kye

That's probably true, but where are the 8K cameras they'll be shooting on? If you're developing 8K cameras for broadcast and you're also in the business of making MILCs then there's quite a lot of synergy there, considering that electronics are basically modular. Of course, for the A7sIII they might opt to keep it 4K and instead of oversampling an image they could do true 4K in hardware by taking a 2x2 grid and combining them to get 4K in 444 colour. If they did this then they could also have two blue pixels and average them in the conversion, which would help noise in the final image, and may keep the A7s range as the low-light leaders.

@webrunner5 Thanks! I think! I haven't tried them, but maybe I should. Much stranger flavours have been edible and if you put enough salt and sugar into something it's normally passable. Besides, if you use those little banana lollies as a benchmark the fake flavouring version of something only has to be vaguely in the same flavour universe to be successful ???

Some follow-up comments on the 8mm SLR Magic: The locking screw is a PITA and there doesn't seem to be any tension you can apply that gives adjustability with some friction - it's either locked or free-floating. I'm thinking I might have a go at grinding/polishing the end of the screw to hopefully smooth it out and allow a 'friction' adjustment, but I'm not optimistic. The distance markings on the MF are useful - I worked out where to set it for 'selfie' focus and that makes it nice and repeatable (I make home videos and am trying to be in them a bit more) It's pretty soft at f16. The difference between 1m and infinity focus is only a few degrees adjustment on the MF control, so I set it to infinity focus and f16 and run with it as a fixed focus lens but reviewing footage it's pretty soft, so later I swapped to f4 which is so much sharper the Focus Assist lines just about double their strength in comparison to f16. If you like a softer look then f16 could be a great option. f4 is FF equivalent of f8, so in a sense you can just think of the aperture ring as a "soften" adjustment It flares quite nicely. The part of the flare with the shapes all in a line isn't that much at all, but the general tint across the whole image (kind of like a light-leak) is quite pleasant and in the footage from Thailand where I move the camera from away from the sun to towards it (but not having the sun in frame) a warm 'wash' appears across the image. No flare: 10 frames later (identical grade): This has the GHAlexa LUTs applied, so the colouration of the tint has been modified - I just looked at the SOOC footage with the contrast pumped up and it's basically orange. It kind of reminds me of the Tiffen BPM filters when the sun is out of frame but still hitting the filter. This is with the naked lens - no filters attached at all.

Yes, Resolve is that smart. If you set the re-time controls to exactly the right number (24/60 = 40% speed) then it will give the same result. Do a test for yourself. Pull in two clips - slow one of them using Clip Attributes and slow the other one on the timeline with a 40% speed adjustment, then step through them one frame at a time and see for yourself what Resolve does. One thing to note - when you have a HFR clip in the timeline and you open up the Clip Speed dialog it shows one box for Percentage and another for Frames Per Second. The FPS box shows the timeline speed (24fps in your case) and when you say 40% it will update this box to show 9.6FPS. It's wrong so basically just ignore it. Incidentally, if you want to have clips conformed to exactly the frames that were shot, I recommend using the Change Clip Speed dialog box instead of the other controls. I remapped the Command-R shortcut to it as well. Entering exact values in there is the easiest way to get the exact value.

This thread is mostly going no-where, but it's not anyones fault. It's actually a symptom of how a nodal editor like Resolve works. Begin PSA In PP or FCPX a person can ask the question "how much power do I need" and someone else can give an answer that is useful because people use the software in basically the same way. Sure, you can have layers and multiple tracks of video, but mostly people edit with one shot visible at a time, and PP and FCPX have such limited colour functionality that you would only use a couple of colour effects. Resolve is different. Resolve is a nodal editor, which means that you can make an infinite arrangement of adjustments that apply to some or all of the image. It's a "how long is a piece of string" question. It's not uncommon for a 'basic' grade in Resolve to include: Lift/gamma/gain/offset level and tint adjustments (to set black and white points exposure and adjust WB) A luminance vs luminance curves adjustment (adjust contrast and finer adjustments like shadows) A luminance vs saturation curves adjustment (desaturate extreme shadows and highlights) A hue vs saturation curve to take one over-saturated item and make it not be distracting A node adjusting mid-tone detail, maybe contrast, and some levels that has a key to only be applied to skintones An oval qualifier with level adjustment for a vignette adjustment This is without any OFX plugins, LUTs, Colour Space Transforms, etc. When you think of Resolve as an EDITOR, you can compare it to PP/FCPX in terms of how it works, but when you think of Resolve as a COLOUR GRADING tool, you should compare it to After Effects. You wouldn't ask "what computer do I need for After Effects" because it depends on what the hell you're using After Effects for - you could be tracking a window or you could be rendering an entire CGI universe. Grading in Resolve is the same as that - you could be setting WB and contrast or you could be doing a day-for-night transformation. End PSA ???

