problem is mostly not the depth of field, but the low light performance, and there is no alternative for something like a 50mm f1.8, or a 24mm f2.0 so in that case, much cheaper to invest into a full frame camera, insted of a fast m4/3 lense
Here is a simple math trick: Divide the focal length by the f-number. What you get is the size of the entrance pupil, the thing that gathers the light. You will see that it's the same for all compared lenses.
The rest is parallax. The thing with your thumb jumping left and right if you open/close your eyes.
Actually a very good point/quick way to illustrate to someone what's "consistent" about lenses of different focal lengths and apertures. Thanks!
I’m so glad you made this video! I’m so sick of hearing people saying “Love that Medium Format Look”. There is no Medium Format look. A 50mm f1.4 is the exact same regardless of which sensor size it’s used on. The only difference is that one can stand closer to the subject with larger sensors vs smaller sensors which allows the larger sensor to get shallower depth of field with the same field of view. But Fuji glass is too slow. Their f4 zooms are only f3.2 equivalent. They can’t even replicate the shallow depth of field as a FF f2.8. Not to mention, they are more expensive and have substantially less zoom range. Even Fuji’s primes are too slow with most being a f2 (f1.6 equiv) and one at f1.7 (1.4 equiv). Sony and Canon are already releasing f1.2 primes. I do believe the medium format sensor does deliver better ISO performance and more color depth, but one is going get a better “Medium Format Look” from a Full Frame system now days over Fuji’s offering.
Great video Chris. I think APS-c comes off very well in that comparison. I was thinking about getting a full frame camera but now I'll stay withy my Fuji X-T30. I wonder how micro 4/3rds would have compared along side of those larger sensors.
FF users about APS-C: ha-ha noisy crop
FF users about Medium Format: size doesn’t matter!!!
Truth! ..lol
As an APS-C shooter, I would like to thank you for snubbing micro four thirds in this video...
Other than what other people pointed I am also interested in the compression a longer lens will give you. I love thé 85mm field of view but i want to get more background compression. The only way to do it is to go on larger sensor and use a longer lens. Same applies for crop sensors. 56mm can give you 85mm but not in the compression of an 85mm lens.
Compression is only effected by spacial relationships between camera, subject and background. You can not take a photo that is the same distance from your subject with the same field of view and have any more or less compression. A 50mm on full frame that is 20 feet away from the subject will have the exact same compression as a 25mm on m43 or any other equivalent in any other format.
A very comprehensive "One time for all" segment about the sensor/lens conversion and final look. It will spare some photographers a lot of time. Thank you !
Great episode! How about color accuracy and the gradation from light to dark areas?
I am considering what my next camera should be. I’m choosing between a Fuji GFX 50s II, a Canon or Sony Full frame camera, and the new Fuji X series camera coming out in 2022. I currently have the XT3 and I love it. But, I’m tempted and just want to make an educated and thoroughly thought out decision.
Despite knowing that FF is the sweet spot technically, I currently use Fuji because I enjoy their colour rendition and user experience. If I were you I'd wait to see what Fuji comes out with, see how X-T5 performs.
Stay with Fuji man, their new 1.4 lenses is all you will ever need. The new 33 1.4 is just a hell of a lens. Plus colors in jpeg on Fuji are super nice. Plus plus look of the cameras. There's no match at Sony with X-Pro 3 or X100V
I've been wanting to switch to a camera with 16-bit color. (GFX 100, any Hasselblad)...
A few months ago, I switched to DxO Photolab 5 Elite and it rendered my X-T{2, 3, 4) photos better than even Capture One. This made me think I don't need a new camera any time soon, as Photolab is bringing life to old photos.
Also, I am looking forward to seeing what those new X Series sensors can do for me.
@@calokid nice to hear that software can you give me some examples of fujifilm shots of your in lowlight especially the new bodies before when I used the fujifilm camera especially the fujifilm X-T2 before I saw noise grain at 3200 and my limit to that camera is 3200 ISO and if I need to take a shot even with noise is 6400 ISO. But right now I don't know the noise performance of fujifilm x100V, XT-4 and newer models and I want also the result you edit from the software you use because I'm Capture One user from the time I use Fujifilm thanks in advance 😁📸👍
@@junichinomura4810 The low-light advantage of some of the newer X-Series cameras really has to do with IBIS. I've shot with the X-T2 and loved the sensor. Every X-Series camera I've used starts to get noticeable noise around 3200, so no real change. On an X-T4, however, I can shoot handheld as low as 1/15th of a second, others, with a steadier hand can shoot as low as 1/8th of a second. The advantage of IBIS photography would only apply to photography in which tripods do not apply.
You could try Photolab yourself, and we could also decide on experiment criteria for exploring how low light looks on X-T4 or we could process one of your RAW files in different software to see the differences.
