In 2003 Canon released the 1Ds, an 11.1 megapixel full frame DSLR that heralded the beginning of the end of film. In just ten short years we have:

  • A 40 MP DSLR,
  • An 80 MP Medium Format Digital Back (MFDB),
  • a 40 MP MILC (Mirrorless Interchangeable Lens Camera),
  • a 40 MP smartphone, and
  • a 20 MP Super 35mm video camera

It won’t be long before we get a 40 MP video camera. Everybody else will just have to keep up. In this article, I look at the number of camera models released over the last eleven years (2003-2013) to find out if there’s any pattern in where we’re headed.

What’s the point? I don’t know. I thought it would be fun. Or serious business. Those who invested heavily in film gear in 2001-2 will have surely felt the paradigm shift by now. Is trying to predict what will happen ten years from now such a bad thing? You decide.

The data

I’ve collected data from various sources, but mainly from Wikipedia. It’s not always easy to pin down exactly when a camera was released. Sometimes, a manufacturer announces a camera and it’s already in shelves or stores. Other times, a manufacturer announces a camera with a shipment date and months pass by.

To make my life easier, I’ve divided camera releases by the quarter. Data is from 2003 to 2013, which makes it eleven years, or 44 quarters.

I have included both still cameras and video cameras, but limited by:

  • Professional level full-frame cameras
  • Professional level APS-C cameras
  • Super 35mm-sized video cameras
  • Only the major manufacturers, but not all of them
  • Professional level micro-four thirds cameras, but not all of them
  • I’ve left out CCD or similar video cameras for ENG, sports, documentaries, etc. The number of professional-grade CCD video cameras released each year has been falling steadily for some time, but the main reason I left this group is because I can’t find reliable data over the past ten years.

There is no concrete method to my madness, but my goal is to find general industry trends rather than make statements about specific models or manufacturers.

The cameras are grouped into classes as designated by their manufacturers. For example, a Canon xD can be a 1Ds, 1DC, 1DX, 5D, 6D, etc. They are all grouped together as Canon xD. Here are the groups:

  • Canon xD FF
  • Canon xxD APS-C (the 7D gets left out)
  • Canon Cinema EOS
  • Sony NEX-FSxx
  • Sony PMW-Fx
  • Sony NEX MILC
  • Nikon Dx FF
  • Nikon Dxx FF
  • Arri S35
  • Red S35
  • Blackmagic Design Cinema Cameras
  • Panasonic GHx
  • Olympus Ex
  • Leica Mx FF

This kind of grouping obviously means some data is left out, misinterpreted or plain wrong. So, don’t take much of it seriously. If you want accurate results, conduct your own research and tests.

The total number of cameras released each year

Here’s the data (click to enlarge):

The total number of professional cameras released is about 70. If you add the brands and models I’ve left out the figure would reach about a 100, released over 11 years. That’s about 10 a year on average, and includes still and video cameras.

Here’s the chart:

As you can clearly see, more models are being released per year than ever before. The growth is exponential. What’s strange is that fewer models have been released in 2013 than in 2012.

If you want to see how still cameras differed from video cameras, here’s the chart:

Till 2006, there’s hardly any correlation between the two industries. After 2007, roughly from the time the Canon 5D Mark II became a legend, there is a strong correlation between the number of still camera models and video camera models released every year. The correlation is greater than 0.9.

This tells us, in no uncertain terms, that the future of stills and video is intertwined.

Which manufacturers release the most models?

Which manufacturer is the most prolific overall? Here’s the chart:

It doesn’t take a rocket scientist to see that Sony and Canon release the most cameras every year, and at regular intervals. Nikon doesn’t lag behind at all, but it doesn’t make video cameras. Here’s the same thing broken down:

There is one major surprise, and that is Red. The number of models Red has been able to release (they only started in 2007) is higher than Canon or Sony over the same period. In fact, Arri has released more Super 35mm cameras than Canon. Now we know who were caught sleeping.

The only manufacturer that has a strong showing in both industries is Sony.

The relationship between DSLRs and MILCs

Did the advent of MILCs ruin the pitch for DSLRs? Sure it did:

Even though the market as a whole dropped this year, it is easy to see that MILCs are the new DSLRs. Electronic Viewfinders are improving by the day, and the extra mirror assembly and size is hard to justify anymore.

The correlation between Super 35mm sensors and RAW footage

Those who shoot Super35mm, do they want RAW as well? The manufacturers certainly think so:

That’s a strong correlation if there ever was one. Expect all future Super 35mm cinema cameras to have RAW mode. Look at the trends:

  • The C500 has raw
  • The F5 and F55 have raw
  • The BMCCs have raw
  • Arri found a way to record raw internally
  • The Phantom shoots raw

You see, RAW video is easier for the manufacturer to dole out, because it reduces the thinking they have to do about color spaces, gammas, and all that crap. You asked for it, now deal with it.

Why are new models released more frequently?

Gottlieb: In the beginning the Kaiju attacks was spaced by twenty-four weeks, then twelve, then six, then every two weeks. The last one in Sydney, was a week. In four days, we could be seeing the Kaiju every eight hours until they are coming every four minutes. There will be a double event, and there shortly thereafter, three and then four…
Stacker Pentecost: And then we’re dead. I get it.

– Pacific Rim

Even a quick glance at the first table and chart should tell you that new models are being released more frequently. But this is not just because of the addition of newer companies, but also because established companies are releasing newer models at a faster pace. Here’s a chart (click to enlarge):

It shows at what rate a new model is added to each group over a quarter. By the nature of the curve (exponential), it is very clear that the rate is increasing for everyone. To find a common graph for the industry as a whole, I selected the highest and the lowest curves in the above chart and arrived at the following (click to enlarge):

Here’s the thing. The formula allows us to calculate how many cameras are released per quarter, per manufacturer (as a simple average). This year, we are at the stage where every manufacturer must release one model per group every year, to stay in the game.

