Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by Policar

  1. This was from the perspective of someone who will only shoot film or Alexa (or occasionally Red), but it applied equally when he was (rarely) asked to use lower end cameras and dSLRs. I definitely notice a weird "locked on" effect with Tamron zooms and even my 70-200mm f2.8 II IS has a bit of a wavy look to the motion that I don't mind, but it's usually fine. I have seen it lock on and glitch out occasionally. Video OIS seems to be programmed a bit different than OIS for stills. That said, at the $1100 price point, most people probably would appreciate the option both for stills and video. Parfocal is pretty awesome, though.
  2. This is really a matter of personal taste. Every professional camera op I've talked with considers IS useless for video because of its unpredictable behavior and motion artifacts. They'd all take a steadicam or balanced shoulder rig over IS any day. Maybe for news gathering or doc I could see the utility, but for "cinema" it's undesirable.
  3. I've seen this, too, even in Gone Girl, where they at least embraced it. The new OLPF looks dramatically better at least. It seems like 95% of people love the Red look. I've never liked it but I'm in the minority and appreciate what the company has done for the industry. It's just a matter of taste. I always found 2k to be enough and preferred rich color over resolution, but it seems the trend is in the opposite direction, and it becomes the artist's job to follow trends or buck them in niche ways, I suppose.
  4. I wasn't posting to a Pewdiepie video intentionally. Someone mentioned losing 100,000 viewers to youtube, so I looked for a youtube video with 100,000 views and equal value on that market, actually more views so more value. Probably about as many people would pay to see that as would pay to see a feature with similar views, at least among youtube's market. It would be hard to find one on Pewdiepie's channel with as few views as he attracts far more viewers, and his content is great for what it is. Exposure does help with sales and at the low end piracy is exposure. 100,000 partial views of your feature on youtube will correlate with some of those people paying to watch your feature at full quality if they like what they see. But it doesn't correlate with 100,000 lost sales, quite the contrary. Asylum was getting exposure for making mock busters, which benefit off the advertising of other features. Their "outrageous" movies now get exposure without piggybacking. That's fairly incredible because no one is going to watch your movie if they haven't heard of it and an press is very expensive. As a business, they're nothing to laugh at, and they employ a lot of excellent craftsmen and do market research. Their content is also better technically than most give it credit for being. A lot of people compulsively torrenting movies do it as a hobby and very few people watch the movies they download more than once. Piracy is an issue, but it's a small issue compared with not knowing your market. Of course Australian horror films do badly stateside; it costs $15 million to launch a distribution and advertising campaign for a small feature. It's going to take something very exciting to justify the expenditure to pick one to compete with Blumhouse content–and piracy has nothing to do with that. If you have 10,000 seeds–if we're feeling generous that correlates with 200 lost sales at 4000 lost rentals at best, but I'm sure tremendously less. That's still maybe $3000 in sale revenue and $20000 in rental revenue, which doesn't even cover a distributor's fee and bring you into the green, but if it did you might see 15% of net points. On the other hand, piracy increases visibility, which drives sales. Either way, it's not a big factor. The Facebook piracy thing is a big deal since it's views being stolen from a similar market. Yes, you'd better make your movie cheap or interesting if you want to make any money. Or good enough to attract major festival play. This would be true whether piracy were a factor or not. It is one, but it's a very small one, and one that it's producers' job to navigate around, for better or worse.
  5. What does luck have to do with it? People don't download torrents by chance. Piracy will be about what it will be. It's fairly predictable, with one exception: If your feature leaks before the rights go up for sale piracy can hurt a lot because it affects word of mouth. Otherwise, it's more a scapegoat than anything. Yes, it can hurt sales a bit. Yes, it's bad. No, it's nothing more than a drop in the bucket. And nothing will stop it. If your video is available commercially, someone is bound to put a torrent or youtube video up. Quality doesn't matter much as regards sales, but perception of quality does. If you can't afford to advertise, word of mouth and IMDb rating (and whatever press and blog quotes you can pay your agent to drum up) matter tremendously. Even festivals can matter provide an impression of quality. But piracy before a feature is released in a major market can hurt a film's perception (at any level from indie to tentpole) because the "quality" perception is no longer under your control; after it's released, it's a pretty predictable but small financial siphon. 100,000 views on youtube, few of which will be completed views, does not equal 100,000 paying customers at $10/apiece. The loss is orders of magnitude lower. This is is a funny video, but I doubt any of its viewers 120,000 would pay to watch it: That's the youtube market. It doesn't correlate with sales, let alone rentals. Piracy is an issue. But it's usually more of a scapegoat.
