kye

I agree, and I've come to think about under/over latitude tests as the measure of useful DR from a camera. I haven't yet gotten an understanding of how this usable DR relates to absolute DR, but early observations are that they seem to be proportional. As someone who incorrectly exposes, almost exclusively, this is the main concern I have. Obviously if you're pulling things up (I don't often overexpose) then noise and the various characters of that noise become significant, but NR is very good these days, so it's the strange colour shifts that I find are what limits this.

I've done several international trips where I was filming with the GH5 and my wife was taking photos with her iPhone - often while standing right next to each other, and when I got home and got copies of all her photos I found that the colour that the iPhone had for its still images was very impressive. Colours looked bright, punchy, giving that clean commercial look where the world looks nicer than it really is. That's a hard look to do in post, and I tried a few times to grade my GH5 HLG footage to match it, but at the time wasn't able to get that close. I've learned a lot about colour grading since then (and still have practically an infinite amount more to learn) but I definitely made a note that I should try and match the two. Apple have done a great job of their image processing, that's for sure.

DR is highly susceptible to processing, like highly highly susceptible, so if there's any change to how the camera processes the image, or there's any change to how it is processed in post then the DR measurement will reflect that. It could be that something like the debayering is different etc based on the metadata from the cameras (ie, not something that he did) or even that it got upgraded int he software in the meantime. I'm starting to lose trust in DR measurements as the more I look at it the more I find that it's sensitive to things that don't matter in the real world, and isn't sensitive to things that really matter in the real world. Not saying that DR itself doesn't matter - far from it - but the way it gets measured I think isn't a good representation of that.

Gimbals and drones.

I'm alway torn between things being functional and being small/light. I would absolutely rig out a camera if it wasn't too large/heavy for what I am doing (shooting personal projects under the radar). The functionality of a rig like yours above is hard to argue with. 8lb though... hand-holding that for hours - my arm would fall off! I've heard other people who shoot minimally use that as a line to help with concerned clients. Just saying "this is how I shot all my work - including <project the client mentioned during the sales process>". Nah, it's cheap LEDs that the problem!

Yes, you only have to add a few items before the size of most camera bodies is visually irrelevant. As someone trying to get the best image quality while actively NOT looking professional, it's extremely inconvenient that this is the case!

New Sirui FF cine lenses... Caleb describes them as being like rehoused vintage lenses, combining the optical properties of vintage glass (flares and haze) with the solid build of rehouses, but for significantly less cost / hassle than the "real thing". More benefits from cheap Chinese manufacturing responding to the vintage lens craze. Good stuff IMHO.

One of my favourite YouTubers is Martijn Doolaard who is renovating a cabin in the Italian Alps and besides the obvious fact that he's a great cinematographer / editor / storyteller, the frequency of fog in the shots I think makes a huge difference. He recently did a Q&A and mentioned his equipment and it's nothing special (A73 and DJI drone) but the footage just looks magical a lot of the time.

I can't remember where I saw it, but somewhere I saw someone from ARRI saying they have been working on the new sensor for years and had one ready to go years ago but the quality wasn't up to their standards and so they went back to R&D. The message was that they weren't in any hurry (as they had 4K+ in the LF) and that quality is their highest priority. It makes sense too - they could easily have released a 4K S35 sensor by now (or a decade ago) if quality wasn't their highest priority. I think ARRI (and their IQ) pairing 4K (and instant Netflix approval) with S35 (and all the lenses that are out there) will be a very popular option. It'll be the "safe" option in a world where 4K is a magical number for cameras (despite not being so for eyeballs).

If you can handle the form-factor.....