Look around you, it’s already happening.

If we look merely two years into the future, we see that, by the end of 2016, a camera manufacturer must release one model every quarter, in each group! That’s only two years away!

Obviously, that kind of output is hardly sustainable. Only Sony comes close to such output, but even their business is not doing very well. The manufacturers know this better than anybody, so this is probably why the number of models released this year has actually dropped by almost 50%!

Manufacturers realize that releasing newer models is madness, so the better strategy is to release a camera that is capable of being upgraded by firmware. The Sony PMW-F5 and F55 are some strong examples. Sony drew an entire firmware update roadmap, which means they knew the potential of the cameras. Rather than putting out a new one every three months, they decided they’d deliver cameras with as much potential as possible, only limited by market pricing. Of course, Sony is able to do this because they also manufacture the sensors. Those who are dependent on third-party sensors cannot always create such long-term roadmaps.

To answer the question posed by the heading of this section, I believe this rapid rate of development is not being caused by the explosion of smartphone and tablet usage worldwide. Almost every handheld device has a camera, and many have two of them. Most modern smartphones are capable of 1080p, and the models released this year even go up to 1080p60 (The iPhone 5S can do 120fps but the resolution looks like it’s 480p). Those who want to shoot snapshots or quick videos no longer need to hunt for a camera as a separate device. However, a professional will not turn to a smartphone anyway, and our data is all about professional grade cameras. So how can handheld devices impact professional camera usage?

In fact, after considering all the data, I can come up with only two reasons that makes some sense:

  • The first is the integration of stills and video on the same device. When this happened in the professional realm (Canon 5D Mark II), all hell broke loose. The results are obvious: Canon 1DC, Red Epic/Dragon, GH4, Leica M240 with video, etc.
  • The second, probably the most important without which the first reason wouldn’t have been possible, is cheaper sensors.

Look at the first table again. Canon had a virtual monopoly until about 2006-7. Its closest rival, Nikon, managed to squeeze out a full-frame DSLR in 2007. Arri introduced the D-20, and the Red One was born, both around the same period. In the very next year, DSLR videography took hold of the entire industry like wildfire. To sum up, here are the reasons for the soup we’re in:

  • Cheaper sensors available to all manufacturers.
  • Integration of video and stills just because it’s possible in a CMOS sensor. If you integrate them, then the sensors become cheaper to manufacture.

The result? Expect all future cameras to shoot both stills and video, and we’ll no longer have separate industries at least as far as hardware is concerned.

Takeaways

What lessons can we learn from the above analyses? I’m not entirely sure, but here are a few of my thoughts:

  • Higher resolution cameras are the future, so I would side with Red in this case. Shoot all footage in 4K and above, if you want to preserve it or earn from it in the future.
  • Learn to live with electronic viewfinders. It’s going to make you Iron Man (haven’t you heard of Google Glass?).
  • Expect to buy a sensor housed in a body, or even piecemeal, and then integrated with small computing devices like smartphones, circuit boards, etc. The SI-2K Mini comes to mind.
  • Smaller and lighter lenses will win in the long run. The Sony E-mount (FZ as well) seems to be the most brilliant solution in this regard. Does this mean the others will be forced to create new mounts?
  • Once manufacturers reach these limits, they’ll turn to internal filters, even higher ISO (Canon is doing this already), higher frame rates, power savings, etc.
  • Better get good at shooting stills and video at the same time.
  • All this will need solid wireless technology.
  • Expect to buy one camera and update it over a two-year period. This is happening already. Don’t buy a camera which will be under-spec’d two years from now.
  • Expect prices to drop quickly from here on. Every manufacturer has reduced prices over the last ten years. A Sony F65 goes for $30,000 used, and a Leica M 240 costs less new than a Leica M9 did when it launched. And that’s not considering the inflation factor. Judging by what Blackmagic Design is capable of putting out, I think any camera over $10,000 will be overpriced.

Guess what’s not going to change? Audio.

What do you think? What’s your take on the above analysis?

One reply on “Camera Trends over the last Decade: What does the Future hold?”

  1. Sorry I haven’t had the time to read the article in depth, but from a first glance, here are my views:

    It’s only a hunch, but I believe more single purpose cameras will appear, instead of more integration. Video and stills are definitely two different areas of expertise. They’re OK merged in consumer devices, but don’t make much sense in professional or even some “prosumer” equipment. I expect more BMCCs-like cameras and I know a number of professional photographers who couldn’t care less about video. It seems to me the reason dual purpose cameras exist is because once your sensor can stay on for long periods (which you need for mirrorless) and your processor is powerful  enough (an area which more or less still follows Moore’s law), adding video is just too easy, but it feels bolted on even on high end Canons and Panasonics.
    I expect Panasonic to fight back the BMPCC sooner than later, maybe Sony too. I can see Nikon focusing more and more on stills-only equipment and leaving high end video behind. I can see many conflicts in Canon extensive line (they need to push video pros to higher priced models and a way to do that is cripple video on stills cameras). The same with Sony.
    Regarding lenses, I expect MFT to rule in video -E-mount second maybe- among third party manufacturers, and also expect motorized zooms for the likes of the G6 or the upcoming GH4.
    In a more distant future (5-10 years?), I foresee huge megapixel counts replacing mechanical and complex zooms. Just have a wide angle high quality prime and do the rest digitally, much like the 40MP Nokias do in the phone industry.

    I’ll re-read your data in depth at a later time and maybe I’ll add something.

Comments are closed.