Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by Policar

  1. How is it better? The 5D RAW has lower dynamic range. Worse resolution. Chunkier noise with much more banding and chroma noise. More skew. Worse color rendering. Lower resolution. What part of it is technically better? Granted, there's the "14 bit" part, which is basically irrelevant because the image is noisy enough in both cameras at base ISO to not quantize to any better dynamic range than can be stored in Canon Log in 8 bits and anything beyond that is just worthless data, but other than that I can't think of anything. Someone did an incompetent test with the 5D in h264 that found it had more dynamic range than the C300 simply because the codec was so strong is wiped out the shadow noise, but competent real world testing shows that in RAW the 5D has worse dynamic range–it's very close, but the 5D has fast loss of color detail in the highlights and noisier shadows. If you're shooting a chart it's similar but the loss of color detail in 5D RAW means it loses about a stop of highlight detail that an informal comparison not accounting for mixed lighting or only shooting in daylight might miss. I'm curious, though, because you make that statement as though it's objectively true and I'm curious why. I could be missing something and there are metrics that I suppose I might not have considered. I also disagree strongly about the workflow, but that's more a matter of opinion and budget. Where I work usually doesn't have the budget to shoot RAW but some people do. 99% Alexa simply because the workflow is the cheapest. Fwiw I have used the 5D RAW a bit on professional work (I directed and shot an opening sequence for NBC and I used my 5D Mark III to grab an angle–it worked great), but I wouldn't use it as an A camera for anything, just my opinion.
  2. Fair enough, the F8 and F4 do have better pre-amps. I thought you meant like an h4n or something, because you referred to low end recorders. My bad though, the F8 is still low end relative to industry standard recorders. I just meant leagues better than the h4n, not better than the F8. That's fair, I must have just misunderstood you. I disagree about the image completely, though. But it's a matter of opinion. I've never needed to deliver 4k and have never had a client who demanded it. Most of my clients who have the option to shoot 4k don't even have the budget to manage the additional expenses in post, which can add many thousands of dollars per project so they just shoot 1080p or 2k. But image and business needs are more subjective and individual. Yours are yours. That's fine. The stuff about banding and DR is simply wrong if you have a clue how to expose, though. But whether you care about clipping highlights is again subjective. I still don't know how you got banding with the C300. What were your settings? Maybe exposing 320 ISO two stops over while shooting a flat blue sky and pulling it down in post or grading badly with a noisy key? While I see it in Canon dSLRS and at A7S all the time I haven't been able to find it with the C100 or C300 and I've shot many hundreds of hours with those cameras and done a lot in post with them, too, as they're our Alexa b cameras of choice. I understand the 1DC exhibits more banding, but it's still minimal. Canon Log 1 is not that flat; it's not true log. To complain about banding because it's possible under extreme circumstances if you expose terribly with the wrong settings and use the wrong post workflow to get a bit (and even then nothing like any dSLR) is like complaining about the Alexa's aliasing and equating it to a 7D. Sure, I've seen aliasing on the Alexa. In fact, I've seen MORE aliasing with the Alexa than banding on the CX00. But I would never complain about either problem with either camera. I respect your opinion and your business needs, but I don't think they're representative of most people's. So to everyone, rent or borrow before you buy. But mind your super whites with the C series. 99% of the reason people think that camera has worse dynamic range than it does is because they people don't know how to expose on set or manage their super whites. The dynamic range is closer to an Alexa or F55 or Dragon than it is to a Canon dSLR, and superior in most cases to an Epic MX, even. But to be fair, the 8-9 stops of DR in the 1dX II, for instance, is enough for most situations. And if your clients need 4k they need 4k. They won't mind the occasional blown highlight if it's within spec. But other than Netflix I can't think of another client that requests 4k and that's only for Netflix-produced, not acquired, content.
  3. The C series has a different over/under (favoring highlights by two stops) from the 1DX and just generally far more DR, on par with the 1DC. The 8 bit argument is wrong, the gamma and color space are appropriate to the codec for Canon Log 1. Also, the C100's pre-amps are way better than a Zoom, even the newer ones. Not as good as Sound Devices, but shockingly good. I wouldn't recommend it to everyone, but I would recommend everyone do their own research. I'm not going to argue with you anymore because it's boring to hit my head against a wall, I just want to state something actually true to provide a counterpoint to such blatant misinformation–simply to encourage people to do their own research (and discover that I'm right). I don't want to argue about this, just to encourage people to do their own research because competent people will agree with me and incompetent people can work with whatever tools suit their incompetence best.
