Nikon Announces D850 Filmmaker's Kit

The D850 filmmaker's kit includes a regular Nikon D850, three lenses, two microphones and an external monitor, plus an extra EN-EL15a battery (so that the kit comes with a total of two EN-EL15a):

 

The Nikon D850 Filmmaker Kit includes the following:

• Nikon D850 DSLR (with supplied accessories)
  
• AF-S NIKKOR 20mm f/1.8G ED lens
  
• AF-S NIKKOR 85mm f/1.8G lens
  
• AF-S NIKKOR 35mm f/1.8G ED lens
  
• Extra EN-EL15A battery
  
• ME-1 Stereo Microphone
  
• ME-W1 Wireless Microphone
  
• Atomos Ninja Flame External Recorder (with supplied accessories)
  
• Custom foam inserts (can be used in hard case for transporting, hard case sold separately)
  

 

The Nikon D850 Filmmaker’s Kit will be available in the U.S. for the suggested retail price (SRP) of $5,499.95 and will go on sale at the end of March 2018. That is roughly a $800 discount from purchasing those items separately. Hopefully this discount indicates that D850 supply is finally catching up with demand.

 

Link on Nikon USA's web site:

Nikon D850 DSLR Filmmaker's Kit

 

Production image copyright Nikon, Inc.

Edited March 20, 2018 by ShunCheung

WOW. That is a stunning deal!

 

That lot in the UK is:

 

Nikon D850 £3500

Nikon 20mm £679

Nikon 80mm £429

Nikon 35mm £450

Nikon EnEL15A £54

Nikon ME-1 £140

Nikon ME-W1 £180

Atomos Ninja Flame £1164

 

= £6596 or at 1.4$:1£ = $9234

 

Makes you wonder what the actual price is going to be if it's released here?

 

Any other price comparisons around in Europe?

As far as I can tell, the D850 Filmmaker's kit seems to be a Nikon USA-only product. The announcement is not on Nikon Japan's world-wide news web site, where new products are always announced:

Nikon | News

 

I recall that previously, there were D810 and D750 filmmaker's kits also. For example, this was a 2014 thread:

Nikon D810 filmakers kit from B&H

G lenses for film-making (AKA video)?

 

There's a reason that serious videographers buy modded MF lenses with geared aperture and focus rings.

 

Another sad misjudgement and lack of real-world know how by Nikon's marketing department?

I am sure you can add geared control of the focus to these lenses but the ease of use may be affected by the slop in the focus mechanism when changing the direction of turning of the focus ring. Even with G lenses you can adjust aperture in 1/3 steps during video recording in the D850. Stepless aperture control is available on Zeiss Milvus (by turning an adjustment knob in the bayonet the aperture clicks disappear) and lenses designed specifically for video. Most Nikon f/1.4 AF-S lenses do have nice manual focus rings but they're not offered in this kit. I wonder why Nikon neglects the implementation of slop-free manual focus in so many of their lenses.

 

I think maybe this kit is intended for people who are doing still photography and video, and to offer a reasonable kit as a starter purchase with some savings. The f/1.8 lenses may have been chosen instead of f/1.4 due to their lower cost. I guess the full-frame 4K recording makes the D850 appealing to some users, personally I find it too much data. The decision to include the 20mm prime could be seen to capitalize on the full frame recording capability (vs. Canon 5D IV which offers only a deep crop in 4K). Nikon may produce something that targets Sony, Panasonic etc. when they make their full frame mirrorless offering later on. Though I suspect Nikon's primary focus will likely remain on still photography.

I think it's reasonably to argue that video on a DSLR is a convenience feature rather than "the best way to do video". I won't be astonished if Nikon decide to push this much harder when they launch something mirrorless (HDR, internal raw recording of 4K, and 8K video are all potentially going begging in the consumer space, especially if they can do something good with autofocus). On the other hand, other than other people adapting their lenses because the mechanical apertures (that they used to have) were convenient for video, Nikon has very little video shooting experience - and I doubt they're about to buy BlackMagic or Red to fill in the technical gap. So we may get nothing, something but not what anyone wants, or the right thing done with some critical flaws. No pressure, Nikon!

The key to professional video is sound, and an on-camera microphone isn't going to cut it. The industry often uses recorders by Sound Design for mobile work, which are about the size of a hard-cover book, and have studio quality sound and time code. I don't often wire up to a master clock, rather use the recorded sound track for synchronization with the recorder. The internal mic on a DSLR is more than adequate for that task.