  6. The tentpoles have by far the highest profit margins. And to a very significant extent their content is dictated by what audiences want to watch. Sequels can save money on promotion because the IP is established. They're safer to green light because the audience is a known quantity. It's fine. TV is serialized; film can be, too. Tentpole budgets lubricate the rest of the studio system and allow for other, oftentimes more interesting features to be profitable. If Hollywood weren't producing tentpoles, they wouldn't produce anything. I suspect that's what some people here would like to see. When something doesn't work out for you or to your expectations, you always want to find an outside source to blame, but for content that's very good and without much marketing, after-the-fact piracy will often only raise its profile and word-of-mouth and sales. Granted, any leak that occurs before a feature opens or a sale occurs can force your hand, almost always a bad thing. Not having control of that and having to answer for content not being good is a difficult thing. Even for the studios--see Wolverine. I have seen one business parter have his business hit hard by piracy. It was over international rights to a $200 million+ Bond movie, which leaked before its theatrical run. I agree it hits hardest at the top. Obviously it's still an issue across the board, but in many of these cases it looks like piracy is the scapegoat, quality or market interest the culprit. A bad movie being widely available won't do anything for its word of mouth.
  7. I might be, too. The sample HDR content was derived from F65 primarily, which doesn't hit the 15 stop mark. It was completely breathtaking nevertheless.
  8. My bad. Weird because I was getting my information from the largest finishing/grading/master distribution company in the world in terms of what Dolby was designating as HDR-spec for cameras, but they heard it second hand. (They're not Dolby, although they have a close connection with them.) Fair enough! My bad. I didn't understand the specifics at the time, to be fair. They also said rec2020 for acquisition spec, which is much larger than rec709, so it sounds like they were wrong about that, too. Maybe they were trying to simplify the information for me.
  9. HDR spec for Dolby Vision acquisition is 15 stops. Not sure how true this is as I'm not an engineer, although I've seen F65 footage successfully presented in Dolby Vision and it isn't an HDR spec sensor.
  10. Camera and tech youtube channels make huge profits off amazon affiliate links. Controversial (or wrong) opinions generate views on youtube as videos are more likely to be shared (as in here) and there's profit in that. But then there's even bigger profit from affiliate links. He's not interested in cameras, he's interested in money. Good information is rarely controversial or simple enough enough to be worth sharing. The more incendiary the criticism, and the less accurate, the more money. I thought you had to at least get the money to rent cameras to start these channels, though. Maybe I should give this a try.
  11. You can use Mandy and ask for reels, but referrals are best. Ask around! Do some culling and assistant editor work yourself unless you have a big budget. $50-$100/hr is pretty standard.
  12. The Fuji Premier needs to reach a 52mm flange distance. It will exciting when this isn't a factor. I still wish there were a C300 speed booster!
  13. I think there'd be a pretty big market if it works better than the current competition and is about the price of an Iscorama or not way too much more. That Kowas are going for $50k/set is telling, so maybe there's room for a higher priced adapter, too. If you could improve close focus over the Iscoramas that would be a pretty big deal, 6' is really prohibitive and large diopters are expensive. The current batch of "affordable" adapters also don't go adequately wide. Blue Panavision-style flares without the need for a streak filter would be nice. A more elegant alignment mechanism would be useful. How's the distortion? The Cooke anamorphics apparently have a weird second order distortion at close focus distances that makes vfx difficult and also just looks bad, but the Arriscope lenses look bad because of the lack of distortion so traditional anamorphic distortion is a big draw for me, not a bad thing at all. There seems to be a trend toward higher resolutions and larger sensors. Even the Amira and Dragon are a bit larger than S35 and sharp enough for a 1.8:1 anamorphic to provide good 2k/1080p. The sample image is really compelling.
  14. What's the price point you're targeting? Is it a screw on adapter with 72/77mm threads?
  15. As all diffusion filters do.
  16. I did try underexposing but it didn't seem to be a much better result. It's just a bit inconsistent and fiddly to work with, whereas bounce board and ND grads just work. All that post workflow stuff is way over my head, though. No idea. Ultra contrast is definitely more a contrast filter and glimmer glass more film look, though.
  17. Cats don't make great ops. Keep it up though.
  18. I think I used a 1 and a 3 and found them both a bit heavy. They lift the shadows but it's similar to shooting log on a camera with a thin codec, you end up with an image that's not much better than the normal image because it's mostly just washing detail out and has worse gradations and tonality once graded back. Big shadow lift, not a huge highlight change. The Alexa has a smooth filter behind the lens (over the sensor) it seems that adds just a very subtle touch. More like glimmer glass. A lot of commercial and narrative DPs are using heavy diffusion to get a film look. Even on the Alexa. I'm combining a UC 3 and a 1/4 BMP to shoot some digital plates to sneak into something that's meant to look like the entire thing was shot on 16mm film and in this case the halation and extra flares are a welcome addition. There's a look to 16mm that almost feels HDR because windows and areas of high contrast flash the image a bit. This is cool: http://www.cinematography.net/edited-pages/digicon-test.htm Not a lot of aliasing and there's a nice look out the window. Seems better than I remember my t2i being at 2500 ISO, less noise and a better texture, I would stick to 800 ISO and lower to get a usable image and that's not too bad. Definitely a step in the right direction, but the shaky camera work isn't doing you any favors.