Hooray! After people started getting their Pocket 4K cameras we only had the Panny S1 to argue about - now we can get excited about the inevitability of 8K vs Sonys ability to quadruple the data rates and not have the camera actually melt! With Japan gearing up for the 2020 Olympics to be broadcast in 8K there is only ~20 months before an entire 8K distribution network (including something like 1000 8K cameras) must be in place, having been constructed, tested, and bug fixed. You'd be mad if you think that Japan will let themselves fail to deliver a technological feat like this while the entire world is watching, Moores law be damned. So, if we're going to have an entirely functional 8K capture, processing, distribution, and consumption network in place in less than two years for the worlds most watched event, then exactly when do you think the tech will be made possible? The day before?

I hear you - it's like flying the space shuttle. I think there are two paths you can take. One is to do as @kaylee suggests and do a small project and search google and YT for the answers for every function. If you learn best by doing then that's a good approach, but I think it's frustrating and you kind of miss out on the big picture about how Resolve works and how it's designed to be used. My understanding is that the overall workflow is different to other packages. The second one is to watch a thorough walk-through of the user interface so you can see how it's organised and what is available. One of the key differences with Resolve is that there is so much stuff in there that is simply absent from PP / FCPX (especially in the Colour tab) and so if you only learn what tools you think you need you'll miss lots of the things that Resolve has that you didn't know existed. If you're going to do this then find the longest and most thorough walk-through possible and just watch that. There's nothing worse than watching a walk-through that isn't thorough and then having to watch another one where they repeat everything you just learned! A couple of points: Unlike almost every other piece of software on earth, the Resolve Manual is absolutely excellent and will answer most of your questions, so look there before asking the internet about something The second place to look is the BM forums where someone might have asked the question before. If you google then other sites pop up but mostly the people who actually know things are at the BM forums. Use YT for learning what buttons do what and what features are available, but don't try and learn colour grading / fusion / audio mixing or mastering from YT folks - they are mostly amateur hacks who use the wrong tool in the wrong way at the wrong time. If you go down this route then you will spend a lot of time unlearning their techniques later on. If you want to hear from people who actually know about colour grading, then liftgammagain.com forums are the best (free) place I've found. There are regular posters there who are polite and helpful who have been grading for decades and really know their stuff. You can also ask here.. Good luck - it's a long process but a worthwhile one IMHO. Put on some music and attack it in bursts with good breaks

I just change the playback speed of the clips in the timeline. Right-click on the clips you want slowed down and it's called something like Clip Speed, set it to whatever percentage you want (ignore the fps box in that pop-up - it's wrong and makes no sense) and choose if you want to ripple the sequence or not, and it's done. I don't know how other people do it, but I think this is one of the great features of Resolve. You can ingest footage of any resolution and frame rate into your timeline, you can change the speed of any clip to any arbitrary speed, and then you can export a file in any other resolution and frame rate and Resolve will handle the whole thing seamlessly. If you're doing any speed changes where you aren't doing a 1:1 of input frames to output frames, you can enable a feature called Optical Flow (in the main page of the project settings, down the bottom) and it enables the Twixtor style of frame interpolating for any sequences in the project that aren't 1:1. I did some tests comparing 4K25p and 1080p50 both at 50% speed in a 1080 timeline (ie, the 4K with the Twixtor effect vs the 1080 just conformed to 25p) and for situations where there isn't a lot of complex motion the 4K looked slightly better or you couldn't tell the difference. The work I do is just for myself and friends and family and isn't critical, but I just do whatever speeds I like and if I want a shot to go a certain duration but I want the shot to start on a certain action and end on another one then I just choose whatever percentage makes that happen. As long as you're not pushing things too far then the effect is normally totally fine.

