Emanuel

Now, about using it in summer... are we talking about a mild summer, or an actual summer that isn’t even properly hot, where the camera doesn’t need to suddenly remember it has thermal limits? Here’s a useful 45-minute guide: And last but not least, one of the best comparison tests I've found so far, if not the most interesting with a unique range of pros and cons is here, as well as probably the smartest use of a fine example for low light is there (serves as tutorial too). Also worth adding: the new Pocket 4P/Pro filters are already starting to appear, along with another comparison around its brand-new, cutting-edge LOFIC image sensor, including the bitrate difference, going up to more 50% i.e. 180 Mbps versus Luna Ultra’s 120 Mbps.

A couple of my cousins did this a couple of decades ago : ) with Kodak disposable 35mm film cameras. No photographer, no wedding videographer at all ;- )

Love the way you put it! : ) As everyone of your posts BTW < 3 Thanks, always a pleasure reading you : ) Great-juicy-post BTW part II ;- )

These two [1*] [2] Korean hands-on videos are probably among the most useful references I have seen so far on the Luna Ultra / Pocket 4P discussion. EDIT — plus this one from another reviewer elsewhere: [3] Not because they end the debate, on the contrary : ) but since they also help show the real operational trade-offs better than most spec-sheet comparisons, while still offering some fairly clear findings on the usual comparison points, such as outcome, colour or dynamic range, whenever those aspects are covered. source *In this 1st video, right from the start, you can see exactly that approach: using this kind of camera as a serious B-cam tool (Osmo Pocket 3) in commercial work, very close to the way I have also been using small capture devices in a similar role, as mentioned before.

Yes, exactly. That was the part I was agreeing with and developing, not trying to claim as a new point. If two people are saying the same thing, there is usually a reason for it. Always better than one person going one way and the other going the other way, just to create confusion, isn’t it? ; ) What interests me is that once convenience starts affecting the rhythm of the set, it is no longer just convenience in a minor sense. It becomes part of how the scene can actually be made. Time, pressure, continuity, the actors’ energy, the crew’s patience... all of that can feed back into the creative result. And I take your point on the 2x/3x terminology when it comes to lens design, range, price and compromise. In that context, of course it tells you something useful. My issue is more with the way it is used as shooting language, especially with phones and small cameras, where “2x” or “3x” often replaces any real sense of focal length, distance or perspective. Your parlor trick is actually a perfect example of what I mean. If you can look at the lens and camera position and already have a good idea of what is probably in the frame, then focal length is not trivia. It is practical spatial knowledge. And yes, Kubrick was right there. The frame may look similar, but once the camera distance changes, the shot has changed.

That Kubrick story is a fine example because it puts the whole thing in very practical terms. It is not about fetishising focal lengths, or saying one lens is more “cinematic” than another. It is about the fact that the moment you decide whether to move the camera or change the lens, you are already making a directing choice, not just a technical one. Much as with the recent example I gave from Leonel Vieira’s A NOITE, which was based around specific focal lengths, the 75mm and 100mm lenses I mentioned. That is not just a preference for a certain look. It is about camera distance, pressure, and the way actors are observed inside the frame. Different names, different contexts, but the same practical truth: once the choice is made consciously, focal length stops being just a number on the lens and becomes part of the mise-en-scène.

Interesting to see this thread again in 2026. Back then the idea sounded a bit futuristic, but now it feels like we are finally getting there from a few different directions. The iPhone LiDAR side is still interesting, especially with things like LidarAC, but what really caught my attention more recently is that we now have dedicated systems actually trying to turn manual lenses into something much closer to AF behavior. DJI Focus Pro is probably the most serious example so far. PDMOVIE is trying something similar too, though it seems a bit more compromised in how it decides what to lock onto. So in a way the original idea in this thread was right, just early. It was never only about using the iPhone as a clever measuring tape. The bigger idea was using LiDAR / distance mapping / external motors to bring some level of autofocus logic to manual lenses and cine setups, and that definitely seems real now. We are still not quite at “perfect native AF for any manual lens”, but compared to where this discussion started, it is no longer science fiction either. Has anyone here actually used one of these newer systems on a real shoot rather than just testing it at home?

I think @kye’s point still stands, and maybe it becomes even clearer once you move away from the usual zooms vs primes argument. It always comes down to how much convenience a zoom actually buys you, especially when time becomes critical. On a controlled set, changing lenses can break rhythm, slow down the crew, cost minutes, and sometimes cost the shot. In uncontrolled environments, it can be even more decisive, because you may simply not have the time, the space, or the chance to change lenses before the moment is gone. So yes, a zoom can solve a very practical problem: time, reaction and continuity. That should not be underestimated. If the situation is changing in front of you, the practical convenience of a zoom can become creatively important too. But I don’t think this cancels the OP’s deeper point. There is also another kind of convenience in truly understanding a given focal length. It teaches you where to stand, how close to be, how to move, how much space to leave, and how the scene changes once your own position changes. A zoom may save the moment, but understanding focal lengths changes how you read the moment in the first place. And this is also why I have never had any patience for the 2x, 3x, 5x terminology. I always prefer the actual focal length value, because that tells me something real about the relationship between camera, subject and space. “2x” may be convenient marketing language, but “35mm”, “50mm” or “85mm” immediately tells you much more about how the image is likely to feel and how you may need to position yourself. Leonel Vieira is a good recent example of this. This June, while shooting the making-of on A NOITE, I was able to observe that very particular relationship he has with focal lengths, especially his liking for 75mm and 100mm lenses, focal lengths he has often favoured and used there. That is not just a technical preference. It affects distance, performance, compression, the way actors are observed, and the emotional space between camera and subject. With those lenses, the focal length becomes part of the staging itself, not just a number attached to the lens.

