kye

Interesting. That's different to the A7Sii where IIRC he had customised one of the profiles. Do you know if he's doing a lot of colouring in post? I'm not familiar with the look SOOC.

Not sure if it's the easiest, but I can recommend Davinci Resolve as an all-in-one NLE. There's a free version that you'll be able to use for basically everything you will need. It's one of the premier colour grading packages used in Hollywood, so the restrictions on the free version aren't aimed at the enthusiasts. FCPX has lots of devotees and PP also does although apparently it isn't that stable and people are starting to leave it to get better reliability.

+1 for getting white balance right in camera. Sony is known to be harder to get good colour from but that just means that you have to be more particular with WB and processing in post. Kraig Adams on YouTube is known for getting great results in camera and is currently using the A7iii. He made a few videos about it that might include his colour profile and settings.

Did those specs say that readouts were at 12-bit (and higher)? Or is that all sensors and the manufacturer just chooses 8-bit regardless?

@leslie did you put an anamorphic lens on your gopro? Or just replicating the aspect ratio? In terms of getting stability I'd suggest doing it right and only having to do it once. If you can look at how rails systems work and learn how they've solved the various design issues while keeping flexibility then fabricate something that applies those principles then I think that would probably be worth the effort.

It's interesting that the 50mm 1.8 is perhaps the most common lens design, and if we take an image circle designed to cover a FF sensor and applied a 1.33x horizontal crunch then that would line up pretty well with an APS-C crop, therefore making a lens that shares (at least some of) its optics with the cheapest and most common lens design ever made. I suspect that's pretty much the recipe for how they made this lens - get a 50mm 1.8 design and apply some optics that give the 1.33x and there you go. This makes Sirui the anamorphic equivalent of Yongnuo, which seems like a great thing for anamorphic shooters everywhere

@mojo43 watched the video - cool stuff! It's definitely a striking result, and you explained it pretty well. Some thoughts if I may.. Yes, please make more videos showing your grades One thing I've noticed from other grading videos is that they start off by talking about the colour, but every time they touch a control they then explain what the control is and how to use it (literally things like which way to drag the mouse to change the value) and I find that really infuriating, so don't fall into the trap of over-explaining the tools (which you didn't) How to use Resolve is covered by many many people, but what isn't covered is the art side of it, both from a perspective of talking about what you're seeing (eg, how to choose a good amount of contrast and what you're looking for) as well as from a perspective of why you want that much contrast and how that relates to the content of your film (eg, how colours link with the emotions of the characters, etc) If you can talk about the art and not how to use each tool then you'll find yourself without that much competition, and the videos will be useful to people who already know how to drive the tools. I'm looking forward to the next one!

I'll watch when I get home, but you're right about not so much on YT. I started a thread about Resolve and colour grading just to collect resources so that we can share them - it's here:

Just found this guy - looks like he knows what he's talking about....

I just conducted a very thorough analysis (I measured the height and width of the thumbnail image above with the screen capture thingy) and it was 3.334 so yes, somewhat wider than normal! The Atlas Orion Anamorphic T2 32mm they used is a 2X, and if you film 16:9 that's 3.55:1 and 16:10 it's 3.2:1 so maybe they did 16:9 with the lens and then cropped the edges a little?

Ah crap. Of all the times to make a typo! Yes, I meant that maybe the Cooke had a lower F-number despite it's T-number being higher than the Xeen. I think that we have a major problem in society because people are discouraged from saying they don't know things. What this means is that they make observations about the world that are useful, but instead of just saying "I see this, but I don't know what causes it" and letting other people take that as interesting information, they say "I see this, and it's because XYZ" which then makes people who know that XYZ isn't how things actually work throw out the observations along with the faulty explanation (and in today's society there's also a growing trend of completely disqualifying anything that a person has ever said just because they got one statement wrong, ever, which is really sad because everyone is wrong about things on a fairly regular basis). From that perspective, I take it that cinematographers who get paid to do high-end work might know a thing or two about the various aesthetic properties of lenses and images in general, and I think there is lots we can learn from them in this regard. However, I'm not going to take everything that any cinematographer says as true without any analysis, or any critical thinking, as that doesn't get us very far except "buy this product because I bought it", which doesn't make us better film-makers even if we could afford the stuff they're talking about. And if you think trying to convey aesthetic impressions of things in words is difficult (which it is) then try getting into high-end hi-fi. The number of times that I've been describing something I'm hearing and someone else on the internet tells me that I'm not hearing it, or that it's not possible, or whatever, is ridiculous, and if I had a dollar every time then I could afford to buy them a system good enough to prove them wrong

and now that you say that, it's obvious! oops ???

