Ditching f mount once Nikon goes mirrorless?

CrossLeaf

Hi guys, so recently i have been doing some research and found out that Nikon lenses are harder to design and more expensive due to its smaller mount diameter. I am just wandering if Nikon would ditch f mount and start a whole new system once it goes mirror less.

PB_PM

Current rumors for the 35mm mirrorless Nikon body are saying it will have an F mount. Nikon's strength is the wide range of lenses available, without that they are no better than the other mirrorless manufactures.

spraynpray

On the other hand, I also read that the mirrorless design needs a different lens to sensor distance and although F mount lenses can be used, a new range of lenses would give better results with an optimised body design. Do I know? NO - but it makes sense to me.

sevencrossing

As s&p says
We dont know
We don't even know, if or when they might go mirrorless with fx and or dx
As I have mentioned before they abandoned it with the Nikon 1

jonnyapple

Andrew, I think the best way to say it is that mirrorless has fewer constraints on lens-sensor distance, which could simplify lens design on lenses with smaller focal lengths.

At issue is that in an SLR you have a mirror that's about 24mm long flipping up between your sensor and your lens, so you can't ever have a lens closer than about 24mm from the sensor or the mirror would crash into it. If you think about it, that makes it very hard to design a lens with, say, a 14mm focal length for an SLR—the lens should sit near the middle of the mirror! They get around this with front lens elements that actually diverge the light rays before the rear elements get them to converge. It's called an (inverted telephoto) retrofocus design, and it adds lens elements exactly where they're expensive (at the front where they're biggest). A mirrorless 14mm lens can have a pancake design because they don't need that divergent front lens group to help dodge the mirror. Longer focal length lenses are already out of the way, so it wouldn't make sense to redesign those for mirrorless. Note the huge diverging lenses in this (KR's) cutaway view of the 14-24mm f/2.8 Nikkor:
http://www.kenrockwell.com/nikon/14-24mm/images/D3R_4636-cutaway-950.jpg
[Edit] Compare those front lens elements to the ones on this 16mm 2.8, which has an APS-C image circle:
http://www.bhphotovideo.com/images/images2000x2000/Sony_SEL16F28_SEL16F28_16mm_f_2_8_Wide_Angle_695859.jpg
To give a sense of scale, this lens is less than an inch thick front to back (compared to the 14-24's 3.8 inches) and weighs a mere 70 g compared to 1000 g for the Nikkor! [/Edit]

I will say that in practice, at least with the Sony mirrorless system that I'm familiar with, the savings in money and size haven't really materialized in a way that preserved image quality. The 16mm 2.8 pancake lens was really unimpressive and I sold it within a month of buying it. The 16-50 kit lens delivers on size but I prefer the older (bigger) 18-55 kit lens for optical performance.

The existing F-mount lenses wouldn't lose performance on a mirrorless F-mount camera (I know because I own one: I have an F-mount adapter for my Sony), but they would be overkill. Nikon would probably have to create some slight difference with any F-mount lenses designed for mirrorless cameras so they couldn't mount to the normal F-mount because it would be a tech support disaster if people smashed up their mirrors with them!

brownie314

If Nikon does produce a mirrorless system I think it would be suicide for them to lock it into F-mount. They should give it a nice mirrorless mount, with all the advantages that go along with that - and make a really good, high quality F-mount adapter.

WestEndFoto

There are a few choices here:

1.
Ditch the f-mount in favour of a new mirrorless mount. But I think a few of us that have invested heavily in the system will be pissed at being forced to use adapters. That would be one of the few things that Nikon could do that would make me switch to Canon.

2.
Add a second mount and invest in a new parallel system. That will keep Nikon busy for several years, is bound to reduce the attention they pay to servicing their current customers that have invested in the F-mount system and cost them in supporting the infrastructure of multiple systems (they already have 2).

3.
Retain the status quo. That way they can focus on what they do really well, which is the current F-mount system. That doesn't mean that they can't do mirrorless. It just has to utilize the f-mount.