Exactly. What you call availablism is central here. But with Wong Kar-wai it never feels merely opportunistic. The available does not remain simply available. It becomes emotional architecture. Pre-handover Hong Kong gave him the lights, the colours, the cramped spaces, the corridors, the streets, the reflections and the limitations. But the art is in turning those given elements into mood, memory and desire. That is the difference between just using a location and allowing the location to become part of the film’s inner life. In that sense, the uncontrolled city is not only a background. It becomes a collaborator. Wong Kar-wai does not simply take what is there. He finds what is hidden inside what is there.

The gyro control itself is not new, correct. But I don’t think the interesting part here is simply that “gyro control exists”. Of course it does. Larger gimbals, phone apps, remote monitors and systems such as the Ronin 4D have already explored that territory. The question for me is not whether the underlying idea existed before, but whether it becomes useful in a different way once the whole system becomes small enough, fast enough, integrated enough and unobtrusive enough. Many ideas in cinema technology existed before they became truly useful. Stabilised camera movement existed before Steadicam became the right combination of body, balance, operation and image. Small cameras existed before 16mm, and later DV, changed the way filmmakers could move through reality. Remote operation existed before it became practical in the hands of a one or two-person crew. So yes, if we reduce this to “a gyro controlling a gimbal”, then it may sound like nothing new. But if we look at it as a pocket-sized 1-inch 10-bit Log gimbal camera, with several focal lengths, proper monitoring, autonomous operation and a dedicated head-tracking accessory, then the proposition changes. It is not only the control method. It is the form factor plus the image pipeline plus the operating mode. And this is also where the remote / detachable screen side becomes important. If this is not a meaningful step forward, then how do we explain that DJI, despite all its experience in this category, still seems one step behind on this specific point? Not necessarily behind in image quality or engineering as a whole, but behind in this particular operational concept. The comparison makes that fairly easy to see. DJI can offer remote control through its ecosystem, apps and accessories, but Luna Ultra brings the detachable screen, remote monitoring and camera control directly into the body concept itself. To be fair, DJI still has a major advantage in ecosystem continuity. If, apart from the optical accessories, the Osmo Pocket 3 accessories remain compatible with the Osmo Pocket 4P/Pro, that is obviously a strong point. It means users are not forced to abandon an existing accessory ecosystem. But that is a different kind of strength. It is backward compatibility and ecosystem maturity, not necessarily a new operating concept. That matters because the remote is not just a convenience feature. It changes the way the camera can be used. There are also small operational trade-offs on the DJI side that are worth noticing. For instance, the dedicated low-light video mode tops out at 4K30p, while Luna Ultra records up to 4K120p in regular video mode and offers PureVideo low-light capture up to 4K60p. Not the whole story, of course, but another sign that Insta360 seems to be moving ahead of DJI* in a few practical areas here. And yes, we have already seen Insta360 explore part of this same logic with the GO line, namely the GO Ultra. With the GO Ultra, for instance, you already have that very useful modular idea of separating the camera unit from the monitoring/control/battery side of the system. In the Luna Ultra, this logic is taken into a different class of camera: the detachable touchscreen remote allows independent monitoring of battery information for both the main unit and the remote, and the system manages charging between both parts. Even if the remote does not appear to support fully independent USB-C charging while detached, the operational concept is still important. The monitoring/control side and the image-capturing side start to behave as distinct elements of the same camera system. That is precisely the point. The more the monitoring, battery handling and control are separated from the visible camera body, the easier it becomes for the device to disappear into the situation. And for documentary, BTS, production diaries or observational work, that can be a much bigger deal than it first appears. A phone strapped to your head controlling a larger gimbal is one thing. A compact dedicated device that can sit inside a BTS, documentary or walk-around setup, become boring after a while, and follow intention without the operator constantly raising, aiming and correcting the camera is another thing. That is where I think the usefulness may appear. Not necessarily for everyone. Not necessarily for controlled narrative setups. And probably not as a replacement for a skilled operator with a proper camera package. But for small crews, making-of work, observational documentary, rehearsals, production diaries, street work and situations where the act of operating the camera visibly changes the behaviour of the people being filmed, I can absolutely see the value. In that sense, I don’t see it as revolutionary because gyro control is new. I see it as potentially revolutionary because a previously awkward idea may finally be arriving in a form factor where it can become natural, invisible and operationally useful. *And not only when compared with DJI or another compact device... ; ) take a look at this higher-end comparison too : X

