kye

Covert Affairs - Season 4 last few episodes. Definitely in english.. Please don't judge me! [edit: and, for absolutely no reason whatsoever, also let me recommend Master With Cracked Fingers as a rather different, but excellent, viewing experience!]

Stupid questions: Does the BMPCC have any kind of auto-controls, like auto WB, and auto SS for exposure? and what is the limit of its SS (do you NEED an ND or can it go down to a small enough exposure time)? I am having mischievous thoughts..

I'm looking for a second / backup camera body for travel. Main setup is GH5, 7.5mm 17.5mm and 42.5mm MFT primes, and the Rode VMP+. Second camera is a Sony X3000 action camera. I shoot live action as well as time lapses with both. My thoughts are that bringing a small MFT body would enable me to have a backup MFT camera should anything happen to the GH5 while travelling, and would also enable me to take time lapses while the X3000 and GH5 are busy. For example, on a cruise ship when you're pulling into port it's great to be able to record time lapses from a couple of angles whilst also using the GH5 to take normal video. A second MFT camera could then record time-lapses (RAW images at native resolution are great quality) and if the GH5 has a heart attack then i'm not in paradise without a camera. I'll also take my 14mm f2.5 lens in case the GH5 / 17.5mm lens (the main lens on the camera most of the time) end up falling out of a plane or whatever. I'd also take a Rode Video Micro as backup audio. Requirements are: Must be MFT as they're the lenses I have 1080 is fine, although I'd prefer higher quality rather than lower quality Built-in time-lapse functionality Still images in RAW would be great Slow-motion NOT required If it can take the same battery as the GH5 that would be spectacular (saves the hassle of extra batteries and chargers) I'm wondering if I should go an old GH1 or something. I know that the later G models have 4K but in all likelihood I won't be using the video functions, just the still images. I have a GF3 but it has no mic in, theres no time-lapse functionality (or way to trigger it remotely), and the 1080 isn't the best. .......and of course, smaller and cheaper is better....

Well, if that's the case, my footage is probably so marinated it'll be liquified! If only I liked editing as much as I like shooting!!

Interesting that the Alexa Mini is right up there. I would have thought that the Mini was a camera with more compromises than the larger bodied models and therefore was more of a specialist camera for lower budgets or more difficult shooting conditions (drones, highly mobile rigs, for mounting on vehicles, etc) but maybe I'm wrong? What limitations does the Mini have over the larger models? and although this is pretty academic as none of us here are really in the market for an Arri... be careful about causality - these films weren't successful because of the cameras that were used and we can't get the equipment and expect it to make us better. It's that the people who are killing it just happen to want to use these cameras. I remember once I was riding my mountain bike up a long and steep hill and I wanted to be going faster and changed down a gear without thinking about it, almost falling over because I could no longer keep up my pace. I realised afterwards that I changed gear because that's what you do when you're going faster, but of course, changing down a gear doesn't make you go faster!

I think there's some discussion of it here: https://www.lensrentals.com/blog/2019/03/just-mtf-charts-zeiss-slr-lenses/ Assuming that's the one, the curves remind me of many of the classic cine glass that they've also tested. Contrasty (the top curve shows this) but not too clinical (the other curves are lower - when all the curves are towards the top then that's when lenses look super sharp) and the lower resolution and astigmatism towards the edges (lower lines = softer and IIRC when the solid and dotted lines diverge that means that the radial blur no longer aligns with the other blur (at 90 degrees to it) so nothing can be completely in focus). Looks like a very promising lens... I look forward to some sample images

That's what I heard (although I'm the last person that would know such things, so apply salt here..) but with some of the quality I'm seeing, my current production standards for my family travel videos have exceeded (a small percentage) of the quality of some shows on there. Still, in the end none of that matters and it's about content. TBH I think that with a very modest setup and the right levels of skill content that looks professional can be created with almost no budget at all. An example might be @DaveAltizers video Porcelainia: Shot on a 7D, L-glass, and graded in FilmConvert it's a great example of how, when done well, content shines through. Think about how many people have access to equipment that meets this level of quality - at this point it would be half the people on earth who own an ILC. And this is exactly what I'm seeing - there is heaps of content on YT that people make for fun or as part of a side hustle, and heaps of content on Netflix where people with skill use the same level of equipment and skill but are more ambitious and take on larger projects and larger distribution goals and do distribution deals that way. The best content on Netflix is obviously better than the best content on YT, but the vast majority of Netflix content has less production value than the best examples on YT. The overlap is huge.

