Yes, Micro Four Thirds is Probably Dead

By Jonathan Timar
Leave a Comment

I’ve started and restarted this post several times. Actually, I’ve finished it a few times too. Mostly in the form of a reply to a comment under some random blog post before reminding myself of the futility of such conversations and scrapping the whole thing.

But even though this topic is entirely unimportant, it does interest me, and serves as a necessary distraction from all the real problems of the world that I can do nothing about. Plus, I think it’s good to record these things for posterity.

So let’s start at the beginning. Many, many years ago, when digital photography was still in its infancy, I used to read a wonderful website called Photo.net. I won’t bother to link because the wonderful website is long gone and in its place is a fairly terrible membership site that is, uh, fairly terrible and barely functional.

But in the good old days it was one of the biggest photography websites on the net, and was run by a man named Philip Greenspun, who I believe taught at MIT. He wrote many fascinating articles about photography, interspersed with his film photos. These were film photos when film was just the way you took photos, there was nothing ironic or hipster about them. I would re-read these articles and live vicariously, as I was still a kid and could not afford photo gear at the time.

One article that forever stuck with me was on the subject of digital SLR cameras. At the time, they had not yet surpassed the quality of 35mm, but they were getting close. Philip Greenspun expressed his disappointment that the trajectory of digital cameras was to ape the size and form factor of 35mm film cameras. He reasoned that digital would allow much higher resolution from a small surface area (sensor) so why wasn’t anyone making a compact interchangeable lens system that could take advantage of this?

Philip’s arguments stuck with me, so when someone finally did make that system, in the form of Micro Four Thirds, I bought into it immediately. I started with the Panasonic GH1. Later, I went all in with an Olympus EM-5 Mark II, and after completing my collection of “Pro” f2.8 zooms, I added an E-M1 Mark II to my kit as well. I was a happy and content photographer!

Let me say this: The newer M43 cameras easily rival the quality of medium format film. Years ago, while the forums at DPReview were full of people whining about the noise performance of this camera or that camera, I’d noticed that even modest compact digital cameras where providing better images overall at 7-8 megapixels than similar film cameras, and the noise was actually less than 35mm film! But no one cared.

Philip Greenspun’s belief that a compact system camera was the mass-market camera of the future relied on the assumption that 35mm film would remain the benchmark, but that’s not what happened.

In the film days, people would take their pictures and then drop the film off at a drug store and come home with a packet of 4X6 prints. Average Joe rarely even looked at the negatives and had no clue how much detail, noise, contrast, or dynamic range they contained. The machine at the drug store took care of it all, and all he saw was the final product.

But suddenly, digital changed all that. Suddenly, the primary way of viewing photos was on a computer screen, and it was possible to zoom into the pixel level and see every last detail, every last speck of noise. People became perpetually dissatisfied and camera companies responded by giving them what they wanted, more and more and more resolution, better and better quality, and the benchmark shifted. 35mm was not the standard against which digital was compared anymore.

Okay, okay, but what does this have to do with Micro Four Third cameras?

A lot, actually. You see, for a long time, large 35mm sized camera sensors were extremely expensive. For that reason, smaller sensors became the most commonly used (APSC) in all but the most expensive cameras. For whatever reason, the camera makers were not interested in building new mounts around smaller sensors at that time, so their solution was to put smaller sensors in large bodies designed for a 35mm frame. This meant that up until very, very recently, most digital SLR cameras were quite a bit bigger than they really needed to be.

Micro Four Thirds was the first camera system built for digital from the ground up. It was the first to dispense with the loud, bulky mirror mechanism. It was the first to offer in-body image stabilization. It was the first to provide full time live view. It was the first at a lot of things, but its central gamble was that it could leverage the ability of digital to pull more detail out of a smaller area, allowing for smaller lenses and a much smaller, lighter system overall.

And for about a decade, it did just that.

But recently, it’s run into a problem that could not have been foreseen.

The first is, as you might have guessed, is smartphones. As rapid as the rise of digital SLR cameras for the masses was, the decline has been even faster. Overnight people went from scrutinizing their family photos and snapshots on their computer screens, to taking and sharing everything on a smartphone with a screen about 1/10th the size. The photo quality of smartphones isn’t nearly as mind-blowing as the hype would suggest. It’s still pretty lousy compared to a good quality stand alone camera, but it is certainly good enough for Instagram. It’s definitely better than the majority of compact digital cameras from a generation ago even though the sensors are even smaller (computational photography works wonders), and in some respects, it beats the previous mass-market consumer format: 35mm film. Once again, most people are only ever seeing the end result or their photos, this time it’s a 5 inch screen instead of a 6 inch print, but the effect is the same (and a bit ironic): regular people are taking more photos than ever before, and never have they cared less about owning a camera.

This means that the only people buying cameras now are pros and enthusiastic hobbyists. The market for low-end consumer cameras is dead, and it’s hard to imagine a scenario where it comes back.