Personally, I hope that they go with #3. I am very sceptical that the advantages of a shorter flange to focal length distance outweigh the benefits of the first two options. I don't think that it presents much of a benefit in smaller size. The D3500 uses an F-mount and it is plenty small.

I think that option #1 is the quick path to bankruptcy and option #2 risks being unprofitable.

I can only think of two scenarios where it makes sense to me to bring in a new system.

1.
Replace the Nikon 1 system with a larger sensor - say DX. I think that the Nikon 1 system is too small be a compelling alternative to smartphones. However, I think it would be even smarter to take a cue from the new DXO camera and produce an adapter and software to use the Nikon 1 lenses on smartphones. Perhaps that is what their project with Apple is about.

2.
Create a medium format (and not a dinky Pentax 645 version) based on a mirrorless design with a selection of wide angle and portrait lenses. I think that a system like this could be really good for landscape and portrait work, where fast auto-focus is not that important and resolution will be sought after. It would be nice to have a system where the eyebrows, eyelashes and iris was super crisp on a full body shot - something that not even a Zeiss Otus does really well (it is merely passable in my view). This would be an expensive system because of the sensor size, but I think that there would be a demand, particularly if Nikon waits for a few years until sensor prices come down a little more.

My humble thoughts.

Msmoto

As has been pointed out, the wide range of "F" mount lenses is attractive. And, while I am guessing, as we all are, my thinking is we will see a mirrorless with a decreased flange distance, but with a fully automatic adapter which will accept the "F" mount lenses.

One advantage of shorter flange to sensor distance is the wide lenses can be less of a "retro focus" design. However, the angle of incidence with shorter lens to sensor distance increases to a point where the sensor may not be able to "see" the light ray as well, thus a smaller angle of incidence may have some advantages, as in keeping the current Nikon flange to sensor distance of 46.5 mm.

This simple fact may also be one of the factors we see the MTF charts of long lenses so flat, unbelievably sharp, especially compared to wide lenses where the angle of incidence is quite large toward the edges of the field, and as a result, resolution falls off toward the edge, in a general way.

This argument is not supported by other manufacturers, however, as in Leica which has a 27.8 mm distance and they seem to be able to produce fairly good images....(the opinion of one who wants one, but cannot afford a Leica...)

So, who knows, but the discussion is always of interest to me...

KnockKnock

@jonnyapple, thanks for this - I've never read it so clearly described before, with the nod to the economics of front elements vs rear.

At issue is that in an SLR you have a mirror that's about 24mm long flipping up between your sensor and your lens, so you can't ever have a lens closer than about 24mm from the sensor or the mirror would crash into it. If you think about it, that makes it very hard to design a lens with, say, a 14mm focal length for an SLR—the lens should sit near the middle of the mirror! They get around this with front lens elements that actually diverge the light rays before the rear elements get them to converge. It's called an (inverted telephoto) retrofocus design, and it adds lens elements exactly where they're expensive (at the front where they're biggest). A mirrorless 14mm lens can have a pancake design because they don't need that divergent front lens group to help dodge the mirror. Longer focal length lenses are already out of the way, so it wouldn't make sense to redesign those for mirrorless.

Spy_Black

I don't think it makes any difference whatsoever to ditch the F-mount on a mirrorless. Lenses would need an entirely different optical design to work so close to the sensor. Personally I don't know if that's an advantage or disadvantage when designing an optical formula for a fixed or variable focal length, but just like the Nikon 1 system, it takes nothing to design an adapter to allow existing lenses to work with the mirrorless design, assuming the adapter isn't deliberately crippled the way it is on the Nikon 1 system.

Starting fresh like that they could make for a larger opening flange that would allow the creation of lenses with apertures under f/1, for instance. That would certainly be advantageous. So to me ditching the F-mount would be a good thing in a full frame mirrorless system, as long as the new mount would allow for designs not possible now, and have an adapter system that would allow the full use of existing lenses.

jonnyapple

@KnockKnock, glad I could help!