Thanks so much Chris!!! Most photographers out there are so confused. All we gain on medium format is larger prints. Thanks again!
i think this is where the misconception comes from. when talking about equivalency, most people are aware that 35 mm in APS-C will have the same look as around 52 mm in FF, but they don't realize that the aperture would need to be adjusted as well. and they just end up comparing 52 mm in f/2 in FF having more bokeh vs 35mm in f/2 in APS-C (while in fact based on the exact equivalency, they should have done 52 mm in f/2.8 -- closest number to f/3 since there is no f/3). the media has not helped either by always mentioning the equal zoom for FF in sentences like, "16 mm lens, or equal to 24 mm in FF", but then not mentioning the equivalent aperture.
and another reason for the misconception is just based on lens availability. say i have a 50 mm f/1.2 lens in FF and i love that look. to get that same look in APS-C, i would need to get a 35 mm f/0.8 and well, good luck finding a lens like that. because of that, people then start saying that FF is better than APS-C since frankly, it can produce images that APS-C just can't. (or maybe it can if that lens is available, but im sure it would be very expensive).
Very fine explanation of the technical part of this mith. Just two (big) points let out of the discussion. The bigger the format, the bigger the lenses the lesser the aberrations, distortions, and much sharper because the properties and qualities of the glass become less relevant than on small format lenses.
Also the bokeh - even if you correctly explained that the d.o.f would be (roughly) the same - would be much creamier on bigger formats while the in focus part will be sharper and with less aberrations well known on very fast lenses.
Last thing: you correctly used (digital) medium format (33x44mm) in the test, which is by the way a "crop" format and quite small compared to much bigger film sizes like 6x7 (56x68mm).
The 100mp Fuji cameras, have the same pixel density as the A7R IV and the 26mp Fuji cameras. Meaning the precision of manufacturing for the optics has to be about the same to have the same level of aberrations and such on a per pixel level.
Theoretically it doesn't matter, as the image exists in a point, not an area. But practically there is a difference in how longer focal length go out of focus "faster".
The really difference will be when you stop down and want deep depth of field. Because diffraction limit is more severe on smaller focal lengths, apertures and sensors. Althaus why people do f/64 on large format and get Soo much better results than f/18 on FF.
"Medium format" as in 6x7 or 6x9 has a negative that's close to the interpupillary distance. So it looks and feels closer to your human perception.
And practically. The fastest lens you can get on MFT is the Voigtländer Super Nokton 29mm f/0.8 - you can't get faster and wider even with focal reducers. Therefore going with a bigger sensor gives you some results you can not reach otherwise.
The interesting thing comes in that to get equivalent depth of field we end up with different apertures, so you can shoot medium format in daytime at a desired depth of field and be less likely to blow out highlights as the aperture will be more closed down to achieve it than apsc. Also, at night, we can gather more light on apsc without being restricted by razor thin depth of field that we would have on a larger format at the same aperture.
At night, you would just shoot a full frame camera with a narrower aperture and higher ISO if you don't want so thin depth of field. The better sensor performance would mean it's no sacrifice. But you do have the *option* of a faster aperture equivalent, letting you get better low-light noise performance or faster shutter speeds.
There's no way to spin APS-C as an advantage in low light.
@@AyeBeAPirate Yeah, fair enough, my comment was likely more acedemic than practical!
Depth of field is only really influenced by focus distance and aperture diameter in mm. Exposure is influenced by the focal ratio, otherwise known as F-stops. Interestingly, focal length directly impacts exposure, requiring a larger aperture to maintain the same F-stop.
In the case of larger formats, longer focal lengths are used to achieve the same field of view as smaller formats. These longer focal lengths reduces exposure, so to maintain exposure, you use a larger diameter aperture. This brings exposure back to parity, and reduce depth of field.
I’d probably consider both APS-C and FF to be kind of the sweet spots for photographers. Sensor advancements allow shooting at darker F-stops at night, while the availability of very fast lenses for both systems allows for extensive flexibility.
For me medium format is less about DOF and more about tonal gradations between light and shadows. Maybe that’s a myth too, or perhaps the MF photographers whose work I admire are just really good at lighting.
I guess if you can afford (and want to buy) a medium format camera, you're more likely to be good at photography and have more experience in that craft.
I think that was only true for film medium format.
Look at the sensor size- there's just not much difference. If you need more dynamic range there is always exposure blending for fixed subjects.
Neither are myths.
It’s very true. This “test” was designed to achieve a desired result. Yes, I can make an xt4 image “look like” a GFX image, but the quality of that image, resolution, ability to edit, shadow and highlight depth, will be significant different.
It’s all the same on Instagram. Much different when you’re printing a large bridal portrait.
There is indeed a difference if you take for example a X1D and a full frame lens like a 50mm 0.95 with large enough coverage for those medium-ish format digital sensors.