I'm aware it was a fog machine. I've been consistently amazed in BTS videos of large productions the sheer quantity of fog that some productions have - and the fact that there are machines capable of generating so much of it. I have vague recollections from before digital of a movie adding fog to several entire hills to get the fog-in-trees look while doing a huge dolly-up on a crane that went from the subjects up through the canopy and out the top to show the hills of trees with fog. Must have been an army of people in the forrest with machines all set to full-blast! Colours in the images look really nice too BTW 🙂

Good post. To clarify I was comparing RAW Linear with compressed Log in the context of ETTR vs exposing for middle grey. ETTR is a great strategy if you're using Linear RAW and don't mind compensating for the level changes in post. The Alexa is a bit special in this way (I believe) as it can output Log in very high-bitrate 12-bit 444, compared to (say) the GH5 200-400Mbps 10-bit 422 log, so the Alexa has an enormous amount more latitude just because of the codec. It's also special in the sense the readout from the sensor is a higher bit depth than the Log codec. On a camera with a 12-bit sensor output, if that was transformed into a 12-bit Log profile then the highlights would lose the effective bit-depth they had when in Linear, but the other stops wouldn't really benefit much because although the 12-bit log has better bit-depth for the lower stops, the limitation would always be the bit-depth from the Linear readout. Regardless, having a 12-bit sensor readout converted to a 10-bit log profile in-camera is probably good enough for almost all applications - I've tried to break the 10-bit HLG on the GH5 by applying curves that were severe beyond reason and the image held up. Of course, the thing missing from most consumer cameras is a decent codec, like Prores HQ or 4444. I'm really looking forward to the GH6 update to see how the 1080p Prores looks. Unfortunately most cameras have really weak internal h264/5 implementations and the in-camera downscaling is often very poor. I suspect the push towards RAW in this segment is partly a way to deliver better image quality without having to design cameras that can scale and compress properly or having to buy a Prores license. Unfortunately it then means they run into the challenges around internal compressed RAW and so we end up with tiny cameras needing an SSD or external recorder strapped to them. Hardly ideal.

@TomTheDP In addition to the above, and I don't know if it's related to DR or not, but the more modern cameras seem to have much better behaviour in their under/over exposure tests. I had the impression on older cameras that when you went under/over significantly that colour shifts and strange / unpleasant behaviour crept in before clipping or noise, whereas now most cameras seem to have only very subtle colour shifts right up until they clip or get overtaken by noise. Maybe I'm getting that wrong, but the DR that exists between noise and clipping is far less relevant if the usable DR between strange colour shifts the dominant restriction.

Living on the west coast and having the beach as a convenient test location, I shoot a lot of sunsets, and my experience with the BM OG Pocket and BMMCC are that there's enough DR to just clip a round section just near the sun and still get shaded skin tones usable (to my tastes anyway). I have no idea if they were exposed "properly" at the right levels, or if they were one (or more) stops under, but the IQ seemed perfectly acceptable. I know the BMMCC is still relatively good in terms of DR compared to all but the current generation of cameras, but with those the exposure challenge is even easier to deal with. I notice really huge differences in practical DR when comparing the GH5 709 modes (that don't have the full DR of the GH5) with the HLG mode (which does - all 9.7 / 10.8 stops of it) and the OG BM cameras with their 11.2 / 12.5 stops. Another stop or two more would obviously be beneficial above those but I think the point at which you're choosing what to be able to include in the shot (ie, what is usable and what isn't) has probably passed for most people, even with the sun in frame.

@IronFilm - great pics but man... what a ride that was!! "we were out in the forrest just minding our own business and playing with cameras" "the bugs were getting annoying so we used some spray" "then out of nowhere the freaking empire showed up!" "It was at least 6 to 1, but those guys are SCARY... he said our bug spray needed a permit and that if we didn't have one, he'd KILL us" "we didn't know what to do.. we didn't have a permit, and the guy was completely nuts, so we mostly just avoided eye contact and tried to let the situation calm down" "he eventually left, probably to go kill more innocent people on a picnic or something, just gave us the evil eye before he left. psychopath!"

