µ4/3 and Medium Format: Dual-System Thoughts [long post]

Started 1 month ago | Discussions thread
ForumParentFirstPrevious
Flat view
Velocity of Sound
Velocity of Sound Contributing Member • Posts: 787
µ4/3 and Medium Format: Dual-System Thoughts [long post]
68

A few months ago I bought a FujiFilm GFX 50S "medium format" camera, and started using it alongside by Olympus E-M1 MkII. It led to some interesting realizations about photography, µ4/3, and our group psychology, which I wanted to share here.

Tired of these comparison-style threads? Wondering why it's posted in this forum?

If I could have anyone read this thread, it would be myself in the past. I'm writing it that way, and hoping that it will be useful to those who feel similarly to the way that I did. My first "real" camera was an Olympus (E-volt E-410, to be exact), and while I loved that camera, I plugged in to various photography communities and very quickly realized three things: that I had become a fan of a "team" with my purchase (or perhaps, forum registration); that I had chosen a brand that was not one of the more common selections; and that my fellow "fans" were both proud of Olympus, and insecure over their decisions to go with Olympus. While I have now owned more Olympus bodies than I can count on one hand, I had never owned a camera of another brand, or sensor size. And while many of my own insecurities washed away when I jumped from my old E-3 to the E-M1 MkII (a shocking jump that really makes you appreciate modern technology), I was always curious about the common sentiment that 4/3 and µ4/3 was somehow worlds behind larger sensors in image quality and performance. Even though I was happy and finally more secure with my E-M1 MkII, I knew that the only way to sate my curiosity would be to eventually buy and use a camera with a larger sensor.

But this goes a bit beyond image quality. I've read numerous complaints about Olympus' menu system, and some complaints about their overall functionality. Having never used a camera body from any other manufacturer, I figured it would be of interest to share my thoughts on what it was like there, too. It's silly to try and compare certain performance aspects of the GFX 50S, a medium format camera that is breaking the mould of medium format cameras as being slow and studio-based, directly against the E-M1 Mk2, which is designed more for action and hails from a system that prides itself on portability. So where I can, I'll try to keep the comparisons more to things that make a bit more sense to compare.

Why medium format?

I was initially between mirrorless "full frame" systems and mirrorless medium format. I didn't want to outright replace my Olympus system, so I wasn't seeking feature parity; rather, I wanted to see for myself what the "ultimate image quality" was like. The common argument that we've all heard is "larger sensors result in better quality," so it seemed that medium format systems - being larger than "full frame" - would be more desirable. And when I looked at various review websites, it seemed that there was indeed a difference between the 50S and "full frame" cameras, although the significance of the difference seemed to vary depending on the review. Fate pushed the decision when I stumbled across a very nice deal on a used GFX 50S system. (And for those who tend to be concerned about how others are dealing with their finances: I could comfortably afford this experiment, so please do not worry.)

A word about equivalence

I really hate equivalence talk. There is nothing else that becomes so boring so quickly, nor that gets people speaking in such a nasty manner to each other, as equivalence talk. There are many different ways to think about equivalence, and I often get the impression that people argue about equivalence from one angle, and don't see that the disagreement comes because their debate partner is coming at it from a different angle. If I could, I'd avoid the topic entirely. But because we're talking about different systems and sensor sizes, it's inevitable.

We're used to talking about the "crop factor" with µ4/3 compared with 35mm ("full frame") as being 2x - that is, double your focal length to figure out what focal length you would need on a "full frame" camera to have the same angle of view. Equivalence for aperture is a bit less clear - some people refer purely to light collection equivalence, while others refer to depth of field. While it's not entirely accurate for depth of field, most calculations I've seen are close enough to use it that way, but I won't be doing that here. Really, I give the equivalence number just to talk about angle of view.

The "crop factor" for the medium format sensor in the GFX 50S - lovingly called "mini-medium format" by some - is 0.79x with "full frame." (Yes, they still use the term "crop factor," even though it's pretty silly to talk about when the other sensor is larger.)

To translate between µ4/3 and medium format, your "crop factor" is somewhere between 2.43 (calculated by sensor diagonals, as both µ4/3 and "mini medium format" are 4:3 ratio) and 2.52 (calculated based off of the fact that 25mm represents your "standard focal length" on µ4/3, whereas 63mm represents your "standard focal length" on the GFX system). I don't want to get hung up on the specifics, and while equivalence will come up again, the exact number you use shouldn't really matter for the purposes of this discussion.

