D850 discounts...

[ShunCheung]

I was actually vaguely thinking that, with the current price of a D700, maybe I should pick one up - since the only backup my D810 has is an F5 (which... doesn't take the same batteries or memory cards

Definitely not. What you want is a D850 to go along with your D810. :) After all, they share the same battery and SD cards.

 

There is probably sufficient space in the back of the F5 to put in an SD card, certainly a micro SD, but that won't do you much good as far as capturing images goes.

[Andrew Garrard]

Well, no, I want a D850 (and grip) to go along with a D5. (Or possibly D4s, which I feel is a better all-rounder.) Sadly, affordable they aren't. Now the D3s is getting affordable, its low-light performance is roughly matched by the D850, so I find myself no longer wanting one.

 

My shopping list still involves the D850 (and grip, and battery, and a flash that will trigger my SB-600s or wireless, and cards, and readers...), a newer 70-200, at least one of the 85mm f/1.4 Art/105mm f/1.4 Nikkor/135mm f/1.8 Art triumverate, and probably a Samyang 24mm, before I get back to saving for a 400 f/2.8. So I'll not be shopping for a backup body any time in at least the next decade, I suspect. :-)

[mike_halliwell]

I've got a D3S that I don't use too much/any more..;)

 

For OOC JPEGs, I'm still pretty impressed with the D500. That may be to do with the internal creation of said JPEG, ie Expeed 5, but they're pretty darned good.

 

Andrew, did you really just put 'Nikon','Cheap' & 'Mirrorless' in the same sentence?

[Andrew Garrard]

Eternal optimism, Mike! (Mirrorless should allow Nikon to make a camera cheaper. Just because it didn't with the 1-series...)

 

I only ever really use OOC JPEGs when I'm in a hurry and want previews, although I still shoot "large" and "fine" just in case. I almost invariably do fairly heavy raw tweaking given the chance.

 

The D3s would be a lot more tempting as a complement to my D810 (even with the controls different) than to a D850, if the 850's really as much better as the DxO scores suggest (but still not up to A7RIII/D5 levels). I'm not sure I'll persuade Mike to do a swap for a TC-16A, sadly!

[ilkka_nissila]

It doesn't in practice seem to play out that way; the development cost of improving EVFs, phase-detect AF embedded in the main sensor, faster readouts to facilitate fast AF, and development of new algorithms as well as a whole new lineup of lenses with dual focusing motors etc. This all costs a lot of money and in practice we should expect that Nikon's future mirrorless full frame camera will be expensive.

 

There seem to be people who think that mechanical systems that move a mirror are expensive (despite the evidence of really inexpensive cameras having this feature on the market) but the investment in semiconductor manufacturing equipment to make the new technology for mirrorless could be in the billions of dollars.

[Andrew Garrard]

It doesn't in practice seem to play out that way; the development cost of improving EVFs, phase-detect AF embedded in the main sensor, faster readouts to facilitate fast AF, and development of new algorithms as well as a whole new lineup of lenses with dual focusing motors etc. This all costs a lot of money and in practice we should expect that Nikon's future mirrorless full frame camera will be expensive.

 

Oh yes, start-up and development costs are expensive - but actual manufacturing costs (assuming economies of scale, which is something I'm increasingly worried about for Nikon) should not be. Assuming there is an EVF (and Nikon don't go down the "rear screen only" approach), a small EVF is relatively cheap and simple to manufacture and calibrate compared with a selection of folding mirrors, ground glass and prisms, plus metering and AF systems. Nikon's first mirrorless bodies could be expensive, but not if they want to be competitive (with, say, current prices for a mk1 A7) and they shouldn't stay that way.

 

There seem to be people who think that mechanical systems that move a mirror are expensive (despite the evidence of really inexpensive cameras having this feature on the market) but the investment in semiconductor manufacturing equipment to make the new technology for mirrorless could be in the billions of dollars.

 

It's the difference between start-up costs and ongoing manufacture. There are fewer moving parts that need separate assembly and calibration in a mirrorless system. A few pieces of plastic or metal in a mirror may not come to much, but I've no idea how fully automated production is, and I'm sure it's more skilled than plugging some bits of mirrorless system together. I doubt Nikon will be manufacturing much of their system from scratch - EVFs are purchasable components (which Nikon was using in the 1 V series), and TowerJazz or Sony have (or will soon have) the sensor tech. There has to be some novelty to be competitive, but I doubt Nikon are going to build their own fab to create everything.

