Caleb Genheimer

Everyone will have a slightly different opinion on this, but I would pretty much say if you’re after anamorphic in 2019, it is for the stylistic optical idiosyncrasies historically associated with Hollywood’s long-standing use of 2X squeeze factor lenses. 1.33X squeeze factor lenses will get you a very mild/watered down version of that look, and in most cases will end up being almost as expensive as 2X options, or will be poor quality optically. 1.5X lenses can be great, but there were never many made, so they fetch a high price. Good 2X adapters are well-regraded at this point, and you’ll be hard pressed to get one for a truly “budget” price. But there are still a decent quantity of quality examples floating around. All this being said, they are priced as they are because the resulting look is absolutely worth it, and they are still priced low enough that they are accessible. Your best compromise is probably a smaller 2X, for example, a Sankor or Kowa 16-s. You won’t be able to use as wide of back lenses without vignetting, but the image quality is still top notch. If you decide to jump right in, the top dogs would be the Kowa Bell and Howell, Kowa 16-H, and the Isco Ultra-Star. If you are seriously shopping, get the @Andrew Reid anamorphic guide, and watch @Tito Ferradans YouTube channel. To use these projection scopes reliably, you need good clamps. Redstan are top of the line for Kowas, and HTN has a great locking front replacement lens body that gives you standard threads while also locking the focus. The budget variable diopter is the SLR Magic Rangefinder. It resolves excellently and increases vignetting minimally, but the coatings produce flaring that some find difficult to work with. The king of variable diopters is the Rectilux Hardcore DNA, but it is priced accordingly and is only made in small batches with a waiting list. You are going to want one of these two. Double focusing is a pain and almost impossible to get perfect. It can be the last item you buy, and you can get used to anamorphic without it, but you will need one. Expect to spend at least $1500 by the time you have a scope, clamps, variable diopter, and a prime taking lens that plays nice with your scope. That’s scraping the bottom of the barrel and getting lucky, but as lens prices go, that’s really not crazy money. $2500 or maybe a couple hundred more should be enough to get you into a good Kowa with a desqueezing monitor and all the extra goodies. There is a new line of 1.8X primes coming out that’s running just over $3000 per lens. I’m planning on transitioning to a set of those, but they’re for m4/3 sizes sensors. There’s a certain economy about the projection adapters, considering the same anamorphic block can be attached to your prime lens of choice for the various focal lengths, but at a certain point, dealing with countless clamps and adapters can get tiresome, and the lens rigs can get long and heavy. Ultimately, projection scopes are not well-suited to a truly professional style production. They’re perfect for passion projects and style-centric work like music videos, short films, etc. A crafty person who has good problem-solving skills can absolutely DIY some of these things and potentially save a little bit of money, but as I’ve pointed out, there are tried-and-true solutions already on the market for reasonable prices. I for one would say you’re lucky, even if the prices are now higher on the actual scopes. when I got into it, Redstan was the only good clamp option even though they didn’t have support brackets back then, and variable diopters weren’t even a thing. There were no affordable monitors with anamorphic desqueeze. now you can pick up a SmallHD Focus (which I would highly suggest you do), and it will serve you well in all your production, not just for anamorphic desqueezing. If I have one serious piece of advice, it is that cheap clamps will make your experience miserable. Get a lens clamp that can clamp down to 15mm rods. Sloppy loose lens rigs make for unusable footage with jolts in it, and clamping to rails ensures that, once aligned, your anamorphic will remain aligned. No “parallelogrammed” footage.

I don’t know that it has an f-stop in a conventional sense, as it is not a complete optical system. I suppose you could still technically define one based on the size of the front element versus the size of the smallest opening within the system, but it doesn’t really lose much light. Your question is similar to asking what the f-stop of a diopter is. The Kowa is an optical device, but it’s not an entire lens on its own. Since it is compressing by 2X, it may even technically be adding almost a stop to whatever lens you put it in front of, but I’m not a technical expert so I am not sure about that.

@Awcine I’ve always understood this, but yours is the most clear explanation I’ve ever come across.

Dang, I didn’t realize they’d all been snatched up! I guess I made a good investment buying a 16-H 7 years ago. I guess it about doubled in value.

