Axel

Yes, I consider it a great advantage to have shot on expensive film, when every second sucks on your wallet. It's offline now, but there was a great german short 'the photographer', in which the said photographer seemingly takes many shots of his model from different angles and in various situations. Then finally he says, let's go, loads a film and shoots ONE photo. As opposed to the MO of the modern digital wedding photographer, who makes 1000 photos to get 20 good ones.   BTW: This short is pretty self-reflective. It has a story that makes a point. Stanley Kubrick once said, every good shot was a recording of a preconceived image. Know what to look for first, then push the button.

      I think he is referring to the "WB shift" Andrew Reid recommends in his GH2 book. It counteracts the green cast you find especially on skin tones. With CC in post, you can tune this further, but all Canon EOS cams have beautiful colors out of the box.

Then I thank you very much. This is great news for me. So the answer is G6.

Then log & transfer, don't use 5D2RGB or the like. You can already reject a lot of clips then.

  I trust your verdict.   What about the OLED-viewfinder though? That kept me from buying the GH3 then. I definitely need to judge focus through the viewfinder, I even adapted a Zacuto eyepiece (an oval ring, hard work) to shove onto the hot shoe:   Background is, I carry around the camera wherever I go, it really is my run&gun companion. I am not for swivel screens or bulky Hoodmans.

Still the (where is the love? Wasn't that a Lenny K. song?) old GH2 with a hack and some ol' glass. G6, from what I've seen so far, is more videoish. In general: Especially if you're starting out, don't search for the cinematic look! Embrace digital. Make it better with innovative lighting (doesn't need to cost you a cent), well scouted and dressed locations, good acting (which always equals good directing) asf. Find a good story strong enough to shake the world. Tell something with all your deepest and honest emotion. The rest will find you and never leave you.

I guess no steadicam. Camera held on a monopod gives enough stability. And the take isn't perfectly steady.

  Moderate, yes?   You can, depending on your NLE, edit natively and just export a ProResHQ master. You can do without render files. Transcoding all clips to ProRes in advance and using it as render codec will produce substantially more uselessly big ProRes files. This workflow is only needed if your computer isn't fast enough to deal with native mpeg4. Qualitywise there is no difference. Often proved.

You are great at discussing details nobody else thought worth mentioning. I enjoy your articles. I doubt that this is true though. The sensor is small, and you get relatively deep focus. Yet it still is not a thumbnail camcorder-chip, and some of the shots I've seen so far seem to be slightly out of focus. As we all know, peaking as focus assistant gets rather useless when you have a big DoF.

  Not necessarily. I admit, there are limits, and that's why nowadays raw is on everyone's lips. But that's not because a codec 'breaks apart', it's because you have a lower range of values. It is important to understand the difference, especially since you are asking for an intelligent compromise.   I always color correct my footage, and I have been doing this in 2002 already, when I knew nothin about 'grading'. The best shot can nonetheless gain from a subtle polishing, always.    In the blog of famous DOP Shane Hurlbut, he admits to be an old school type of photographer, who tries to get everything right in-camera. What does it mean? He thoroughly lights the set, he carefully measures exposure and he sets the white balance manually to fit to the mood he tries to achieve.   The last step introduces a problem. A white balance that is not set to white will shift all values in the image and leave a cast, similar to that of a color filter (though of course a color filter brightens it's own color and darkens the opposite). With secondary color correction, one can do much more subtle changes, and, for example, protect the skin tones, which in most cases will result in a better image. You can apply a correction to every color separately, stack secondaries upon each other - but only if you start with a perfectly balanced, evenly exposed image in the first place (both the goals of primary cc).   And if you do these extreme changes, will the 8-bit, heavily compressed stuff fall apart?   This depends on these things:   1. The accuracy of the application you grade in. It has to be 32-bit floating point computing, which means there will be no banding or other artifacts caused by your new mapping of tones.   2. The codec and it's implementation. HDV (or XDCAM EX) often can't even be brightened without reaching it's limits. Mpeg4 of the EOSs or the GHs usually works well (the latter particularly with higher bitrates).   3. If your changes stay within the range you recorded. This is the important point. This is what you need to be aware of when shooting. It's also true for less compressed video like 10-bit ProRes or raw, only those are more forgiving. This is what you can do in-camera.    Is post a pita? I don't think so. It's creative fun. But that's a matter of personal preferences.

