kye

Great post - thanks! As a GH5 user I can see lots of interesting little improvements here that are against things that irk me about the GH5 so will be real-life improvements. A few thoughts: V-Log clipping at 88 isn't a big deal - for 10-bit footage the difference between 88 and 100 is inconsequential and it's designed to provide compatibility with V-Log cameras with larger DR on the GH5 the HLG mode contains the whole DR of the camera (unlike any of the other picture profiles) so if HLG is in the camera then that's a full-DR alternative to V-Log 13 modes rather than 5 is huge, and especially that there's 4 on the physical dial - I will definitely be using more than 5 of these modes Sleep function forgetting things was definitely a PITA so refinements to that is huge, and the extra custom modes help with that too I'll be particularly interested to see how the 1080p (or 2K?) Prores mode is integrated into the camera. In addition to image quality, it'll be interesting to see if it requires the newer batteries or not, etc. Can you give a bit more information about how well the current (h26x) 1080p ALL-I mode works? Particularly, I'm interested in how the footage looks from these modes: 1080p24 mode 1080p24 mode with 2X digital zoom 1080p60 mode 1080p60 mode with 2X digital zoom On the GH5 all those modes are downsampled and the footage looks flawless. Is the GH6 the same?

When you say "workflow" what kinds of things are you talking about? I notice that each discipline (cinematographer, DIT, editor, colourist, etc) uses this word to mean different things, which is pretty obvious when talking about an Alexa, but cameras like the GH6 are often used by people that don't fit into such nice neat boxes 🙂

My first experience feeling very old (apart from interactions with young kids) was seeing a medical specialist for something and thinking "are they old enough to have graduated from medical school yet?". I suspect that most of life as (say) an octogenarian would be simply looking around and seeing kids running everything and being amazed the world doesn't fall apart.

I saw another FP video yesterday, and even though I think the person didn't do a great job on the grade, you could see immediately from the footage that the camera isn't overpowered by high-contrast scenes, but seems to render them quite neutrally without it being a "stretch". This is perfect as it gives a really solid base for your grade and if you are wanting to really push/pull the footage then it gives you extra leeway to push/pull it further from neutral before the image starts to suffer. The more I see images from this the more I like what I see. I wish CineD would revise their DR tests, as the one they did was on the first firmware version and had some odd qualities to it that I think might have been improved upon in subsequent updates. Can anyone comment on the DR and image quality in V4 vs V1 firmware??

Thanks for your reply - very useful, and also heartening that it's not the footage that is the challenge but the camera itself. I'm also further heartened that it seems the challenges are just in working out the camera rather than challenges that come up shoot to shoot (like fiddly buttons or confusing menus etc). I've done many (many...) of these types of tests in the past to learn to get the most from the GH5 (and other cameras). I've shot test scenes in different modes, uploaded to YT, downloaded back from YT at different resolutions and then studied the resulting images to see what is visible at the end of the whole pipeline and what is not (4K vs 1080p source material is essentially imperceptibly different when uploaded to YT at the same bitrate/resolution). etc etc. This means that those considerations are, for me at least, kind of inconsequential. I'd do some tests, evaluate my options, and then just work out my rules for shooting. On first glance, and knowing what I know about lighting levels (which I've talked about in another thread) I'd be tempted to have DR+ on all the time, have a fixed ND that I put on when outside during day, and then just let it expose with auto-ISO. I tend to shoot with aperture only varying a few stops anyway, so that sort-of doesn't factor in that much. Every time 100 people say "FF" on an MFT thread, one less review gets published. We're voting with every comment, and most people are voting against themselves. Go back in this thread and re-read the last 10-20 pages - you'll be left with the distinct impression that no-one is in the market for this camera and that AF has killed MFT. It might not be true, but it's how it sounds with all the moronic comments that people make.

I'm quite familiar with how the "feel" of footage from different cameras can be very different, and was heartened by how the S1 footage felt similar to the GH5. What does the footage feel like to you? Obviously this is a highly subjective question, but keen to hear why you agree with their assessment about the footage being difficult to work with.

