kye

The T-stop of a lens is often just a little slower than its F-stop, so I would suggest these are the cine version of F2.8 lenses. Maybe this one rehoused 16-35/2.8? The extra space and size would allow for the power-zoom functionality which doesn't seem to be present on the still lens, as well as equalise filters and weight across the range.

A couple of weeks ago a lady noticed my GH5 / VMP rig and asked me about it. Turns out she runs a small studio and is looking to replace her PD170 and "go HD". We spoke for some time and it was obvious from the conversation that she hasn't been looking at the tech for quite some time. Her three concerns were 1) no-one was offering any money at all for selling her PD170 (people were offering $20 and 'novelty' amounts like that), 2) she wanted to digitise all her tapes, and 3) she wanted to get new equipment and was admiring how small the GH5 was. Really goes to show how we on these forums are in a little bubble of 4k/6k 10-bit etc, when there are people out there who haven't gone HD yet, and we did the $200 challenge with HD cameras!

I don't know as I haven't done it myself, but that is normally quote as one of the reasons you shoot raw. I think the difference is the compression. If you are shooting RAW then there will be a pixel just to one side of an edge and it should be completely the colour of your green screen, there will be an edge pixel which probably has some green and some object colour, and the next pixel should have no green in it whatsoever. I think with most forms of compression you end up with that transition line (where the colour of the pixel is a mix of green and the subject colour) being a lot wider than one pixel, which creates problems getting a clean edge mask, so you end up with the subject having a green halo, or you have to crop into the subject to eliminate all the green but now the subject edges have all been chopped off.

The $200 challenge was fun, and I've been thinking about the next one, and what might be some good rules. I'm travelling and have had slow internet for the last week, and the highest quality I could watch YT without buffering was 480p, and it has been an interesting experience. Obviously the compression basically hammers the video quality into the dust, but some traits remain, and the aesthetic is interesting. I've been watching on my 13" laptop screen, and although things are notably fuzzy and lots of movement has lots of artefacts, some of the really important things still come through, like composition, colour, DoF, editing pace, sound, music, etc. So, I have a challenge and a question.... I challenge you to watch YT in 480p for at least 30 minutes. The idea is to watch some of your favourite films, and to watch long enough to get used to the quality, so that you get a solid impression of the feel of having very low resolution, rather than just a first impression. You should watch for long enough to forget you're watching for video quality and see how much enjoyment you still get. The question is, if I make a next challenge that you have to make a film and one of the criteria is that it must be uploaded to YT in 480p (720x480) would that be of interest to anyone?

Great stuff. I'm reminded of this article on shooting with a single focal length sent to me by @mercer https://noamkroll.com/many-iconic-directors-have-shot-their-feature-films-with-just-a-single-prime-lens-heres-why/ There are lots of advantages to having limitations and I've incorporated many of them into my setup, kind of unknowingly lol. I also like the idea of a 40mm 2x lens - if you frame someone up vertically the way you would normally (head, mid, etc) then it keeps it closer to a 40mm lens, but just gives you a wider background, so although it's a 20mm from the horizontal view, that's typically not how you frame up shots of people. You also 'learn' the lens and can see the framing without looking through it which helps creative vision, you don't need to change lenses so are faster on set, matching in post is much easier, so there are many advantages. I use a 35mm equivalent lens as my default, potentially cropping to 2.35:1 which would make it a 46mm equivalent lens in vertical FoV, so we're operating in the same territory Sure! Just put a handle on the bonnet of @IronFilms truck and if you look confident striding into the airport maybe they'll let you take it as carry-on! Seriously though, lots of the peripheral stuff is getting smaller and smaller - I've seen a battery-powered smoke machine product that was half the size of a VHS cassette, we've got tiny LED lights now, so things are moving in your favour.

I'm in the same boat as you - it's a lovely lens but carrying it around all day would do horrible things to the ligaments in my wrist!

@IronFilm Nice calls on the Fujinon zooms - very nice image from those.. and also the Sigma f1.8 zooms that @barefoot_dp also mentioned. I especially like... ??? and when you say "I'll note the title said VIDEOGRAPHY." I'll give you permission - what film-making kit would you have??

My dream kit is (mostly) my actual kit because even if I was super-rich there aren't other options that have the combination of features I want. Setup One Things I already own: GH5, Rode VMP+, Voigtlander 17.5mm f0.95, Canon FD 70-200mm F4 + Canon 2X TC Things I would upgrade: I would replace my SLR Magic 8mm F4 with the Laowa 7.5mm F2 for ergonomics and low-light ability I would consider replacing my Konica Hexanon 40mm F1.8 with the Voigtlander 42.5mm f0.95 I would replace my cheap monopod with a nice monopod, maybe a Sirui one with the three little feet on the bottom Setup Two Things I already own: Sony X3000 action camera with remote control monitor thingy, finger cage thingy Both setups use Area-Swiss style QR mounts and also the Peak Design Capture quick release plates for ease of use. The reason I choose this setup is that I shoot travel videos of my family as well as sports of my kid playing football. The considerations here are that the camera can't be big enough to attract negative attention (from other parents, or from security at private places that don't allow 'commercial' photography) and can't be heavy enough to not hand-hold all day (many places don't allow tripods or are too busy / cramped to use them anyway, and we're normally on the move and often on-off boats or busses or whatever so I need to be able to move quickly). I like the low-light of the fast apertures and combined with MFT they don't have too shallow a DoF to use. Having a 16-35mm fast zoom would be great, but no-one makes a f0.95 zoom, and if I went FF and got a 16-35mm f2.8 it would be slower, have deeper DoF than my 17.5mm (which is equivalent of a 35mm F2), and would be too heavy to hand-hold. I don't change between 16mm and 35mm too often and if I needed to then I can use the X3000 which has a wide on it. Wide shots in low light are either worth swapping lenses for or aren't worth getting. I really feel like most people would love to buy an ARRI LF and Master Anamorphic Primes or something, but my setup is the best compromise for me regardless of price that yet exists.