Also 8 weddings, spread over 5 years (so hardly a 'profession'). Depending on the scenes (Family anecdotes, bachelor party, preparation, office, church, courtyard posing of the newlyweds, big party), between 60 minutes and a little over 3 hours. Depending also on the degree of planning/scripting.

I made the test quickly, so that it looked okay. Got a '16'. Must repeat it, it rankles me. I don't think it has to do with the monitor. If it had, there was the excuse: Mine is just too old and can't be calibrated properly. Probably just two or three squares in wrong order ...       I think it's crucial to know one's handicap. Does a person with no perfect color vision ever become a good colorist? The eye is unreliable in more than one way. Perhaps it's better to stay aware of that. Colorists rely on scopes primarily, anyway. Listen to Steve Hullfish: http://vimeo.com/55139561

Probably indeed a problem with the hack.       So it looks the camera was perhaps put off after a longer clip at a high data rate, and the index was not finished.   After a long recording, make a very short recording (just 1 or 2 seconds).   Probably QT will not open the file. I once had such a clip (with FCP 7 then). Premiere also didn't recognize the clip. But Toast did. Open Toast in the media conversion window (or so, I have a german version). Drag & drop your clip into it, export as QT, ProRes (gear icon lets you set export codecs). My problem was, I could only export the audio as an extra clip (make it Aiff) and had to synchronize both in the NLE.    If this doesn't work, either upgrade to 10.8 (AVCHD is played back by QT without conversion, the upgrade is safe) - but I bet it won't work either, because still OSX doesn't like .mts without or with wrong metadata.   Or take your card to a windows computer, where you export both clips with highest quality settings of H.264.

Here are the facts: > To be physically able to distinguish between 2k and 4k, the image has to be really big. Your point of view must form a equilateral triangle with the left and right borders of the screen (what John Galt calls: Sit before row six!). > This is so close, that the image gets noticably distorted, and few will remain there if there are any free seats behind this. > If you do it right and let yourself be almost surrounded by this giant image, you won't see a better quality than the folks from the middle. Unfortunately, it's the other way around: Had it only been a 2k image, you had now the chance to see the resolution's limit: pixels. > The equation reads "higher resolution = bigger image". > Quality is no factor in this. > Resolution @ 2k or 4k is also no longer about "detail" or loss of it. On the contrary, it's about being able to render more organic, smoother surfaces. > A sunset that suddenly fills your whole field of vision looks dramatically different than the same sunset on your smartphone. I already said that the size of the bigger image should have higher resolution, but that alone won't do, because there are two conditions yet to be fulfilled. > Color resolution. The rule, that we can skip 1:1 chroma sampling, because our own eyes only read 4:2:0, only is valid for the fovea, when we see something small that we can stare at. A 4k image (from close) will resemble a black&white film that was manually colored. It will look "dead". > Color depth. Gradients that have perfect pixel resolution can't spread only a few HSL values. You might expect banding artifacts, and they may indeed show up, but not always. What you will always get is an artificial looking, flat image without vivid colors, ugly. About upscaling and downscaling: > In both cases, every pixel of the image is an interpolation. > It's a myth that a downscaled image from 4k to 1080 contains more detail, it contains more interpolation-artifacts. > There IS sophisticated downscaling ... > ,,, as there is ever better upscaling. > 4k needs to be 10-bit 4:2:2. At least.