Have many camera operators gone through all the necessary certifications and permits to be able to fly drones commercially? I haven't seen much discussion of it in those circles, but maybe I missed it.

Yes, I saw that and it's hugely impressive. Not being into FPV and not being familiar with it, the thing that surprised me most was the low-quality and drop-outs of the video signal that the pilot was using. It makes sense logically that it would be low-resolution, but the noise seemed pretty extreme. It is a long way though. Drone pilot seems to be a profession that is (temporarily at least) booming, with both the introduction of compulsory licensing as well as the introduction of FPV-style operation which seems to be hugely more difficult. These may disappear once the AI gets good enough but in the meantime it's a whole new career in film-making. While not being in the context of the entertainment industry, there's also a huge market emerging for infrastructure inspection by drones (flying a drone around buildings or the infrastructure) laden with all manner of sensors that can automatically detect various issues. It's an expensive endeavour, no doubt, but much preferable to enormous scaffolding or hanging with ropes and doing inspections visually or by hauling huge equipment with you, which is the current option.

Yes, quite interesting and much more useful than the CineD type tests that only measure DR at native ISOs etc. It also reminds us not to compare DR results between different reviewers, which only adds to the value of this test as the methodology is (assumed to be) directly comparable between the four cameras they tested.

Very interesting! While things like this feature don't mean much to the full-manual folks, I think that implementation of little features like this can really influence the experience and even the overall capability of a camera, especially when shooting fast in uncontrolled conditions. I use the 2x digital zoom on the GH5 all the time, which when in the 1080p mode, still oversampled from ~2.5K sensor area to the 1080p, giving a high-quality image and extending the focal range significantly. Considering the size of these cameras, and the hand-held target market, these are the things that really matter. I'll be looking forward to this comparison! I'd imagine I won't be the only one - everyone who hasn't been tempted to the dark side by FF will be very keen to see how they compare.

1.5kg is pushing what you'd want to carry around, that's for sure. My GH5 / Voigtlander f0.95 prime / Rode VMP+ combo is approaching that and is close to the limit of what I can comfortably carry around for long periods. In terms of the FP-L the main consideration is codecs. I'd inspect the images with a fine tooth comb - if it's anything like the FP the uncompressed RAW is a tsunami of data and the internal compressed formats were really not what you'd hope for. I'm considering the FP but would only use it with external RAW recording - and that's only for my own personal work.

Thanks, that's really useful. I would have been very surprised if I could do 4K60 but not 1080p120! I have seen various references over the years of bit-depths being reported incorrectly - 12-bit being recorded as 14 or 16 bit, and 14 bit as 16 bit so it might be a common thing. As soon as you go to 10-bit then you need (in theory) two bytes to contain the value, which is 16 bits, so maybe it's saying that the container is 16-bits, but of course the data will be limited to whatever bit depth the sensor/camera is capturing. If I go this way I would prefer a BM solution as the integration with Resolve will be perfect and the bitrates will be much more manageable. One article said that 3:1 BRAW was the same as Prores HQ, so I'll be able to get equivalent bitrates but with greater bit depth, plus on the 120p shots I can use as little as 12:1 and get roughly the same bitrates as 24p. This means 3:1 1080p would be ~180Mbps vs FP internal 610Mbps, and for 100p I can go from the FP 2,530Mbps to as low as ~180Mbps. You can often get away with more compression on slow-motion footage as things don't change much from frame to frame but it's nice to have the option. I don't think that Prores RAW has as many compression ratios available - just the two. I note the VA can record to SSD via USB, so that might be handy too.

I'm trying to find a 1080p 12-bit workflow up to 120p from the FP into Resolve, but using a compressed codec to limit file sizes. The FP lists 12-bit 1080p up to 119.88p on the HDMI Output, so that seems good: The BM Video Assist units don't list any frame rates beyond 60p, even in lower resolutions: Also rather troubling is that they list the HDMI as 10-bit. Does that mean that BRAW is 10-bit from these units? The Atomos Ninja lists 12-bit so I'm assuming we can get 12-bit into the unit: Then the codec challenge is there because Resolve doesn't support Prores RAW, and the Ninja doesn't support the 12-bit Prores flavours. It does support DNxHR HQX which I believe is 12-bit (link) and I believe Resolve supports the format (it will render proxies in the format so seems good?) but DNxHR is UHD only - the 1080p version is DNxHD (the last letter is different) and those seem to be limited to 10-bit. There is a converter to convert from Prores RAW to CinemaDNG which Resolve will read, but I'd like to avoid the transcoding step if possible. Any advice would be appreciated.....