That's really common, and why many photographers have someone else help them in selecting their best prints. Gary Winogrand was a famous street photographer and used to deliberately not develop the rolls of film he shot for something like a year, so that when he was looking at the images he wouldn't remember the situations associated with them but would only judge the image based on the image itself. We're all guilty of preferencing a shot because it was hard to get, rather than the finished product being good

I'd be very surprised. Considering that I have very little to do with being famous, with people who are famous, or with people who know people who are famous! My Bacon score isn't 6, but it's not 2 either [Edit: after a bit of searching (considering I'm not in IMDB) I got it down to 4. Not bad.]

I was watching a TV show last night and they had some cool spy vs spy stuff through the streets of Hong Kong, and it was shot well, had great colour science, the story was engaging, but it felt like a home movie. I've watched several seasons of the show and other sequences didn't have the same feel. I worked out that it's because I'd been to where they were filming. I've increasingly noticed this trend - if I've been somewhere then it doesn't seem so exotic anymore, it doesn't have that 'other world / smooth / detached' feeling that something being 'cinematic' feels like to me. Some of my favourite travel videos have transformed over the years from spectacular experiences to slightly awkward sequences of shots of places I've been. Does anyone else find this?

I agree about them being visible if you're working with different colour spaces, but after the conversion then I find it's less relevant. My workflow is typically: Node to do things prior to conversion (mainly WB and exposure) CST General 709 grading nodes My thoughts around having a node in different colour/gamma spaces would be if I could have one before the CST to simulate things being done in-camera, and after the CST as custom grading things like having nodes in YUV etc. Obviously the nodes before the CST would have to be tailored to the colour/gamma of that clip, but if I just saved a power grade for each camera then I could just add them from presets then adjust away. More experimentation is required. I agree with your sentiment, but I'd actually say they're not clueless about mounts, they see them as the boundaries between entirely different universes that cannot be breached. @BTM_Pix The amount of "oooh, the P6K has EF mount - finally there's a camera that exists in my universe! All the other ones have been in a different language in a parallel universe, and if an EF lens was ever in the same room as a non-EF camera then it would turn our solar system into a black hole". and by the way, to most people there are two lens mounts - "the one I use" and "not the one I use" which has the internal description of "nope nope nope nope.. back away now.... alien creatures are amongst us... they didn't choose my lens mount soooooooo.................. WARNING!!!!". The Canon vs Nikon debate places these two crowds so far apart that the curvature of the earth prevents even radio communication between them. The MFT mount may as well be an underwater song made of colours.

Wow, the E1 sure is reasonably priced... I think subconsciously I was thinking the E1 was in the same pricing category as the E2. So many cameras to keep track of!

That works From what I'm seeing, there are still a lot of shows that seem to have smaller budgets. I'm not certain, but I wouldn't be surprised if lots of shows are made and then bought by Netflix, rather than being made for Netflix and having that kind of budget. Mind you, I don't only watch the big shows, I'm watching a bunch of stuff, some quite niche.

Absolutely. There are some situations where the physical size of a GoPro is required, but the majority would be fine with any of those cameras. It's not like GoPros are really that much cheaper these days either! I think that vlogging really benefits from a very wide angle lens, as it makes it look like the person is a sensible distance away from the camera, rather than them in that uncomfortable spot where they might be talking to you but might also be leaning in hoping for a quick kiss In this sense I think that the Osmo Pocket (at 26mm equiv) and RX0 (similar FOV) missed the mark for this. A GoPro is a more flexible tool IMHO being that you can crop in for a smaller FOV at a lower resolution, but it's pretty hard to crop-out and get wider than the lens!! Especially considering that very few people with either of those cameras won't also have a phone with a 24-28mm lens. yeah, the second-hand market is basically a graveyard for older models. That's why I don't really feel bad about doing this to my Hero3 but it not surviving the process: In terms of vlogging with an F3, sure you can.... Just hit the gym first!!