  4. I bought a C100 and a 5D Mark III right after they came out. Sold the 5D pretty quickly although the RAW looked good (not as good as the C100 other than its shallow depth of field "FF" look, which was impossible to pull focus with so whatever). The idea that the 5D III's RAW is technically a better image is a misconception fueled by people who are incompetent on set or in post. Both have great images, though, and the quality is really pretty close. I've used almost everything on the market and it's the combination of great ergonomics and a good image without much work in post that's led me to not replace the aging C100. I feel like nothing else on the market has an image that's better in a meaningful way without some sacrifice. Sony is technically a bit better (a stop better highlight detail, a stop faster native ISO) but the ergonomics and workflows are dodgy and the color is hard to work with on the high end and on the low end it's a real nightmare with overheating and SLOG 2 having awful color (the Kodak emulation LUT on the F5 is decent, however) and the ergonomics are awful. The Red is expensive and difficult to work with on set and in post, awful in low light, and actually had slightly less dynamic range than the C300 etc. until the Dragon and the new color processing. The Red M had like 8-10 stops of DR, MX about 11-12, and color wasn't great then. And still it's behind Canon and Arri but it does become subjective because its looks is digital, not film emulation. The GH4 is not bad IMO but its 1080p is surprisingly soft (the 4k is fine but you get a bit of crop) and the ergonomics I don't love and it's not super reliable. But it's not bad at all. But they're all fine. I guess for me if I want a better image the next meaningful step up from the c series is renting an Alexa, but that's also a pain to use. I sort of worried about minor differences until I tried something that was actually different and now for me it's Alexa or bust (though the C300 Mk II look pretty nice after the firmware update–haven't tried it with the new firmware though!). I don't care for the "FF" look, but if you want shallow depth of field get some f1.4 or f1.8 lenses and an 80D to go with the C100 I'd say. The 80D is easy to use for video and the autofocus is useful for close ups (where sharpness isn't important) and you can get APS-C lenses like the new Sigmas that are sharp and fast enough to match f2.8 on FF and for cheaper. And the DR is RAW is good on the 80D for stills. Also the Canon RAW workflow is the exact opposite of the cinema series workflow–a nightmare, and that's really why I abandoned it. The 5D II is nice for stills, but has poor DR in RAW compared with the 80D. I'd get an 80D and a C100 Mk II, but I think part of it is that I'm lazy and I do care about color more than most people (consistently score off the charts in color vision tests). I wouldn't bother with a 1DXII as a video camera. Same very limited dynamic range as other Canon dSLRs. 1DC could be nice, but it lacks the efficient low bitrate codec.
  5. Wow, that's pretty good if it's representative of its quality. Again, the DR-60MKII isn't bad.... it's just not great. That's pretty good. That's the first I've heard of zipper noise on the F8. I'm surprised, but I believe it, and it's definitely a deal killer. Tascam and a pre-amp sounds like the way to go on the low end, I guess (and Sound Devices everything if you have the money).
  6. How much better are they than on the DR-60 mk II? I found those to be bad, but still better than a Zoom h4n by a significant margin and definitely usable for home use, just underwhelming. For the price, bordering Sound Devices grade (as with the F8 and Zaxcom) pre-amps would be nice, even if the build quality is lower.
  7. I don't generally watch vlogs, but the above video makes a pretty good case for them. I think if you go back to Bazin or even look at some of Lévi-Strauss and Sanders Peirce's ideas on semiotics that inform him and the Cahiers, you begin to reexamine what film is uniquely good at and why it became such a popular medium in the first place–and its strengths lie in its ability to record a convincing record of life, a recording of something that actually happened that feels real. Film shares this with audio recording and photography, but film takes it to another level. Unlike literature or painting or animation, it's not a symbol or a drawing, it's a record of something that actually happened in front of a camera. So way before you even worry about lighting or blocking or editing or even storytelling, what film does that's unique and remarkable is that it provides a lifelike and moving record of an event. Modern blockbusters move away from this tendency because they rely so heavily on CGI and animation and compositing that it begins to feel like a video game (which they're imitating; they share a common audience). I think you feel that modern blockbusters are different, and I definitely prefer action movies from before CGI became so commonplace because they feel more physical to me. Mad Max got a good reception because it was a bit more physical. Maybe part of the increasing appeal of MMA fighting (and historically the appeal of sports) is the you get back to that physicality. Now we're at that stage where the eight year old girl (or whatever Coppola said) finally has the resources to tell a story affordably with film. We'e got dSLRs and iPhones and we're making movies instead of just watching them. And I think two tendencies are emerging from that. One is to imitate what we're watching in theaters, and the other is sort of to break off from it. This website is definitely more for imitators, trying to get something that looks expensive for cheap. And I don't find any of the work I've seen posted on this website to be any more interesting than what it's imitating. Some is more technically adept than the rest, but