What's with the photos?

My intention is to have a mixture of photos taken with my E-M1 Mk2 and GFX 50S in this thread. My original goal was to try and point out how the photos are more similar than they are different, but it's a bit silly to do that on a web forum. DPReview won't allow me to post the full-resolution images from the GFX 50S unless they've been cropped a bit, and it seems rare that the forum shows images at the true quality they've been uploaded at. And even if I could show off the images from both cameras at full quality, there's a big difference between what's presented and what it feels like to take the photo in the setting, and then edit it. So really, they're here to add a bit of color and break up some of the text. Wish I could say that I chose my masterpieces for this post, but it's a bit more random than that.

So without further ado... let's move on.

Build quality

Occasional failed grip and broken lens hood from the 40-150mm f/2.8 aside, while nobody has ever faulted Olympus for their build quality, I was very interested to see how Olympus compared with Fujifilm - particularly when Fujifilm's offering costs more than double that of Olympus. The spoiler for this section is that I encountered another forum user who used both the GFX and µ4/3 (Olympus) systems, who stated his preference for Olympus build quality, and I have to agree.

Unboxing the GFX 50S, I was taken with how nice the camera strap was. It smelled of some leather-like material, had nice padding, and had "Fujifilm" embroidered in a dark text. No loud declaration of the camera model, no lining of a strap with a material that offers more resistance than cushion, the strap exuded quality. And I make a big deal about the camera strap largely because the rest of this section will sound like Olympus fanboyism otherwise.

The GFX camera and lenses are well-built overall, but they don't feel that way... especially compared to Olympus' Pro-line equipment. I suspect it's because Fujifilm did a good job with keeping the weight of their gear down, as well as keeping the size of their lenses relatively small. But they can't get away from the fact that their lens mount is one of the largest available, meaning that even their smallest lens balloons outward toward the back. Without that added heft, the gear almost feels toy-like compared with Olympus' solid lenses.

Giving a rough sense of scale: the Fujifilm 63mm f/2.8 on the left, with the Olympus 25mm f/1.2 on the right. The Olympus is actually a bit longer than the 63mm, although to be fair, that lens utilizes an external focusing design and extends in use. Regardless, you could fit the µ4/3 lens mount into the G-mount if you really wanted to.

If you're not doing a direct comparison, the lenses and camera body really feel fine. Yet the lens hoods are an area that receive a fair bit of criticism, even from other Fujifilm users. They're made of a plastic that feels flimsy, and I've already noted some pock marks and scrapes on them. By comparison, the Olympus lens hoods feel so solid that when I have my camera slung over my shoulder and I accidentally bump it against a wall, I check the paint on the wall for signs of damage first, and then check the lens to ensure that it didn't take a bite out of the paint. It's a testament to Olympus, and a bit surprising; none of the GFX-system lenses cost less than $1,000, and many are over $2,000. While it's true that Olympus' Pro-grade lenses are the most expensive on the system, many hover near the $1,000 mark; at this time I can only think of one that costs over $2,000.

Ergonomics

This one was a bit shocking to me: there are some features that I can only describe as being design errors with the GFX 50S.

First, the good things: the 50S employs a "joystick" for selecting autofocus, although you can still push the directional buttons if you really want. They also have their power switch located around the shutter button, and the front control wheel is placed on the front grip in such a location that it doesn't feel terribly different from where my finger would hit the front control wheel on my E-M1 MkII. The joystick was a feature that I missed immediately upon picking up my E-M1 MkII again, and the placement of the on/off switch was so good that even my wife - after using the GFX 50S only briefly - did the same thing that I did and instinctively tried to switch the Olympus on/off around the shutter button. There's just something about being able to activate that switch from the hand that's already positioned to grip the camera that feels like it makes sense, particularly with mirrorless cameras, where the on/off switch is used more frequently than a traditional DSLR.

