The Next Full Frame Cameras

Oh, the headline got your attention, did it? 

There's only one "known" full frame mirrorless camera coming that we can talk about: the Sigma L. That's going to be a 20.3mp full frame, L-mount camera that arrives some time in 2020. 

Sigma will tell you that it's a 60.9 million something sensor, but in terms of the output, it's really 20.3 million final pixels. The thing is, it's not a Bayer sensor, but a Foveon design—and one that returns to the original Foveon three layer approach—so basically you're getting full luminance and color data at 20.3 million positions. 

From there, we have one "somewhat likely" full frame candidate in 2019: Sony has yet to introduce a Mark III version of the A7S. What that would be is anyone's guess, as the A7S sensor was an unexpected one-off and that's likely to be true again. 

I use the words "somewhat likely" for the A7S Mark III for a reason: keeping an S model distinguished from the new A7m3 on the video side is going to be a tough deal to accomplish. The 12mp of the A7S basically means you're using every photosite for 4K video, and thus you're getting Bayer impacts when you do (i.e., you're interpreting luminance at every other position). The 24mp of the A7m3 has better luminance integrity for detail but worse luminance noise. From a pragmatic approach, though, Sony would now be shoehorning the A7S into a more narrow space than before, and it was already in a narrow space.

Thus, any A7S update has to find a place where it delivers "better" video than the A7m3, which is a tall order. I suppose it's possible to do a 24mp 8K video sensor. Plus there's the chance for a global shutter, removing all rolling shutter issues. But you can see how Sony's got themselves into a fairly tight window here given the A7m3's highly credible capabilities. And if the true target is a "better stills" camera, then the sensor pretty much has to stay well below 24mp, or else Sony has to have invented some new technology that truly goes well beyond what current Exmor sensors can do and obsoletes their A7m3 in doing so.

So, "somewhat likely." I'm sure Sony would like to continue to have a mostly video-oriented A7 model, but they also have to steer clear of getting into the FS XDCAM professional range, too. It's a narrow, tricky, window for them to get through. 

The curious thing is how much Sony is pushing new abilities and performance into Mark III (and A9) firmware this year. That would tend to tell me not to expect a new A7R or A9 this year, but that their lives will be extended by firmware updates. 

Rumors abound about Canon introducing late in 2019 what essentially would be the followup to the 5Ds/r, but in mirrorless form: a high-megapixel count R model (technically, that should be named RP, for R Professional, but Canon's making marketing mistakes galore as they race to mirrorless; the camera they named RP should have been named RC, for R consumer). 

Given Canon's mismatched RF lens launches—high end lenses for lower end bodies—it makes strong sense to believe that those high-megapixel count camera rumors are credible, and that it's coming as soon as possible. But what's the sensor tech? Sure, it'll be dual pixel, as everything Canon fabs now sports that. But will it be something beyond? Canon has a patent on a "dual well" type approach: they really need to do something to bring forward their dynamic range capabilities, and particularly on a smaller photosite, high megapixel count, high-end camera. Put another way: Canon needs more than more pixels, it needs better pixels. 

Finally, images of Nikon early prototypes for the Z6 and Z7 emerged and generated false rumors about Nikon creating a lower-end model to take on the Canon RP. No doubt that will happen, but those photos of "prototypes" aren't what that low-end camera will look like. Nikon certainly would love to replicate the D610, D750, D850 lineup in Z, which would be a Z5, Z6, and Z7. And Nikon would love to eventually add the D5 equivalent, which I'll call a Z9 since it would sit at the top of the lineup and give them a complete line. 

But there's nothing that indicates that Nikon is at all ready to do so or that it would happen in 2019. Indeed, to a large degree, Nikon would have the same problem Canon currently has if a Z5 were to appear any time soon: no suitable lenses for the consumer camera. 

It's not difficult to guess what a Z5 would look like, though: re-use the Z6 image sensor, but lose the top OLED; use a lessor EVF and LCD; opt for SD card over XQD; lose some of the weatherstripping (but not the seam overlaps); perhaps even get rid of the tilting LCD mechanism. As I noted elsewhere, the question is whether Nikon could strip out US$200 worth of parts and manufacturing costs, though. Because that's what they'd need to do to get to Canon's RP pricing. So we also start losing things like the headphone jack, the USB 3.0 port (back to 2.0), the thumb stick, and maybe even replace some of the metal frame internals with plastic/carbon fiber.

If you're keeping track, that's a lot of stuff to rework, test, document, and set up for robotic manufacturing. Again, while I expect Nikon will get around to making a Z5, I don't think it likely to happen in 2019. 

As I wrote earlier this year, we're in a very good spot with full frame now. DSLRs built out their full frame lines earlier, and now mirrorless has mostly built out, with the next steps being reasonably predictable.

So with that in mind, here's the full list of what you can buy today (in bold) and anticipate tomorrow (non-bold). All models in basically ascending order of price/sophistication, with truly speculative models listed in brackets:

  • Canon RP, R, RS, [RX?]
  • Leica M, M10, Monochrom, SL, SLm2
  • Nikon Z5, Z6, Z7, [Z9?]
  • Panasonic S1, S1R
  • Sigma L
  • Sony A7, A7m2, A7m3, A7Rm2, A7Rm3, A7S, A7Sm2, A7Sm3, A9

So again, if mirrorless full frame is something you aspire to, there's plenty or product to choose from already, and there's enough information and informed speculation to understand what might be added to the near term choices. 

I also went through the Canon RF and Nikon Z full frame lens lineups earlier this week. The current and forthcoming Leica, Panasonic, and Sigma L lenses are relatively well known at this point, too. The Sony FE lens lineup is pretty extensive with only a few gaps left to fill. 

So. Even if you aren't going to spring for it in 2019, you should be able to begin making a decision as to whether full frame mirrorless is a place you're likely to go, and further have a pretty good idea of which of the four mounts you're likely to opt for. 

Addendum: You might have noticed that Sony has eight current and potentially nine models in their lineup. I would say that's unsustainable, and will ultimately cause Sony the same issues it caused Nikon when Nikon ran into the declining market size squeeze. 

For the big three players, there are probably only four models they need to make:

  1. True consumer (RP, Z5, ?)
  2. Sophisticated consumer/prosumer (R, Z6, A7)
  3. High megapixel/high prosumer (RS, Z7, A7R)
  4. Pro (RX?, Z9?, A9)

Pushing any further than that in a small volume market—which is what full frame has been, is, and will continue to be—starts to be extremely inefficient. Moreover, if you don't get strong parts re-use across models—and Canon is not lined up properly for that at the moment—you destroy your gross profit margin very quickly. 

Meanwhile, the temptation to keep older models on the market to offer more choices and add some parts use volume—typically driven by the image sensor commitment—is an old Japanese CES trick, but one that tends to put you in a long-term bind. All you're really doing is pushing the rock further down the path. But meanwhile, the path is getting smaller and smaller and tougher to navigate in the camera market ;~). It's a little like an addiction at some point: you finally got your Mark I numbers to where you wanted them, but that was expense of the Mark II numbers. When you get the Mark II numbers where you wanted them it was at the expense of the Mark III. But each "hit" produces less high in a contracting market. So...

Memo to Sony: get the Mark I and Mark II models off the market ASAP. Define a new entry camera (A6). Push frequent and strong firmware updates to keep the Mark III models "current." 

text and images © 2019 Thom Hogan
portions Copyright 1999-2018 Thom Hogan-- All Rights Reserved
Follow us on Twitter: @bythom, hashtags #bythom, #sansmirror