mostly it's a bunch of music videos and montages meant to showcase a new lens or something and it's basically a bunch of camera tests. Which is cool, that's a cool hobby, and it's fun to engage with and it's good to know what gear is out there so you can do your thing–make art, money whatever with it, once you get bored with camera tests. So the work doesn't interest me, but it's still a worthy topic. Neither does vlogging interest me, and in fact it interests me less, but I still think it deserves respect because it's doing something new and unique and compelling. Hence the enormous emerging audience... Vlogging goes in the exact opposite direction, back toward cinema verité and away from blockbusters. And verité, unlike direct cinema, acknowledges the camera, which I think is sort of the film step in presenting "reality." Which, if you'd ask Bazin, is the point. Vlogging brings that to the next level. And yeah it's obnoxious like Michael Moore is obnoxious because the filmmaker becomes his own protagonist and you might not like him. But it's cool to watch stuff people are doing and imagine you're doing it. It's even cooler to imagine you're also the filmmaker recording it. And vlogs let you engage with content on both of those levels. Even something like Rocket Jump (which is more in the "imitating mainstream media" category than vloggers are) makes the entire process from funding to production to distribution transparent to the viewer and encourages the viewer to do it, too. So you're watching their content but separate from that you're relating to it as a potential filmmaker. To me, this is really cool. Most kickstarter campaigns are dumb money grabs, but with something like Rocket Jump it becomes an alternative form of financing that's communicating directly with the audience and that's cool. All the BTS elements there are cool. (The photography tutorials and stuff that pollute YouTube aren't–because 99% of them are just promoting horrible information and lowest common denominator aesthetics.) The BTS elements of vlogging remind me of a film nerd sneaking on set and reading Fangoria or American Cinematographer or watching the dvd deleted scenes... except it's going even way further than that. YouTube kind of fulfills the promise of verité, and to some extent realizes the potential of cinema itself on a very very basic level–even if the content is generally not my thing and I'd argue usually pretty awful. Snapchat and Vine do, too. More than that, they allow you to be the consumer and the producer, so you get a real community. But most people are boring. And most content is boring. If it's democratized, more of it will be boring. Painting wasn't great during the renaissance because it was cheap, you know? Painting got worse when it was democratized. So while I think your Snapchat or YouTube channel can be really banal and millennial and shitty and most of them are... that's the content, not the medium. The medium itself is really cool and there are some YouTube channels and Vimeo channels I enjoy and to trash the medium because most of it is garbage would make me a hypocrite because I really love some of it. I think if we feel alienated by these media it's a pretty boring response to just imitate an outdated one instead unless you really commit and say–okay, I'm holding myself to the standard of my heroes. I'm not content shooting with a camera that maybe they used or has the same resolution of one they used, I don't really care about that at all. I'm going to challenge myself to do with my resources better than they could do with them, or if not better than more personal to my vision. And that's the approach successful vloggers are taking. Like it or not, the cutting edge of documentary is YouTube and Snapchat. I'm not going to say vlogging is a better pursuit than shooting a documentary for the festival scene. I will say I think how you evaluate each has more to do with how you feel about its audience than anything else, and at that point it's a social issue, not a technical or theoretical or aesthetic one. Edit: I think this website produces some good camera tests, however. The "feel" of the image is more than its specs and going out and shooting with a given camera or set of LUTs gives you an interesting window into their potential that specs alone can't.
  8. The only thing you need to be a professional is a client.
  9. All I suggested was basing your decision around your needs. If your needs require shooting in mixed light (and yours, specifically, seem not to, which is fine), the Alexa on the high end and Canon dSLRs on the low end are best suited to those needs. Obviously no one who's vlogging is using an Alexa. I don't get what's "ironic" about a vlogger having similar needs to a documentarian (or "film documentary producer," whatever that means). Vlogging is an offshoot of documentary, and a popular one. But I agree almost any camera would be suitable from an image quality-perspective for both. The focus on image over content is a little silly here sometimes, and it seems you'd agree.
  10. Agree to disagree. Performance under mixed light is what attracted many of my coworkers to the Alexa, so it clearly matters to them. My friend who had his work graded by Stefan Sonnenfeld at CO3 bore witness to a long rant about chroma clipping on the low end cameras that were used on the project so it matters to colorists, too. I agree that for most users, almost anything will work for vlogging. But most users aren't going out of their way to ask about it. Most weddings and corporate (or even festival films) would be fine on a 7D and an iLav, but users here are more likely to skew higher end. I don't see vlogging as any more or less noble than weddings and corporate or short films. But I do think it requires a bit more in the way of automation and a small form factor and less in the way of technical polish.