Actually, I think now is the perfect time for manufacturers to separate AF from face-recognition. On my GH5 it keeps the two together and so if you have AE on and enable AF then face-recognition turns on, sees the face, then adjusts the exposure to properly expose the skin-tones, but then if you turn AF off then it goes back to exposing the scene like a landscape in some generic way. Yet another way to say "if you use auto then you don't care about quality". If we're going to have the camera do something automatically then it should do it how we would do it - which is to expose the skintones correctly. If you want to adjust for a particular situation then that's what Exposure Compensation is for, right? If BM were to release a box-style camera then I'm not sure if I'd lean to the 12K sensor, or the 6K sensor from the P6K due to its much better low-light and dual-native ISO (the UMP12K is limited to 3200 ISO in firmware I believe). If they were to expand their range, it would make sense to me that they'd go MF/LF by splicing 2 or 3 of the 12K sensors together to go head-to-head with the Alexa LF, and that for the second one they'd go small and put the P6K sensor in a small box camera for drones / vehicle mounts / gimbals / light or b-camera duties.

Yes, it's unfortunate the pricing gap between the PCI solutions and the USB/Thunderbolt solutions, considering that the differences are basically putting a metal box around it. I've been annoyed at this in the past, being a laptop user. Another reason to standardise on 1080p workflow and give the middle finger to tech for its own sake with little-to-no visual difference (and get 4x the processing power for free!) 🙂

Yeah. The overall impressions are that BM only releases cameras when they've got stock ready to go, and with COVID that is a challenge, so likely delayed by the supply chain challenges. Who knows.

I think I'm just sick of the polarised way we talk about specifications. It seems that every specification that gets mentioned 1) spawns an entire tangent where we MUST descend down into a rabbit hole talking about the spec rather than letting it be part of a larger context, and 2) we MUST express only opinions that it is absolutely critical or irrelevant, and nothing in-between is mentioned or even tolerated, with usefulness comments treated as particularly heinous. This all-or-nothing mentality applies to resolution, AF, low-light, stabilisation, etc. People jump down that rabbit hole with all the nuance of Dr-Jekyll/Mr-Hyde. I'm talking about DR because it's an active problem I have on my GH5, for the real-life situations I shoot in, and it's something that the GH6 improves upon with a headline feature. Specifically it's something I've noticed that is dramatically less of a problem on the BMMCC than on the GH5, and they only have a stop or two of difference, so it's not an Alexa-or-bust type of situation - it's a situation where a small improvement would give a very large improvement in how many situations aren't capturable. In terms of people talking about usage and thoughts and experiences, yes, I really enjoy those discussions, but they're almost never held here. There's no reason they couldn't, but there are a bunch of subtle factors at play that discourage these discussions from happening more often. I've tried to start more than my fair share of them, and mostly they just die with people typing, frankly, ridiculous replies that almost couldn't be designed better to take the conversation off-topic or to discourage meaningful discourse. I think that's why a lot of great people leave these forums - they get sick of endlessly talking about specs without context and move into other places where the higher-order topics can be discussed, or stop talking about them at all. It's definitely a challenge to get people on here to talk about anything outside the camera department (have a look at the level of engagement in the colour grading and editing threads I've tried to maintain) but we don't even talk about lighting here, and even the lenses section is relegated to a subforum where we know that topics go to die. This is a camera forum at best, and a camera specification forum most of the time. And most of the discussion is in threads about cameras that no-one here will ever own, and probably can't afford (Z9, R5C, etc etc). Then a thread about the GH6 appears, and everyone makes fun of the technology behind the AF, and when a video comes out they don't like they write the entire camera off entirely, like somehow a poor image is the cameras fault, when others have previously posted great looking images. The more I concentrate on everything other than the camera box (lenses, colour grading, editing.... story *gasp*) the more I realise that the camera plays a very small part in the overall picture. The whole "camera social media echo chamber" is perfectly designed to make you concentrate on the camera. The image from a camera is more noticeable the better it is, great colour grading is credited to the camera not the colourist, great editing becomes invisible rather than standing out, great audio is completely invisible, great sound design is at-best subliminal if not invisible, and story is so deep a subject that there's no way it's going to get discussed here. Cameras are where the action is from a specs point of view because the specs are really obvious. The best camera videos actually have really rudimentary editing - compared to what I'm seeing in high-end TV the camera videos might be 10% on the editing scale, maybe 5%, or even less, sound design is similar - may 0-20%, story is often non-existent, but these camera videos are delivering cinematography at 80% or occasionally more of professional standards. This teaches you to think of the camera as being the star of the show. If the images looked poor then these videos would be laughed at almost universally, and yet the image quality of every award-winning anything from 20+ years ago would be worse. I think the echo chamber is just losing its hold on me and I'm getting more frustrated by the specs-first polarised way that these things are discussed. Sorry if I'm a bit rude amount it, but the more I learn about film-making, the more I realise these conversations are fundamentally missing the point, and wasting the opportunities to have a meaningful discussion. As such, let's change it up from "no-one cares about DR" and "MFT is doomed due to its low DR compared to FF".