[mike_halliwell]

I'm thinking of trading in my D3S and 70-200mm VRII to somewhere that might 'overlook' the 400000 frames....;)

 

For all it's faults the 1 Series wasn't so bad. Trouble was it was way too expensive and they stopped development just as they were getting somewhere. I know a couple of people who liked the J5 and V3 and wanted the hybrid of ideas. They finally put a SPAM dial on it and they then crippled long lens use and you could only tether with the V3. They removed the IR remote from everything past the J3 so you had to use the crap App. Grrrrr......

 

Foot with bullet holes in....

 

The way it seems to be regarding camera size is not the mirror/no mirror, it's the size of the lens. You cannot make a small, fast DX lens. Sure the body isn't as deep, but the lens is still 3" long and at right-angles to the body. You've only got to look at the long telezoom for the 1 series to see the body isn't the issue. Sensor size might be, but that's insignificant too.

Edited February 9, 2018 by mike_halliwell

[Andrew Garrard]

Well, the UI on the V1 was abominable, but what it could do technically was relatively acceptable. But I got mine only once the price dropped to compact camera levels. It's way less useful (because it's less portable) than an RX100 with the same sensor size. Later versions improved the UI, but still didn't really fit in a pocket, and didn't get the huge price drop.

 

For telephoto lenses, size is determined by sensor size (unless you shove a teleconverter or focal reducer in the way). You can't make a small, fast DX lens with the same light capturing ability and DoF control as an FX lens - at least, the diameter is inherently the same, and the weight typically depends way more on this than the length. Mirrorless has the chance to make for smaller wide-angles, but only if the sensor likes having light incoming at obtuse angles; you can't really do the Leica trick of having the wide-angle almost touching the film plane. For anything beyond a moderate wide angle, having a mirror box-sized gap at the back of the lens doesn't seem to make much difference optically.

 

If the lenses are already fairly small (and I stand by my argument for collapsible options), a small body makes a difference. I don't believe anyone cares much about the size of a camera in the hand, within reason - it's whether you can store it away that matters. Getting rid of the prism (both for height and weight) has potential benefits (which is why I was arguing for a pentamirror FX body); getting rid of the AF module in the base of the body helps with height too. Still, people like having a decent grip, and mirrorless means higher battery demands, which implies somewhere to put the battery. They can't get that small - as Pentax proved with the tiny sensor in the Q not making that much difference to body size.

 

At some point, my 70-200 VR II will probably also get exchanged for a mk3. I'm still determining what else happens during that process; currently my 24-120 is in the firing line, with some consternation about options for my 77mm LPR filter (if I can find it). Now, if only I could make a vast amount of money...

[ilkka_nissila]

Irrespective of who makes the EVF or sensor, processor etc. the cost of its development is carried out to the purchase price (multiplied). I would be surprised if Nikon didn’t try to achieve a competitive edge by developing their own EVFs. They might even make the equipment that is used to manufacture the EVF.

[c_watson1]

It doesn't in practice seem to play out that way; the development cost of improving EVFs, phase-detect AF embedded in the main sensor, faster readouts to facilitate fast AF, and development of new algorithms as well as a whole new lineup of lenses with dual focusing motors etc. This all costs a lot of money and in practice we should expect that Nikon's future mirrorless full frame camera will be expensive.

 

There seem to be people who think that mechanical systems that move a mirror are expensive (despite the evidence of really inexpensive cameras having this feature on the market) but the investment in semiconductor manufacturing equipment to make the new technology for mirrorless could be in the billions of dollars.

 

Or not. Think you're assuming Nikon would peevishly(and pointlessly) reinvent the wheel to produce a MILC system. Doing some serious headhunting at Fujifilm might save a few yen!

[Andrew Garrard]

I'd say "divided" rather than "multiplied". The cost certainly doesn't go away, but the combination of what we must hope will be economies of scale, the fact that many of the technologies are already well-known (depending on how much Nikon intends to innovate), and the availability of off-the-shelf components from manufacturers who have already sunk the cost of development means that I would hope a manufacturer of Nikon's size and experience would be able to make a mirrorless item cost-effectively.

 

Maybe the big question is about innovation. A survey on a rumor site suggested a surprising number of people would buy a Nikon mirrorless system no matter what form it took. I doubt that will extend to the rest of the market: Nikon missed their chance to be first to the market with the basic concept, so they now have to compete not only with relatively refined products that have been through many iterations, but need some way to differentiate; an F-mount adaptor is a tenuous selling point.

