Just Published - Mirrorless wars

[Kevin Raber]

I just published a short Rant/article about the looming Mirrorless Wars.  It's going to get interesting.  Check it out HERE

[jaapv]

Interesting read - but where are Leica SL and CL?

[Kevin Raber]

Did you watch the video?

[rdonson]

Good job with the video!!!   There's a number of people who think/hope that Canon will use the EOS mount on the new mirrorless 35mm camera.  That seems absurd to me.

[jaapv]

Did you watch the video?

Will do

[Jonathan Cross]

Thank you Kevin - a really interesting video with some very possible predictions.  To me the elephant in the room is cost.  With respect to Canon and Nikon, I suspect that a significant number of people have already jumped ship and gone mirrorless elsewhere.  So what will be their market, professional probably, but will that generate the volume?  If the volume is not there they may try to recover the development costs through high margins and prices.  That may cause even more to think very seriously about changing.


I still have a Canon 5D3 and mainly use it with a macro lens and a 100-400 Mk2 lens.  My use of the macro is dropping as I am finding Fuji's focus stacking on my X-T2 exciting.  For travel, the 5D3 is too heavy and bulky, so I use my Fuji, and others may well have gone to Sony, Olympus etc already.  I print up to A3+ (13x19") and 24MP is enough for me.  I cannot justify 40MP or more.

You point out at the end of your video, have fun.  I am, so Canon and Nikon will have to do something remarkable to get me to spend money on new kit rather than on taking me places to have fun taking images.

[Paulo Bizarro]

I am very comfortable with Sony A7 system, that I have been using for nearly 4 years now. For landscape, travel, I don't need anything else: a couple of cameras, plus 3 lenses (21, 35. 70-200), all fit in a smaller backpack.

Whatever Canon comes up with, or Nikon, will not make me change.

[kers]

I read:'
Canon is still quiet, and I believe they are working on something big. Wisely, they haven’t said anything yet. But Canon will have to use what I call the nuclear option to enter this market successfully. They will need to top everyone else in features, lenses, and performance."

The problem with Canon is- for years- that the sensors are not as good as Sony's.
It would be a big surpriose if they can compete with sony's BSI sensors especially since they are designed with mirrorless camera in mind.

[michaelsh]

I am very comfortable with Sony A7 system, that I have been using for nearly 4 years now. For landscape, travel, I don't need anything else: a couple of cameras, plus 3 lenses (21, 35. 70-200), all fit in a smaller backpack.

Whatever Canon comes up with, or Nikon, will not make me change.

For me it's the DP2 and the DP3 Merrill - two cameras with two outstanding lenses already attached. Not to mention the Merrill (Foveon) sensor. And at bargain prices too - around 500 Euros each. And I don't need a backpack (or tripod) 

[32BT]

I read:'
Canon is still quiet, and I believe they are working on something big. Wisely, they haven’t said anything yet. But Canon will have to use what I call the nuclear option to enter this market successfully. They will need to top everyone else in features, lenses, and performance."

The problem with Canon is- for years- that the sensors are not as good as Sony's.
It would be a big surpriose if they can compete with sony's BSI sensors especially since they are designed with mirrorless camera in mind.

Plus Canon already tiptoeed in the mirrorless market which sort of negates a nuclear option.

They probably need a nuclear tech jump although i find it hard to imagine what tech jump that might be considering the diminishing returns of for example DR relative to current pro offerings. They also need to open up their mirrorless lensmount.

[ErikKaffehr]

Hi Kevin,

Some nice speculations. Some comments:

16 bits are not needed as long as the individual pixels don't have DR exceeding 14 EV. The DR is defined as full well capacity divided readout noise. Most modern sensors are close to that limit. But if we make the pixels smaller, per pixel DR will decrease.

So, keeping pixel size, we would need 15 bits, sooner or later. Going to smaller pixels, 14 bits will be enough. The next generation of MF sensors are 14 bit wide, according to Sony spec sheets. There may be artificial ways of making them 16-bit, of course.

Interestingly the sensor used in the Phase One IQ3100 MP has a DR close to 15EV and the IQ3100MP actually have a 16 bit raw format, older Phase One backs were just 14 bits. That was a good engineering choice, but quite a bit of fake marketing.