  19. It looks badly shot to me–not incompetent, just not great–but we're all allowed our opinions and that's mine. Even the interiors are badly overexposed and the lighting and composition on the faces are poor. Again, composition is subjective but this is pretty bad by any standards. "Sharp" is consistent with C300. I've never heard of sharpening in post on a commercial project except to fix misfocused shots. If this is from freelance shooters in LA it's almost certainly C300, that camera is everywhere and practically what you're expected to own. I think the Vice documentaries shot with the C300 (and dSLRs) look nicer. It's definitely the camera I see most in LA, with the Amira and C100 coming in about equally tied for second. This is generally doc work. I don't work on set so it's what I see in the street. When I'm on set I see the Epic Dragon a lot, too.
  20. Poorly lit. Poorly exposed. Poorly framed.
  21. Whatever it is it's very poorly shot. C300 is believable.
  22. A lot of Alexa's magic (and the film magic it's emulating) is halation/diffusion. After further use of the ultra contrast, it alone isn't enough for me. The HDTV/FX filter might be the thing, but my wallet can't afford the hit. Let us know how it goes.
  23. Looks like a nice camera for stills.
  24. It's not. It appears on external recorders, too. 4:2:0 "artifacting" is basically invisible except as slight pixellation around highly saturated high contrast edges and would be invisible at 50% resolution. This is aliasing, there's a moire pattern, visible no matter how much you downscale the image. Fwiw I've seen aliasing on the Alexa under similar circumstances, and it's surprisingly pretty bad at times. Canon is worse, however, and looks different. If you think about Nyquist, the AA filter would have to reduce resolution to mtf0 up until 960X540 on a 4k sensor to eliminate aliasing under similar circumstances, so I doubt anything will be immune. But the C300/C100 is a little worse under certain circumstances and I assume it has to do with its imagining pipeline. I would say you'll see this in one shot out of 10,000, though. The Mk II seems to have a color profile modified closer to Arri's. More magenta rather than yellow in the skin tones. Different blue, less teal. Slightly different red, less orange. Better overall, a bit more accurate. But not a big deal to most people, whereas the better viewfinder, better codec, better low light, and 60p make the $2000 price difference justifiable. I wouldn't buy two cameras if you only need one. Nothing depreciates in value faster. If you prefer the FS5's image for your own work and can figure out a good workflow for corporate videos or bread and butter work (shoot AVCHD) just do that.
  25. I'm not sure I agree. The way the image is constructed is pretty similar, with each pixel over-sampled from multiple photo sites rather than interpolated 1:1–it's just that Sigma has the three photosites stacked rather than four adjacent and no AA filter. So there's a bit more sharpness and a bit less error. Both have some aliasing, but aliasing is far more acceptable for stills. But the per-pixel sharpness and look are similar. As regards sharpness, the C300 has the sharpest and close to the most detailed 1080p image available, assuming your lens is sharp, including the Alexa (which is fairly soft as things go) and 4k downscaled from other cameras. That said, it does exhibit aliasing with red and blue fabrics and the tonality isn't great. And if you downscale a 4k image with nearest neighbor interpolation from another camera, you've got something super sharp, too, but again–not as great as regards tonality. The anti-aliasing filter is also weak on Canon's range. I'm not saying the image is better than the competition–it's worse overall than 4k debayered at 4k and downscaled intelligently–just that per-pixel, it's super sharp. And looks like what you'd see 1:1 on a Foveon-derived image. I'm not sure in what respect the C300's image isn't more Foveon-like than anything that isn't Foveon-derived. Both images are constructed similarly and oversampled and have good subjective color if slightly inaccurate color, and greater per-pixel sharpness than debayered and/or downscaled images especially when viewed at their native resolution, but fewer pixels in total than the competition. I get that you don't like Canon's business model, but the objective qualities of their first generation cinema cameras are what they are, for better (low light, color matrix and lack of chroma clipping) or worse (codec, dynamic range compared with the current generation, skew, very limited slow motion options). I'm not saying it's better overall–it isn't. Nor is sharpness that important compared with tonality, which is why I prefer the Alexa to anything else. Just that it's what the image looks closest to, and the closest imagining pipeline for sake of comparison.
  • Create New...