Simultaneously, the high-end of the market has been getting cheaper and smaller. A few years back, Sony dropped full-frame mirrorless cameras on the market, and packed a 35mm sensor into bodies that were not much bigger, and in some cases smaller than M43 cameras. Yes, the lenses were still bigger, but for many it seemed not to matter. Now both Nikon and Canon have jumped into the mirrorless world with both feet.

    Keep In Touch

    Get the latest posts by email, plus exclusive content, offers, and free stuff just for subscribers.

    (required)
    (required)

    The result is that Micro Four Thirds is facing pressure from the top and the bottom. Smartphones all but erased the market for a compact and portable carry-around camera, and cheaper, smaller “full-frame” cameras are erasing much of the appeal of Micro Four Thirds for the high end as well.

    Let’s use a picture to illustrate the problem, because a picture is worth a thousand words, especially when something is tricky to explain.

    Olympus Micro 43 Versus 35mm
    Olympus Micro 43 Versus 35mm

    On the left you have an Olympus OMD E-M1 Mark II with a 40-150 f2.8 lens attached. It’s an awesome combo that I own. On the right, you have the upcoming Canon EOS R5, which is to be Canon’s first 35mm mirrorless camera with in body stabilization as well as some mindbogglingly awesome video features. Attached is the Canon RF 70-200 L lens. According to CameraSize, the EOS R5 is 99mm high by 138mm wide by 84mm thick. This makes it only 8mm taller, and 4mm wider than the Olympus. Practically nothing. It is about 15mm thicker, so there’s that. As of now, we don’t know the weight, but the 5D Mark IV that the R5 replaces weighs 800g and the little brother EOS R weighs 660g, so let’s split the difference and guess 730g. That makes it only 70g heavier than the Olympus. Uh oh!

    What about the lenses? The Olympus 40-150 weighs 760g. The Canon RF 70-200 2.8 L weighs 1070g. That’s a significant 30% or so heavier, but it include OIS whereas the Olympus does not, and, as you can see by the picture, it’s a bit thicker, but also a bit shorter. Take take it all together as the Canon combo is only 25% heavier than the Olympus combo, and either combination is actually going to take up about the same space in your bag. Whoa!

    Now you might be thinking that 25% is a significant amount, and I’d agree if it only represented a 25% increase in image quality, but the truth is the Canon combo is capable of much more than 30% better image quality because of that dreaded word: equivalence.

    I hate talking equivalence when it come to exposing pictures, because it doesn’t matter from that perspective. Exposure is the same no matter what because every camera is designed to produce an image of similar brightness from the same settings. But when it comes to comparing cameras it’s a different story because what the camera does behind the scenes to achieve that brightness varies greatly.

    An M43 sensor is roughly 1/4 the image area of a 35mm sensor, which means it gathers 1/4 as much light. This is physics, there is no way around this. I’ll forgo a further technical explanation here because others can and have explained it so much better. Just realize that all else being equal, the Canon combo will give you 1/4 the noise at the same ISO and twice as much control over depth of field.

    All else is not equal because the R5 is expected to have 45 megapixels of resolution, which is more than double what the best M43 cameras offer, and perhaps ever will offer given the recent announcements by Olympus and the precarious position of Panasonic.

    Why does this matter? Because as is often the case in the photography world, the conventional wisdom is flat out wrong: higher megapixel cameras actually produce less noise, not more. Anyone who says otherwise is failing to compensate for output size. All else being equal, yes, a 45 megapixel image at full size will have more apparent noise than a 20 megapixel image from the same size sensor BUT if you resize the 45 megapixel image to 20 megapixel, it will actually have less noise and also be much sharper than the native 20 megapixel image. Incidentally, this extra resolution also makes up for the difference in field of view between the Canon 70-200 and the Olympus 40(80)-150(300), because you can crop down to 20 megapixels and have the same effective view.

    So for a negligible size penalty, and a only a 25-30% weight penalty, you gain four to five times better noise performance (remember the benefit of down-sampling extra megapixels), twice the creative control over depth of field, and a lot more resolution.

    But wait, there’s more. You can also put some of the high quality, slower lenses on these cameras and the size and weight penalty drops even more. The Canon 24-105 f4-7.1 weighs only 395g. It’s 88mm long and its diameter is 76mm. Its M43 equivalent would be a 12-52mm f 2-3.5. It’s half the price and basically the same size and weight as the closest equivalent lens, the 12-40 f 2.8 from Olympus.

    So you have extreme flexibility with the Canon RF, you can take a 25-30% weight penalty and gain 400-500% the image quality and creative control, or you can pick a slower lens and have something that’s practically the same size as an M43 setup with equivalent or better results.

    I have loved my M43 system to death. I think my Olympus OMD cameras are some of the highest quality products I’ve ever owned. I certainly think they are better looking than Canons, whose smooth, rounded design aesthetic I’ve never cared for.

    But when I actually run the numbers and do a fair comparison, the M43 advantage doesn’t look so advantageous anymore. In summary, the low-end of M43 is dead because of smartphones and consumers that don’t care about standalone cameras anymore, and the high-end of M43 its size and weight advantage has been all but eliminated, and even its cost advantage is looking unsustainable.

    If you agree or disagree, please leave a comment below.