 

Pulling focus is hard to do when using a DSLR in a cage. It's usually done on a heavy tripod with handgrip controls, or by a separate operator (focus puller). The frame for this feature can cost as much as this D850 kit. It's not hard to spend $2k on a fluid head alone.

 

With the advent of Super-35 (APS-C) and FF sensors in affordable video cameras, like the Canon Cxxx series and Sony FS5 and FS7, using a DSLR (or ILC) seems like a novelty approach.

 

I think even for filming your children, events or weddings, a zoom lens would be more useful than a handful of primes.

Agreed - I was duly amused by Nikon putting an on-camera mic in the kit. I'm very much an amateur, but I have at least dedicated audio (Zoom H5, in my case), lapel mics and such rather than believing that putting the microphone in the same place as the camera is a good idea. They have their uses for run-and-gun journalism (or maybe candid shooting at a wedding, at a push), but this is a lot of money to spend on something with only marginally better audio than a cell phone. Unless the D850 has a much better audio digitiser than I'm expecting - and even then, it's not really the obvious place to do it.

 

Micro 4/3 seems to be the better balance sensor size for video - if you really need the aperture advantage from large sensors, hitting focus starts to be troublesome (as I found while watching the Les Mis film on IMAX, often shot at a large aperture, and seeing how often focus was in the wrong place). I don't know how good the video industry has got with autofocus, but good video autofocus in consumer devices - which seems to be not quite there yet even for Canon and Sony - feels like it's a necessary next step.

Nikon would be better off including the gold cardboard lens boxes with the kit, rather than a hard case. At least then their customers could get a good return selling the useless surplus lenses on eBay, and pocket a fair discount on the kit price.

 

They should also leave the mic business to AKG, Audio Technica and Rode, who know what they're doing in the audio field.

The whole point of using large sensors for video is a shallow DOF. Professional geared focusers have a scribble strip which can be marked for various focus points in the scene. Everyone has to be in the same place each take. With autofocus, you risk hunting, and are generally limited to the closest subject. That said, it works pretty well for mid-level cameras and above. For cinematic purposes, control (i.e., selective focus) trumps automation.

 

Zoom makes a relatively inexpensive stereo shotgun mic which fits the H5 (and other) recorder. It's better than the built-in mic, but picks up way too much "room" beyond about 5' from the speaker. The industry standard is a Sennheiser MKE-416 ($1k), but used on a boom, just out of camera view. I use an H6 for interviews. Rather than fuss with wireless, scanning for interference and such, and always fresh 9V batteries, I have a set of wired lavalier mics. Zoom sound quality is very good, and the AA batteries last a very long time in their H series recorders.

The whole point of using large sensors for video is a shallow DOF.

 

And the capture of low light (when the large sensor is actually using all its photosites and not skipping most of them). But I agree, shallow depth of field is a reason to use a larger sensor. And it's useful both for isolating the subject and for making it easier to avoid spending a fortune on a set design by ensuring that it's usually out of focus - which seems to be what Les Mis was trying to do.

 

Professional geared focusers have a scribble strip which can be marked for various focus points in the scene.

 

Which, with the greatest respect to the film industry and generations of focus pullers, is a horrible hack that just about has a chance of working with the relatively large depth of field associated with smaller sensors and when we had iffy 35mm film and standard def television, but pretty much doesn't work at large apertures and 4K. Hence my experience watching Les Mis and being distracted from "Empty Chairs at Empty Tables" by the focus being on Eddie Redmayne's left shoulder, not his face. Hitting a mark is one thing, but when being off by two inches is enough to be distracting, moving the actors to the focus points is never going to be reliable.

 

With autofocus, you risk hunting, and are generally limited to the closest subject. [...] For cinematic purposes, control (i.e., selective focus) trumps automation.

 

Hence my argument that the technology needs to improve. Pulling focus should, in this day and age, involve moving between subjects (not distances) at an appropriate point in the scene - and letting the electronics in the camera track the subject movement. Since the subject movement is usually a bit less erratic than a DSLR trying to track sports, I get the impression that it should be possible to solve it better than current solutions - but since my experience of video is mostly with a DSLR (or something with enormous depth of field) and a Nikon one at that, I don't really know the state of the art, other than people complaining about it.

 

I've also seen TV shot with what was clearly a tilt-shift lens. I maintain that it should be possible to autofocus those, too, but that's a separate feature request.