Yeah, there's something to be said for ignoring these relatively small differences... as in, a side-by-side difference that is obvious might not be obvious if in two separate shots with a shot of b-roll between them, and probably isn't noticeable when used in different productions. Certainly, if one is ten times the price of the other then you have to question the value you're getting from something, although the rental cost wouldn't be a 10X factor once you take into account the insurance overheads etc. You can absolutely make incremental improvements to get from where you are to nirvana, but the cost involved is pretty darn high! The Dog Schidt lens in the main lens test in my OP is a modified version of the Helios 58/2 which can be had for $100 or less on ebay. It's not as good as the $10,000+ lenses in the test, but it's not 100 times worse either. The laws of diminishing returns really kick in once you have a camera with interchangeable lenses and you get a few relatively fast primes in your kit.

Certainly the IBIS distinguishes the GH5 from its competition more today than it used to, but before the GH5S, A73, P6K, P4K, Z6, etc, lots of people were buying it and using it on tripods exclusively. If Panasonic can make another “solid reliable work-horse” camera that had modes that other cameras only dreamt of in the GH6 then i’m not sure how important improving the IBIS will be. Of course, how you create a workhorse with new features is a bit of a tricky one.. (8K? 6K60? 4K240? 1080p960? or 12-bit h.265 in a variety of aspect ratios and bitrates perhaps?). Certainly I want higher bit-depths, higher DR and dual-ISO for shooting in difficult available-light conditions, but the file sizes of RAW make my blood pressure rise beyond acceptable levels, plus the lack of stabilisation for vintage manual primes is a deal-breaker anyway. Smaller sensor isn’t a trade-off when you compare it to 10x8 style film cameras, it just seems to be sub-optimal because the shallow DoF and wide angle lenses people want aren’t quite so available. I look at the S1H and apart from the ridiculous cost and the fact I’d have to re-buy all my lenses, I just look at how large and heavy it is and rule it out immediately based purely on that. Everything is relative. FF is a trade-off that isn’t worth it for me because it’s too large.

I've also seen mixed reviews. I think it works great in some situations and not in others. It would probably work very well for non-IBIS cameras like the GH5S and P4K/P6K. In terms of working with the GH5, the IBIS becomes a problem. The motion recorder will record what motion there is, but it won't know what residual motion there is left after the IBIS has done it's thing. You could potentially try and record that too, but that adds another level of complexity to things. I think it could be made to work, at least to the point of it being better than EIS, but it would require R&D, the willpower from Panasonic (or other brands), and they'd have to see a clear return-on-investment. It's definitely true that the GH5 owes part of its excellent reputation to the IBIS, but I'm not sure how much of a factor that is (as the GH5 is also an excellent camera in many other ways) and beyond the IBIS and the EIS I'm not sure how many people would be in the "shut up and take my money" camp if Panasonic improved things on top of the existing performance.

The footage looked pretty poor in shadows (watching in 4K on a 32 inch monitor) but that's pretty normal for YouTube compression in my experience. It's also common practice to add grain in post in order to stop the YT compression from making horrific banding too. I've seen videos where people de-noised the shadows in their video and the banding was so extreme that it looked more like a psychedelic fractal-trip video than poor video quality. I heartily recommend taking some test clips with increasing levels of grain applied and upload that to YT then watch it and see what it looks like once the YT compression has absolutely crushed it. I suggest trying 3-5 clips in varying lighting / contrast conditions and having a couple of seconds each, then having about 10 repeats of this sequence of images with gradually increasing levels of grain applied. I'd suggest a processing pipeline of NR followed by added grain, either from a plugin or just noise, it won't matter much for this test. And don't make the mistake of not adding enough grain, go from zero grain (so you can see what that looks like) to about three-times as much grain as you think is even sensible. It might be good to put some text in the video with your settings so that you have a future reference. I did this test, with the Resolve grain plugin at defaults except the strength pushed up to somethings ridiculous, and just ramped up the opacity of the effect. The full effect was still far too much, but the sweet spot was about three-quarters of the way up, and the first two thirds was basically identical because the YT compression absolutely killed it. It's a pity that the P6K video wasn't uploaded to YT in 6K. Screw that... I can't believe people go to so much trouble with 2X anamorphic just to get the aspect ratio they could get by just cropping!! ???