Hi Chris. You are talking about background blur which may vary but depth of field remains equal at any given aperture regardless of sensor size. I love this channel and have been following you guys since camera store
Thanks. The only reason I would buy a medium format camera is image quality for huge enlargements. Which I don't do.
One thing I struggle with on Micro Four Thirds is ISO, as you would expect with a smaller sensor. My f2.8 lenses are more like f5.6 on full frame in terms of depth of field. However, since sensor technology is constantly changing and improving, I wonder how much further you can push the ISO on, say, a Sony a7 IV vs. my OM-D E-M1 Mark II and get similar noise levels. I usually think of two stops as a rule of thumb, but is that really the case?
On my camera, I consider anything above ISO 6400 for emergencies only, and I wouldn't want to do much cropping above 1600. Unfortunately, I find myself reaching 3200 and 6400 more than I would like to when I'm shooting wildlife due to the fast shutter speeds.
I don't use M43 but use(d) FF, APSC, and 1". Crop factor works as you'd expect, and FF will look like the ISO is 1/4 of M43. That said, M43 has really, really good stabilization, so the difference will mostly appear with moving subjects.
My issue with M43 (I sold all my cameras and lenses (Lumix G9, etc.) was ISO, DR, and color noise, but mostly pixel density. The size of the G9 M43 sensor and the density of its pixels gives M43 a more digital look (less organic). This is also seen in some newer APC-S and FF cameras. Even though some look for pixel perfection, for me it was almost resembling iPhone quality. Raws were like artificially sharpened and I could easily tell the size of sensor by the look of the images.
@@borispradel1037 but can u imagine 10 years down the road when gh9 comes what can happen? Iso performance will be atleast twice or thrice better. And computational photography technology is expected to come to m43 before apsc due to smaller size.
Dang, that’s really sad. I will happily shoot at 12,800 for professional work, this is one reason I only use full frame bodies.
@@josh_boak yup 12800 ISO is the usable for lowlight scenario in emergency use also for colored photographs of any fullframe camera bodies right now 👍😁📸
That was superb! Excellent video. Thank you!
Interesting video, Chris however, though I know your "focus" was on depth of field equivalence, I think you glossed over the major advantage of medium format when compared to the smaller formats...the fact that when you print very large the MF image requires less enlargement hence more sharpness and detail are retained in the final print. And just FYI, I used to develop large format b/w sheet film, and a contact print from an 8x10 negative is the sharpest, most detailed image one can get, since there is Zero enlargement. Thanks.
Most important factor in depth of field is final image size relative to viewing distance. Look at the same print from further away and the depth of field will be greater. Hence popularity of incredibly shallow depth of field when images are only viewed really small on Instagram. Same depth of field does not work for image to be viewed on phone and one to be viewed as 20" print (unless you view it from far away). So strangly this justifies fast, heavy, expensive glass just to post on Instagram.
Having had fast lenses in the past I looked through my lightroom catalogue and discovered I more often ruined shots wide open than not so now shoot with an f4 zoom in FF and it is great. I guess on crop I'd need an f2.8. Generally fast glass is a fetish as it is no longer needed to compensate for slow films.
Dont agree, the test is flawed and here is why ua-cam.com/video/fHCtop_yfvY/v-deo.html
@@de-graftasafo-adjei4646 Edit of what I posted on that video: DPReview definitely triggered you didn't they! Depth of field is actually ZERO for all lenses and can only defined in practical terms by what an acceptable size for the circles of confusion on the sensor/film plane is (en.wikipedia.org/wiki/Circle_of_confusion) and that depends on the angle subtended at the eye of the *final image*. Literally all you have to do is look at a image with shallow depth of field on your monitor at a normal viewing distance and then look at it from the other side of the room and you will see that the depth of field appears to increase. Take a portrait where the ears are slightly out of focus as an example. From the other side of the room they will look in focus. In that case there is nothing to do with the camera at all :)
I shoot 8x10 (and whole plate) and contact print them and they look great but I know if I were to enlarge them the depth of field would be way too shallow - unless the prints were viewed from far away (Father Ted reference). So does the 8x10 have shallower depth of field than my APS-C with an f/2 lens on it? Well it all depends on the print size! (also subject distance which is a biggie).
It's so funny seeing people get wound up about this. Naughty boys at DPreview provoking everyone. I'm just stoking the flames.
In my experience, going from canon 5d III(fullframe) to fuji xt3(aps-c) I just do not have the same tools at hand regarding field of view and depth of field. My concern is mostly getting a wide angle with great separation. That is not as simple(if impossible) with my fuji as it was with canon. For example saying that fuji's 16-55mm f2.8 is same as canon's 24-70mm f2.8 is just rubbish. 16mm just does not produce the same separation as 24mm. + 16mm gets a lot bendier(barreling).