  11. It doesn't have to be. Watch Casey Neistat's channel.
  12. Couldn't disagree more strongly. Features like image stabilization, miniaturized ergonomics, low light ability, autofocus, and ability to shoot under mixed and poor light are far more important to a blogger than to someone on a controlled set where they have a gaffer, AC, etc. to take care of it. I haven't used the M5, but the feature set sounds good for vloggers.
  13. Canon love/hate aside, consider the workflow and ergonomics, as they are worse for the C500 than any other Canon cinema camera. You need an external recorder to shoot 4k, the ergonomics are really odd for a high end cinema camera, there's some weird color stuff with RAW and external recorders (that has led to some bad experiences with color and unreliable recording on high end projects I can't mention and for which the internal video or video from competing cameras compared favorably in surprising respects as a result). And the C300 Mk II is better in every way for a lot cheaper although its color is Arri-like rather than Canon-like. (And not as good as Arri.) But then look at what Cronenweth did (and consider what it says about the much-vaunted Red that Social Network and Gone Girl look worse if anything at such a high budget) and consider you're getting a C300 plus occasional 4k for clients who need it... and it seems like a good deal. Then again, a low end commercial rate is about $4k/day plus camera rental, so the camera will pay itself off in a day if you need 4k. It is the least "Canon" of the Canon cinema cameras, though, for better or worse. Nice color but be careful about chroma clipping when using an external recorder and make sure it's reliable, be careful about rigging, and realize that the C300 Mk II offers so much more for less than twice the price. The C500 is insanely sharp per pixel btw. I wouldn't touch an FS5 with a ten foot A7S. The FS7 is nice, however, though I think it looks more like video. Low light is awesome. Canon does analogue gain like Panasonic does with the Varicam, but Panasonic only offers 800 and 5000; Canon offers 850-20,000 ISO with analogue gain so the low light and DR and internal NDs and whatever are a dream. So low light should be about as good as the new Varicam, which is to say far better than the Alexa and Epic.
  14. Policar

    Light Meter

    Yeah I agree. I don't necessarily use a meter to expose the shot anymore because I expose digitally different from film (with more concern for highlights than shadows and sometimes just ETTR with the Alexa) whereas with film I just exposed 1/3 stop over most of the time. But for everything else–ratios, vfx, going between set ups multiple times, etc. I always use a meter. Also, most DPs I see without meters are secretly getting shade from their gaffers. I know some people can read ratios by knowing which IRE values correlate with what on which camera at a certain gamma setting and looking at the face in the scopes, but that doesn't work for me because there's so much variation there between cameras and gamma settings nor am I able to pre light a set without a meter whereas with one there's absolutely no need for a camera and you and your crew can get way ahead. If you can expose and light purely by eye as some can (Slocombe etc.) then you don't need a meter at all, of course, but your gaffer should still carry one. Unless you're that good, someone on set should really really really have a meter. Even if you're not lighting, it'll dictate how you use fill and negative fill. Just because you don't need it to expose, doesn't mean you don't need it to light. If you don't know how to light and always leave it to your gaffer, that's fine, but make sure your gaffer has a meter then. I agree that a cheap meter (spectra is the industry standard and a lot cheaper than a dual meter) is all you need, though some DPs will use incident meters and others will use spot meters. Elswit and a lot of AFI guys use spot meters primarily. I use an incident meter for everything except green screens (which should always read at 18% gray and which are almost always overexposed or sloppily lit these days) and occasionally looking to see if highlights will blow out.
  15. Policar

    Light Meter

    I use a 758 cine. Whether you use a spot or incident meter you should use something, but different cameramen prefer different things. So I bought a dual meter as I'm pretty inexperienced. I couldn't imagine shooting without it, but if you find you don't need it, there's no reason to have one.
  16. I thought don't breathe looked really bad, surprisingly for a company 3 grade. But film grain is a bit "affected."
  17. We use a Flanders Scientific monitor when I work on footage for broadcast. I think there are about $2k+. I use an HP Dream Color calibrated for rec709 at home. It's about $500 and just okay. Some issues, but calibration and gamut are nice. Where I work that does web content (high end web content, too), they use the regular display (I think on an iMac) and calibrate it using a probe, because most of their work is for web. There will always be major differences. For tv you can hire out a colorist, buy a generation 9 Panasonic panel and have it calibrated (these are still used on major network tv series), buy a cheaper/smaller Flanders panel, etc. For web just get a good screen and calibrate it and you should be fine. Also avoid grading too aggressively, rarely a good idea, it looks coked out.