I agree - if they take feedback and accommodate what they can (technology permitting) they could end up with a really interesting camera that accommodates some niches that the other manufacturers didn't realise existed. Getting the extra bit-depth by recording less pixels is a great idea, and you could get that by either not using pixels at the sides (for an anamorphic mode) or by not using pixels at the top/bottom for a 2.35:1 aspect ratio, both of which are useful in different situations. That's one of the things I really like about ML RAW - the options to lower the data rate through aspect ratios. Assuming the viewfinder previews it well (which I would assume they would do) then it's great to be able to shoot like that. That's one thing I miss on my GH5, if I want to shoot 2.35:1 then it gives me these tiny/weak guidelines that are practically invisible, but doesn't 1) black out the other parts, or 2) not record the other parts. Getting a smooth 2.35 shooting experience would be really good, both by framing things easily, as well as allowing you to compose with the right aspect ratio. I find myself composing according to what I see in the viewfinder, as well as using the other eye to see the world and what is out of frame before it goes into frame, and when combined with the other eye and also thinking about sound as well as not falling over something while I'm walking or whatever, the aspect ratio guidelines get lost and I compose for the 16:9 and then have difficulty cropping in post because I've framed using the top and bottom parts of the 16:9 image.

It seems counterintuitive to me and that makes me think it's probably a sensor thing (which I don't understand and know very little about). I wonder if @androidlad can shed any light on this particular aspect?

One advantage of RAW is shooting green-screen and getting clean edges. Most forms of compression crunch edges and give you halo problems, so this would be an advantage, especially if you lit and exposed well to get the image right in-camera. In a sense this becomes really great for film students and small-budget productions where you're cutting out the LOG profile / colour grading in post headaches but still retain many RAW benefits. I wonder if there is still a bit of a gap in the 1080p RAW camera lineup? The BMPCCv1 had bad audio preamps, the BMMCC needed a rig, ML can be fiddly for some people. I don't know if the P4K / P6K do 1080 RAW without any major drawbacks? I also don't know if there are many people in the market for shooting 1080p RAW? But if there is a gap in that market then maybe this will find a niche there too? That handle looks very cool BTW, combined with a hardware to lock in the connectors this would be quite an elegant way to get the 12-bit RAW 4K. Or (unlikely) have it do 1080 RAW internal and 4K RAW external simultaneously? No idea on how that would work, but having 1080 RAW internal and 4K RAW master files would allow full grading to be done on the 1080 12-bit RAW before rendering. Things requiring tracking like stabilisation or compositing would need to be done on the 4K files, but for productions not requiring those, that would be interesting.

Do you think that they will release a video-centric flagship as well? That seems to be a common approach now (S1 vs S1H, A7R vs A7S, etc). That would be fascinating to see considering they don't have a cinema line to protect, so could go all-in. Yes, I didn't think you were knocking IBIS. I haven't seen the patent but I'm very aware that if you dig deep enough everything electronic happens physically. I read somewhere that there's a limit to OIS / IBIS of something like 6.5 stops because at that point the sensors start compensating for the rotation of the earth. I think they'll likely exceed that 'limit' eventually, but I find it amazing that technology based on mechanical parts can be made to be so accurate and in such a cheap and reliable way. I've studied the whole stack of IT in terms of algorithms and the logic of data processing, below that is the machine language that the chips use, below that is how analog circuitry implements digital behaviour, and below that is the physics of transistors and how electricity behaves in silicone and other materials. I'm no expert in these, but I know enough to understand that modern technology truly is a marvel and we are incredibly lucky to be alive at such a time in history to be able to see all these things and play with them. I'm not that old, but I remember as a kid not having a TV and when we got one only having a tiny black&white set with the rabbit ear antennas when we got the first one. In this sense I've seen a complete technological revolution within my lifetime, and I'm planning to live to be at least twice as old as I am now, and I can't imagine the progress that will happen over that time, but I'm really looking forward to it.

There's some interesting ideas in there. It will be interesting to see how it ends up. If they respond to user feedback it will be something special, although if they're talking to people that want to use a $2k-3k camera as a crash cam then they might not be quite the same things we're chasing.