My experience as well
The analogy I like to use is gears on a bicycle. You certainly can pedal up a steep incline in a high gear, but the amount of work required to do that would be more than most people would like. It’s a rough analogy, granted. So, while you certainly can match looks amongst these different sensor sizes, there’s clearly a “comfort zone” for each format. We saw this in the early days of prosumer digital cinema cameras when customers were seeking out ultrawide lenses for their first generation blackmagic cameras. Not disagreeing at all with the comparisons presented here… but, as a thought experiment, imagine if micro-4/3 had been included in this roundup.
now all the pros are riding 34x32 in the alps haha
Dont agree, the test is flawed and here is why ua-cam.com/video/fHCtop_yfvY/v-deo.html
Thank you for sorting this out. It's often difficult to explain - from now on I can just point to your video ... 🙂
Dont agree, the test is flawed and here is why ua-cam.com/video/fHCtop_yfvY/v-deo.html
the real difference can be found in colour rendition. medium format sensors capture a 16bit image that reduces banding resulting in more lifelike colour/contrast transitions.
The logic of the test is strange in my opinion: "let we adjust the aperture for the medium format camera so it has the same depth of field as an APS-C camera. Now you see, it has the same depth of field!" ))
Basically the medium format makes it easier to make shallower depth of field lenses. GF 110mm f/2.0 is equivalent in terms of depth of field to a 55mm f/1.0 on an APS-C camera. And the latter lens will be definitely inferior in terms of quality wide open. And GF 80mm f/1.7 simply doesn't have any equivalent in the XF world. Let alone zoom lenses.
And this difference is even more pronounced if you compare larger film formats to a "full frame". Basic f/3.5 lens for 6x9 will produce the same depth of field as f/0.75 on an APS-C camera.
Whenever I hear arguments like this video, the diminishing returns you suffer trying to match the “medium format look” with anything but is never addressed, especially in the comment section. This is the first I’ve seen addressing just that.
Dont agree, the test is flawed and here is why ua-cam.com/video/fHCtop_yfvY/v-deo.html
@@kapapa3764 yes we do. And in fairness Fuji GFX (and any other relatively affordable "medium format") is not so much of a "medium format" if compared to FF.
Actually - there is a difference (although, most people will not care too much - but for my shooting, it is important to get a certain look). It's about the depth of field you can get at a specific distance with a specific focal length. That's what can make medium format images look "special" or maybe the real word is "different". Everything else is correct for the general case, it's just that there are situations where it is not possible to frame a subject at a specific distance and get the same fall off. For example, it can be very difficult to get a wide angle full bodied portrait with shallow depth of field on an APS-C camera, without either long zooms or exotic lenses - but even then, it's not the same look because you don't get the same distortion characteristics.
The thing is - a 24mm f1.4 IS a 24m f1.4 no matter what the sensor size. It doesn't change. But obviously the framing will be different at the same distance. So that's where the FOV and DOF equivalency comes in - you need to change the distance to the subject to get the same framing, and therefore DOF necessarily looks like it is different when on different sensor sizes.
The laws of physics have not been repealed with the advent of digital photography.
But the equivalency is the key.
I can very well have 1.2f on 35mm for full frame and I won't be able to buy equivalent lens (0.75f) for ASP-C, hence bigger sensor => shallower DOF possible.
First of all, another awesome video, even though I am like 1 year late to it...BUt what you are confirming is hat in order to maintain the same DoF in smaller sensors you need larger apertures! So I would argue that definitely it is easier to get shallower DoF with the MF than with the FF and with the FF it's easier than with the ASPC...Might be just semantics, but using the same lens with the three sensors, while the angle of view or apparent focal distance changes, aperture does not. F2 is F2 in any sensor. Now, the larger the sensor the most difficult/expensive to build large apperture lenses. An APS-C F1.6 lens is cheaper than an F3.5 MF lens. And yet there would be no gain in terms of DoF. In orther to have a gain you would need a F2.8 or F2...So, looking at this way, yes, nothing magical or special about MF.
Is shallow depth of field still a big trend in photography ?
Yes thank you this is why micro four thirds with 25mm f/1.4 works for me! Gives the same photos as full frame 50mm f/2. Unless you need lots of megapickles, which I do not. I gotta say, the reason MFT died is sensor development stalled and why weren't they putting f/1.2 apertures on all their lenses from day one? Even the Voigtlander lenses with f/0.95 are not too narrow depth of field, they're totally usable wide open on MFT.
But you need megapickles so you can crop 2x when using a shorter but much heavier telephoto lens!
The 6 year old sensor on the EM1m2 already outperfomed, and still does, the best FF body by more than 1 stop of DR shooting at quivalent ISO settings from ISO200 and above. This M43 sensor punches way above its weight.