I’ll be saving up but waiting for the 28mm to start the kit off with the wide. Definitely in that magic “sub-$4K” price bracket that nobody has done well yet. That’s the price of a flagship DSLR body, which really is pretty reachable/attainable. I’d bash them for not being S35 compatible except for the recent explosion of the Pocket 4K. Truly a cinematic beast of a camera, and a set of (near) 2X anamorphics would punch it WAY above its weight class. I’m sure the smaller sensor size is one big factor in the reasonable pricing. If these are actually decent cine glass, they’ll keep m4/3 alive in my heart pretty much indefinitely. Don't get me wrong, medium format is inevitably coming, and it’s glorious. It definitely will have its place in the spherical realm... but smaller sensors work better for front element anamorphics. The glass doesn’t have to be ginormous to work.

Thanks!

Does anyone have experience on how these two compare in terms of vignetting with wider angle taking lenses? The HCDNA looks to be bigger diameter but that’s not the only factor I’m sure. I have a Rangefinder and am trying to figure out what I might gain by switching.

The one and only Redstan has a host of new products out on their website which I haven’t seen mentioned anywhere yet. For those that don’t know, Redstan have always been the OG maker of snug slip fit projection Ana clamps. Especially exciting to me are the clamps with support feet, as they allow you to tighten things down to a set of rods and eliminate mount/adapter slop. Some things are listed as limited runs, so I’d snag em quick if you’re interested. http://www.redstan.com/index.php?route=common/home

I will! Got a lot going on this week but I’m hoping to squeeze in a test video.

Hey all! I’ve been doing some soul searching lately... they say the camera lens is the window to the soul, right? Anyway, seeing as my current cameras both have multi-aspect micro four thirds sensors (GH5S and Pocket 4K), I’ve been planning to update my taking lenses for anamorphic. I use a Kowa 16-H which is fantastic, and the SLR Magic Rangefinder, which works well in most cases. I messed around some with old primes that I have kicking about, and determined that the right 28mm lens just MIGHT squeak by in anamorphic mode on the GH5S, and would probably work perfect on the Pocket. I was also after a native mount pancake, because let’s be honest. On gimbals, the typically long and front-heavy anamorphic setups just don’t work well. I have a Ronin M (not a small gimbal by 2019 standards), and I have to add a lot of ballast to BARELY make the Kowa work. The gimbal strains hard and can’t tilt to vertical. Well, a short google later, and I ordered a Meke 28mm pancake on Amazon. What the heck, it’s really cheap, doesn’t hurt to try it right? Well, it just showed up, and it works PERFECT. If you have a Kowa 16-H/8Z, I highly recommend snagging one of these. As with anything on the wide end, it vignettes pretty hard with the Rangefinder mounted, but with just the Kowa, it’s clean as a whistle (cropped to 2.39 of course.) This will be my new go-to for gimbal work and for wide establishing shots. I have yet to test it on faces, but it’s probably perfect for that near fisheye mumpy up close look. If I get in a tight corner, I’ll be relieved to have this (literally) in my back pocket. My only gripe is that they don’t also have a 35mm/40mm/50mm pancake set, because a range of pancakes this small would be gamechanging for anamorphic adapter users. I never thought I’d have a Kowa setup as small as this. It’s crazy. Until a new anamorphic beats the Kowa for width and quality, I’m sticking with it.

The main challenge is the length and front-heavy nature of projection anamorphic setups. You have a modern lightweight camera body, with a vintage prime on front. Off of that, you hang a clamp, and a heavy anamorphic adapter . . . and then on the front of that, a heavy single focus solution and a wireless follow focus next to it, all the way out front. Most gimbals might be able to handle the weight in compact distribution, but with it elongated and front heavy, the torque demand on the tilt axis (and a lesser degree the pan axis) are pretty hefty. I would shy away from single handle gimbals and get a decent mid-tier two handle gimbal, like the Ronin MX or something.

