With the release of the Fujifilm GFX 50R, the Hasselblad X1D II 50C, as well as a burgeoning used market, digital medium format has become more attainable than ever by professional photographers wanting to step up to the next level in image quality. However, the full frame market is firing back on all cylinders, producing cameras that claim to rival medium format, such as the Sony a7R IV. Many medium format users are quick to point out that there is a medium format look that these high-end full-frame cameras are lacking. So, what is the medium format look? Is it real? Why, yes. Yes, it is.
Marketing Mojo Versus Real Life
Perhaps the best place to start with this topic is to look back at an earlier technological marvel: APS-C. Why? Well, back when digital SLRs first started becoming popular, making full-sized sensors was cost-prohibitive. In order to have a shot in bringing this new product to market, the cost could not be astronomical. APS-C, or cropped sensors, were born as a compromise between economy and quality. Even the Nikon D1, back in 1999, equated its new cropped sensor product to the F5 and F100, their 35mm film flagships at the time. Nikon was eager to sell bodies, and marketing the new cameras as cropped sensors wouldn't have helped their cause. Of course, as time went on, sensor technology became cheaper, and the first commercially available full-frame DSLR was born as the Contax N. It was a flop, but immediately after it came the Canon 1Ds. From then on, digital sensor size fever has run rampant. We can see it on this site every day! Some say full frame has a look that APS-C can't touch. Some say APS-C is just as good as full frame.
6x45, the smallest medium format size
Fast-forward a few years and "medium format" digital backs begin to show up for commercial photographers. Of course, in the nineties, there were scanning backs, but they were so different they barely factor into this discussion. "Medium format" backs are so named because they work by attaching to medium format film cameras. However, up until recently, the sensors have always been significantly smaller than a film-sized medium format frame. To obtain a sensor that is close to true medium format size, you can expect to pay north of $50,000! The new GFX and Hasselblad cameras, although amazing, do not have true medium format size sensors. Not even close. However, "Full Frame Plus" just doesn't have that same ring to it, does it? So, here we are.
The True Advantages of Medium Format
Back in the film days (let me pull out my cane and wag my finger for a bit), the advantages attained by jumping from 35mm film to 120 film formats were huge and obvious. When you needed a step up in quality, typically in portraiture and landscapes, moving to medium format was the logical step up. Even 6x4.5, the smallest medium format size, showed smoother tonality and a large jump in resolution when compared to 35mm.
6x7. Nicknamed the ideal format for its closeness to 8x10 in ratio. Mostly marketing mumbo-jumbo, but I love it!
Notice I haven't said anything about depth of field. Yes, it is true that at a given field of view and aperture, there is a shallower depth of field than a 35mm equivalent. However, 35mm formats, digital and film, have access to faster glass, easily able to compensate for a smaller sensor/frame size if you want shallower depth. And here is the point a lot of photographers get hung up on: they believe that the medium format look is solely a function of that shallower depth of field. That simply isn't true.
The advantages of medium format are greater resolution potential and better, smoother tonality. Sure, we can say that shallower depth is an advantage as well if that's your bag, but that never really was the purpose of medium format. In the grand scheme, this whole shallow depth thing is a fad. Shallow depth was mainly seen as a liability. It was something to overcome by stopping down. Now, bokeh mania has taken over and medium format has been equated to that shallow look. If you define medium format by that shallow look, then yes, you may believe that the idea of a medium format look is a hoax.
6x7 as demonstrated by my best model.
Resolution is an easy advantage to overcome with digital. As the pixels get smaller, more and more can fit in a given space. That's why you can have 40+ megapixel cameras in cell phones. Loads of resolution, garbage image quality. But what about tonality?
Tonality is the big one that photographers seem to forget about, and yet it is the greatest strength of larger formats. Because the frame is larger, there is more space to make a tonal transition than on 35mm. Therefore, the transition can be smoother. Period. The larger the format, the better the tonality can potentially be. That's not my opinion. That's science. Think about it this way: You have to go from white to black within 2 inches. Now, make the same transition from white to black within 6 inches. You can place more tones in 6 inches than in 2. It's that simple. This greater space for tonal changes creates truer, more lifelike images.