No, 50 mm f/2.8. 25 / 1.4 = 50 / 2.8
I feel like this video kinda proved the opposite. It's a situation where a standard medium format lens gets you this baseline depth of field but meanwhile on the APSC you have to blow a bunch of money on their very high end super wide aperture lenses to try and replicate the DoF benefits of a larger sensor. I'm not a "gear matters" kind of person with gear acquisition syndrome or anything but when you have to make the other cameras work a Lil harder to replicate something that the other one does naturally, the natural one seems the winner?
Yeah but what about the cost of the medium format bodies? Doesn't really seem like you are saving a lot of money by going with that option, if any at all.
@@alexbustamante6532 yeah i thought about that too, but that also depends on the body you're getting. Fuji GFX 50R retails 4.5k USD, almost always on sale for 3.5 or even 3k. My A7RIV retails 3.5 as well, so it's effectively the same cost. Only time cost really becomes a factor i feel is if you're genuinely debating an APSC vs Full Frame vs Medium Format, but depending on the body, your price range between FF and MF is negligible
The 56mm is $1000, the Sony $1600, the 110mm is $2200. In APS-C you can go up to 50 F1 at $1500 (not coincidentally similar to the Sony) or down to 50 F2 at $450, making it a very flexible system for both size and price.
@@problemat1que yeah if we're using the price argument obviously APSC is far superior at being cost effective but the video isn't as much focused on the price end, moreso focused on the final image. Obviously price matters, it's just a secondary subject here
A very clear and refreshing explanation. Thnx.
so its just in my head ? amazing work guys !!!
This is only applicable in the context of shooting on native lenses to each system. The beauty of mirrorless is having the ability to adapt. 44x33 is actually a great compromise because in many cases you can still get full (or close to full) sensor coverage with FF lenses. Pairing digital medium format with those larger apertures gives an insane amount of compression. The biggest advantage is compression at wider focal lengths. This also doesn’t take into account natural dynamic range and detail retention in post processing.
compression comes from camera distance, not the lens. the lens limits fov, as does the sensor, so moving to medium format gives you more fov within the same focal length, but doesn't do anything to perspective.
This video literally disproves the "compression" thing, how did you miss that?
Is there a difference, yes, super noticeable, nope, replicatable, not really. lemme explain:
* Lets assume parallel rays come in and exit the rear lens at an angle to meet in the center of the sensor... (most diagrams are drawn like this)
* Now, notice the rays form a cone if you will, like this: >
* Whats important is that flange distance sometimes in medium format, of the distance from the back lens to the folcal plane is longer, then the cone angle is smaller.
* Ok, when something goes out of focus with a longer cone (focus in front or behind the focal plane) there is now a little circle on the sensor.
* the size of this circle also changes with the angle of that cone, hence the flange distance.
* if you have a long flange distance than something far out of focus doesnt seam to be that out of focus..
* when the flange distance is short, that cone angle is big and things can jump out of focus faster and even more out of focus the further things are.
This was a tough one for me when I went to medium format, I couldnt get super out of focus fall off, the out of focus was so gradual I kept going back to my full frame for some shots. Dont misunderstand me, the medium format had more out of focus, and thinner focus plane where I was shooting beauty at f/16-f/22.
Overall what I saw the major difference was thinner focal plane and smoother transitions in the focal range
best comparative video ever
I feel like this video misses the point. No one is concerned about how to make every image look the same across these cameras. People want to know how medium format CAN look different.
Take a 50mm f1.4 lens and try it on every one of these cameras with the same subject and compare them. Do the same with a 135mm lens, and with an ultra wide lens and compare the differences.
There is a difference in how each sensor size affects the image, but you will never see that if you are only focused on how to make them look identical.
Also it is important to show the larger sensor medium format cameras, regardless of how expensive they are, since we are specifically looking at how sensor size affects the images produced.
Great video!!!
A true comparison would be to put a Hasselblad V mount lens on all 3 cameras and take a shot at same aperture, same focal length on all 3. Then tell me there is no difference. Nobody shoots medium format to emulate APS-C! Depth of field is just scratching the surface of what a large sensor is all about. Larger sensor = better low light at any given resolution vs same resolution on smaller sensor. There's a reason 100 megapixel has come to a GFX camera before a Sony A7R. Larger sensor = wider compatibility with more lenses. You can crop to GFX to FF and use adapters to vintage glass designed for SLRs or cinema cameras, as well as using medium format lenses from Fuji, Hasselblad, and so on. To see the full potential of a medium format lens on FF you need a focal reducer and that in itself has compromises, won't work with mirrorless medium format lenses. To say the video is a gross oversimplification of photography and why people choose medium format or larger sensors than APS-C is an understatement. Why try to reduce everything down to the most simple spec imaginable and use different apertures, different focal lengths to achieve that? What does it achieve? Nothing!
Excellent examples. And hey, you guys finally got some sunshine up there :)
and that will be very idiotic, people buy full frame/medium format for larger pixels, better low light, tonal gradations, bla, bla, bla, not for your mathematical deep of field calculations, what have no relevance whatsoever...what next you will put a phone sensor and demonstrate how good it is ? Damn that was so shallow...............