There certainly seems to be a new focus on anamorphic in the cinematic YT echochamber of late. Personally I'm not that enamoured with the look. The flares are cool as a special effect, but I think they'd get old quickly, and even after re-watching all three seasons of The Expanse recently, which is sci-fi at its finest and also included more than its fair share of people in spacesuits exploring the unknown whilst their headlamps create absolutely spectacular (circular) lens flares, even then part of me was thinking how cool it looked and the other part of me was wondering WTH it had to do with space adventuring. I'm more inclined to think that the oval bokeh and flares are more of a sentimental association with cinema rather than cool in an absolute sense. Of course, the wider aspect ratio does look epic to me, and I think that's because you're forced to compose with people fitting into the frame vertically and therefore you're getting more panoramic effect, taking a small step towards immersion. I was thinking last night that 2.35 is 1.32 times wider than 16x9, so I suspect (pending verification of the maths) that my 7.5mm wide cropped to 2.35:1 is actually the FF equivalent of a 20mm anamorphic setup, and cropped to 2.66:1 it's also the equivalent of a 22mm anamorphic setup. They're approaching the classic 21/24/28mm FOV that Hollywood loves so much, right? @heart0less your spreadsheet seems to be calculating FOV as an angle, but doesn't convert that back into equivalent focal length, is that right? That doesn't align with my brain, which thinks about FOV in FF equivalent lens terms.

If small amounts of barrel distortion are associated with better 3D, does anyone add this in post? In a literal sense, if the wider front element was able to "look around" an object then the edges of what it sees would be further behind an object. I'm pretty sure that's what having a larger aperture is, and why wider apertures have larger bokeh. Ah! These lenses are specified in T-stops and not F-stops, so maybe the Cooke at T2 actually has a larger aperture than the Xeen at F1.9? T-stop is related to exposure value and also takes into account the light losses in the optical path, so maybe the glass has less transmission but the aperture is wider? It's kind of hard to tell from the video as they have different minimum focus distances and slightly different focal lengths, so a direct comparison of the size of the bokeh balls doesn't seem to be available. I've noticed in my lens comparisons that when comparing lenses of similar but not identical focal length (eg, 55 vs 58mm) that the same F-stop consistently gives more 3D pop on the longer lens, because the background is slightly more out of focus.

So, 50mm with a 1.33 factor will have the width of a 37.6mm lens, but on the P4K that becomes a 71mm horizontal FOV?

Video from BM (with lots of low-light shots): Cinematography: Dima Kalenda Blackmagic Pocket Cinema Camera 6K Blackmagic RAW Constant Bitrate 5:1 ISO 3200 6K (6144 x 3456) 50 fps Atlas Orion Anamorphic T2 32mm @ T4 EF mount The IQ from the taks is really nice, definitely good lenses.

Yep, Xeen doesn't look too bad there! At 1:30 he claims that the larger front element on the Cooke means that the lens "looks around" the model and "It's going to take the background and push it way back and it's going to bring her forward". This is something I've never heard before, does anyone know what he's talking about here?

Not sure. The only thing I'm certain of is that a 50mm anamorphic is wider across the horizontal FOV than a 50mm circular lens would be

Doesn't the math work differently for anamorphic? I thought it was 50mm becomes 100mm, but applying the 1.33 factor that 100mm to the width gives you 75mm. So it's as wide as a 75mm equivalent, but is obviously only as tall as a 100mm. Maybe I'm wrong though.

Of course I am... considering this is the budget lens thread The spherical cine lenses aren't that much cheaper!