 

a zoom lens would be more useful than a handful of primes

 

I'd meant to say: For video kits, Nikon's problem is that they don't, as far as I know, make a zoom (parfocal) lens. They make varifocal lenses, which change their focus distance as you zoom. If the body had good enough video AF tracking, this wouldn't actually matter - you could zoom the lens and the AF system would adjust to track the subject; in theory you could at least do this electronically even for an optically varifocal lens just by detecting the focal length change, without AF (hmm, that one goes on the ideas list). For now, Nikon probably went with primes because people would have grumbled that the zoom wasn't really a zoom. Of course, Nikon could make one, and that would be an interesting differentiator for a video-friendly mirrorless system.

Nikon's problem is that they don't, as far as I know, make a zoom (parfocal) lens

 

I had to say that, didn't I? Ephotozine's review of the 180-400 is just out, and claims that it's at least approximately parfocal. That said, I suspect there's a reason it's not included in this kit!

I'm not sure why the word 'parfocal' has entered the language. A true zoom lens, by definition, should hold focus throughout its focal length range - otherwise it's a varifocal lens.

 

I expect it's due to years of lens-makers incorrectly attributing 'zoom' to lenses that are in reality just varifocal. Same as the word 'macro' is used with abandon to describe any lens that's vaguely close-focussing.

 

FWIW, I don't think gearing a G lens focus ring would be of the slightest use. The internal clutch mechanism of the lens could easily slip and render the gear position way off. And as for stepping the aperture in 1/3rd stops; I think that might be even more distracting than a single near-instantaneous change of exposure.

 

Say a pan from distant scene to shadowed face required a 3 stop opening. The visual 'judder, judder, judder' of an aperture opening in 1/3rd stop steps would almost certainly be unacceptable, compared to a stepless aperture being pulled manually.

 

In any case, aren't the days of manual pulling over? Isn't there a box you simply program with focus and aperture settings that magically whizzes the lens from one setting to another via motors these days?

I'm not sure why the word 'parfocal' has entered the language. A true zoom lens, by definition, should hold focus throughout its focal length range - otherwise it's a varifocal lens.

 

I expect it's due to years of lens-makers incorrectly attributing 'zoom' to lenses that are in reality just varifocal. Same as the word 'macro' is used with abandon to describe any lens that's vaguely close-focussing.

 

And "prime" didn't used to mean "single focal length", "telephoto" doesn't mean "long", "normal" is a relative concept, and Canon and Nikon don't show the same f-stop in the viewfinder. All photographic terminology gets abused (don't get me started on "wide-angle distortion"). Nikon, of course, make the macro/micro distinction, but this just confuses everyone else. We have to rely on context and seek clarification when necessary; I'm just happy when "zoom" isn't used to mean "long". I'm a fan of precision in languages (in the land of the grammar nazi, I am the fewerer, as I like to say - but haven't quite dared put on a T-shirt), but we're not going to get the photographic community to be consistent; my personal fight is to keep f-stops described as f/2 etc. rather than f-2, in the hope of explaining fractions to the next person who wants to understand depth of field.

 

And as for stepping the aperture in 1/3rd stops; I think that might be even more distracting than a single near-instantaneous change of exposure.

 

I was under the impression that bodies with "power aperture" had finer control over the aperture setting - something like 1/10 of a stop? You'd probably notice that if did it slowly enough, but I suspect it's moderately subtle for amateur use. A fully stepless aperture could be smoother (and more reliable), but automating exposure might not be easy. Of course, you can fix exposure relatively easily in grading; 1/10 of a stop (if I'm not imagining it) is probably enough to control bokeh non-intrusively. How well the automated systems adjust to a variable-aperture zoom, or even light fall-off with close focus, is another matter.

 

Given the gears on video lenses, I assume everything can be electronic, although I'm sure people wind remote handles for precise control as well. It wouldn't surprise me if there were still usually people sitting at the electronics trying to tune the focus as the subject moves around, though. I'm not good enough at manual focus to do that, but some people are. Still, I do think it's an automatable skill - but I have precious little knowledge of the professional video industry.

A true zoom lens, by definition, should hold focus throughout its focal length range - otherwise it's a varifocal lens.