What that means for mirrorless? It means that we may see 16 bit pipelines on large pixel cameras, like the Sony A7s#, but I would think that the A7s is sort of dead, as I guess that 8K is around the corner...

Of course, not everything that makes engineering sense translates to sales/marketing sense, but I would think that engineers still make a lot of the calls at Sony.

Would Nikon or anyone else make non 3:2 aspect ratio 43 mm diagonal sensor, that would be really nice!

Best regards
Erik

I just published a short Rant/article about the looming Mirrorless Wars.  It's going to get interesting.  Check it out HERE

[TommyWeir]

The thing with Canon is they're not just a camera company.  That's one thing Nikon has in it's favor, it has more skin in the game proportionally.  I've a feeling that out of the two of them, Nikon has the biggest risks to take, and as such could well leapfrog to the top of the pile.   A mirrorless equivalent to the D850?  With an intelligent lens adaptor as well as a solid roadmap of new lenses?  That would be a thing to see.  The large lens mount is intriguing...

Happy with my X-T2 and exploring the Fuji lenses.  I'll abide till I jump to MF.

[farbschlurf]

Yes, interesting times. I think we are also going to see many things regarding the "software" of the cameras. EyeAF for sure only is a beginning. AI of any kind in cameras will be the next big thing.

[Alan Goldhammer]

It will be really interesting to see what Nikon does in terms of new lens designs vs. support for existing lenses.  To me the whole point of mirrorless is to reduce both size and weight of the camera/lens combination.  The longer Nikon teles and some of the fixed aperture zooms are quite heavy as well as large and this would potentially negate the use of a mirrorless system. 

[Rory]

16 bits are not needed as long as the individual pixels don't have DR exceeding 14 EV.

Is this true?  What about Nikon's trick of scaling the amount of photons/electrons in the A/D process?  I thought it was the number of steps, not the range, that defined the bits per pixel required.

[amolitor]

Technophiles of all stripes tend to confuse "what it would take to make ME buy it" with "what it would take for the product to be successful" and this is completely wrong. Technophiles (most people on LuLa and other photo forums) are early adopters, not majority buyers. Sony has made excellent headway into the early adopter market, and it is clear that both Canon and Nikon have decided that now is the time mirrorless goes mainstream and they intend to cut Sony off at the knees.

The challenge for Canon and Nikon is not about out-technologying Sony, but to translate their brand loyalty to a somewhat different product. They need to make people believe that it's still a Nikon camera, still a Canon, with everything it is that makes those cameras desirable, useful, good quality, good value, whatever characteristics it is that makes people buy a Nikon or a Canon today, but with a new and more modern idea of picture-making.

Trying to out-feature, out-tech the other guys is a mistake, and the opposite of what you want to do in this case. You end up with a product that is buggy and unreliable, because you let your engineers get ahead of themselves, and at the same time you lose some of the brand familiarity. Canon and Nikon are the stalwarts, for the most part their products Just Work and are both relatively easy and enjoyable to use.

The story is that this technology is now mature, proven out, ready for You the ordinary person, and We're The Company to bring it to you, because we are the stalwart.

[32BT]

Technophiles of all stripes tend to confuse "what it would take to make ME buy it" with "what it would take for the product to be successful" and this is completely wrong. Technophiles (most people on LuLa and other photo forums) are early adopters, not majority buyers. Sony has made excellent headway into the early adopter market, and it is clear that both Canon and Nikon have decided that now is the time mirrorless goes mainstream and they intend to cut Sony off at the knees.

The challenge for Canon and Nikon is not about out-technologying Sony, but to translate their brand loyalty to a somewhat different product. They need to make people believe that it's still a Nikon camera, still a Canon, with everything it is that makes those cameras desirable, useful, good quality, good value, whatever characteristics it is that makes people buy a Nikon or a Canon today, but with a new and more modern idea of picture-making.

Trying to out-feature, out-tech the other guys is a mistake, and the opposite of what you want to do in this case. You end up with a product that is buggy and unreliable, because you let your engineers get ahead of themselves, and at the same time you lose some of the brand familiarity. Canon and Nikon are the stalwarts, for the most part their products Just Work and are both relatively easy and enjoyable to use.

The story is that this technology is now mature, proven out, ready for You the ordinary person, and We're The Company to bring it to you, because we are the stalwart.