The shock to me was the way that the control wheels were not useful. The 50S has a very deep groove on the rear, providing ample support for your thumb to grip the camera from the back. The trouble is, the rear dial is too close to that groove. I'm used to frantically using the control wheels on my Olympus to make fast, large adjustments, but that's impossible with the 50S: your thumb hits against the groove even when the wheel has only moved 2-3 clicks. And even looking at the other buttons, some of them are incredibly small. I faulted my E-M1 MkII for giving me some trouble when I tried using it with gloves on - an issue I don't recall with my old, larger E-3 - but some of the buttons of the GFX 50S would be downright impossible to hit with a gloved hand. In fact, some are so tiny that I initially thought that they were nubs, rather than buttons. Funny thing for a larger camera body.

Fujifilm's philosophy seems a bit different. While the camera is fairly heavily customizable, many switches and dials have defined functions. All lenses have an aperture ring, and while the ring does have a setting to move control over to a control wheel, I figured I'd try using it and now understand why the Panasonic users like it so much. (Unlike the autofocus joystick and on/off switch, however, I don't find myself missing it when I return to my Olympus.) There is a large control dial for ISO and a separate, equally large dial for shutter speed at the top of the camera, and I dislike them. Whereas I have set one of my control wheels to allow me to control the ISO on my Olympus, having it dedicated to a dial as Fujifilm did feels like a waste. The dial, unlike a control wheel, can't be adjusted rapidly in an easy manner, and it practically forces you to look away from the viewfinder. Adding insult to injury, if you set the dial to have one of the control wheels handle the settings, you lose the ability to have ISO set automatically. Combined with the fact that the camera doesn't show you what ISO it would automatically choose until you half-press the shutter button, I can say that I vastly prefer Olympus' way of doing things.

I guess it's worth noting that this revelation was interesting in itself. I've often wondered whether the complaints over Olympus' system resulted from people already being used to one system, or whether the other systems were truly superior. I started in digital photography with Olympus, and know its way of doing things well; it seems most people who were coming to 4/3, or who come to ยต4/3, started with another system first. My complaints over Fujifilm's way of doing things - and seeing Fujifilm users criticizing other systems (particularly Sony) as not being very user-friendly - makes me think that most complaints are probably just people being less adaptable once they've already gotten used to a certain way of doing something. (I suppose it's still possible that Olympus' system is truly superior, but eh )

Firmware

Fujifilm did something really nice that Olympus seems to be on the road to introducing with the E-M1X: customization of menus. Specifically, there's a section in the Settings that allows you to choose other settings to be filed there. No more diving through menus to find some frequently-used settings to change anymore! Auto-ISO also allows you to customize three presets to quickly change between, and you can save seven customized profiles for quick recall. (My E-M1 MkII only allows three customized profiles, but the caveat is that they can be activated by switch, whereas they need to be chosen through menus on the Fujifilm.). In general, it feels like there's less to dive through on the Fujifilm... but then, there also aren't quite as many specialty features to go through, either.

Fujifilm doesn't have Olympus' "super control panel," but they do have something that looks a bit similar in their Q Menu system. Oddly, I haven't found a way to customize this. I'm sure there must be, but it seems a waste. Olympus' "super control panel" offers options relevant to both JPEG and RAW shooters, as some of the settings influence things like the drive mode and metering mode. By comparison, the Q Menu seems to only influence JPEG settings - useless to me as a RAW shooter.

Fujifilm also implemented some other odd choices. For example, if you have face detection on, then you can't change the metering mode ("photometry," as Fujifilm calls it) - you need to disable face detection, change your metering mode, and then re-enable face detection. In some ways, it makes sense: when a face is detected, the camera meters for the face. You don't want to imply otherwise, right? But it's a pain in practice, because you still want to know (and possibly adjust) your metering mode as a fall-back for the shots where face detection isn't working. I've remarked on it already, but if you don't want to adjust the ISO dial and would rather use a control wheel, there's no way to lock the ISO to "auto" through the wheel - you have to set it to "auto" on the dial.

As far as face detection goes, while Fujifilm isn't regarded as a leader in this field, it's an interesting comparison with Olympus. I'd say the biggest difference is that Olympus locks onto faces more easily, but it also loses them more easily. Once the GFX 50S has found the face, it tends to hang on to it more - even when the person turns their head. But in general, I find myself waiting for the face detection block to come up in situations where I know the Olympus would have already found it. Like the Olympus, the Fuji doesn't seem to have a way to direct the camera as to which face to focus on when multiple are detected. Unlike the Olympus, when a face is detected your usual autofocus block disappears. It reappears (and replaces the face box) if you move it, but the camera quickly brings the face detection box back, which can be a bit jarring. Again, I like to know where my autofocus point is as a fallback in case the face is no longer detected; this is another point in Olympus' favor.