I agree. In fact, now I think about it, if I had to choose between 4K 8-bit and 1080p 10-bit I'd be torn, but might go 1080 with 10bit just because of the flexibility in terms of doing radical grades in post, which I do quite a bit when in less-than-ideal available lighting conditions. I see GoPros (or at least, action cameras) quite a bit on Netflix shows. They're always for shots mounted on the outside of a car, or other difficult placement situations, but what gives them away is the over-sharpened over-contrasty image, not the FOV or image quality more generally. I don't know why people don't just apply a blur filter over the top to match them better, even if it's not about matching, just making them not look like the image has been embossed onto something would be a great step forwards! If I vlogged I'd be pretty tempted by an action camera. Wide FOV, portable, you can get decent stabilisation with more recent ones, some have 100Mbps files, etc. Not bad for that particular use-case. There aren't many other use-cases where they'd be a good fit, but for some I'd imagine they're a reasonable choice, but that wouldn't be that big a segment of the market..

Yeah, I was contemplating the better IBIS of Olympus but the 10-bit internal of the GH5 won out. I was emailing with one of the Olympus ambassadors about options and when I said that I had gone GH5 for the 10-bit even they conceded that 10-bit internal was a hard thing to pass up.

Ah, I misread your comment and thought you were calling BS in the other direction lol. Yes, if you're putting a priority on accuracy then the IDTs don't work. I tried both and confirmed it for myself some time ago. Taking two clips in HLG and exposing them one stop apart, then bringing them into Resolve, taking one of them and doing an X to Linear conversion, scaling up the brightness, then converting Linear to X again. I tried every combination of X I could find (2100, 2020, etc) with each way to raise the luma (Offset, curves, Gain, etc) and no combination was a perfect match. In the end I realised that many of the conversions were similar, thus my earlier comments about log profiles being similar. When I finally wrapped my head around the idea that grading isn't about accuracy, it's about what looks nice, then I stopped worrying about profiles. Many professional / high-end colourists don't bother with CST / ACES / RCM and just take the log files and adjust colour balance with LGG, adjust contrast and primaries and are done, often only taking 10-20s per shot for the bulk of the work before starting the polishing and 'look' adjustments. Instead of doing a CST -> adjustment -> CST can you just use a single node and set the CS of the node (it's in the menu you get via right-clicking the node) and would it do the same thing? My understanding is that the CST is simply a user interface to the same RCM functionality that applies to clips in the media pool, timelines, etc.. I've been meaning to try it, but just never got around to it. It would save a lot of time and simplify the node structure somewhat.

As people are posting their whole rigs, mine is the GH5 and Sony X3000 action camera. GH5 for "proper" shots, and the X3000 for random travel sequences between locations, and if I suddenly need a super-wide angle. For the GH5 I have the FF equivalent of 15mm F4, 35mm F2, and 85mm F2. I'll also take a longer lens on a trip if there's animals or sports, but it's not a core lens in my kit.

Are you using the Panasonic IDT and finding that it does a really good job? If so, I wouldn't be surprised. Anyone who has spent any time grading (or colour matching) different gamma/colour spaces will know how similar they really are to each other. Both Log and Linear are mathematical terms used to describe certain functions. Sensors 'see' in Linear, and then encode to Log (assuming that's how you're recording) and although they do differ in subtle ways, both in deviating from a mathematical Log curve as well as doing 'nice' things to the colours, I'd recommend people to get some Log footage and then convert it to 709 using different input profiles and see what differences there are. Many of the settings are so similar that when you click on the new one there is so little change that you wonder if you hit the right buttons... For the GH5 I've been playing with both the HLG.2020 and HLG.2100 profiles and getting a decent look from both in various circumstances.