During the shooting of IN THE MOOD FOR LOVE, the lead actress, Maggie Cheung, complained to her co-star Tony Leung that she had absolutely no clue what was going on. He told her something along the lines of not to worry, because they would end up with a masterpiece. That uncertainty was not accidental, but part of Wong Kar-wai’s working method. Rather than starting from a fixed, fully locked script, he often finds the film during the shooting itself, shaping scenes, characters, rhythms and emotional tensions through repetition, variation and discovery. The actors may not always know exactly where the story is going, but the final film emerges from that process: less as a pre-planned construction than as something gradually found in the performances, the silences, the gestures and the atmosphere. It remains one of my favourite films, and Wong Kar-wai one of my fave filmmakers BTW : ) On a personal note, my own career in Asia as a filmmaker began there. Not only because of the beauty of his images, but because of the way he turns uncertainty, restraint and emotional incompletion into cinema. Few films are so precise while seeming so elusive, or so deeply romantic while refusing almost every conventional gesture of romance. With Wong Kar-wai, desire always seems to exist somewhere beyond the frame. The way the inner universe of the male character is developed makes cinema an internal experience: from the male character into the viewer. An absolute gem. Wong Kar-wai shoots the soul from within, and the inner world through the soul. Remarkable stuff, and yet it is there.

No. Remember: the First Amendment ; ) Secondly, we now have AI, and AI can be made to justify almost anything ;- ) The Soviet Union came to an end on Christmas Day, 1991.

WOW What a thread! :- )

Exactly, that is the interesting part for me too. I don’t think the best use of this will necessarily be obvious “POV footage” in the usual action-camera sense. The more interesting uses may be the almost invisible ones: situations where the camera is there, alive, following attention, but not constantly being operated in a way that announces itself to everyone in the room. Your point about parallax and offset is completely fair. If the camera is too far from the head, then the head direction and the actual camera perspective will never match perfectly. But I am not sure that perfect matching is always the goal. In many documentary or making-of situations, what matters is not to reproduce the exact optical perspective of the eye, but to reduce the burden of manual operation and preserve a more natural relation with the subject. That is why I find it interesting for BTS, documentary and observational work. Not as a perfect substitute for a shoulder camera, not as a Steadicam either, even though, as a gimbal camera, it belongs to a lineage that goes much further back, including the invention that so impressed Kubrick that Garrett Brown ended up operating his own system on THE SHINING. I see it more as a small autonomous B-camera, or a reactive camera system, able to follow intention without forcing the operator to constantly lift, aim, correct and therefore disturb the situation. In a film set, for instance, the value may be precisely in allowing the device to become boring after a while. If people stop reacting to it, or react to it less than to a conventional camera operator moving around with a visible setup, then it starts doing something useful. So yes, I think the strongest uses for it may be almost invisible. Not spectacular POV shots, but the kind of footage that exists because the tool reduced the amount of intervention needed to capture it.

Yes, but I don’t think it is exactly the same thing. A GO clipped to a hat gives you a small head-mounted action camera. Useful, of course, and probably better for some pure POV situations. And I say this as someone who actually owns two Insta360 GO 2 units and the GO Ultra. In fact, the GO Ultra was also used in the making-of situation I mentioned above, so I am not dismissing that kind of camera at all. But the interesting part of the Luna accessory, at least to me, is not simply “put a camera where the head is”. It is the possibility of using the head as a control input while the camera itself can remain somewhere else. And that is a whole different thing. With a hat-mounted camera, the camera position, lens, sensor, stabilisation, codec, monitoring and point of view are all physically tied to the head. With something like Luna Ultra, you are potentially dealing with a much larger sensor, better latitude/dynamic range, 10-bit colour, log recording, stronger codec options, proper monitoring, several focal lengths, a real gimbal system and a camera body that can be placed where the shot actually makes sense. So it is not just “POV versus POV”. It is the difference between a tiny self-contained action camera recording from the head, and a more complete image acquisition system where the head becomes one possible way of controlling the frame. The gaze and the camera body become partially decoupled. The head can guide the framing, but the camera can still benefit from a better sensor, 10-bit recording, log capture, stronger codec, better lens choice, better stabilisation, better monitoring, better audio possibilities and a more deliberate production setup. So I agree that an Insta360 GO on a hat is a valid alternative for some situations. But I see this as potentially more than that: not just a POV camera, but a different way of operating a small gimbal camera with less visible intervention from the operator, while still keeping the image quality, 10-bit colour depth, focal-length flexibility and production control of a more serious camera system. And the outcome is already starting to speak for itself. Even if the example comes from the competing Osmo Pocket 4P/Pro side, the IQ benchmark is becoming very clear. D-Log 2 10-bit colour: https://www.facebook.com/reel/937883892382597 Small pocket gimbal cameras are no longer just convenience tools or “good enough” secondary cameras. They are beginning to produce images that can stand on their own, with a cinematic quality that would have been unthinkable in this form factor only a few years ago. Back to the subject and to my eyes, this is not just a clever accessory. It is part of a real revolution already in motion.