My advice is to ignore the titles of videos, and to only watch the ones that seem interesting. They're a stills photography channel, I wouldn't bother watching if they had a video about something that only applied to a type of stills photography that didn't interest me. In terms of squirrels, memory is different to attention span, but sure, I'll stop vilifying them publicly

Yeah.. ain't that the truth. Probably buried deep in my hdd somewhere, but the overall lesson was that the recommended solutions (like a steadicam, shoulder-rig, or other rigs with three points of contact) are the best ways. I had fun experimenting, but I eventually came to the same conclusions as the rest of the industry.

I realised that when I posted, but if only you could buy skills and just install them! I do have my eyes on a couple of courses, but in reality I know better than I actually do, and that's mostly just down to habits and practice. Going from not knowing the theory to knowing it and barely being able to do it given infinite time is one thing, but doing it properly, reliably, and quickly on location as well as everything else you have to do is something else entirely. My brain knows quite a lot, but it's my muscle memory that makes footage

Further to the above comments, weight at distance is one part, but the other is how and where you hold it. I spent ages running around the house with various rigs of PVC piping (modular for travel) and even went so far as to fill it with water (readily available at your destination!) and quickly came to the conclusion that the less firmly I gripped it the less hand-shake it got, and basically replicated a steadicam by making little handles that could swivel and were mounted at the balance point of the whole rig.

Definitely suits the aesthetic, nice work It's a ton of work to learn a new package so I can understand not moving from FCPX, but what I would suggest is getting familiar with the colour tab and what you can do there. Watching a good 10-20 minute walkthrough video showing you all the controls for colour will pay dividends almost immediately. For example, did you know that in the Colour tab you can highlight any number of clips, and then right-click on a graded clip and select Apply Grade (I think that's what it's called) and it copies the entire grade onto the highlighted clips. Also worth noting is from a Still you can choose Append to node graph, so you can have stills that have parts of a grade and you can just build a grade from there bit by bit. I've messed around in the past having Stills that had different converter nodes, nodes to desaturate the shadows/highlights, the Glow OFX plugin, etc etc.

Adding weight is one thing (that people already mentioned) but what they didn't mention is adding weight at a distance. A steadicam works because the weights are at a distance from the camera. Here's the physics of the situation: https://en.wikipedia.org/wiki/Moment_of_inertia Something you can try immediately is to use a tripod. Mount your camera on a tripod and try the following tests: Use the tripod with legs compacted and folded together and hold it underneath the camera allowing it to work like a steadicam Same as above but extend the legs most of the way to the floor Same as above but fold the legs out You will find that the first test gives some smoothing in tilt and roll but little stabilisation in pan, the second will be more of the same, and the third will add in some stabilisation to the panning as well because the legs are not all on the same axis. With physics you need a combination of mass and distance, you can't get around it. If you don't want either then do what I did and invest in IBIS.

It sounds to me like you guys are reacting to the title a lot more than the content? But even if you aren't, I actually happen to think he presented a reasonable argument that MFT will die. I think the sticking point might actually be timescales. We'd probably all agree that MFT will die in the next 100 years, and not in the next 6 months, so it's a question of degrees. The Northrups talk about long-term trends in a world that has the attention span of a squirrel and I think that mis-match of contexts is often a factor in people disagreeing about this stuff. In terms of people "making mistakes", or in this example I'd say it's closer to "missing the mark", everyone does that. But then again, if you gauge things by the internet comments then basically everyone should go kill themselves immediately and companies should give up on everything and hand out fantasies for free. Plus everyone seems to fall in love with what they own, which is a strange phenomena. And if you want to get a sense of how level-headed the Northrups can be, check out their reply to Jared Polin around RAW vs JPG. Tony was so level-headed the video was almost boring to watch - not only did he not capitalise on the drama but his reply was designed to put the whole thing to bed instead of create more. I think most of their videos are akin to "The Pocket4K is quite decent really" and that's why they have such a following. They gave the GoPro 360 camera a pretty bad review (for many good reasons) but still presented its good points, and even more recently they still listed its pros and cons when comparing it to the more recent insta360 camera. They're definitely not perfect, but I think on average they are ahead of 99.99% and criticising them based on the title of a video rather than the myriad of points within the video isn't that helpful.