So basically, focal length and aperture stays the same on all format. the only thing that changes is the size of image circle and sensor size. is that right ?
Depth of field is determined by the f stops on your lenses, not the format of the camera. The lower the f stop on your camera, the lower the depth of field. The higher the f stop, the higher the depth of field.
When we say “DOF” what we really mean is background blur. “DOF” is about what is in focus and background blur is about what is out of focus. How much blur an out of focus area will have is dependent on the size of the aperture opening. On a 110mm lens at f/2 the aperture opening is 50mm where as on the 56 1.2 the aperture opening is 46mm. That’s almost a 10% difference. This is why when comparing size of the bokeh balls on the indoor comparison the MF took the cake over the APS-C. The further away the background is from the subject the more this will become apparent because the lens with the lager aperture opening will have a larger maximum blur spot potential.
When comparing to the full frame tricky part is, yes, the MF isn’t that much larger than FF so the advantage isn’t all that noticeable
Larger apertures often have more aberrations visible, which can affect the perception of quality. The greater resolution also allows you to better discern what is and isn't in clear focus, giving the illusion of "more depth of field". There are some differences at larger scales but not because of the format but rather the lenses. Matching the DoF of say a f/5.6 lens on 8x10 would require a full frame f/0.7 lens, and that doesn't really exist.
Seems strange trying to match the images by stopping down on the MF. If you shoot wide open on the fastest lenses the difference is immediate.
Something I noticed pretty straight away when I started with a Nikon D3300 several years back, some companies are much better at supporting the APSC format than others. Fuji ofcourse knocks it out of the park, while I think Nikon and Canon string you along with some decent APSC lenses here and there, hoping you will 'upgrade' to full frame. If you do your research and want to commit to a smaller/lighter kit, it's very much worthwhile researching what APSC-specific lenses said company offers first. Thankfully many third parties fill the APSC-starved gaps of Nikon and Canon on the DSLR side. Things are rolling out a little more slowly with the RF and Z mount. Here's hoping we see more quality 3rd party autofocus lenses make their way to those mounts in the coming years.
Ultimately is about lens and the intent of your photography. Sensor size doesn't truly matter.
To me the sensor size of medium format is at least 645.
Thats what he mentioned. But I‘m with you, 44x33mm isn’t medium format for me. It’s some kind of KB-H, like Canons APS-H sensors back in the days.
Medium format is defined as being larger than Full Frame 35mm, and smaller than a 4x5, so 33x44mm sensors fit within that definition.
@@Bayonet1809 No, medium format are fx the 120 film and the smallest image size was 4.5x6 !
@@cameraprepper7938 Film roll size is not relevant to the definition.
@@Bayonet1809 It‘s defined beeing between 4,5cm x 6cm and up to 6cm x 9cm and thats far away of 3,3cm x 4,4cm. This is a non existing film format, so it can’t be Medium Format what is defined by film.
Hey DP, I know you’ve already covered the iPhone 13 as a standalone video, but what I think is lacking is a iPhone vs entry level camera review. There’s many iPhone vs $5000 dslr videos but none are comparing the same price points and would be very interesting to see how they would compare, either be a old dslr or a new dslr with a decent lens?
I will only go back to medium format when I can get 6x6 or 6x7 sensor, and of course not before the price is acceptable.
There's also nothing magical about full frame depth of field compared to apsc but everyone acts like full frame is the bare minimum in photography.
I think the term “magical” is overused. There’s a bigger difference between APSC and FF than FF and Medium Format. You don’t need FF but it’s definitely better to me. But to each their own.
@@JET-Photo Copium
@@jamesburns679 There's nothing magic about it if you compare them with the same >equivalent< apertures, sure. But FF has more lenses with larger equivalent aperture lenses than APS-C, and that's where the 'magic' happens. There are tons of 1.4 lenses for FF, but an APS-C lens would need to be f0.95 to get the same DOF, with everything else equivalent. And the ironic thing then is that the APS-C 0.95 lens will most likely be more expensive than the FF 1.4 lens. Look for example at how expensive the Fujifilm 56mm 1.2 is, compared to equivalent FF 85mm 1.8 lenses. No APS-C system has anything equivalent to an FF 85 1.4.
So, no-one claims that the FF magic is in the sensor itself. Rather, it's in the lens selection. FF really is the sweet spot between sensor size and lens selection with large equivalent apertures. And I'm saying that as a current Fuji user. I love their user experience and colour rendition, but there are times where I really miss my Sigma 35 1.4 ART, since Fuji doesn't have anything properly equivalent.
@@nikoolix Copium
Thank you for trying to educate people on this topic!