Also, if you're going to test the cards by recording with the camera, make sure to record something with intense movement, not just a still scene. The bitrates can be a lot lower in reality if nothing much is moving. Trees blowing in the wind, fountains (zooming into just the water spraying part), or design your own. I got five still images that were completely different, put them into a 1080p60 timeline and exported them as a prores file. Then play the video on loop so it just sits and kind of flickers as the video loops. Then point the camera at it, set a smaller aperture to get everything in focus, and record it at 24/25p for a few minutes. I analysed the footage I took to confirm that this test works and even if the shutter is open between frames no frame is anything like the previous one.

Agreed. There is a huge debate about log in 8-bit vs 10-bit, and there are instances where people do grades and show minor issues, and there are tests where people deliberately film difficult situations and then try to break the 8-bit footage and can't do it. There are many more complexities involved, as @KnightsFan mentioned, and compression is the biggest one. Trying to have a serious debate about 10-bit vs 12-bit is fine, but just don't try and push the angle that 10-bit is anything less than 99% as usable in 99% of the situations. If you're after ultimate quality then sure, shoot RAW that's totally fine, and yes it does give ultimate freedom, but that's like comparing a Ferrari and a Lamborghini - you can compare and one might be faster than the other but you can seriously claim that the slower one isn't fast enough to be used for everything except the smallest of situations. See my above shot of the HK harbour and tell me how 10-bit would somehow have been better when there aren't really serious artefacts in a grade that no-one would ever do in any situation in the real world. In terms of WB, yes, the colour science makes it trickier, but if you know what you're doing and have a half-decent software tool at your disposal then the only thing stopping you from getting an excellent balance is skill. and I would know - I shoot available light and often mixed lighting all the time and the thing that limits me is my ability to adjust WB in post. Check out these images.... An ungraded frame from a shoot I did, note the horrendous green/magenta lighting: and the two grades I got back from (the very gracious and much more experienced) members over at LiftGammaGain.com.. The first from Szilard Totszegi: and second from Cary Knoop: and the thread is here: https://liftgammagain.com/forum/index.php?threads/advice-for-grading-mixed-green-magenta-light-sources.12727/ This was after I'd battled with the video for hours, and gotten no-where near what they managed to do. Even after seeing what was possible and outright trying to copy their efforts to learn from their examples, I still didn't get it as good.

Indeed it is We went on a special trip to a beach where you can see it over the ocean and all we saw was clouds. Admit defeat, run back to the bus, then get back to the port and the thing is in full view and the sun set over it. Life is funny like that sometimes!

That's how I shoot all the time, grabbing a quick shot then running to catch up. Or getting shots of the family while they're doing their thing. For me the strengths of the GH5 really help with that. IBIS for walking while shooting (trying to simultaneously do the ninja walk and also not walk into a pole or anything) or for standing and grabbing a faux tripod shot. The 10-bit for being able to significantly push the image due to no control of lighting and little to no time to work the scene. Etc... My other approach is just to get volume, as not only does it mean you get lucky more often, but the practice increases your hit rate too. Here's some random GH5 shots from filming at the speed of life...

I wasn't. That makes sense. I guess it just depends on what lenses you have on it. I have a GF3, and when paired with the 14mm f2.5 pancake it's (just..) pocketable, but you put the kit lens on it, or anything with a longer focal length or wider aperture and the size advantage disappears pretty quickly, but whatever works for you. I thought for a second there your reference of people who aren't so 'patient' might be to people in fast cars, maybe a street racing reference, illegal street racing, small camera doesn't attract attention etc... for street videos, or maybe undercover law enforcement. Or alternatively, referencing some kind of meme I'm unaware of, or humour that I wasn't getting, or..... or.... ???

I'm confused. It was a genuine question.. like, either I learn something about the GX9, I learn something about the GH5, or I can offer advice.