I've never really shot manually, but am now contemplating it, and needed to get my head around the various situations and how bright they are, what NDs would be required, etc. It's a basic thing but no-one seems to talk about it, and when it comes up in other threads knowledge seems to be a bit patchy, so I figured I'd do a bit of testing. My first test was just to work out what the exposure range is of how I shoot. I shoot in available light, so have to adapt to whatever is there. I shot with a colour checker, just exposing the brightest patch so it wasn't clipping, across the full range of situations I shoot in. These include in full-midday-sun, outside in the shade, inside in a brightly lit room (lit from large windows of natural light), in a small room lit by a single LED bulb (8W 800 lumens), and lit by an iPhone 13 Mini with completely white screen at full brightness at 30cm from the checker (this is a useful reference for me as I shoot a lot in blue hour or darker and often peoples faces are lit by their phones). What I found was this: All measurements were within a stop of being right - if a setting clipped the square I went down a whole stop. Yes, I'm aware 1/50s isn't exactly one stop lower than 1/125s (if you care about this then you've missed the point, please immediately proceed to the DPreview forums). My goal is to create natural looking images across these lighting levels, and was thinking that perhaps it could be done using a combination of a single vND, aperture changes, and a dual-native ISO camera. The eye naturally closes the iris in bright lighting and opens it up in low lighting, so there is an argument that to be more natural you would also do this, so this is aesthetically relevant and also helps us out technically. It looks like the dual-native ISOs on cameras tend to be around 3 stops apart (P4K 3-stops, P6K 3-stops, S5 2.66-stops, R5C 2-stops) although the Sigma FP seems to have 6 stops of difference between them, which seems the most useful. Then if we said that we could use F1.4 and F5.6 then that's another 4 stops, bringing us to 7. That would leave the vND to take up the remaining 7 stops, but in extreme low light situations like the iPhone then using the ISO above the higher native ISO seems reasonable. Most cameras will have a usable noise level between their base-ISO and upper-native ISO, so if they have those 3 stops apart then we can probably go about 3 stops above the higher native ISO and stay within what the manufacturer deemed was an acceptable level of noise. Especially considering these cameras all have enough resolution / codec to handle a bit of NR. In a sense I'd like the ability to shoot a bit wider open in brighter conditions than 5.6, but I could "buy" that by pushing the ISO at the darker end. How do you manual-shooters who shoot in available light configure your setups with NDs? Obviously if you're working with lights then you can just adjust them instead, but that's a completely different way of working, and you can use fixed NDs too, as you're not working fast.