Absolutely - this is a dominant factor in most online discussions. One of the things that I think contributes to this is that there is no right or wrong in colour - the colourists are fond of the phrase "if it looks good, it is good". However, as you well know, the preferred workflow for anyone working with a professional colourist is for the director/dp to light and expose according to the directors vision, then the colourist can pull everything into a timeline, apply a global look, and then tweak from there, but mostly it's about respecting the choices made on-set. When it comes to the prosumers, there is no pro colourist, so there doesn't need to be a convention or agreed relationship established, and people can do whatever they want, and do, and "get away with it" because "if it looks good it is good". Real users disagree... I think you've gotten yourself turned around. RAW works with ETTR and the cameras that "There are too many cameras out now that you just pick up and shoot and bingo, 90% of what you wanted. One and done as they say." THOSE are the ones that you should think of as 0-100. When I shoot with the BMMCC (either in RAW or Prores - they feel the same I'm post) you ETTR and that's the best quality. I'm talking about recording run-n-gun in very high DR situations here too - sunsets and people standing in front of them, etc. It's when you get to shoot LOG that you want to avoid the areas close to 0 and to 100. I tried ETTR on the GH5 in HLG and WOW - if you want plastic skintones then they're there and they're plastic.... The reason is that LOG profiles designed by manufacturers take the bit-depth of the file and allocate more of it to the middle region of the luma scale, and really skimp on the highlights and shadows. I learned that the hard way by putting skintones in the highlights and got the plastic look. You can think of LOG formats as compressing the highlights and shadows - that's why they require very careful exposure for your skintones, and why you would want to adjust your ISO to a value where the skintones are at the right level (in the sweet spot in the middle where the good IQ is) and the DR of the camera is distributed so you get the highlights or shadows that you're interested in for that scene. In RAW, it's Linear, so the brightest stop of light gets literally half of the luma values, the second brightest stop gets one quarter, the next one eighth, etc etc.. The shadows get almost nothing. In that sense, the best image quality is in the highlights, so that's why ETTR was a thing - it makes sure that everything in your scene is exposed as bright as it can be (without clipping whatever it is that you want to keep in the scene) so everything gets the best quality.

Yeah, but by the time that a $100 SSD can record 4 hours of 8K60 RAW people will be saying "24K is the minimum required to shoot videos of my cats". It sounds ridiculous now, but people were making jokes about 4K being more than anyone would ever need. I don't use an external monitor on the GH5, no. I have one for my BMMCC and do find it to be cumbersome, and before I bought any FP setup I'd definitely be taking that setup out and shooting a dozen or so test videos just to really confirm that such a rig would be suitable. For the work I do I really find that a tilt/flip screen is invaluable, especially as I place particular emphasis when shooting of trying to shoot from interesting angles to make things more visually interesting, so I'm often shooting from above or below. Interestingly enough if you shoot a wider follow-shot looking down from as high up as you can reach with your arms and you stabilise it heavily in post it actually looks like a close drone shot, so that's a fun thing to include. In terms of the FP I am watching closely the discussion about exposing on it, but I would probably just set it to auto-ISO and let it expose for me, while having everything else in manual, and adding a fixed ND while outside during the day. I only use a 1080p workflow so the downscaling already does a lot of noise suppression, so the acceptable ISO range is greatly expanded for my purposes. I used to use a workflow where I had to render proxies before editing and it was a complete PITA, so I went to the 1080p ALL-I mode on the GH5 to essentially do that for me in-camera. I wouldn't be delighted if I had to go back to transcoding footage again. Lens choices would probably be an unstabilised 16mm and a 24-105/4 OIS. OIS would be a must for me as I often shoot in less than stable conditions (hand held while cold, low-blood-sugar, tired with shaky muscles, etc). Overall it wouldn't be a cheap setup, so it would have to really excel in other aspects to be a better option than the GH6. We'll see. I'm deeply aware of this "big camera = pro" dynamic because it works against me. I go places where pros aren't allowed and do things pros aren't allowed to do, so having a big camera is a problem for me, but the industry seems to think that if you care about image quality then the camera size is irrelevant. It's like making all cars that can do over 50kmh 30mph the size of a semi-trailer.... "hi, I'd like a small car"............."ok, here's a sedan - I like driving slow too!" "um. no, I want to be able to drive on the freeway" .............. "oh, I thought you said you wanted a small car.. here's our big rig - it can haul 20 tonnes!" "um, no, I don't need to haul anything more than some passengers around" ................ "I thought you said you wanted to drive on the freeway? I'm confused."