It's All Relative
Of course, there are a few potential pitfalls stopping us laypeople from readily seeing the difference on these fancy new cameras. First is print size. The human eye is only so perceptive, and trying to discern smoother tonal changes on Instagram just ain't gonna happen. You really need to print large and be close to see the difference. And before you bring up the viewing distance argument, go to any museum and watch normal people look at art. They look from close, far, and everything in-between. They couldn't care less about appropriate viewing distances.
Next, because most of these medium format sensors are nowhere near true medium format, it's much more difficult to see the difference from full frame. For example, going from full-frame to the GFX sensor, there is a 1.7x increase in size. That's actually less than the difference between APS-C and full-frame. If most people can't tell the difference between APS-C and full-frame, how in the world would they be able to tell between full-frame and digital medium format?
Film is a much easier medium in which to see the difference. The smallest true medium format size, 6x4.5, is a full 2.7x larger than full-frame, while 6x7 has an astounding 4.76x more area! That's why you can easily see the difference in tonality on film. I imagine if I was using a Hasselblad H6D I'd also be able to see the difference, since it's full-frame medium format digital, but I haven't got 50K lying around to purchase one and I have no need to rent one.
6x7 of Noelle
Does It All Matter?
In the end, what truly matters is the preferences of the photographer, their needs, and their ability to justify the purchase of the camera they use. If they work in an environment where having a big medium format camera is part of their image as a professional, great! If they want to be able to show maximum tonality with large, in your face prints, also great! If you don't need it, that's fine as well. But distilling the medium format look down to depth of field is just silly, and it doesn't reflect the true reasons that the format exists as an option for photographers.
What do you think? Still not convinced? I can't say I blame you! As technology gets better and better, it's easier for full frame to catch up to the advantages of medium format. But that doesn't mean the differences aren't there.
You managed to write an entire article about a meaningless buzzword. Congratulations! "Tonality" can depend on resolution (of both the sensor and the lens) and bit depth. It has nothing to do with physical distance, as we aren't exactly getting close to the planck length here. What a stupid, illogical notion.
I think the point is the following. Imagine you have a sensor that is 2" square and a sensor that is 1" square, each has a resolution of 1,000 x 1,000 pixels. Now imagine we project an even tonal gradient across each sensor. Each pixel on the larger sensor is twice the width and height of the smaller sensor and, therefore, has four times the area and collects four times the amount of light. That's a two stop improvement in light gathering at each pixel. Each additional stop of light reduces the shot noise by a factor of sqrt(2). So two stops decreases the noise, the standard deviation about the mean, by a factor of 2. The number of different tonal increments that can be distinguished depends on the magnitude of that noise so if we decrease the noise by a factor of 2, the number of tonal increments that can be resolved goes up by a factor of 2. This is a little bit of an oversimplification since as the brightness varies across the sensor so will the effective noise level. But you get the idea. Bigger wells means lower noise means an ability to resolve smaller tonal increments. Obviously you also need to have enough bit depth to encode those tonal increments or they're wasted.
Nope, that's a confused mess of a comment, mostly because "tonality" is not a real word and means something different for everyone. Obviously the larger sensor is less noisy at the same iso. You don't need to write a novel about it, it's obvious. The "increments" however are a function of the bit depth and dynamic range. "Tonality" usually refers to a combination of this and resolution, except that nobody knows what it refers to exactly. More importantly, smooth gradation is not only the easiest thing to interpolate digitally, but also cannot really be displayed on our 8-10 bit screens, much less in print.
@Brent Daniel Actually that's not necessarily true, let's say the smaller sensor is 1/2 the bigger one, the amount of light reaching each of those 1000 pixels depends uniquely on the lens you are using, if you are using a lens that gives you the same FoV but is 1 stop faster on the smaller sensor you will end up with the same amount of light on each pixel = same shot noise.
That's why despite the bigger size MF (be it GFX, Phase One, Hassy etc) are not recomended for low light shooting since there aren't lenses that are faster than the equivalent fullframe FOV.
The advantage of medium format however is the fact that if the surface is bigger first you get more dynamic range, and second the lens you need to use to capture a certain amount of detail works at lower MTF compared to the one you would need for that same detail on a smaller sensor.
That means that a given lens gives you more detail the bigger the sensor is (as long as it can cover it of course), and that's why FF lenses perform better on FF than on APS-C bodies.