So to me ditching the F-mount would be a good thing in a full frame mirrorless system, as long as the new mount would allow for designs not possible now, and have an adapter system that would allow the full use of existing lenses.

Yes, that's the idea, I think. But it would be nice to people already invest in Nikkors to have an F+ mount that would take both legacy F-mount lenses and mirrorless-designed lenses without bothering with an adapter.

Increasing flange throat size—the original question of this thread—I don't see as a very pressing need. f/1.2 is possible at 50mm with the F-mount and that's kind of at the fringe of the market. I owned the 50 1.2 and only rarely shot it wide open because the depth of field is razor thin! I found that f/2 was a better compromise as cameras got better at low light, so I own the 50 f/1.8 now.

Spy_Black

"Increasing flange throat size—the original question of this thread—I don't see as a very pressing need. f/1.2 is possible at 50mm with the F-mount and that's kind of at the fringe of the market."

Says who? If you're gonna start from scratch, why not allow to expand your horizons? Imagine a FF f/0.75 50mm optic. Why not? how 'bout a 400mm f/1.4? 600mm f/2.8?

If you're going to start from scratch, do it right. ;-)

CrossLeaf

can anyone kind of explain the advantage of having a wider mount diameter? i heard that having wider mount diameter also makes it easier to design lenses and this is one of the reasons why nikon lenses are usually more expensive than canon

WestEndFoto

A lens designer can be a little more lazy, that is about it. If the throat diameter was wider, the benefits of making a larger sensor (medium format) would be too great to resist and then we would be back to where we started.

DX has a massive throat diameter compared to the format size and I don't see DX stealing the show.

And I am sure that Nikon could find a way to make any F-stop work with the current throat diameter, or even a throat diameter of half the size. Some bright boy came up with retrofocus after all.

Not saying that there is no benefit, but I suspect that throat diameter is not what keeps Nikon lens designers up at night.

jonnyapple

Okay, @SpyBlack, I'll bite because this is kind of fun to think about. ;-)

My first response to the exotics suggestion (400 f/1.4 and 600 f/2.8) was that weight and cost were the limits and the flange had nothing to do with it, but I'm willing to work to prove myself wrong. This is a geometry problem. We need to draw the worst case scenario light ray and see if the flange gets in the way. Worst case would be a single lens element 400 mm from the sensor, and at f/1.4 it has a 286mm diameter (143mm radius). The light ray that comes in from the lower right of the field of view and hits the upper left of the lens has to be redirected to hit the upper left corner of the sensor, which is 21.6mm (the hypotenuse of a 12mm x 18mm triangle) from the center. The F-mount flange throat is 46.5mm from the sensor and has a radius of 22mm. Using similar triangles you can find the distance behind the sensor this worst-case ray would cross the lens axis. Call that distance x:
143mm/21.6mm = (400mm+x)/x

x turns out to be 71.3mm. Now we can use similar triangles again to see how far from the center this ray is at the flange. Call that r:
21.6mm/r=71.3mm/(71.3mm+46.5mm)

r is 35.7mm, so it does get blocked by the flange, so you're right that it is a limitation. (For reference, this means we're talking about a new flange that would take almost double the area of the current one if you want to avoid this.)

Remember that this is worst case, though. It would be interesting to see what fraction of the rays are blocked instead of doing worst case because that would tell you how much vignetting you would see because of the throat diameter. I'm not going to, though, because I think @WestEndPhoto is right that engineers could get around it with multiple elements and I still stand by my other concern with a 400 f/1.4 you have a huge front lens element (~11"). That is going to cost you a lot in cash upfront and physical therapy after using it. Sigma makes a 35 pound, $26,000 500mm f/2.8 monster that you can buy right now. I wonder how many they sell...