 

We'll see what happens, but the problem with not shipping a product early (if the 1 series doesn't count - and it shouldn't) is that you set a higher standard when you eventually do ship something. Nikon at least have to get it right, and ideally have to make it special. I wish them the best, but that's a lot of pressure, and I'm not giving them great odds.

 

In the meantime, the DSLRs aren't going away - but mean that others have cheaper production pipelines. And the D610/D750 are overdue replacing. Mirrorless isn't a "bet the company" thing yet, but it's close enough that I think Nikon should have been acting more decisively and sooner.

 

Still, Nikon saved their market position with the D1 series and then D3. They've pulled off surprises before (and not just the Df). And they currently have some very good products on the market. Interesting times.

[ilkka_nissila]

Or not. Think you're assuming Nikon would peevishly(and pointlessly) reinvent the wheel to produce a MILC system. Doing some serious headhunting at Fujifilm might save a few yen!

 

No that's not it. To sell its mirrorless system, Nikon need to offer a much better implementation than existing systems (since those systems are established and Nikon will be starting from scratch in the market). So they will be developing technology (faster & higher resolution EVFs with greater dynamic range, better AF etc.) that allows them to offer a better implementation. It has nothing to do with reinventing the wheel. Copying other people's tech or providing similar performance will not lead to sales; as a latecomer to the market Nikon has to provide a much better implementation to have any chance of a significant presence in the market.

[Andrew Garrard]

Agreed with their problem, Ilkka, but I think they need to pick an area or two in which to innovate - not replace every component with a custom version. It's unlikely that Nikon will come up with custom technology that can improve on every product on the market - at best they should simply buy the best off the shelf versions available except for some key components; more likely they'll cut corners for cost where practical. Very likely at least some of their "advances" will be obtained via an exclusive licence from a third party (as with the D800 sensor) - TowerJazz being an obvious supplier - wishing to take advantage of the volume Nikon will hope to ship.

 

To the best of my knowledge, Nikon have no history in developing custom EVFs (or LCDs in general), and their history of PDOS sensors is limited (since I believe the 1-series sensors were from Aptina). Nikon may well have custom algorithms to tune the AF from the processor side, but how much custom sensor tech is involved is another matter.

 

Nikon can absolutely innovate; they may do so with an exceptional EVF (although the technology has certainly advanced in existing implementations), they may do so with stacked sensor tech (though whether their options are better than Sony's is TBD), they may introduce exceptional AF (though, knowing Nikon, they'll try not to compete with their SLR options too much). They might even provide a state of the art video option, although that would be a significant departure. They could do something novel like providing rear movements at the sensor. There are plenty of areas where camera technology can be advanced (many I'd like to see in an SLR). Being as responsive as a DSLR would be an advance over many mirrorless options currently on the market - but the competition is clearly improving, and current low hanging fruit may have fallen before Nikon can harvest them.

 

But I doubt Nikon will try to innovate in more than a couple of areas. To do otherwise would be incredibly risky, and Nikon have had a number of high-end cameras with high-profile (if practically minor) problems. However they innovate, Nikon really have to have a product that works. Solid is better than novel but flaky (unless Fuji have sold a lot more X100s than I think).

[c_watson1]

However they innovate, Nikon really have to have a product that works. Solid is better than novel but flaky (unless Fuji have sold a lot more X100s than I think).

 

Nikon's buggy product cavalcade and indifference to its DX base damaged brand loyalty. Fuji's CRM and kaizen policies aren't exactly hurting sales. Any Nikon MILC system will have to amount to a departure from the conventional I'm so far unconvinced the company can successfully pull off. Flaky? Fuji fixes glitches at warp speed relative to Nikon. Recall the difficulties Nikon ignored with the D600?

[Andrew Garrard]

Oh, absolutely. I wish Nikon would even approach Fuji's record for retrospective feature enhancement. Nonetheless, even with fixes, the original X100 was a mess (later ones are substantially improved, if still out of my casual price range). Nikon has to do better.

 

The D600 had one issue, that Nikon managed terribly. But the D800 had an issue, and so did the D750, and D810, and D500. They're improving, but with a new platform, they can't afford a high-profile slip-up. But nor can they afford a delay. Tricky, but that's why they make big bucks...

[mike_halliwell]

Guess they're worried about the a repeat of the cancelled DL fiasco .... co-incidentally 1 year ago tomorrow.

 

I actually wouldn't mind a 3 dial camera...:cool:

[Andrew Garrard]

I'm not sure which digit I'd put on a third dial. I'd take the press-and-turn secondary functions on dials, though. (Obviously as a configurable option - I'm not thinking to impose this on people as a default UI.)