A short film from Germany, shot with the Pocket. They speak english, because Germany isn't exactly the home of cinematographic talent, and the latter has to be invited from elsewhere. However, it's graded rather extremely, but it has a story and not just landscape or cats. One of the marks of cinema. http://www.youtube.com/watch?v=tZ_INEKUQpA#t=22

  Save the $33, it's all built-in in FCP X. But it requires originals. Not orphaned, .mts- or .mp4 - clips you stumbled across on some forgotten drive from the attic.   As Michael Cioni puts it, FCP X is a media assets manager with an attached timeline, wheras all old-fashioned NLEs are timelines with an attached media-path-list. Use this advantage, don't import dumb video clips.   See what makes FCP X special (see all or start at 13'42"): http://vimeo.com/73797466

    Don't be irritated by hypes. I still soldier on with my reliable GH2. Never before in my life had I made so many videos.      That's it. The more better graded raw videos will flood the net, the more will all those 4k-downscalers (I call them 'Pro-Dads' for the amateur dad's software Mercalli) realize how "so last century" and how desperately videoish their Son-Pan-Can-4k pixel-puddles look     Good question. The downside of ProRes is that it's not particularly state of the art in terms of efficiency, it's like carrying around a kettle of soup, though a bouillon cube would do and some hot water at home. Compression is not the devil, and Panasonic did always well with highly compressed codecs. Just think about a clever way to avoid the high data rates and big files that come with raw.    Also, Panasonic always tried to be compatible. Probably they provide a solution for every big NLE. I wouldn't worry about that.

Sorry, I didn't mean to confuse you. I had some Nikon glasses from my F3 and at first adapted them to my 7D. The fastest, the 35mm and the 50mm, both f1.4, were not perfectly sharp with the 7D (they were on the F3). I forgot this, when I again used them on the GH2, with the said Novoflex. It seems there is no point in buying full frame lenses for MFT faster than f2.0., I heard this about other lenses too. But since, due to the crop factor, a 50mm is like a 100mm on a GH2, 2.0 is more than enough. Has nothing to do with the Novoflex.

The lens' aperture ring clicks, the bigger blue one of the Novoflex doesn't. It is a short way from close to open though (and wide open doesn't work anyway, I also read this about other adapted lenses), and I doubt I could fine-tune the exposure during recording, pan away from the window for example. So if there was no such ring, this was no dealbreaker for me. If i had different kinds of mounts to adapt, I'd rather buy cheaper ones and let them on their lenses.

I have a Nikon to MFT Novoflex. It makes lens and adapter one unit, the aperture ring is big and smooth. A friend of mine has some enjoyyourcamera adapter, very cheap. Okay, unless you compare them directly.

Yeah, that's the idea. If there was such a thing. Reality has everybody sitting in front of a worst-case-monitor, but every single one for different reasons. Some are too dark, some too bright, some are blue, some are green, some are yellow. I am trying to get my monitors as neutral as possible, without spending too much money ('cause I am a bloody amateur). I have two Syncmasters, calibrated regularly with Spyder Pro. Unfornutaley, they are now two years old, and the values are already out of range. I had and have access to digital projection, and I saw quite a lot of my stuff on the big screen. In former times, the experience was badly disappointing. There were sooo many mistakes I only saw in this size. I was so embarrassed, I wanted to signal my co-worker to close the douser, immediately. Then I decided to do something about it. A video graded on a neutral monitor will look best on every average shitty consumers' monitor. A DVD of a Hollywood film, upscaled with the excellent hardware of the projection room, will look way better than any fullHD video, graded on an average shitty monitor.   The false assumption of the thousands of MagicBullet users is that a colorist applies a stylish look and little else. He makes the image shine, he optimizes and balances before and after he applies any look. He couldn't do with a monitor that, for instance, has a slight blue cast, because he would counterbalance this with yellow, and the result would show a yellow cast on most other displays. These are the principles. Besides, BM designs a cinema workflow. Wouldn't it be nice, if you didn't have to second-guess too much? Are we all happy with 8-bit @ 2 mbps for Vimeo?     Again, I can't afford the real stuff. I have a Beyerdynamic DT 770 Pro, recommended by a friend. I hear things I never heard with other systems. Test this: Audionamix extracted a real 'modern' clear and powerful 5.1 sound out of a 50 year old optical mono track from Psycho (Bluray). In the Extras-section about the restoration the audio engineer said, they could hear the birds inside the studio chirp (because the old studios were like barns). As a demo, they ran a silent part, where the detective enters the house. I could hear the birds only with my headphones. The part with the birds is missing in the shortened youtube-version here: http://www.youtube.com/watch?v=d4KZqK6YYjE   Same thing: You can change something for the better only, if you know what's there in the first place. If what you see or hear is heavily distorted by your hardware, you'll make things worse. Do I need a 5000 $ external monitor? Prices rise exponentially, equipment that gives you one percent more accuracy will cost you twenty times as much. I think the unvaluable virtue is to stay aware of the limitations, to know how easily the eye is deceived. There are cheap methods to get from, say, 85% reliability to 98%, painting the walls in a neutral grey, calibrating your monitors, use balanced 5600K backlight, always judge by the scopes too. Then you can insert your own clip as DCP with 12-bit 2k cinema trailers and be relieved to see they look as you expect them to look.