Close focus

A brief mention here: Olympus lenses seem have about half the minimum focus distance of the GFX lenses. I haven't done an exhaustive search of other systems, but it seems that Olympus regularly has the shortest minimum focusing distance. I've almost never had my camera indicate that something was too close - and that includes my son, who loves to run up to me and stand squarely in front of my camera when I crouch down for a shot. With my GFX 50S, on the other hand, there have been more than a few times when I needed to back up... or tell my son to stay where he was, as I quickly moved backward to get the shot of him. It's not a feature that most people would use on a daily basis, I'm guessing, but it's another area where it feels as if there's a surprising compromise that I didn't expect to run up against with another camera system.

Maybe this isn't the most flattering way to take a portrait, but it's sharply in focus! His forehead was practically hitting the lens hood.

Silence in shooting

One of the biggest benefits in moving to mirrorless, to me, was the silence. My loud, whiney-sounding 4/3 lenses were replaced with much faster lenses. In fact, the µ4/3 lenses are even quieter than the 4/3 lenses bearing ultrasonic motors. In a dead-silent room I can hear a gentle sound if I place my ear near the lens, but that's it. Some of my lenses have a slightly more audible click of the aperture, but on most, it's silent. I can take my E-M1 MkII into a museum and feel confident that even someone standing next to me wouldn't know that I was taking a photo.

The Fujinon lenses are pretty quiet overall, but they're loud compared to the Olympus lenses. So far, two types of autofocus drives are used in the GFX lenses: older, stepper motors, and newer, "linear motor" (from what I can gather, magnetic) drive systems. The stepper motors are still worlds quieter than my old 4/3 lenses, but they - or at least, the one on my 63mm f/2.8 - are still audible, again sounding like a bit of a whisper. The linear motors are a bit quieter, but they produce a higher-pitched sound. I don't think that someone at the opposite end of a quiet museum room would be able to hear them, but someone standing next to me would.

Sound in operation always mattered, but it didn't become critical to me until I had children and wanted to be able to take photos without waking them.

However, it wouldn't matter if the lenses were silent because the GFX 50S, despite being mirrorless, is relatively useless in full silent mode.

This is something I didn't realize when I bought the E-M1 MkII, and had assumed about all mirrorless cameras: they're not all the same when it comes to operating silently. We're used to the old paradigm of mirror-bearing cameras, where the mirror would flip out of the way and then the shutter curtain would open - either fully, or in a sliver that quickly ran down - to expose film or a digital sensor. With mirrorless, the sensor was exposed all the time and there's no mirror to flip. There are essentially three was to handle an exposure: to have the shutter come up, and then go back down; to do a full electronic exposure; or to combine the two, using an "electronic first curtain shutter", where the exposure is initiated with the electronic shutter but the mechanical shutter still comes down to complete the exposure. Olympus' standard, non-silent shooting mode works using the electronic first-curtain, from what I understand. However, the electronic shutter relies on sensor readout speed to avoid artifacts (most notably, the the "jello" effect). That is, the sensor reads the pixels from left to right (or right to left, I don't know the exact orientation) from the top row, and then from one below that, and then below that... until it reaches the bottom of the sensor, and during that time the resulting image can be influenced by changes in the scene. Surprisingly, there is wide variety of sensor read-out speeds among mirrorless cameras.

Sensor readout speeds don't seem to be published by manufacturers, but people measure them. The GFX 50S' sensor readout speed was measured at around 1/4 of a second. In other words, if you would have any motion in your scene that would be notable at that exposure time, then you'd see the "jello" stretch effect. By comparison, the E-M1 MkII has a read-out speed of 1/100 - still slow enough to cause motion artifacts, but fast enough that unless you're capturing action, you probably won't see it. The E-M1 MkII is a bit special, though; lower-end µ4/3 cameras have varied readout speeds that tend to be slower. Many "full frame" cameras seem to be measured at 1/30. The Sony A9, their sports-focused model, has the fastest readout speed: 1/160. So while the GFX 50S is easily the slowest of the bunch, forcing me to use it in electronic first-curtain mode (with accompanying sound from the shutter), it goes to show that not all mirrorless cameras can do what I'm used to with the E-M1 MkII.

