Canon is finally making it’s first steps toward producing a professional, mirrorless option for those dedicated to their brand. While the EOS M5 and EOS M6 marked their first trip into the mirrorless universe, their highly anticipated professional mirrorless option will be full-frame and utilize a fully dedicated sensor.
Based on an "internal presentation of Canon’s 2018 roadmap" Canon's first full-frame Mirrorless Camera will feature an entirely new, dedicated sensor. It may not be earth shattering news that the camera will feature a sensor created specifically for that body, but it does bring up a few questions and a few possibilities that make this news intriguing.
First of all, Canon isn’t a brand that takes these new ventures lightly. For all of the criticism we throw their way about being “behind the times” or slow to develop, they’ve made a name for themselves as the brand that releases cameras only when they’re ready. They’ve never been pressured into releasing anything early.
To me this signals that the camera might be quite a bit of a ways away from a real announcement/release. Developing a sensor, firmware, and processing to seamlessly communicate is possibly the most complicated part of an engineer's job. It takes time, and Canon is known for taking time. This is something that's supported by Canon saying that there won't be an announcement until at least August 2018. And if I'm being honest, that’s probably a good thing.
Also, this probably means that the features and specifications of this mirrorless camera will actually compete with the competition (Sony a7SII). That may seem like a fairly obvious assertion, but I think that idea would have been undercut if Canon had decided to instead adapt a DSLR sensor to this use. Mirrorless shooters are often looking for different things in a camera than DSLR shooters and this sensor should reflect that different perspective. This, in my mind, is only a good thing.
Finally, and in a very general way, this makes it seem like Canon is taking the mirrorless market seriously (FINALLY). The EOS M5 is a great little camera that takes good images and the EOS M6 is fine. There’s nothing wrong with either of them. But, as a mirrorless cinematographer, it never felt like Canon really took me seriously. It always seemed like they were pushing their cinema-style bodies on me, instead of adapting to the changing landscape. Maybe it’s a bit premature, but if Canon is making a dedicated sensor for this camera, I think it’s a great sign that maybe they’re actually going to make a splash in the mirrorless market.
Maybe.
I hope.
[via Canon Rumors]
If it were a curved sensor, that could be significant. It could allow for some noticeably smaller lenses. The downside, of course, is that it would render all lenses in the Canon ecosystem, totally useless (from my understanding).
Bob... I'm going to need you to cease and desist with regards to using my last name. Thank you, sir.
:-D
That said, I agree, Bob with the awesome last name. It would be nice to see a company really embrace the notion of SUPER high quality lenses that are small and light and do so at a price the average consumer for that sensor size can afford. The irony could end up being that the system ends up the same size/weight as the APSC M line... OOOOHHHH... Gimme a FF mirrorless with a 35/1.4 that's the size of the EF-M 22/2! If a curved sensor can do that... WOW! (by the way, I realize that's probably not possible - but a guy can dream)
I'd love to see some more curved sensor experimentation, I just don't believe Canon will be the one to really embrace it before it's proven. For better or worse, they're slow to change.
Curved sensor makes a zoom lens almost impossible to make. And Canon sure won't do that. Right now, out of their 7 or 8 lenses for ef-m, all but two are primes.
Various curved sensor reports have understated logical implications that comport with physics; 1. Any curved sensor is optimal for one focal length, problematic for others so, 2. it introduces lens design problems especially for zooms but, 3. is enticing for a fixed focal length lens especially where the sensor is very small because the curve is then very small.
The goal (respecting physics) would therefore likely be better results in phone cameras, or other fixed lens small sensor cameras. What is missing in the various reports is a fair sense of the limits, hype being the stock & trade of a lot of online publishing.
~~
I'll play Devil's Advocate against your wish just for fun, and offer that I'd love to see less "curved sensor experimentation," because physics says it's not the area to be doing research for high end cameras with big sensors and huge lens systems. In a world of budgets with limits, I'd prefer available research be put to better use. For instance a waterproof 100 megapixel EF mount mirrorless Canon 5DSr II.
But the entire reason Canon makes cameras is to sell lenses! The lenses are where they make the money. So my suspicion is they'll do what ever they can to keep EF compatibility. Maybe I'm wrong and they want to sell a whole new set of lenses :)
Yeah, what would sell MORE lenses than releasing a whole new mount that's completely incompatible with everything you've ever made? EVERYONE has to start from scratch!
So... mirrorless users are often looking for different things in a camera... in terms of sensor technology? I'm not sure how a Sony A7SII sensor would perform differently in a DSLR body than it does in its current state.
I think it really comes down to size and processing. Clearly a sensor can perform separate from the processor, but if they work and are engineered in tandem, it stands to reason they'll be more effective. The a7Rii and the a99 ii seem to share the same 42mp sensor, yet they do perform very differently.
Of course... the processor (duh me). So it's the sensor and processor being developed in tandem that would make the difference. Got it :)
"They’ve never been pressured into releasing anything early."
(spits coffee all over computer)
How long have you been into photography, sir?
"While the EOS M5 and EOS M6 marked their first trip into the mirrorless universe,"
...Not sure where the EOS-M, M-3, and M-10 fall on the timeline, but I'm pretty sure they all came before the M-5 and M-6.
Haha yeah, pretty much this.
You're absolutely right that they exist. I was more speaking in terms of professionally usable options, but I can see how that's not clear. I do appreciate you keeping me honest though hah
Yea but the 1080p video on this thing is gonna be EPIC!!
With a built-in "softening filter", like the 6D II :-). Chicks love it, man! No wrinkles, no eye begs! EPIC indeed...
As Tony Northrup is especially astute in pointing out when he reviews the various top end cameras; a stop or slightly more of extra dynamic range is only a piece of the image puzzle. One should not make tech stats the single issue, because it's not. Minor differences in stats are sales tools, while above average photography is a complex practice concerning optical variables multiplied by artistic vision. What you know about art history should be more useful to a photographer than some extra DR.
There are respectable reasons for buying into a number of brands, a good example being Northrup's interesting choice (last I checked) of a Fuji (I forget the model) when he wants a walking around camera for traveling, because it is more instinctive to use. This aids shooting the street more than a spec. (For that reason people loved Leica.)
If you found yourself in an extraordinary landscape and all you had was a lousy 5DSr and a damned version II 24mm TS, you'd be lacking a tad of DR, but amiss to not work the scene telling yourself you'd get the 1.5 stops back in Photoshop if you had to.
I'm so excited. I'm hoping it will only come in 2019 so there will be no excuse not to include low bit rate, soft 1080p video. The kind the 6D II shoots.
You see this camera will NOT be DESIGNED FOR video. This will be DESIGNED FOR stills. If you want video, log around a $10K cinema body as well. That's what's DESIGNED FOR video, damn it!
"While the EOS M5 and EOS M6 marked their first trip into the mirrorless universe"
I guess the M, M2, M3, and M10 don't count for some reason. Of course, they were mostly terrible, while the M5 and M6 were half decent.