P.S. if the sensor is twice, the size you get twice the light on each pixel, not 4 times, but I guess you just forgot to quadruple the surface ;)
Hi Paolo. Yes, I was assuming that the light intensity per unit area incident on the sensors was the same. The choice of lens/aperture certainly impacts that. If each well of the larger sensor is twice as high and twice as wide as that of the smaller sensor, it has four times the area and collects four times as much light (2"x2" versus 1"x1", both with the same resolution), right? I hadn't thought about the effective sharpness. Good point.
If each pixel is twice as wide and as high yes, the are is 4 times larger, is just that at the beginning you wrote "imagine you have a sensor that is 2" square and a sensor that is 1" square", guess the 2 instead of the 4 slipped :P
http://largesense.com/products/8x10-large-format-digital-back-ls911/
I feel that large sensors have better tones. It is hard to explain why. I enjoy medium format more than small format. 8x10 and 20x24 are pretty awesome and are easier to see the large effect.
Heh. Clearly you've never shot film. Or if you have, you've only ever used one format.
We are not talking about film...
Here that, Hans Rosemond! You're not talking about film!
I agree with Rk K.
If somebody wants to test this, please take m43-camera with a 25 mm lens and use, lets say, f/2.8. Take a shot and compare that to a "full frame" camera with 50 mm lens and f/5.6 and two stops higher ISO. If you keep the shutter speed same, you will get exactly the same result.
There is a factor of 2 difference in sensor widths which should give you the "full frame look" compared to the smaller sensor. The relative difference between 6x7 and 35mm film is roughly the same.
So the same testing but keep identical camera settings, so same shutter speed like 1/60, same F-stop like f/2.8 and same ISO value like ISO 400. Use a focal length that gives you same framing in widest (3:2 vs 4:3) like 17mm and 35mm and then take photos from same tripod. So you take a shot, swap camera and take a shot. Take a landscape, portrait, still life and architecture photos.
Give the files to a good image editor who task is to get best out of the files. And then produce various prints.
- from 2-3 common consumer photo services and 2-3 high grade professional services.
- produce the common and larger than common prints like 6x4, 8x10, 11x16 and 30x20.
Put those prints on front of northern window and ask 4-5 friends to come by to tell their opinion about image quality.
80-90% shouldn't see any difference, and rest who does, shouldn't care the difference.
I made so, don't care the pixel peeping when I know a 1x1 cm area in a 30x20" print where to look at the difference that would reveal it by comparing two prints side by side multiple times.
I hang my prints, I sell those and no one ever goes to pixel peep anything, even if they would, they wouldn't see the difference between GH4 and A7r III.
Not really.
For “Same image,” one must use the same shutter speed, (motion blur), and same aperture diameter, (DoF). Since the focal length has changed, then too has the f-number. Thus one must change the exposure index.
Bear in mind that the exposure index is NOT a camera setting, but a light meter and development setting.
Just a minor correction, but your point is well made.
There is no MF Look. Medium Format uses larger sensors and more depth from 16 Bits than DSLR 14 Bits. This is most valuable in 4 color or greater magazine reproduction printing.
You're right about the bit depth it does make a difference. The issue is that most MF cameras are also14 bit.
16bit is relevant to color rendition only insofar as one is making large color adjustments where lower bit counts have less data to interpolate to a new color. In the real world it has no real effect.
I'm not so sure that I can completely agree. In many comparisons, I've done between 16bit medium format cameras vs full frame in areas where you have a bunch of colours meeting you could clearly see more colour detail and more vibrant colours. It really does make a difference but whether it's worth it or not is a different matter.
I would presume this would be more apparent because of higher res rather than 16bit. The comparison would need to be made between sensors of relatively the same res.
The best test would be to photograph a test scene with a GFX 50 at 14bit and then with a GFX-100 at 16bit, resize down to the 50MP size and compare. There would still be differences but I would suggest that they are more a result of filters and profiling than bit depth.
I was looking at some comparisons between 14 and 16 bit both with the GFX-100 and there didn't seem to be much difference. The phase detect pixels might be causing some noise and banding making the extra 2 bits not as useful.
If you dudes don’t know 4 color offsett printing .. you don’t know squat about color depth and bits from quality 16 bit sensors.
If you don’t shoot for Magazines with primo equipment .. you won’t know.
If only I could reply with a gif...
Considering that even humidity changes would have a material effect on the reproduction of ink on paper I think maybe you were trying to say that ink on paper repro is miserable compared to photo (which is only 8bit) but I couldn't hear properly when you speak through your hat.