I don't think that out-featuring or ubertechnology is quite what we mean here. Certainly one doesn't need the big players to outdo one another, they need to be ergonomically superior, durable, and relyable. But a new mount means buying into a new set of more modern lenses, and considering that Canon's offerings are perceived as, well, long in the tooth for lack of a better term, people will perceive this as a decision moment.

Obviously, the low end just runs on the perceived superiority on the high-end which means that if Canon is gradually thought of as losing the technology race (which one way or another this game has become) the low-end will switch.

Whether pros will swich is obviously dependent on an entirely different set of criteria.

[Les Sparks]

Many of us, I know I do, have much more invested in lens glass than we do in cameras. So when I looked to upgrade my camera a few years back, I ended st sticking with Canon because going to whole new lens system was too expensive. I accepted a few less features that I wanted but didn't really need. Now if I have to change lens systems, camera features dominate my choice, assuming that the lens system choices meet my needs. The economics of staying with Canon are gone.
Right now the one thing that Canon has going for it is the huge line of lens it has. A whole new camera with new mount will mean s whole new lens system which will take awhile to develop and make.
I know that Canon will probably come out with adapters for the old lens systems, but that's a sub-optimal solution because it'll have to add one more thing to stick on the lens and probably will compromise some of the lens features (auto focus speed, accuracy, or whatever). 
I assume that the same things apply to Nikon.

[digitaldog]

Excellent video, I learned a lot.

I'm quite new to mirrorless, a few months with a new Sony to test the waters (RX-10 IV) coming from Canon 5DMII and lots of lens. I'll keep the Canon for awhile but I'm not missing it. My only 'learning curve' is getting used to the viewfinder. As Kevin talks about, there are advantages of it being based off the chip but after 40+ years of viewfinders that are not 'video' output, I'm finding it an adjustment in what I see, get and wrestling a bit with how that image seen there updates as I move the camera around for framing etc. I'm a bit too old in my ways to even use the LCD but I can see how in some cases, it's going to be much easier to shoot than using a viewfinder. Anyway, I'm happy with the very modest investment in this new technology. I may go full frame in the future and I can see just dumping all the Canon lens for newer technology that's better 'mated' to a new body.

As for the 14 bit vs. 16-bit, my understanding is that outside of massively more DR coming (wouldn't that be awesome compared to more MP), more bits here is mostly marketing than useful but I'm open to a better understanding of all this.


I also have to wonder how many need a 50MP let alone 100MP camera but again, I can see where it is necessary for those making huge prints. 

[D Fuller]

It will be really interesting to see what Nikon does in terms of new lens designs vs. support for existing lenses.  To me the whole point of mirrorless is to reduce both size and weight of the camera/lens combination.  The longer Nikon teles and some of the fixed aperture zooms are quite heavy as well as large and this would potentially negate the use of a mirrorless system.

For me, size and weight are much less important. A little smaller than DSLRs is nice, but too small is not great either.

Excellent performance and ergonomics are more important to me. On-sensor focussing weems to have the potential of providing excellent auto-focus, but great lenses have to be big enough to accomodate the optical elements and motors. For professional cameras, that's OK with me.

I expect that Nikon's first mirrorless will be a tour-de-force mirrorless D850-class camera: high res, high dynamic range, lots of computational tricks like focus stacking and eye-focus. But what will the second and third be? I have a couple of thoughts (but absolutely no what Nikon will do.)

One possibility would be a less-expensive, smaller, lighter camera targeted at travel. Think full-frame Leica CL. Not really designed for ultimate performance, not really matched to long lenses, but very well-balanced with wides and the shorter zooms people most often travel with. Perfect for street photography, but only adequate for Formula One.

Another possibility, if Nikon wanted to make a splash in the video market, would be a low-light mirrorless that bests the Sony A7S. I don't expect this, but if they wanted to move aggressively into that space, a camera like that would do it. If it had a PL adapter and played nicely with non-Nikon wide lenses, that would be even better.

A third would be an expensive, medium-res, high-performance camera in the D5/A9 class with excellent weather sealing, class-leading data rates and autofocus and the like.

In the mediate-term, a lot depends on how clearly they set their design goals for the first several models. One camera for all doesn't seem like a strategy for success.