WestEndFoto

And what practical benefit would a 400 1.4 be? Is there not enough bokeh for you at 2.8? Also, remember that the thinner the lenses maximum depth of field, the more challenges are placed on the focus system at all apertures. This would not be a very good BIF lens I suspect.

Spy_Black

I don't think anyone bought a 400mm f/2.8 for the bokeh. ;-) There's a reason high-speed lenses exist.

Look at the new 58mm. What is it's maximum aperture? Why didn't Nikon improve on the old lens while keeping the old maximum aperture? There's more than glass inside a lens.

WestEndFoto

I am a little uncertain as to your point.

Compression and bokeh is why a 400mm 2.8 is on my list for portraits. Why is a 400mm 1.4 on yours?

MaxBerlin

Chief ray angle is one good reason Nikon performs so well across the board. A sensor optimized for WA with 'lenses' (like the Sony A7r) compromises edge and corner performance for telephoto lenses. There is always a tradeoff.

PB_PM

I am a little uncertain as to your point.

Compression and bokeh is why a 400mm 2.8 is on my list for portraits. Why is a 400mm 1.4 on yours?

Indeed. Even at F8 a good 400mm lens's the bokeh makes a 50mm F1.4 lens look like a crap in comparison. Why? The depth of field is so narrow at 400mm it's not even funny.

I don't think anyone bought a 400mm f/2.8 for the bokeh. ;-) There's a reason high-speed lenses exist.

Indeed, there are two. One enabling faster shutter speeds, and second to create separation between the background and the subject (aka bokeh). Of course there is a vast difference between making a F1.4 85mm and a 400mm F1.4 lens. Technically it is possible, but it would be so big and heavy that you'd need the back of pickup truck just to transport the thing (okay exaggerating, yes, but not by much). I'm no math wiz, but I'm betting the front element would be bigger than a 4 D4 bodies. I guess if you want to shoot from a golf cart that is fine, but for real world use? Haha, yeah no comment.

WestEndFoto

And it would be an expensive lens. Certainly over $50,000 and possibly $100,000. The weight would be about 30 kilos, my rough guess.

Ironheart

The front element would be 286mm or 11.25 inches in diameter. This means the lens body and hood would be over 300mm or almost a foot in diameter.

Spy_Black

The point is having options, not necessarily making a lens like that. ;-)

Ironheart

We're just having a bit of fun, I love exotic lenses. I guess I should have been a billionaire )

sevencrossing

And it would be an expensive lens. Certainly over $50,000 and possibly $100,000. The weight would be about 30 kilos, my rough guess.

Seems an excellent reason to ditch the f mount, to me
Currently anyone buying a Dx or Fx Nikon, only have 87 F mount Nikkor lenses to choose from.
The range of second hand F mount lenses is only about 50,000,000
The chances of Nikon dropping the F mount are about the same as Porsche dropping the 911
Nether are perfect

Msmoto

As there is some discussion on the blog regarding the Nikon mirrorless and potentially a new mount, I will bring this thread up and offer my thoughts. For whatever they are worth.

A new mount might be supported, intellectually at least, by the fact the throat of Nikon lenses is rather tight (44mm), and the mount to focal plane is fairly long at 46.5mm.

As to the actual camera body becoming a reality, I suspect this will occur, but only after both Nikon and Canon are willing to sacrifice sales on the DSLR line. It is my opinion, once the mirrorless is refined to the point all functions are as good as the top pro bodies currently available, and the viewfinder looks to one's eye exactly like a ground glass surface, then we are seeing the beginning of the end of the DSLR as the mirror would be more of a liability than an asset.

As Nikon's F mount is used by a large number of dedicated amateurs and professionals, I am pretty certain the new mirrorless body would be able to use the F mount lenses, albeit with an adapter if the new body has a markedly shorter flange to focal plane distance. The blog will be interesting as Admin offers up the inside stories which will be slipping out in the future.

And, I am really waiting for all this to happen.... maybe before I reach 80... LOL