That may be true now, but the way I understand it is that it's a shrinking market and that kind of split may not be the recipe for sustainability. Who knows of course, those with crystal balls (or steel balls) will be putting all their savings on the winning horse. Every successful YouTuber is clickbait royalty - that's just the reality of getting new subs and keeping ahead of the others - look at the UK tabloids (and sensationalist media the world over) and tell me that it isn't a successful strategy. What I like about Tony is that he isn't afraid to say things, isn't afraid to predict things, but in addition to those traits (that are shared with nearly all YouTubers) he's also level-headed, admits when he's wrong, and explains his thinking so that you can understand why he came to the conclusions he did. Even if he gets things wrong he definitely adds to the conversation rather than just taking up your time and contributing nothing.

I'd like to be a better DOP, audio technician, editor, and colourist.. ??? Unfortunately, skills aren't on sale!

I recall some issues that Dave Dugdale had with the a6xxx cameras that looked like IBIS and IS fighting each other. I don't recall if it ever got resolved, but I think that the only way to work out what is going on is to do a logical evaluation through testing. Shoot a shot with everything plain settings and verify that you can get a smooth pan like that. After you've established a baseline start adding in the settings you use one by one and see which one 'creates' the problem. Then leave that setting on and start reverting the previously adjusted settings back to your baseline and see if it comes good again, and if so then it will be a combination of multiple settings.

I'm yet to make the move, but I think I'll have to eventually. Slightly OT, but don't they have new OS's free for a certain period of time and after that you have to pay for the upgrade? I think I had to pay for one when the version I was running didn't support the software I wanted to run and the only free version was the one that had buggy wifi so I paid to get the second newest version.

Absolutely! Think of all the awful products out there, and take a minute to order why no reviewer has ever looked at any of them. It's simply incredible that the people who make awful products never contact reviewers, or if they do then the reviewers aren't interested, or if they are then the product gets lost in the post. That postal system must be crazy good at assessing how good products are - it's the final gatekeeper in the entire system!!

I'd have a look at what the YouTube community uses (especially people who do talking to camera) because they shoot a ton, are hard on their equipment and value reliability and durability very highly. Kai Wong and Lok for example.

Even more fail!!

I've used it with a key to beautify skin a few times, but I'm not sure that it's the best way to go as I haven't compared it to other methods. Also worth checking out is the Sharpen and Soften OFX plugin which gives you independent sliders for small, medium and large textures so you have more control, and it let's you sharpen some and soften others so it's powerful for skin tones. If you want to use a dedicated sharpen or blur tool you can use the Edge Detection OFX plugin (I think that's what it's called) in the mode to display the edges and then connect the main output from that to the key input of the node with your effect in it. This gives you the ability to apply any filter to edges (and presumably also the inverse) in a very controllable way. There are so many ways to get the job done in Resolve!

@Sage I have shot a bunch of footage at night with HLG at various exposure levels (ETTR as well as ETTL to keep highlights) and I'm wondering how to pre-process the footage in post to get the right levels for the LUTs. At what step in the Resolve node graph would you suggest adjusting the exposure back so skin tones are in the recommended 40-60 IRE, and which controls would you suggest (perhaps either the Gamma wheel or the Offset wheel?). Thanks!

My timecode keeps resetting - does anyone know what settings to use so that it only counts up while recording but doesn't reset? It reset at 00:13:11:01 so it didn't "wrap-around".