Good thing then that to the best of my knowledge, none of the "non-cinema" zoom lenses (and likely not even those even though they probably are) are labelled "true zoom" but just zoom :D

Not too long ago in another forum, the question whether the Nikon 70-200/4G VR is parfocal was posed. I tried mine and got a fairly clear "no" as an answer. Another photographer tried his and arrived at an equally clear "yes". Apparently the difference in opinion came from me testing at closer distance and he at intermediate focus distances (with possibly DOF masking the change in focus). As a general rule, I always assume that any "variable focal length lens" is varifocal until I have determined otherwise. FWIW, the Sigma Art 24-105/4 definitely is not parfocal.

 

but we're not going to get the photographic community to be consistent

the confusion around the use and meaning of the words "bokeh" and "DOF" immediately spring to mind ;) Probably wouldn't have to think too hard to come up with more examples where "precision in language" is sorely lacking (or is that "precision in understanding"?) "High rate of speed" pursuit anyone?

Edited March 22, 2018 by Dieter Schaefer

"I was under the impression that bodies with "power aperture" had finer control over the aperture setting - something like 1/10 of a stop?"

 

- Unless Nikon have sneaked this feature into the D850 unannounced, the aperture on DSLR bodies can only be set to 1/3rd or 1/2 stop steps. So, yes, 1/10th stop intervals would most likely go nearly unnoticed, but Nikon bodies just don't allow the aperture to be 'pulled' with that degree of refinement.

 

As for relying on AF during a video take? We're a long way from that being a useful reality I fear. If you think manual focus-pulling is a bit hit-and-miss, just be thankful the makers of Les Mis didn't rely on AF.

 

Given sufficient reverse-engineering, I'm pretty sure that G lenses could be made operable by a programmable 'puller' box, and in a refined manner. However that would require intercepting the connection between camera and lens. And of course a shallow mirrorless body would make that a relatively easy task. Not so much with a bloated DSLR.

Edited March 22, 2018 by rodeo_joe|1

"I was under the impression that bodies with "power aperture" had finer control over the aperture setting - something like 1/10 of a stop?"

 

- Unless Nikon have sneaked this feature into the D850 unannounced, the aperture on DSLR bodies can only be set to 1/3rd or 1/2 stop steps. So, yes, 1/10th stop intervals would most likely go nearly unnoticed, but Nikon bodies just don't allow the aperture to be 'pulled' with that degree of refinement.

 

I was wrong, it's 1/8 of a stop. I believe this applies only to video shooting - you definitely have 1/3 or 1/2 stops for stills. (A pedant would wonder how they can do 1/3 of a stop when this isn't divisible into 1/8, and why there aren't sixths going on somewhere. But it's probably down to stops being approximated pretty much everywhere.)

 

As for relying on AF during a video take? We're a long way from that being a useful reality I fear. If you think manual focus-pulling is a bit hit-and-miss, just be thankful the makers of Les Mis didn't rely on AF.

 

My optimism is showing! Given that some video cameras already have separate optical paths for different colours (although I don't know how common that still is), I'd have thought getting phase detect AF working properly in video should have been a solvable problem. Since I don't shoot with a large sensor Canon or Sony, I can't say how good they've got, but it does feel like it should be a matter of time. Until Nikon go past contrast-detect video autofocus (unless they're going to try to fit in an extra focus pass during an inter-frame period) they're not going to compete here. I don't hugely mind, since I care about stills way more, but it's an area where they could improve.

 

Given sufficient reverse-engineering, I'm pretty sure that G lenses could be made operable by a programmable 'puller' box, and in a refined manner. However that would require intercepting the connection between camera and lens. And of course a shallow mirrorless body would make that a relatively easy task. Not so much with a bloated DSLR.

 

I'm not sure how much engineering would actually be needed. There are adaptors for mirrorless formats that take G lenses, and simply have something that mechanically moves the aperture lever. E lenses are a bit harder, but then Canon lenses have had the same issue since 1987.

 

Anyway, while I'm happy to do the occasional video shot with my DSLR, I'm unlikely to be pulled away from the idea of getting something with a micro 4/3 mount should I want to get decent video. YMMV.

 

One thing I've failed to find out: How many of the sensor sites does the D850 use for video capture? Not areal coverage, but is it skipping? I believe previous bodies would achieve video by reading, say, one pixel in four (I've not done the maths, forgive me on the exact numbers) in each direction, so the full frame sensor was only contributing as much light to the image as one 1/16th the size. That's one reason that smaller sensors with less downsampling held up quite well, and why the cameras with low-res sensors (like the A7S) which didn't downsample do so well for video. I've not paid attention to how the D850 behaves here.