I would almost argue that the Pocket 4K is $1K, considering it comes with a software license worth $300. The mind-blowing practicality of BMRaw cannot be overstated either. I dare say it cuts faster than h.264. People dock it points for having deficiencies that require it to be kitted out for serious use, but that’s just the pedigree of a typical cinema camera experience. If anything, it is easier to rig than bigger, more expensive cine cams. I have a cine cam now, and money left over for other things like wireless follow focus and monitoring solutions that can quickly make professional production methods a reality, even on small projects. My plan is to get a quality cage with lens adapter support, and a positive-locking EF Speedbooster to transform the Pocket 4K into a rock solid full frame RAW sensor block. The image quality I’ve been pulling out of this little thing without even trying is already punching obnoxiously close to Alexa territory... and we’re talking hand-held selfie vlog shots with an SSD dangling off the grip. Borderline idiocy. This camera gets me into the cine environment on the cheap, and there’s a whole lot more to that besides the end image quality. The UI and general handling are a whole other workflow. I can kit up the necessary accessories, and almost all of them will be useable in the future with yet-unreleased camera bodies. Gear like v-mount batteries, Nucleus-M, Small-HD Bolt, rods, quick release systems, mattebox and filters, shoulder rigs... Sure, you have to kit it out, but once you do, you have that kit forever. Resolve 16 is a beast. I’ve spent 10X as much time in Resolve the past month compared to time spent in my old haunt, FCPX. The NR is magical and very natural if you know how to dial it in. The wait for the new FilmConvert is agony right now. I’m ready to leave the baked in film print emulation of the current version in the dust. I’ve been using the status-m workaround lately, and while cumbersome, it makes FilmConvert pure magic, doubly so when you give it BMRaw to work with.

I call it Arrival Syndrome. Good lord, to intentionally choose to limit your whole film to about 50% of the available tonal range?! It’s moody on an Instagram photo sure, but oppressively fatiguing on a 2 hour cinematic. I’m of two minds on the audio. People absolutely have TV systems capable of producing cinema dynamics in a clear manner. It would be better rather if manufacturers of audio equipment would integrate dynamic volume of some sort.... If you have your volume down, compress everything more so you can hear quiet stuff, and bump the center channel a touch for dialogue retention. When you crank it, use the dynamics of the source. It is a bit hypocritical to bash for wasting dynamic range on one hand, and complain about extreme use of range on the other. Device-aware playback is the answer. Or, Devices which analyze the content and adjust to reproduce it satisfactorily. If you’re on an iPad, the pad should be smart enough to detect high dynamic audio.

If you’re delivering to cinema, it hardly matters. Most festivals still ask for 2K DCPs or even Blu-Ray. On a good cinema projector, a 2K DCP looks exceptional if you master it properly. I’ve been shooting 2X for almost a decade, before there were affordable monitors, 4K/high DR DSLRs, or single focus solutions. Composition rules apply even if the image is squashed. Wether you crop the sides or not, stretch 2X or 1.5X, the vertical lines of resolution remain the same. Perceived resolution is a funny thing. It’s infinitely more important to get a sharp scope and a taking lens that works well with it... and then to nail your focus. I prefer a 2X scope for the look, and the potential inconveniences are easily solved. The single most helpful tool with any anamorphic is a good monitor that can desqueeze and crop your image. Do yourself a favor and grab a SmallHD that runs the latest firmware. It’ll desqueeze and crop to any configuration. Anamorphic is just something you have to jump into and learn for yourself. It’s comparatively easy to do these days. Pick a scope and start experimenting, learn your rig and adjust your setup as you encounter things you want to improve.

Yeah, voight 40 or Konica Hexanon 40... and ditto on the Kowa 16-H/8Z. It’s the best of the 2X projection scopes.

Andrew, I think that your asessment of H.265’s adoption roadmap versus the arrival of new RAW codecs is very astute. I was having a conversation with one of my company’s Videographers about exactly that: the timing is too late. H.265 gave me a huge leg up in the NX1 for several years (especially after the hack), but now it’s too little too late. I predict it will be the “Pro” codec in everyone’s flagship models for maybe two more product cycles, then everyone will one by one start to one-up each other with RAW integration. What that RAW format ultimately proves to be remains to be seen. ProRes compression, while tried and tested, is a bit old hat, and while I don’t doubt that applying it to a Bayer pattern signal produces fantastic image quality (ProRes RAW), there probably are more efficient possibilities out there at this juncture. In the limited time I’ve had to play with BlackMagic RAW on the Pocket 4K, I’d say it is more closely representative of what may be possible. Perhaps we will see H.265 compression applied to a bayer pattern signal. That would probably be a very flexible codec for the size indeed! The emphasis shift that I hope to see in the long run is away from codec flexibility and dynamic range, towards quality of color science. In two years if the dynamic range of most cameras is high enough to be a non-issue, and RAW in various forms also exists across the board, other issues like color shift over the range of exposure, and dual pixel autofocus will hopefully be pushed to the forefront. The Alexa remains unthroned for a reason, and it’s not just that it shoots RAW. Canon protects its Cinema cameras with that great AF. RED has had a monopoly on good compressed RAW with REDCODE, but not for much longer. The FS5 has that brilliant internal electronic variable ND! There is a lot of tech out there right now that single manufacturers provide. Hopefully it all spreads around soon just like RAW.