This will be a non-issue if and when we get the "electronic global shutter" developed.

Image Quality: Overall

I saved this one for last because it's probably the first one that most people think of, and I think it's the most interesting. This is where psychology comes into play.

I used the GFX 50S for about a week and then downloaded the images to begin processing them. I was blown away by the first image that I saw: it was so lifelike! If equivalence stirs up nasty discussions in the µ4/3 forum, the topic of consternation in the medium format forum is about a "medium format look" - it's something that some people swear exists (and they argue against skeptics), yet nobody can seem to pin down what it is. And so I found myself wondering, was I witnessing "the look"? And then I began to get nervous, because I had gone into this thinking that I'd just become a dual-system user, but what if the internet was right? Was I going to have to sell all of my Olympus gear, and replace it with a "full frame" system to have acceptable image quality alongside a camera that had all the features of my E-M1 MkII?

Over the next two or so weeks I shot with both cameras, taking one on one outing, and another the next day. At the end of the two weeks I began sorting through the photos, all of which were put in chronological order. Most of these were simply walks with my son, and none of the walks stood out in such a way as to make it obvious which camera had been used based on the scenes alone. Without looking at data like the focal lengths and apertures, or even the file names, I began going through the photos, culling and pixel-peeping. "That one is clearly the GFX 50S," I'd think to myself, again trying to figure out what features I was seeing that made it seem so life-like... only to find that it was actually taken with my Olympus.

While I'm making light a little bit with this story, it's truly what happened. That's not to say that there is no difference in image quality: my guess rate wasn't exactly 50/50. However, I incorrectly attributed photos to the GFX 50S enough times that I felt satisfied with my Olympus. The differences are there, but the significance of those differences is in my head. The fact that blinding myself meant I couldn't easily see it is proof enough, and some of you have already heard me talk about that study in which group consensus caused people to perceive things that weren't real (basic things like incorrect shapes that were displayed). While it was rare, I did also have one or two photos where I thought to myself, "that must be the Olympus" (usually due to burned highlights), and it was actually the GFX 50S.

Let's dive into some of the specifics.

Image Quality: Dynamic Range

The GFX 50S is regarded as having some of the highest dynamic range, with some claiming 14 eV of range. In practice, this comes in the form of less post-processing. With my E-M1 MkII I'm almost always recovering at least a few highlights, and I can almost always get what I want that way... but with the GFX 50S, I rarely feel the need to adjust any highlights. And when it comes to lifting shadows, I can raise shadows enormously without generating much noise - something where the E-M1 MkII begins to fall apart at fairly quickly.

Comparing a few camera models, the GFX 50S leads the pack, although the current hot mirrorless full frame camera, the Sony A7III, isn't far behind. The E-M1 MkII trails behind; the GFX 50S needs to reach ISO 800 before its dynamic range falls to the E-M1 MkII's best around ISO 200. I've included two older cameras: the D700 (which we were all lusting over in the 4/3 days), and my old E-3, which burned highlights like crazy. While the D700 is now an old camera, notice that the E-M1 MkII exceeds its dynamic range... even though we were always told that it was "just physics" that larger sensors would somehow inherently be better than smaller ones.

While the GFX 50S is impressive here, modern technology may be making this something of a moot point. I can easily switch the E-M1 MkII over to exposure bracketing, and produce a HDR image with a greater dynamic range than a single exposure from the 50S would have. Modern HDR software is very impressive, and handles motion in a scene very well; I've taken a photo of colleagues actively moving, and the software still accurately gave me a result that required only a little bit of touching up to correct. Additionally, it's not like the 50S is immune to burned highlights or lost shadows. I've still taken photos indoors where the scene outside couldn't fully be recovered.

Simply put, I like the added dynamic range quite a bit, but it's not a feature that makes me pause when grabbing for my E-M1 MkII. While I'd like to try exposure bracketing to make HDRs off of the GFX 50S, without IBIS I'm a bit skeptical about how some of the results will come out... but you'll never know until you try.