I'm not even sure WTF we're talking about here any more.. Sure, phones are "real cameras", no-one uses "real cameras" any more, people don't care about digital clipping as long as they can tell who is in the photo, and if people don't know what the specification is called then the effects in the image don't matter or are somehow invisible until you can memorise the acronym and then they magically become visible. I now feel much more educated, and it's definitely all very relevant to the GH6 and it's dual-gain technology.

We're simply not the target market - that doesn't mean it's a bad product or anything. In a sense, you offering your opinion on this is kind of like offering it on language classes that teach Japanese to people who speak Spanish - it's just not aimed at you. PCs (of which smartphones are just another version in a smaller box) are jack-of-all-trades, but also master-of-none. There are things that you will have a seriously hard time doing, if they're possible at all. Anything that absolutely has to be done real-time is one of those - computers are devices designed for no guarantees because there's always the chance that Windows or OSX or whatever will decide it's time to update the driver for the mouse and video playback will cease for half a second or something. Good luck getting a video signal out of Resolve that hasn't been "made better" by the operating systems colour management processes. Good luck getting a specific resolution and frame rate out of OSX - at all. This is a box that does one thing absolutely reliably, where your computer is something that does a bunch of stuff, which is most of what you want, unreliably. That's why BM sell hardware, and why people buy it.

BM have an entire ecosystem of video hardware that is designed for broadcast. I suspect you could probably run a small TV station almost completely on equipment they make. A setup consisting of a camera and a PC for editing and colour grading is a very small part of such an enterprise - we just don't talk about the hardware side of the industry that much on here. Saying a hardware device that plays/records media should also be a software controller because it has a knob and some buttons is like saying that a cinema camera should also be a dash camera because they both have sensors and write to a media card 🙂

Well, I covered that in my description - people don't care about DR.........IF they're shown ONLY images where it's not a problem for the story. I think I've personally seen about a third of the people I know struggling to take photos of their kids against a bright background and maybe a quarter of them have sworn out loud while struggling with the DR challenge. I'm not that social a person, but just check your Facebook feed to see how many parents of school-age kids are taking pictures of them on their first day back to school in the new year. If you think that CAMERA = WATCHING TV then sure, no-one cares about DR, but if you think that CAMERA = CAMERA then millions upon millions of people care about it. Ask people if they think that the quality of digital cameras has improved over the last 20 years and I'm pretty sure you'll find people perk up and talk about the photos they can take, rather than talk about the TV they watch. When I was travelling in India with a charity we went out to the villages of the rural poor and they lived in the middle of nowhere down dirt roads and had dirt floors. However, they had old cellphones and were very happy to show photos of their kids. These people did not have a TV. (or running water, or access to proper sanitation) Perhaps you're forgetting, the TV show you're making is the most important thing for you but for your audience they couldn't give a stuff about it compared to the images they want to make of their own friends and family. They say that we need to make things where the audience identifies with the characters because people want to watch TV and movies that are "about them"... well, the photos they take are ACTUALLY of them. You can't beat that.

The false colour seems like a game-changer if you want to shoot RAW in manual mode without an external monitor. Great update!

Think of it like diffusion. I fundamentally disagree - people will notice and will complain HARD when there isn't enough DR. "Why is the sunset white?" "Why are the people in front of the sunset pitch black?" What you are saying above is essentially "People don't care about DR, assuming that they're not operating the camera, and assuming that whoever did operate the camera either controlled the scene to not exceed the DR of the camera, or filtered out all the shots without enough DR so that the person only saw shots where the camera had enough DR". That's a VERY different statement than they don't care.

I thought this was interesting: One significant point he makes is that because DR hasn't got any new huge headline features that they'll spend time consolidating what they have, making more incremental improvements, and that it's a sign that the package is maturing. To be clear, for anyone that isn't familiar with Resolve, in the last 5 or so years when they make a huge upgrade it's on a level that is completely beyond what the other NLE's would ever do - one year they added an editor, another year they added Fusion, another year they added Fairlight, etc. Think of Resolve as more like the entire Adobe video suite but just in tabs 🙂 However..... Resolve 18 is not what I was hinting at for this thread. The rumour is that they'll drop a new camera on the first day of the show. Which is still in a few days time.