You should be able to make it work. Same rules apply when selecting taking lenses that don’t vignette though. Get your anamorphic and then figure out the taking lens fov that is compatible. Find out which focal length matches that fov on your sensor size.

They really need to add a GH5s equivalent 4:3 mode to the Pocket

Because the factor (0.64 or 0.71) is not a whole number, you divide on your calculator... pretty darn sure that’s the correct way to do the math, but if I’m wrong I’ll gladly eat my words.

Finally, zombie filmmakers can rejoice! Undead skin tones in stunning 8K H.265!!!!!!!!

In my experience, one can throw math at a forum until the cows come home, but until you actually set up a particular focuser/scope/prime/speedbooster/sensor combination, there’s no real way to know. If the sensor size is that important to you, start there. Get the camera that you want. then get the scope you want, and a focus solution. finally, start auditioning prime lenses until you find something you like. the factors/equivalencies/etc. are ultimately crude ways of approximating the complex physics of the optics. They’re great for making a strong educated guess, but won’t really give you a definitive answer as to the “ultimate” setup. If you want a real scope for s35, just save up and snag a LOMO. Or sell your car and get a Hawk. There really is no quick path to assembling something that will perform exactly like a real cine anamorphic. same goes for s35 cameras. Get a Red, Pretty sure they save stills. The math... 1. look up the sensor dimensions 2. multiply by the focal reducer’s factor 3. Profit

Yes we agree... but the suggested focal lengths are incorrect. I can BARELY use a 37mm natively on my GH5s with my Kowa 16-H (the most forgiving 2X projection lens out there). That means at least a 75mm lens is needed on fullframe, and this is all assuming you are shooting video and cropping off the sides of the 16:9 for a final stretched output ratio of 2.39:1. I’ve shot the Kowa on a full frame canon in stills mode, and to cover the whole sensor I had to use an 85mm. It still had little tiny vignettes in the corners. You probably get around 16 degrees of vertical angle of view, and 38mm degrees of horizontal angle of view. In fullframe non-Anamorphic terms, that’s equivalent to an 85mm in the vertical and a 35mm in the horizontal. You might squeeze a tad more depending on the lens. But single focus adapters, filters and diopters on the front of the scope will all further constrict the vignette. I’ve never heard of anyone using a 40mm on fullframe with projection anamorphics, but if I’m wrong someone should speak up.

Don't confuse focal length with field of view. There is NO advantage/disadvantage on ANY sensor size when it comes to field of view/angle of view (how “wide” of a view you get) through a scope. It will vignette at the EXACT SAME field of view no matter what size sensor you use. The change in useable focal length (in mm) on various sensor sizes in indeed due to “equivalency”, also referrers to as “crop factor.” There is something to be said for larger sensors with regards to depth of field. If you want it shallower, larger sensors will be the ticket.

Extra resolution is helpful when doing any transformation of the image in post. A great example is Anamorphic. I’ve been developing a method for reducing some of the mumps in the middle and extreme compression at the edges of my anamorphic footage, and pushing those pixels around can definitely start to muddy the image. Downscaling 8k to 4K or 2K will also reduce noise by summing the data from multiple photosites. I’ve recently finished a short film, got a DCP authored at 2K, and it looked fantastic in the theater. There’s even 1080p footage in there. You can’t tell. There’s definitely an edge seeing the 4K master file on a good monitor though, and I DID do mild reframing on about a third of the film. Having thrown something up on the “big screen”, I can absolutely see the benefit that “overshooting” (with respect to delivery resolution) would have. That being said, I don’t give a rat’s rear end about Sony cameras. Their look is the most unpleasant thing to look at IMO. Hopefully Panasonic sticks one of these in their mkII FF body, or in a fullframe EVA-2.