FYI .. Offsett 4 color printing is a whole differ world than injet stinkjet and apparently you are in the dark on high quality repro. You could ask an art director why they want and expect high res 16 bit MF, primo printers, expensive ink, premium paper or just stay in dark like a Photographer. It’s an intense process control that few photographers understand in both printing and plate making. .
Thanks for the laugh dudes . keep your clicker clicking and commodify writing on things you don’t understand.
Lol, haha, you're like the bad guy out of a crappy romantic film. Do you ever get tired of sniffing your own scent?
Everyone:...
EL PIC: I know everything I'm better than you.
Just bigger AH with no estate wealth or proper education.
I love you too AH ..
When two AH meet :-p.
Ahem.
And depending on your client their baseline requirements can change.
We used to produce pre-pro for high end clients and Disney print reps will/would actually bring micrometers to measure line and dot repro. Don’t even get me started on their color requirements!
Yes I know Disney and how requirements change with them and other high quality clients.
On Food shots I get ADs that insist they live view and approve before we take final shot. But this dude above does not.
A fussy client is a given in high end repro. So too are the limits of the technology. Mickey Mouse himself cannot give CMYK more gamut even if he threatens to bite the printer.
The discussion of CMYK here is foolishness as it has no chance of even coming within laughing distance of other processes in terms of anything other than cost at scale.
"Image quality is a multidimensional thing, some of which can be quantified and some not. Still, by no measure of image quality does a good Micro 4/3 camera and lens perform more poorly than a good medium format film rig, and by some measures it performs considerably better. My overall subjective evaluation is that the aggregate image quality of Micro 4/3 today, in film terms, falls midway between 6x7 medium format and 4x5-inch large format.
"Almost everyone you can find who is still arguing that Micro 4/3 can't match up to professional film has not done substantial amounts of serious work in both media. I believe the technical term is 'talking through one's hat.'"
https://theonlinephotographer.typepad.com/the_online_photographer/2016/0...
Do you happen to know who that guy is? He knows more about printing and image quality than anyone here will ever do.
So, m4/3 system cameras already had 2012 the "medium format" look and quality... Why are the other small format users so naive thinking they get better by going larger sensors?
Keep that hat on your head It will cover the gaps
Does medium format have a "look"
Digital: Nope
Film: Absolutely.
Haha I figured you would contribute to this one. Glad to have you participate!
You know I can't help myself haha
Yep, completely agree. In the film days, there was nothing like shooting my portrait stuff on MF, and likewise, pulling out the 4x5 for landscape stuff. Now, my FF DSLR does just as nice on landscape stuff as my 4x5 did. Crazy as hell, honestly. People who have never shot film will never understand this.
Let the comments begin. I am gonna keep refreshing this page every few hours today and every few days
It’s like a form of Voyeurism lol.
Yeah, meaningless fights doesn't serve anything but entertainment
Despite the outcome (if at all any), I going to enjoy my GFX-50S and X-T3. Maybe gfx-100s in an year (first let me be done with gf 45-100mm when it gets released).
You need that "I'm just here for the comments" MJ popcorn meme.
It is all relative! I've started shooting 4x5 film a bit, and the shallow DOF can be one huge pain for closer-up work. Getting the tape measure and DOF app out doesn't seem to really work, and I'm thinking of just sicking with MF and its much greater DOF for that kind of work.
haha yeah I just did 6 portraits with 4x5 film and on f/11.0 and still scared to send them to lab :D because prob will be out of focus
You make a point about tonality relating to size. While this is absolutely correct, it has to do with the fact that film has grain which a small format negative/transparency have fewer grains to stretch out over a large print. For small prints I have made a lot of 8x10B&W prints that look just like my 6x6 prints.
With digital a pixel becomes that grain. Thus the number of pixels equalizes the differences between formats presuming all other things (noise, artifacts) being equal. So, arguably, an APS-C sensor of 32MP will render equal tonality of a MF sensor of 32MP.
So the short answer, for me, is that the MF look is no longer the sole province of 6x7 film but the rendering I get from appropriate res for the reproduction.
I'm with Hans, it's real. To my eye, MF and LF have more actual depth.
What is "depth"?
Not flat.
Unless you're working with a 3-D medium, not flat is contrasty, and contrasty is not hard to achieve with any camera.
I guess MF and LF are more like a 3-D medium than not then.