Wow, Andrew, I was completely unaware that such 1/8 stop adjustments are possible in video using Nikon DSLRs. I guess it is time to study the manual ;-) In any case since the actual control lever in G lenses is analog and not digital, it could be adjusted in arbitrarily small steps to create a continuous smooth effect, but the motor in the camera may or may not be capable of this.

 

I know that movies are generally shot with manual focus because of the degree of control required and focus adjustments are used as creative effects, and action scenes usually have a lot of depth of field partly to make manual focus possible. However, I was just watching figure skating world championship ladies singles short program broadcast from Milan and in the tight close-up super slow motion sequences the depth of field is fairly shallow (i.e. background is virtually homogeneous and only the athlete is seen in focus) and usually it looks quite good but occasionally the focus drifts off for a moment before coming back. I don't know if they auto or manual focus but I would assume it is auto since manual focus doesn't just start drifting off. In some wider view moments also when the skater is relatively far away and the audience is seen, the camera focuses on the audience. So this behavior is likely a result of the use of autofocus. I don't really know if anyone can manual focus on an approaching skater using a narrow angle of view lens in such a way that the subject stays continuously in focus. The videographer has a lot to do, they have to choose what to include in the shot, follow the fast movement of the skater, capture the right parts of the program as slow-motion etc. Focusing manually may be just unrealistic a lot of the time.

 

Canon's dual pixel AF seems to work well during video recording and holds focus quite well so that it doesn't drift in and out of focus. However, there must be some limits to where it can be trusted i.e. how fast can the subject move towards the camera or away, how much lateral movement there can be, what about when the lens focal length is adjusted during the sequence? I think this technology is becoming more and more realistic to use also with shallow depth of field. However, movies shot for theatrical viewing may be a different matter.

 

getting phase detect AF working properly in video should have been a solvable problem.

 

Right, Nikon's recent patents suggest they're working on some kind of dual pixel or quadruple pixel (cross-type) phase-detect AF for their future sensors most likely for mirrorless in mind, but obviously video is another reason to pursue this. I don't know if these implementations produce the kind of horizontal lines when shooting into the light with a fast lens wide open that have been reported by dpreview.com in e.g. Sony cameras with on-sensor PDAF sensors.

How many of the sensor sites does the D850 use for video capture?

 

I believe it uses all the pixels within the area used for video and performs some kind of averaging to come up with the 4K image. Not necessarily the best quality interpolation but no line skipping is used.

Thanks, Ilkka.

 

The 1/8 stop is, I suspect, a trade-off between granularity and nothing happening when you press the button. With the aperture lever, I've no idea how precise things are anyway - but I suspect not very. An external mechanism moving the aperture can be more precise, but I suspect there are diminishing returns at some point.

 

Interesting about the skater (although it could just be someone very good at moving focus manually). It strikes me that film crews could consider using IR dot projectors (like Kinect) to help make AF more reliable; this isn't so practical for amateurs in the field. Having a separate person doing the focus pulling likely helps, but it's a relatively mechanical skill. I'd missed the horizontal line effect; I wonder whether there are a limited number of AF sites on Sony sensors (improving the overall DR by not mangling all the sensor sites), whereas Canon may avoid this by splitting every pixel in two. Both Canon and Sony have clearly been trying to get this right for a while, and it would make a massive difference to usability for amateurs wanting wide-aperture video.

 

Good to hear about the D850. Actually being able to use the light gathering ability of the sensor is a significant improvement for a large sensor DSLR.

Hmm, I wouldn't class sports coverage as 'film making'.

 

Film making, in my book, involves a pre-conceived storyline, storyboard and controlled shooting environment, with rehearsed camera positioning, focus and exposure control. Not just leaving it to luck and the whims of an AF motor!

 

If Nikon had billed this as a part-time or wedding videographer's outfit - fair enough. But 'filmmaker's'? - Not likely.

Edited March 26, 2018 by rodeo_joe|1

I guess my argument is that the "whim of the AF motor" is something that could be worked on. There's no doubt that the D850 isn't something you'd want to leave in AF mode for professional video, but something with PDoS might be if it were done well enough. The film industry has historically relied on making the subject land in the focal plane rather than making the focal plane land in the subject - and I would argue that, especially if the subject is human or under human control, that's not exactly reliable either. For the image quality we used to find acceptable, that's probably no big deal; as we shoot more 4K, IMAX and even 8K with digital sensors (so much less roll-off of local contrast with frequency) and with more use of larger apertures (partly to offset the increased set production cost you need if you want everything to look good at 8K and it's all sharp!) it's going to be harder to get the focal plane locked in: the depth of field is narrower. The same has happened for stills - we used to say "f/8 and be there", but many modern photographers would think that's a recipe for softness.