  18. Just curious, and not that I don't believe you, but what evidence do you have of this? The only thing that doesn't add up is that 1824X1026 would only resolve about 500 lines of resolution without aliasing according to nyquist, or 700-800 lines or so with efficient debayering algorithms that do allow for some aliasing (which are typical to most cameras). The 1000+ figure assumes a level of efficiency that's totally unprecedented. And that the 5D Mark III is producing a 1920X1080 RAW according to magic lantern, and its (very soft, let's be honest) image is no softer per-pixel than a JPEG straight from the camera. Upscaling would suffer from even softer per-pixel performance. Also, why make a sensor that's 3X HD only to bin to 3/5 of that? Just curious where you're getting the 5x5 figure because it doesn't correlate with real-world performance in any way. The above figures (the post before yours) make a lot more sense to me. So does the mix of line skipping and pixel binning in different dimensions for the earlier generations correlate with ISO performance being about 1.5 stops worse in video than in JPEG before the 5D III. The result is still super soft and not exactly great... but we all know that. I haven't done much research into this and it sounds like you have, but the figures posted before what you wrote correlate closely with camera performance and what you've quoted doesn't. Of course no one knows for sure except Canon engineers, but I suppose maybe someone at Magic Lantern has a pretty good idea, and I was curious if you were quoting someone from Magic Lantern or if you'd figured this out on your own and how.
  19. I'd never read that term until Red used it to justify their (at the time) bad colors. (Which are much nicer now.) Since it's totally subjective anyway just buy what looks good to you. If you look at a still JPEG from a Canon dSLR and compare it to a RAW image debayered in ACR the RAW file looks WAY sharper. Same difference you'd see when comparing h264 to RAW video. So basically, yeah. Is the fine detail picture style available in video mode? I think it was introduced on the last few cameras. That has a better debarring algorithm for JPEGs at least. Still fairly limited DR, but whatever.
  20. That sounds about right. The Alexa also seems under-represented on this site, you'd think everyone here would embrace it. The 6k MF alexa gets love, understandably. I haven't used it but I get why you guys here who have used it would love it, my budget is just limited to the original Alexa usually, and since I don't shoot for theatrical releases about 2k, 2k is enough for me. So unlike the majority here, I like shooting on the original Alexa, despite the crop and low res 2k and 1080p (which are softer than the C300 let alone the A6300, the sharpest of all). I like other Canons cameras for having the best looking color and image even in terms of chroma clipping, flawed in the A7 series and unfixable in the grade. But my eyes are going soft so maybe I don't appreciate resolution above all else as younger more talented eyes do and I'm used to Super 35 (like the 5d 4, which is 1.64 crop aka S35, not 1.74 crop, as is erroneously quoted for no reason) because it's what I grew up on and thus I forgive those faults like a crop to the ideal standard format and a resolution that's only on par with the highest end format of distribution currently available for its other merits like the best lenses available and best color available and while we can still use the 18-35mm sigma and 14-20mm Tokina we are missing out on like five overpriced underperforming Canon-brand EF-S lenses so sure this is a tragedy. To be honest, I'd pick the 1DC for stills. But for stills I'd still buy it the 5D and if I had it for stills I'd still use it for video. But I get that S35, the cinema standard for 100 years, and 4k–more resolution than 35mm film has ever had–isn't enough for most here. The Revenant wasn't good looking enough. Resolution and image capture size good enough for an Oscar win in photography and picture... no... too soft. Which is why I treasure this forum. Forget content. Forget art. Let's bicker over meaningless specs and inconsequential price changes (if you're charging less than four figures a day wet hire to shoot, what are you even doing–and four figures a day pays gear off fast).
  21. Attach it to a huge, heavy (but balanced) rig.
  22. What's interesting is that they've introduced the "fine detail" picture style, which uses (or seems to use) a better debayering algorithm than their standard one. If you look at 5D Mark 3 JPEGs at 100% the per-pixel detail is poor... same as the video detail at 100%. That's one reason both perform much better in raw, but it's less noticeable in stills because by that point you're pushing the lens's resolving limits. Anyhow, there are definitely reasons to go with another brand, particularly if you need 4k in a small camera. But I think Canon's thinking is if you're a working professional you can easily afford a $10k-$30k investment in your camera system.
  23. Could be line skipping. It seems likely there will be a crop factor since Nikon is on board with it, too. But I think it's premature to freak out. 6720/4096 is 1.64. If it's per-pixel then it's a 1.64 crop, not far from super35 at all. 1.7 for quad HD maybe. I don't get why the number 1.9 keeps coming up, though.
  • Create New...