There's another "magical" thing: APS-C/MFT lenses that "have bright f/1.2-2.8 aperture" are magically lighter compared to FF lenses with such apertures. Well, they are just darker. I don't mean having darker lenses is bad, you just shouldn't pretend that physics and geometry don't exist :)
Also, folks say oh medium format has better low light sensitivity, but you have to take into account the fact that you can shoot at f/1.8 on APS-C easily, and that aperture would be ridiculous and unusable on medium format, so you can compensate for low light with the lens rather than increasing the ISO.
So many people and so many videos make this too difficult. The only factor affecting your image is magnification, that’s it.
The only critical physics benefit of larger formats (technically) is angular resolution when tilting or swinging lens or film planes.
Ok, well I want the look of a FF 35mm lens 1.4 on APS-C... care to explain how to achieve this?
@@Frontigenics
a "look" is a very subjective thing, but if you mean equal depth of field, It's not possible due to the magnification difference you'll experience using an APS-C sized sensor. You can't have the same field of view and same depth of field because the magnification will be different, and therefore either one of them will have to change.
Yes… but the compression is different… which is what people confuse with D.O.P.
MF looks different in this way… but not for the reasons people think
Another example of marketing, as you basically say at the start. Most digital MF isn't anything like the size of MF, But the perception to the consumer, or rather to the enthusiast, is that MF represents a huge difference in size/quality like it did in the film days.
This is not medium format)))) not even 6x4,5)))
Fuji is a crop of medium format))))
But all the magic of medium format is not a dop! All magic is about perspective!!!
Perspective of 80mm but view of 50mm!! Or even 45mm!! This is a true magic!
I think MF would be good for large prints, where detail is part of the presentation. APS-C and M-43 have the advantage of having a large DOF, which comes in handy for some applications. When the smaller sensors can stack images with sensor shift, you can get some great results on static subjects that need the larger DOF advantage. My desire for MF is for the ability to photograph moving subjects with the ability to capture great detail for large prints (which is entirely a stylistic choice, not a technical one). Currently I have a Canon RP, and the advantage from moving to a larger sensor from M-43 is the reduction of noise for indoor shots, and 6 more megapixels. Plus, I love the JPG colors canon makes. I tend to work in more from a purest perspective, "cropping inside the camera" and making color adjustments "in the camera". I appreciate it's a slower process, but again that's a stylistic choice. There's a camera out there for everyones style and technical workflow, and they are all valid! Heck, even using a phone camera is valid!
I use the mantra "you can make a technically perfect photograph, but if it's a boring image, it's just a boring image". Have skill in your technique, but remember that content is king.
These types of videos are extremely helpful, because they tell us photographers the abilities and limitations of each camera. It helps us decide if the camera is going to aid our style or hinder it. Plus, the presenters are just entertaining to watch! Thanks DP Review, Jordan, and Chris!!
Side note, I think it would be entirely reasonable to evaluate a lens using "angle of view" and not the millimeters of the physical design (because there is variation on angle of view between lenses of the same millimeter). But, just like I'll never learn the metric system, I'm sure it would be like pulling teeth to try and get people to start using AOV.
Good lecture! Lots of people indeed don't konw this effect.
I would be more interested in what's best for large depth of field (landscape photography) with cosideration of lens diffraction :)
This was great. What about DR? Anything significantly different because of sensor size difference?
I recently purchased a 45mm Pentax 6x7 lens and attached it to my Pentax 645z. Was told the 35mm equivalent of that lens would be 23mm but it looked exactly like a 45mm 645 lens I already had for my Pentax 645z. Finally I decided to place that 45mm 6x7 lens on my Pentax K-1 full frame and aside for the different frame sizes (pentax 645 has a 4x5 frame) it looked closer to a 55mm equivalent than a 23mm lens I had that was native to 35mm full frame.
In other words I saw almost zero difference in terms of what was included in the frame when mounting the 45mm 6x7 lens to the 645 as the 45mm 645 lens mounted on the 645. What am I missing aside from the money I spent thinking this 45mm 6x7 would give me the same focal length of a 23mm lens native to full frame?
This just illustrates the lenses make the most difference in terms of image quality
I think it's related only to modern digital 'medium' format. On 6x9 film difference is more obvious. But there is no sensor of that size. I don't even mention large format.
it's the same on film too, but there are other benefits of a MUCH larger image capturing medium than DoF. All MF film still has a tonal range advantage I think over digital...and still handles highlights in a much more pleasing manner than digital.
Medium Format is magical for enough reasons to be better than FF. It doesn’t mean I don’t like all the sensor sizes. I just know which one I like the most MF, period.
There's no difference... besides all the differences. I mean, everything technically is as pointed out, but put someone who doesn't know the technicals on the street with a different sensor size than they normally use and let them take photos the same way they always have. They'll tell you there darn well is a massive difference.
I found this out the hard way when just starting to use a Minolta Autocord while in Santa Fe, photographing my family in front of the church. The most beautiful family portrait ever taken _with the guys in the second row obviously out of focus!_ In practical use, there's a difference.