I wouldn't class sports coverage as 'film making'.

 

I see now that I interpreted your words out of context and assumed it to be about video in general and not specifically cinema. Although Nikon USA named this the filmmaker's kit I am sure they're not targeting films shot in large studios, but something that is made for a customer or personal project and is to have a more "cinematic feel" than if shot with a dedicated (typically small sensor) video camera. I.e. someone working on their own, to have the ability to make something that looks a bit like cinema. (I don't know how they think the customer will focus these lenses though.)

 

Anyway, if we assume a one-person crew, something that could be event coverage, news, an interview, or a personal documentary project rather than a scripted movie (it is debatable by what is meant by "film" since film itself isn't typically used any more), I think autofocus has its applications. I get it that poorly functioning AF during video is horrible to watch but examples I've seen from Canon's dual pixel AF during video suggest it is able to handle many situations quite well. For example if a person is talking they naturally sway forwards and back as their body doesn't like to stay in a fixed spot. If you have someone watching focus you can adjust it but there will be some residual wiggle in sharpness. Dual pixel AF seems to be able to handle this kind of a situation automatically so that you don't notice the transitions. A friend of mine who does video interviews a lot said he stops down to something like f/22 to compensate for the subject movement during the interview. I can't believe that this is the ideal solution. It may be the focus transitions are aesthetically pleasing when done by an experienced operator, and small errors may not matter so much, but towards higher resolutions, I believe AF with subject recognition will be better than manual focus in many situations (already in stills photography many portrait photographers have started to use Sony because of their eye AF, because conventional AF can't quite nail it with the same precision and consistency). It may not be today but some time in the not-so-distant future that this is also used for video. I also think that if the actors don't have to make an effort to stay still for the purposes of easier photography, then the result can be that they can take more freedom in how take the space and act, and the result can look more lively.

 

Also even in cinema (not just in sports coverage), there are action scenes and usually they are shot with almost everything in focus. With autofocus it might be possible to use shallow depth of field also for such scenes and this might be interesting for storytelling effect (sometimes you want to hide something and show other things). I like to shoot action at f/1.4 in my still photography, crazy as it may sound, but as the technology evolves, and I get more practice, the focus misses are getting more rare. I believe also in video action this can look good if done right. It may not be here today but I think it won't take too many years.

 

Anyway I think the key here is what is actually meant by a "filmmaker" in the context of this kit. We may be thinking of different things.

it is debatable by what is meant by "film" since film itself isn't typically used any more

 

Just to say, I'm not sure that's true yet. A little rummage on IMDB suggests that The Cloverfield Paradox (2018) seems to have been shot on film, and (maybe) so was some of Ready Player One. But I don't know how definitive that information is.

An actor's 'mark' needn't only be dictated by focus.

 

As a superb example, watch Johnny Depp in the prologue to 'The Libertine'. He struggles to keep on his mark within the skeletal lighting that's been carefully set up.

 

It's not just about keeping focus sharp.

 

It's all very well speculating about what AF might achieve in future, or what other companies' AF systems are capable of. The point is that Nikon's AF just isn't up to snuff at present.

Good point - although the focus is something that theoretically can be dealt with! (And in many cases it might be the most sensitive aspect.)

 

Nikon's AF certainly isn't adequate for a moving subject in video. I imagine this would be within their capacity to resolve when they have a viable mirrorless option, unless it relies solely on CDAF; I promoted it as something they might like to use to differentiate themselves. In the context of this package, either they need lenses suitable for manual focus and possibly automated external focus drive (as Thom Hogan suggests, Rokinon have some), they need to have technology other than what the D850 currently offers for video AF, the customer has to rely on a deep DoF and small aperture which negates the advantage of a large sensor (unless the subject is distant), or the subject has to be very static. Unfortunately, this means that the D850 as a videography kit is a lot less compelling than it might have been, and this looks more appealing as a way to get a bunch of primes on a budget.

 

But I'm not speaking as a videographer, so this may be more popular and useful than it looks.