This isn't something I do much, but I know that if you get a hollow sound then it's likely to be phase cancellation. Phase cancellation works based on the wave-length, and lower frequencies have a longer wavelength so tend to cancel easier / under more circumstances, which is why a hollow sound is a common side-effect. I'd suggest that using the in-built features might be useful because either they will get it right and you're done, or if they get it wrong they might get it almost right, then you can zoom in and just fine-tune it by lining it up by eye. Happy to hear from others if there's a better way though!

When they work out how to un-blur images in post IT WILL BE THE BEST THING THAT HAS EVER HAPPENED TO FILM-MAKING. We can record in 1080 and then un-blur to get 4K, or 8K. All those police shows will come true where they take a blurry video and just keep saying "enhance..... enhance..... enhance..... enhance....." until you can tell where he ate dinner by looking at the individual molecules in the crumbs on his chin. In IT, once information is gone, it's gone. That requires much better ISO performance than we have currently, plus it would require that the next exposure start immediately after the last exposure stopped, otherwise you'd get gaps in your motion blurs. I get that with a high enough frame rate and a bit of AI you could close those gaps pretty easily, but I'd prefer IBIS and the fantastic low-light performance. Also, this high-framerate plus AI processing doesn't seem to give the smaller-camera plus better battery life that @Snowfun was talking about

If you have a gimbal then why not leave it on there the whole time and not need IBIS? It means you don't have to bother mounting and unmounting the camera from the gimbal, and it also gives you more options in what camera to buy.

That's why I like the IBIS in the GH5 - it stabilises my fully-manual vintage lenses. Fast, cheap, great MF, great ergonomics and stabilised is a combo that only IBIS provides. This is something that I think people don't understand... OIS and IBIS stabilise DURING the exposure of each frame - Electronic stabilisation stabilises AFTER the exposure of each frame. If you're shaking the camera at all, the fuzzy light-trails and motion blur in each frame can't be fixed by electronic stabilisation. Every camera with electronic stabilisation is designed for very short exposures - action cameras. As soon as you want a 180 shutter or work in low light then electronic stabilisation isn't much use. It means that you throw away all your ND filters, or always use the camera on a gimbal. [Edit: this applies to any motion of the camera or any motion in the frame]

6K requires at least 6000 pixels across, so in a 4:3 sensor that would equate to ~27MP. Alexa shoots 3.2K and upscales to 4K, which is a 1.25x upscale, which if Canon used the same logic would require a ~17MP sensor. If it's true, which sounds feasible, then it will be interesting to see what they do. There's a famous saying "yesterday is the best predictor of tomorrow" so I wouldn't bet on a surprising move from Canon.

What do you think makes IBIS obsolete? Isn't it simply OIS but applied to the sensor? And please don't say electronic stabilisation has replaced it!!

And actually, if @zerocool22 went with a good quality 1080 camera then maybe he could get two bodies [edit: for the same money by buying second-hand] and have a lens on each and swap between them for different shots. Maybe a 24-70 equivalent on the camera on the gimbal and another with a fast tele prime for extra reach and 'hero' shots with crazy bokeh or whatever. Also, having two bodies means that you've got complete redundancy in case something goes wrong.

I see both sides as well. The way I view cameras is as a combination of features or attributes, so any camera will be good in some situations / for some types of work, but bad in others. If you happen to do work that the camera is good for then the camera may be a good value for you, but if it doesn't align to your requirements then it wouldn't be a good value. People are critical of Canon cameras, but they provide a set of features and traits that are valuable enough to some people that they're worth the money and Canon makes a sale. I maintain that the reason people get so critical of cameras or of the choices of others is that they don't realise that the other person or camera has a different set of priorities, and people underestimate how different we are to each other. I find my GH5 to be a good fit for me because the weaknesses of the camera aren't high priorities for what I do, but if you have a high priority on AF then it won't be a good choice for you and acknowledging that means I don't need to get upset when someone chooses differently to me.

I think that we're not too far away from Resolve / PP / FCPX offering a feature that will review your footage and create a 'rough cut' for you. Maybe some kind of wizard where it does face recognition and asks you who are the most important people, maybe has some preset types of video so it knows what types of shots to prioritise (happy highlight reel, travelling with people, all about the location, etc) then it creates a timeline with the edit in there, maybe in some form that's easily editable. I think the first film-making job to be completely replaced will be the Assistant Editor positions who catalogue and edit and tag the footage and create daily and scene assembly timelines ready for the senior editor to work on a final edit.

If this can be done in a 'legit' way - ie, with some assurance of quality of workmanship and support from someone that customers know and trust then this is quite a cool thing. Having options for customisation is quite cool.

Great stuff, and truly exciting to see a company pushing forward with such a leap forwards in technology. It will start expensively, but trickle-down always occurs as the people who value it most repay the R&D and then the quantity in the mass-market will drive the price even lower. The 1/120 is an interesting thing. I wonder if there is some kind of time component that prevents it from working over longer exposures?

I'd suggest anyone stacking filters try different orders and see which order gives the least reflections. Setup the camera on a tripod and shine a light source into the lens to really accentuate the effect and then compare images and see which has the least undesirable side-effects.