Pretty sure that the Lift/Gamma/Gain controls are designed for use on 709 footage, and the Contrast/Pivot/Offset are designed for use on Log footage. This might (I'm guessing here) have implications for how the out-of-bounds values are handled (eg, the curves might do predictable things and the LGG might not). To make a more general comment in the context of the bewildering complexity of the various colour spaces and gammas that are being discussed in this thread, Resolve is now getting to be a very sophisticated engine when it is put into a "colour managed" mode (RCM or ACES) and in the most recent versions some of the controls are now "Colour Space Aware" so will act differently depending on what colour space you have told Resolve that the project is in. To expand on this somewhat, you might have footage that's shot in one LOG format, and if you tell Resolve to manage the colour space in the same log format that matches the footage, and you adjust the Offset wheel +1 then you will get the same results as if you shot the footage exposing one stop brighter. However, if you do the same but don't match the log format of the project to the footage the same +1 adjustment will do strange things because Resolve is doing complex things underneath the surface to tailor its behaviour to the specific colour space you've told it to work in. I've confirmed this to be the case with the HLG files from my GH5, which (almost!) line up perfectly with rec2100. In previous versions of Resolve without colour management (I never tried ACES) I shot two clips one stop different and tried to duplicate one shot by modifying the other shot, but I failed to find a combination of colour space and tool that would do the correction. Now in Resolve 17 in Resolve Colour Managed mode it's dead easy to do it and the +1 control on the Offset wheel does an almost perfect job. (On footage that is directly aligned to a colour space in Resolve, the transformation is perfect - HLG on the GH5 isn't). I suspect that this means that some tools effectively have a CST prior to their adjustment and then one after it to put the footage back into the mode that your project is in. This isn't the only place that Resolve does "hidden" CSTs in - you can program individual nodes to do that too I think. I had a long and rather frustrating conversation with a pro colourist when I asked about configuring projects to use RCM and they essentially advised me to do everything manually, despite using the RCM modes (of CSTs) being the proper way to do things. I think the advice was well intended and was simply coming from a place of concern for the depth of complexity actually involved in the tool, and their (probably frequent) experience of non-post-pros going down the rabbit hole and essentially getting lost, so putting up a fence to protect people from the journey makes sense considering how many of them never make it to the other side. (I was also trying to do something a bit horrible that complicated the workflow beyond how that mode is designed to be used, so my example was even worse than the standard situation). To make an even broader comment about this whole thread I am reminded of a long conversation / debate about if there was any magic in the Alexa ARRIRAW files, and it was pretty clearly showed from some under/over tests that the RAW from the Alexa was a completely neutral linear capture - just like how sensors are designed. This should be true of the RAW output from any sensor because it's literally how sensors work. The challenge is how to process it in a way that gives you the results you want. It's true there are potentially slight differences between the frequency response of various R G B filters on the sensor, but they're likely to be relatively similar and can easily be adjusted with an RGB matrix if you're really keen on matching cameras (that's the basis of the BM RAW conversions from Juan Melara IIRC). Otherwise, you can simply take the RAW Linear files from any RAW camera and apply whatever conversions you like - if you like the ARRI colour science then apply their LUTs or transforms, etc. It's also worth stating that many heavy colour grades will completely obliterate any sense of colour accuracy so I would imagine the pursuit of it is really only relevant in situations where a natural or hyper-natural (eg, commercial look where everything has to be cleaner than real life) look is required.

Or does BRAW / Prores RAW also allocate more bit-depth / bandwidth to the middle part of the DR?

