I think were seeing a bit of a shift in the MF market though Shadow always from the traditional advertising/fashion bent and more towards something that might be used by landscape shooters. Cameras like the Pentax 645z and Fuji GFX are pushing more into UWA range with there lens selection, Fuji is going to have an 18mm equivalent soon for example.
I think fighting a pure resolution war against the larger format is a questionable tactic personally as I think you are going to be hitting serious limitations of a lot of 35mm lenses when you get above 50 MP. Right now these MF bodies aren't showing there true potential at 50 MP though and if a 100 MP 44x33mm sensor did happen as rumoured would clearly win that war.
44mm isn't all that much bigger than 36mm. You'd gain more if you usually shot a squarer format, but not if you typically live at 3:2 or wider.
Add to that the completely new lens lineup required (no sharing lenses between your action and your non-action body) and the likely poor AF of such cameras and it's unlikely to be anything more than a niche option.
The centre of most good lenses - zooms and primes - can already handle up to 150-200MP, or potentially more. The edges are obviously a lot weaker, but good primes (Otus, some of the Sigma Arts, various other primes from Canon, Nikon and Sony) can handle a much higher resolution sensor than is currently available.
Then there's the usefulness of having more pixels. 54MP is a significant step up from 36MP, provided all the pixels are good (and not just the same smear spread over more pixels) - it gets you to 150ppi at a 60x40" print size, or 100ppi for a 90"-wide print. 70MP, or even 100-120MP, would be a significant step up from 54MP, since it gives you more usable detail at realistic, if large, print sizes (121.5MP would give you 150ppi at 60x90", while 96MP would give you 200ppi at 40x60"). But, beyond that, 150-200MP of real detail probably just wouldn't be that useful. Sensors of that resolution would be useful, even if they outresolved the lenses, as that would reduce aliasing artifacts without the need for AA filters, but that would not require a lens that resolves more than 120MP or so.
This is achievable corner-to-corner with lenses designed for 35mm format. They just need to be no-holds-barred designs, like Otus lenses, rather than run-of-the-mill lenses designed with production cost in mind. Or they could get a huge helping hand with curved sensor technology. And these are probably still going to be cheaper to make (and lighter to carry) than lenses with larger elements needed for MF sensors.
The actual number of people who want resolution that high is though I would say not massive, perhaps significantly larger than the old MF advertising market in the past but not big on the scale of Nikon's normal sales targets. Theres IMHO significantly more to be gained from a camera that is seen to do everything well which I think the D850 is in with a good shot of being if all the rumours are true.
That was also the D750 and the 5D4. The D850 merely takes it one step further. It's a big step, though, since there now appears to be no compromise made in AF capability for those who need more resolution, but may not need the same frame rate. Previous generations always required you to choose between AF and resolution (when there's no technical connection between the two), not just frame rate and resolution (which are connected by limited bandwidth and processing speed). Now you can choose between super-fast and low-resolution, fast and mid-high-resolution and possibly (if Nikon completes the triad with a higher-resolution offering) slow and super-high resolution, all with the same AF system.
I think this camera could actually help them pickup a significant part of the higher end market personally as they seem like they might have got the drop on Canon who have stuck to the old mind-set with the 5D line of somewhat crippling to drive future upgrades and flagship sales which as they look like they've thrown absolutely everything they could at the D850.
Definitely. The 5D5 will need similar specs to compete, even if it were released in the next six months. Given that it is likely 2 years away, it will need to be even better. But Canon, being their usual arrogant selves (and only looking at the Japanese market) probably won't bite.
The 5Ds2, on the other hand, is in no danger, so long as they push the resolution/DR/IQ side of things to put it in a completely separate product category (dedicated to resolution rather than general-purpose).
Sony is probably more likely than Canon to release a competing product. Currently, they have a 24MP/20fps camera with a top-tier AF system. Continuing on from the A7r2, and incorporating improvements in on-sensor AF made since then, they will probably release a 60-80MP/5fps camera, likely with the same top-tier system. These bracket the D850 at both ends, but leaves a huge void in the middle for those who need more than 24MP, but also more than 5fps. It wouldn't surprise me if they ended the A7s line, kept the A9 as the speed/high-ISO version, kept the 'r' line as the resolution version and introduced something in the middle (around 48MP/10fps). So one model dedicated to speed/ISO, another to resolution/low ISO IQ, and a do-everything body in the middle, all with the same AF system. Their lens lineup is nowhere near competitive yet (particularly for the sports/wildlife users most likely to use a 48MP/10fps body), but the basics are there, and it can only improve.
