Leaderboard

Somehow I missed this promo short film, its kinda brilliant in how it showcases a lot of the features in a pretty creative narrative, really liking the overall image from this new sensor:

Technically a sensor by itself doesn’t have a “look,” but bigger sensors change how lenses behave and how we can use the image circle, which is where the cinematic signature comes from. FF already gives shallower depth of field for equivalent framing and a FoV closer to classic 35mm photography. Open gate modes on 3:2/4:3 sensors take that further by giving you the full vertical readout, letting you reframe, use anamorphic, or extract stills without losing resolution. Medium format pushes this even more, however Fuji’s GFX sensors aren’t “true” MF like Alexa 65 or Hasselblad backs, but it’s about 1.7× the area of FF, which is enough to reveal more lens character, smoother falloff, and a slightly different perspective than FF or S35. So when you pair the sensor with fast primes and open gate capture, the format really does offer creative possibilities that smaller sensors simply can’t match. I'm loving what I'm seeing from the Eterna so far by the way!

Not trying to jump into your apparently long discussion, but EXACT look of Full Frame can be achieved on M43, but you need faster lens, and once you go that fast, the lens becomes so big and heavy that the size advantage of the M43 format loses its relevance at that point. But the difference between 44mm wide sensor and 36mm wide sensor is far less than FF and M43. Its basically f/1.2 vs. f/1.4. And at the same time, lens makers of the two systems went opposite direction. FF lenses are now bigger and faster than 44mm MF lenses, cause FF systems want to differentiate themselves from low end cameras, even at the expense of size/weight, while MF systems want to break the collective mindset that MF=Bulky/Inconvenient.

@eatstoomuchjam Funny enough, the Greg Fraser quote you referenced actually underlines my point. He’s saying outright that moving to a larger format opens up creative possibilities precisely because you’re seeing parts of the lens you wouldn’t on a smaller sensor. The lens itself hasn’t changed, but the relationship between the sensor size and the lens’ image circle absolutely changes what’s captured. That’s all I was getting at: bigger formats don’t alter lens physics, but they alter what part of the lens’ projection you get to play with, which translates to a different look/feel on screen. This is similar to when you have open-gate (true 3:2 or 4:3, not just 17:9), you’re also opening up different aspect ratios and lens visibility. Whether you see the extra image area as flaws or character is subjective, but the distinction Fraser is making is exactly the one I was trying to point to. I know many GFX users like to put certain FF lenses for identical reasons, and the GFXRF & Eterna 55 use that high MP large sensor to open up multi aspect ratios which is why I don't think its irrelevant to the discussion.

Sorry, but this sentence makes no sense. A bigger sensor doesn't change how a lens behaves. The lens always behaves the same and projects the exact same image circle. And "cinematic" is a meaningless term so "cinematic signature" is equally meaningless. This is untrue. Assuming that you mean compared with S35, to be more specific, FF gives shallower depth of field for equivalent framing at the same focal range and aperture. You can get an identical image by using a wider S35 lens with a bigger aperture. This is exactly what focal reducers do when they focus the image circle of a FF lens down onto an S35 sensor. In fact, when using a focal reducer, a Komdoo or Komodo-X has slightly less DOF at equivalent framing than a natively FF camera (the equivalent crop factor at that point is something like 1.05x). This also becomes largely an academic distinction if you don't insist on shooting fast lenses wide open 24x7. And even if you do, do you need shallower DOF than Army of the Dead, that Zack Snyder movie from 2021 that was shot entirely with the Canon 50/0.95 dream lens wide open? This is true, but irrelevant to a discussion of whether there is or isn't a medium format look. It only reveals more lens character in the sense that for some lenses, you see the worse parts of the image circle outside of the standard 35mm film size. In some cases, it also just means you can't use the lens at all - for example, my Noctilux-M 50/1 only barely covers 24x36mm and already has dark corners and edges. On a GF sensor, you just get the edge of the image circle surrounded by blackness. That lens has plenty of character already on FF. On film, yes, but this is related to the inherent resolution limits of film. On the 100 megapixel GF sensor, this is technically true vs 35mm format cameras that have 61 megapixels, but it's largely an academic distinction that is barely noticeable in practical terms. But when you're using line skipped/binned 4K off of that sensor, you have less smooth falloff than off of a 35mm sensor recording 8K. This is also academic and can barely be seen. At the same focal length, sure. Luckily, we can change lenses. It would be true to say that you can capture a different image with the same lens as a smaller format. So if your goal is to get a different look out of your vintage Nikkor 200mm lens and if that lens has an image circle big enough to nominally cover a sensor that's 44mm wide. If you're shooting 16/17:9, will that difference be substantially different than the image from a V-Raptor XE with a 41mm wide VV sensor (that costs $1,500 less)? No. That or if 3:2 capture isn't a concern, go rent the Ursa Cine 17K 65 with a 51x24mm sensor. Yeah, footage from the GFX 100 II looks nice too. But so does footage from the GH7, a camera with a much smaller sensor. For video, I prefer footage from my UC12K LF to what I get from the GFX 100 II.

Really funny indeed, it seems with long takes the only way to trim and save only important parts of footage is through Resolve’s Media management. Tried to rename clips to R3D first in CFExpress, but exporting trimmed parts from Resolve did not work. Compared one old clip and other than R3D’s chroma NR there was not much difference to NEV. Have to test this more with clips that are saved straight from card as NEV.

It would be funny if turns out RED and Nikon couldn't push Davinci developers to address the issue with nraw rendering and decided to fix it themselves by a metadata trick.

Yes, but as long as some people still buy VW and some people buy the upgrade, other car manufacturers will be inspired to do the same. This is how the enshittification race to the bottom works. It used to be that if I bought a plane ticket on any airline, I got a checked bag for free. Not only that, but overhead space was plentiful and I could be a bit chill about boarding the plane. Then one airline started charging for checked bags. When people didn't stop flying with them, the other airlines followed suit. After all, people are apparently willing to accept a shittier flying experience. Now checked bags cost money and overhead space is crammed full which makes me one of the annoying boarding buzzards circling around the boarding stanchions so that I can be sure to get on in time to put my camera bag above me and not get into an argument with the boarding agent about whether I'm going to gate check it (and no, gate agent, I cannot relocate all of the lithium batteries into my other much smaller "personal item" that goes under the seat in front of me). Anyway, the shrugging and accepting it is basically what is at the core of enshittification.