Image Quality: Pixel Peeping, Cropping, and Printing

A lot of the medium format guys remark about how the system, and Fujinon lenses, far outclass their old "full frame" gear. Whether it's true or whether it's a case of them trying to justify their purchases, I can't say; I've never used "full frame" equipment. However, I feel that I can pixel peep to about as well (perhaps slightly better?) compared with my E-M1 MkII and Pro-grade lenses. I can go to 200% zoom and still feel pretty good about what I'm seeing.

Granted, there's a big difference between 50 megapixels and 20 megapixels, particularly when it comes to cropping. Some have even suggested that the 100-megapixel GFX 100 allows you to take multiple photos from one exposure, thanks to the massive cropping capability, which is an interesting idea. I don't need the 50 megapixels, but I do like them.

So far, I haven't done any huge prints out of the GFX 50S. 16x20 is the largest I've gone so far, and I haven't noticed any big improvement in image quality (although admittedly, I've only printed through Shutterfly so far, which isn't known for having the absolute best print quality there is). However, viewing photos at full size on my 27" "retina" display - larger and more demanding than 16x20 prints - is very satisfying at 50 megapixels, whereas the 20 megapixel files can struggle a bit.

Image Quality: High ISO Performance

I suspect I'm a bit less sensitive to image noise than most people: I'll happily use my E-M1 MkII to ISO 3200, and also don't fear ISO 6400 (although I'll try to avoid it, if I can). I use the GFX 50S up to ISO 12800, and feel like ISO 12800 looks like somewhere between ISO 1600 and ISO 3200 on the Olympus. It's somewhere around a two-stop difference. ISO 1600 looks clean.

But things get a bit interesting here, and I'll admit something to you all: my E-M1 MkII is my low-light camera.

ISO 5000 looks pretty clean to me, not that this tiny version would show the noise well.

One thing gets back to what I said about the GFX 50S' files being more "life-like." I noticed this even when I did head-to-head photos of the same subject. What I found were two things: the Fuji files seemed to have more contrast (which could be the sensor and/or lenses... or it could be that Capture One, my preferred RAW-processing software, just has better defaults for Fuji's files), which was easily corrected; and the Fuji seemed to underexpose compared to the Olympus. Whether that was due to differences in metering modes, or whether it's intentional, I can't say. I could believe that it's intentional, because the Olympus handles highlight recovery well but suffers a bit with shadow lifting, whereas the GFX 50S can lift shadows like a champ... but what it means is that I can knock down the exposure adjustment by -0.7 to -1 with the Olympus to have even more favorable exposure settings.

IBIS plays a large role in this, too. While it doesn't help for freezing motion, I can hand-hold down to about two seconds with my 17mm f/1.2, which means that I can still keep the ISO low. While the newer GFX 100 has IBIS, the GFX 50S does not - and while I wish it did, µ4/3 cameras and Olympus are still regarded as being leaders in IBIS performance.

Lastly, there's the lenses. While the widest native lens for the GFX system is f/2, and the widest at normal angles of view are f/2.8, I'm left to believe that the depth of field would be so ridiculously shallow at something like f/1.4 that I'm not even sure how much I'd want to use lenses like that in darker areas. Which leads into the next section:

Image Quality: Depth of Field

The medium format forum is unique amongst photography forums because while the rest of us seem to be desiring shallower depths of field, the medium format group usually discusses methods to increase depth of field.

Here's a confession: I used to be an absolute shallow DoF junkie. Don't really know why; maybe I was just following the trends, or maybe it's easier to blur out a background than to try and compose in a skilled manner. Regardless, after my son was born I found that I preferred a bit more context to my backgrounds. Blur things out so that the subject of focus is clearly visible, but have enough context so that you can tell what was going on, and what the surrounding scene was.

The background here is a blurred mess that will probably be totally unidentifiable to me in a few years, once I've forgotten the location. I have another shot at f/5.6, but it's still horribly blurred even then. It doesn't look horrible in small thumbnail form, but believe me - view this one at the full size of your screen, and the amount of blur in the background looks ridiculous, to my eyes. And to think that the DoF could be even more shallow if I adapted some f/1.4 lenses.