I suspect that this will derail the conversation completely but I'm going to ask it anyway... if you're shooting RAW then does middle-grey even matter? It definitely does when shooting LOG, as many LOG formats allocated less bit-depth to the brightest and darkest stops so you really did want to put your skintones in the middle where there was better IQ, but considering that Linear RAW gives the most bit-depth to the highlights and the least to the shadows, does the concept even make sense any more? I understand that if you're working with an editor/colourist then you want to expose things so that you can put on a display LUT for the editor and the colourist has a good starting point, but if you're trying to eek out the most quality and are willing to change exposure between shots, why wouldn't you just ETTR?

Yeah, I know there are but it seems to be the exception. I frequently find that discussions with "pros" assume you'll be able to operate fully manually, that you can have a rig the size of a fridge (or truck) if you need it, that light can be provided / modified to suit the composition, and that normal camera accessories will all be fine (up to the point where it makes the operator resemble a Borg drone). Any mention of situations where the speed of working is a factor or there are any limitations that might impact what you can do seem to be considered an exception. Those situations do seem to be handled pretty well when they come up, so I'm not suggesting a limitation on their behalf, merely assumptions around what is considered normal. I laughed pretty hard when the guy below mentioned that he had to work "incredibly fast" (6:25) to get a few random shots of his GF sitting on a train, or to get someone in focus on an escalator (essentially it's a stationary portrait when both of you are moving at the same speed). If he found that filming a subject who was motionless and would respond to direction to be "incredibly fast" then we've basically run out of words! "Extreme" seems suitable language to indicate to people that they should set aside their normal mode of thinking. The GH5 is still a really great offering at the intersection of the various considerations needed for working in highly unpredictable and fast-moving situations, despite showing its age in a number of ways (DR, colour, codecs, low-light). I'm not going to be in the market for a GH5 replacement until I start travelling again after COVID is actually gone (rather than the current wishful thinking that's going on in the PR departments of most governments), but I'm considering the GH6 and also the FP as potential replacements. The GH6 would keep the strengths of the GH5 but doesn't completely bridge the gap between the GH5 and current FF cameras in terms of low-light, DR, and perhaps colour (jury is still out on that one I think). The FP would limit me to OIS lenses (and likely less stabilisation compared to the GH5 with IBIS on unstabilised primes) but I think (when combined with a BM Video Assist to get BRAW and its sensible bitrates) it would buy me considerably better codec, DR, colour, and perhaps low-light too, so it might be a sacrifice worth making. I've also done a bunch of work during COVID around my editing process and style and have changed my requirements a bit because of that. Neither of these options is anywhere near the price of something like the R5C, which is competing in another league (8K RAW, DPAF, etc). I must admit that if I was willing to invest significantly more into a camera system, and was chasing something with PDAF then the R5C would be a contender, along with things like the FX3.

After years of telling people that I have different needs to what they think I do (or should have) I have now realised that I basically shoot in extreme situations. I shoot in caves, out of light aircraft, while walking/running, through glass, in high DR situations, in nasty weather and at crazy long focal lengths (>800mm FF equivalent). The fact that I do these things as part of home videos is irrelevant, and as soon as you say that, people just instantly think I should use a potato and be happy about it. I'd suggest you stop talking about "pro" and start using the phrase "extreme". It's true, after all, most pros shoot in situations where the world revolves around the camera, rather than the camera needing to fit into the world, and let alone like us where the camera needs to be able to survive a world that is actively hostile to both good images as well as equipment failure.

Keen to hear your thoughts on the camera, image, and this rather strange exposure behaviour... You mean a test like this?

Nope. Sex sells, but fear sells more. Look at the cover of any newspaper, basically ever, and tell me if it's "good news" or "bad news" 🙂

Wow - that would be why no-one has posted sample footage of it yet. Maybe it'll come in the firmware update where they add 4K and 1080p for Prores. There's absolutely no way they'd let people shoot their highest quality codec with non-oversampled lower resolutions. Maybe, I was just saying that it has the full DR. Same for me. Plus it gives the flexibility (once you know how to set it up) to adjust WB and exp in LOG before applying the LUT / CST.