Dont agree, the test is flawed and here is why ua-cam.com/video/fHCtop_yfvY/v-deo.html
Great video as Always!!!!!!
Well the aperture is equivalent throughout the three , no crap the depth of field is the same/similar. If you use the same equivalent focal lengths with the same aperture for each lens , then youll see the difference. For instance all at ,F2 and equivalent focal length. I always looks at depth of field per a given aperture. If all lenses theoretically have a f1.8 with equivalent focal length then i think you would have a obvious winner.
So for the lenses tested there is little difference. Sure. What you did not test is something like the 80f1.7 on GF. This sort of lens and the mitakon 65 f1.4 allow environmental portraits that would be hard to replicate on 35 and basically a non starter on smaller formats.
At uni anyone using less than 6X7 was teased in the dark rooms, that's the real power.
back in the film days there was a real quality gain for large hand prints, today I assume it's the same for large prints.
As ever it also says "look at me, my camera is so big and costs so much so pay me more for work" which I assume is a big part of it today.
the title of the video ought to have been "There's nothing magical about DIGITAL medium format depth of field"
I have a mamiya RB67, and as Chris pointed out, the old school film camera NEGATIVES really were that much bigger than the 35mm film cameras, especially 6x7 and upwards - so a noticeable difference in LOOK could easy be seen in the pictures taken with a 6x7 v/s one taken with a 35mm film camera.
for digital medium format, yes, Chris is on point that there really isn't much of a difference compared to 35mm full frame digital - nonetheless, there is still that much of fine detail, color rendition, beautiful shadows and highlights that medium format STILL delivers "slightly" better than the 35mm full frame cameras.
to each his own i guess, I use a 16mp Olympus EM1 now, seems very good for the kind of photography i do - but then there's those who can see the difference in digital medium format camera pictures or are required to use medium format cameras for one reason or the other.
good video, as usual !
On a sad note.. I have a Fuji XPro 3 and XT2 and find myself taking photos and videos on my iPhone way more often that my Fuji cameras
it's a different story on film. Photos from the gw690 at f/3.5 do have a very very different feel. We are talking about a 6 x 9 vs a 35mm.
I think that's tonal range more than DoF, and still an advantage film has since most digital MF sensors aren't that much bigger than 35mm
This video is going to do Tony Northup's head in!
My issue with MF is the ugly 4:3 AR (no, crop to 3:2 doesn’t do it). I’m APS-C and FF all the way until a 3:2 MF is available.
DPReview is trying to put a title to stir people emotion. My oversimplified take from the video is that "There's Nothing Magical About Different Formats of DOF - it's about equivalency". I'm surprised Chris and the team put out a video and the title like that.
If you shot all camera formats at f2 for example all the Depth of fields would be completely different, idk if this comparison works since you’re shooting all camera f stops at the same equivalence.
I think Medium format would have a shallower DoF since it has a larger sensor, f 1.8 fullframe is still a f 1.8 on crop sensor, the only thing that changes is depth of field and focal length.
The aperture is still as open no matter what format.
@Hardware Software if you shoot all formats at the same aperture. Your results will look different.
A 1.8 fullframe is the same as a 1.8 crop sensor.
The thing that changes is depth of field and the crop of the focal length.
This test is incorrect because they’re compensating by putting the apertures to the equivalence of “depth of field”
Which isn’t an honest test.
@Hardware Software the only thing that needs to be compared is equivalence to focal length and set at the same aperture.
If i shoot a 56(85 equiv.)f1.2 on a crop sensor and shot 85 1.2 on a fullframe the depth of field would be vastly different, the fullframe would have better background blur 100%
So if you shoot on a fujifilm MF 110mm which is a 87mm equivalent, lets hypothetically say it had an aperture of f1.2
The depth of field would be more immensely shallower then a FF
Achieving the MF look ?
@@djstuc exactly, the whole point is show us the difference in DOF for each format. Set at the same aperture, with the same field of view.
Im not sure, if this guy gets what we’re trying to say?
@Hardware Software bro lol
If fujifilm made a 110mm f1.2 tomorrow
Theres no way a fullframe manufacturer can make a lens that is 85mm f0.94 with autofocus.
You know how expensive that would be to purchase? And absurd that would sound. lol
The larger sensor will have better DOF, better dynamic range, better picture quality because the sensor is larger
@@djstuc honestly have a feeling that the GFX would look more creamy. I watched a ton of comparisons vids on crop sensor vs FF when i was deciding to buy a camera. And you can always tell which camera was which, off of depth of field. The fullframe always had the best DoF
So if they compared MF to FF with similar apertures for sure the MF would smoke the FF
Let’s see micro four thirds.
You can kinda figure out for yourself how that would work. This video literally proves that the math works out. So by that same principle, you can do the math yourself to get an understanding.
@@nikoolix whoosh