Because the GFX system doesn't yet have many wide-aperture options, it's possible to go even more shallow with "full frame" (although some, but not all "full frame" lenses can be adapted to the GFX system with minimal vignetting). However, you get the impression that the GFX system is really shallow; I did a few shots at f/22 and even f/32, apertures we never really touch on µ4/3 (not that any of our lenses go to f/32), and still things weren't fully in focus. That's shallow... right?

Looking at it a bit differently, it makes me think that we have the whole thing backwards when it comes to what we call "depth of field control."

I've seen a few equations for calculating depth of field, but I like the one of Wikipedia best. It states that your depth of field is determined by your aperture; your distance to subject (squared); and the circle of confusion (which is a bit more complicated than I want to explain here), all divided by your focal length (squared). For those of you who really, really hated math, squaring something means that you're multiplying it times itself. From what, you can deduce a hierarchy of sorts: distance to subject and focal length have a heavier impact on depth of field than does aperture. Anyone who has had the pleasure of shooting telephoto lenses and ultra wide lenses probably has a gut feel for this, even if they didn't know the equation.

Here's why that equation matters, and why I think we have it all wrong when we talk about depth of field "control" - larger sensors use longer focal lengths for the same angle of view that we do. Or, put another way, we on µ4/3 can get away with shorter focal lengths for the same angle of view.

Let's talk some concrete numbers to prove the point. Compared the m.Zuiko 25mm f/1.2 with the Fujinon 63mm f/2.8 - both of which have about the same angle of view - we get these results, calculated for focusing on a subject five feet away:

µ4/3 25mm f/1.2: total DoF of 0.43 feet

GFX 63mm at f/2.8: 0.4 feet

And comparing them at their smallest apertures:

µ4/3 25mm f/16: total DoF of 8.61 feet.

GFX 63mm at f/32: 5.6 feet

There's less than one inch in the depth of field when both lenses are at their widest, but we have a difference of three feet when they're at their smallest. Put another way, if you subtract the two, I have a total range of 8.18 feet of possible depth of field that I can manipulate with my Olympus, and "only" 5.2 feet with the GFX. The numbers for "full frame" will be different and somewhere between the two (closer to the GFX), but the principle will be the same.

In other words, on the larger format, I think it's fair to say that there is finer depth of field control, because the DoF is not adjusted quite as much between each aperture step... but when it comes to full and total control, µ4/3 offers us more total control. This may be arguing semantics, but I think it's important. It's definitely a feature that is overlooked and under-appreciated.

When I realized that, I had a new appreciation for Olympus' vision for the 4/3 format. It's true that there's a shallow DoF craze these days, brought about by many people viewing photos small, probably among many other things. But if what I've heard from the older photographers is true, up until recently there was generally a struggle to get more depth of field, not less. µ4/3 gives you that capability. While you can still go more shallow with other systems, it's not like µ4/3 is a slouch in this regard... with the added benefit that even shooting wide open can still yield some nice backgrounds (depending on what you're after).

It goes a bit beyond the depth of field, though. When you think about the "crop factors," the angle of view changes less per millimeter of focal length with the larger sensors compared with µ4/3. My Fujinon 32-64mm f/4 - currently one of my favorite lenses on the system - is essentially a 12-25mm lens. (The µ4/3 Leica DG 10-25mm f/1.7 compares favorably in size, and adds a bit extra in the aperture and focal length.). A lens like the m.Zuiko 12-100mm f/4, which is a bit of an anomaly in the photography world, would be a 32-250mm on the GFX system. And Olympus' 12-200mm? 32-500mm on the GFX to match those angles of view. The µ4/3 claims of smaller and lighter lenses didn't exactly pan out, particularly when considering the wide-aperture lenses, but that versatility has led to some interesting developments that just wouldn't be practical on other systems.

The psychology of the masses

Here's something I found fascinating: how would other people perceive the GFX system? Having been a 4/3 and µ4/3 shooter for my entire photography life (a little over a decade, now), I was used to hearing 4/3 dismissed because larger sensors are better. Surely the GFX would be exalted and admired by everyone, right?

Amazingly, it wasn't the case. I'd read comments on reviews from "full frame" shooters who claimed that "there's not a big enough difference in size" to result in a significant difference, among some other justifications that sounded somewhat insecure. But consider this: APS-C is 164% larger than 4/3, and while it's less common than it used to be, people still claim that APS-C is superior because... well, size I suppose. Yet "mini medium format" is 167% larger than "full frame."