  And more. As you know, Resolve now exports JPEG2000, the standard codec for cinema packages. They decided to be consequent and added this. Not that there were no free alternatives already, but Fraunhofer are like the masters of codecs, sounds terrific.   @Brellivids Sure about the AOC? Are you making anamorphotic films?

  I read it will be a free upgrade for Resolve 9 owners. I also read the Resolve Lite beta is not restricted to 1080. I also read the Dell monitor above (and later models of the series) is no real 10-bit despite the billion colors. Sigh.

  I can't recommend it, but I am as you evaluating my options. The Dell Ultrasharp U2711 with 2,5k resolution also features 10.7 billion colors instead of measly 16,7 millions. It's 10-bit (EDIT: It isn't, see below). Though we are living in an 8-bit environment predominantly, I am confident that this is changig fast. And how about actually seeing the colors you manipulate in 12-bit, at least for 10-bit? The bigger resolution helps insofar, as Resolve has a one-monitor GUI.The monitor gets sold used for ~300 € at ebay, sometimes 400 € with a dealer's warranty. I'd like to hear opinions on that.   Note, that Resolve 10 Beta is available for Windows as download (The Mac download page says "0kb" still). With Resolve 10, I guess our lives will become easier. It states to handle variable speed changes with optical flow and compound clips. The latter two FCP X terms, which lets me hope the download will work soon.

When you are just starting, lesson 1, let raw be lesson 101, second volume. You can start with just about every camera. If I lost my whole equipment in a fire and just had so much money, I'd be torn between those choices:   1. GH2, used (latest biddings in € on ebay), 500 €, with SLR magic 12mm and SLR 25mm, cheap tripod, cheap but well chosen shoulder rig, Tascam audio recorder or the like. 2. BlackMagic Pocket, new, 880 €, SLR magic 12mm (wait for other aproved lenses), cheap tripod, cheap but well chosen shoulder rig, Tascam audio recorder or the like.   and, if I was into photography in earnest: 3. 5D M2, used, with kit lens 24-105, ~1800 €, external EVF via HDMI, able to zoom in the 5D's output, peaking, that means Zacuto or this price class, ~800 €, Sachtler ACE, 550 €, relatively cheap, but VERY well chosen shoulder rig (let's say, 300 €). Tascam or the like.   Why not the ACE in any case?    Because, I wrote this before, the main duty of a tripod is to stand motionless. Many tripods under 100 € will do with light cameras. Only if you wish to make very good, smooth pans, the fluid head needs to be of high quality. And still it's a craft only mastered by few, most pans have either the wrong speed or are unnessesary und uncinematic.   But why the ACE for the 5D then? Because it is relatively small and lightweight, yet built well enough to carry the 5D. You get heavy weight, low duty junk for ~150 €, but it's not worth the shipping costs. And besides this, the head allows for good pans.