To be fair, I don't think I've noticed the same person making both claims and trying to have it both ways. Such a thing doesn't defy reality, though. It does lead me to believe that for what ever reason, the photography community at large has settled on "full frame" as the general standard of what constitutes the best, and many excuses and reasons are made to justify that - even if they become a bit illogical, as someone who has now heard it from the ends of camera systems with sensor sizes both smaller and larger than "full frame."

Ah, and one more thing: it seems that a lot of people who haven't used it before criticize the 4:3 aspect ratio. For me, coming from µ4/3, it was one of the draws of medium format over "full frame" and APS-C...

HDR processed in Aurora HDR, range of -1 to +1 (limitation of the camera at the time), taken with an old Olympus E-PL1 and the Zuiko Digital 7-14mm f/4 (on a crummy Panasonic adapter that didn't do the lens justice). And with this photo, I've hit the 10-photo limit per posts.

The post is long enough. Can we just get out and shoot already?

In the week or so before deciding to write this post, I came across a bunch of old photos that I hadn't gotten around to processing. They were taken years ago with my E-3 and the camera with the then-best sensor, the E-PL1 (yes, marketed as the absolute bottom entry-level camera). Processing those old files with modern software was interesting. I could get more out of the RAWs than I would have in the past, but I also recognized that my E-M1 MkII (and yes, GFX 50S) offered much, much more headroom for highlight and shadow recovery. And yet the old photos weren't bad; in fact, while I had repeated it enough times that I fully expected every single sky to be one big, blown highlight, I was pretty impressed with the colors and tones that I was seeing. Not enough to go back, and not enough to not recognize just how far photography gear has come, but enough that I lingered on a number of those old photos, and spent time processing them.

So here's the conclusion, for me: there's a difference in image quality, but it's not the earth-shattering difference that I was led to believe I'd find. (And if people were asking me about how to spend their own money, I'd say that it's not worth the $3,000-4,000 difference.) I favor the GFX 50S at present, but I have absolutely no reservations about picking up my E-M1 MkII and leaving the 50S at home. Occasionally I find myself being lead to believe that there's a huge difference, but I can usually bring myself back to Earth by using my Olympus a bit more and proving to myself that the differences are really not as big as everyone seems to think.

The caveat to this is that I have read opinions from others who do see a big difference. Maybe it's the way they process; maybe it's the screen they're using; or maybe it's just the way that they're tuned into certain things that I am not. I absolutely don't mean to demean anyone who tried out a different system and reached a different conclusion. But again, I wonder about the psychology of it all...

My advice to anyone wondering about what system is best, and feeling agony over perhaps only knowing Olympus? Life is short, and if you can afford it, why not try out a system with a larger sensor? See for yourself what you're missing, if anything. If you can't afford it, or if you have tried another but you keep finding yourself feeling disgruntled, then my advice would be to limit your time on photography forums - specifically, gear forums. There's no greater way to feel pleased with your camera than to get out and use it.

And with this, perhaps I've finally reached the point where instead of spending money on bodies and lenses, I'll spend more on prints. Escaping the cycle of gear lust... it's a bit like photographer's nirvana, don't you think? But I'm still looking forward to the upcoming Fujinon 45-100mm f/4 OIS, and the m.Zuiko 150-400mm f/4.5 Pro. So perhaps I still have some more to go before I reach enlightenment...

Hope you all enjoyed reading, that you like your cameras at least a little more than before (especially for those who were disgruntled), and that we can all appreciate that there's a lot more to the camera than just the sensor.  May you have time to get out and take some photos this weekend!

 Velocity of Sound's gear list:Velocity of Sound's gear list
Olympus E-M1 II Fujifilm GFX 50S Olympus Zuiko Digital ED 50mm 1:2.0 Macro Olympus Zuiko Digital ED 35-100mm 1:2.0 Sigma 150mm F2.8 EX DG Macro HSM +10 more
Nikon D700 Olympus E-3 Olympus E-M1 Olympus PEN E-PL1 Sony a9
If you believe there are incorrect tags, please send us this post using our feedback form.
ForumParentFirstPrevious
Flat view
Post (hide subjects) Posted by
NCV
NCV
NCV
NCV
NCV
NCV
NCV
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow