Axel

Dear Matt, if you like to become a true fanboy, here are the guides (from Urban Dictionary):

  Indeed, it could be much easier. With FCP X. Just about every workflow you can think of is possible. And if your system is too old or too weak, you can transcode in the background (either as proxies or ProRes/lite/HQ/4444), you can then relink to original media for top quality renders (one click). Or you stay native during editing and export a high quality ProRes master.    And there can be an easier way than dragging clips off the memory card: If the clips are no longer inside folders (>Private>asf), if they are not represented by a useless clip-icon, but by a thumbnail showing a frame from the clip, if you don't have to double-click to see the video, if you don't have to click at all, because the thumbnail is a mini timeline that you can "scratch through" (EDIT1:scratching is the wrong term, because it suggests dragging an old-fashioned playhead, EDIT2: It's wrong anyway, I mean "scrubbing") with your cursor (which becomes a skimmer spontaneously when over the tumbnail). If you can overview the content of all your clips without one click by diagonally skimming over all rows of clips (if this is to work directly from various cards, a Thunderbolt reader would come in handy, and it eats a lot of RAM, otherwise you can copy the card to an assigned harddrive, one likes to have a backup or two anyway), even before importing anything. Also before import, you can view and edit (multiple edits in one clip) to your needs. The differences to the old NLEs are in every detail, and this is just the first stage.   It's dangerous to write something like that, because though it is true, it is not accepted, and the threads tend to turn into 'warzones', as someone has put it. One can edit with every NLE, the results count. Full stop. Fanboys or haters out!   Now it looks there is a new party of fanboys setting up their manifest: The raws. They look down on ordinary AVCHD users. All of a sudden they were elevated to higher beings.   Let's nip this in the bud, right?

  Short film 'Choices'. The hopeful junior filmmaker imagines a battle of the top low budget camera toys, filmed in stop-motion-like animation, using AAEs puppet tool. Dialogs from Rocky? Boxring represented by Andrews table? The Pocket beats the GH3 and the 7D to pieces. The 5D says: 'We won, it's over.' The Pocket answers: 'It ain't over til it's over' - and kicks the 5D over the edge. It falls with the Wilhelms scream, hits the ground and falls apart. The filmmaker takes the Pocket and smiles, the Pocket smiles back. They leave, iris fade-out, THE END. Must find a better ending. 

  In theory, yes. A sensor that size with only 2 megapixels should have a favourable signal-to-noise-ratio.

  Yes, but.   You saw the Sofofly clip. You have to admit that it's more than just 'okay video'. The camera compresses allegedly up to 11 stops of internal raw into the 8-bit file it finally writes to the card, and that's not bad at all. If you grade a raw image for 8-bit, you do the same, only that you have more flexibility.    But I agree. I am looking forward to receive my Pocket and start with raw 'videography'.   In my view, it's a good thing that the camera is so user-unfriendly (didn't find a better term in english, help me out). It is not for wedding filmers and that kind of video. I am prepared to treat it like a big fat A-camera, with awareness of all the factors that contribute to an image that is as well technically a.g.a.i.g. and aesthetically stunning and that tells something meaningful about it's motif. I find myself being extremely bored by all enchanting beauty shots on the net that just show off. And  I start looking at photography as an art with a capital A, That's where a recording differs from an image.    Not that this approach was prevented by, say, a GH3. 

  I know a place where such tests are performed with almost every new camera.    But test charts don't show aesthetic aspects, which to be honest bother me more than scientific ones. Had the naked facts been of real relevance ever, the classic Mark 2 could never have become famous, because it didn't perform well resolutionwise. This was published from the start.   Isn't it funny that we got a bunch of affordable cameras that all have amazing image quality? They are more different in their requirements for individual usability. My non-scientific ranking:   1. The BMPCC for reducing all advanced 'features' to one: Think cinema with raw, for under 900 €. 1. The 5D MIII raw, still for the aesthetically most pleasing images. 1. The GH3 for ergonomics. And clearly it looks as if it had the best resolution of them all.

Absolutely, and I hate wisenheimers as well, must be my shadow.   The poor threadstarter has given up long ago.   That was my first experience with greenscreen: I read a book about Final Cut 2 in 2001 and chroma keying was mentioned. It looked easy. I had a bright green folder on my desk which I taped on the door. I put my VX1000 (DV PAL) on the tripod. I painted red color over my wrist, focussed on my hand and made a strangling gesture. Then I filmed myself, mimicking being strangled. Of course I had 'the hand of horror' in mind. Compositing was a matter of trial and error. I had to deinterlace before I scaled the hand, I had to get rid of the arm, I had to add the cut area with the bone (Photoshop) and animate it. It was far from perfect, but it was big fun. It was diffuse daylight, and though my hand was close to the folder, I only had a few frames with spill. I simply cut them out, that looked even more creepy (funny, actually).   I recommend a playful, respectless start. If you haven't done it and only read about all things that let it go wrong, you'll be discouraged.

  I reckon you are pissed a bit. Accept my excuse and let me repeat where we agree: Recording in higher resolution is good. Let's set aside if this is because of anti-aliasing through intelligent interpolation by the software you scale with or because the average of 4:2:0 equals true 4:4:4 when downsampled. Life is too short.

I can't see the video (due to copyright-laws in my country), but your comment makes no sense. Andy surely didn't perform the keying in 480p. Your explanation remains wrong. If a 1080 clip has visible artifacts, they get worse when you downsample the video. Noise is more prominent, jagged edges are magnified, every other artifacts become more visible. Also true for the new 'issue' with the BMPCC: You see almost nothing @1080, but once you downscale the image, the grids pop out. 

@ D.L.Watson Nice clip. Good pans. Which tripod do you use?

Czesc Marek, yes, that's what I understood as well.   @Andrew Very interesting. For staying in 8-bit and avoiding the need to handle raw, the GH3 (and of course the 5D) are still in the game.   Dapper still life. You seem to like a victorian club atmosphere that looks as if lit by gaslight, somewhat inappropriate for technical stuff. I wonder if you smoke a pipe and play a fiddle at night ;-)   We also have some posessions in common: The 2001 paperback, the Bolex Paillard with spring mechanism and a rusty film can ;-)

  You are absolutely right. And yet, in the seperate review video, you see - or rather hear - other problems with a camera that pricy. First of all, hear the echoes from within the thin plastic body when he touches anything. Acceptable for a 500 € G6, but not for 1200 €. You can tell by this clip alone that you can forget shooting handheld, even with the aid of a pistol grip. Crackles will resonate and be read by the internal mic as rumbles, and most probably also by any mic on the hotshoe that has no good shockmount. Then hear the noise of the zoom servo, it's like ED209 (by the way, a zoom lever/dial, but obviously no ease-in-ease-out like in 90% of the 15-year-old camcorders, fuzzy logic for zoom transitions was actually introduced in the 70's)! Hear the manual zoom hissing 'shshshh', plastic friction like that of a lomo toy lens.   In the hands of a more experienced operator, the video will be quite okay, but judging by the official clips it also says video, not only in 60p.

  You are not an idiot, I guess my english isn't very good. With 'unless you work in a set like this, you can do with 4:2:0' I was trying to say that it is not good for 4:2:0, because of the overall spill. The sentence may have been grammatically wrong.   If there are roughly 4 pixels on the edges where green and foreground mix in 4:2:0 without any spill, you can imagine how many there will be with a lot of spill ...   As human being, with eyesight limited to 4:2:0, you probably see purple dots if you close your eyes in this studio. Personally, I have never been on such a set. But I can't believe that color keying will suffice to get a clean matte for the Peter Jackson team. There will still be some trainees who spend hours and hours on rotoscoping additionally. And you can detect another trick in all LOTR movies: The actors are always backlit. Read jghardings post here.

  As I wrote, scaling down is about anti-aliasing the edges through interpolation of pixels. What you stated before, that 4:2:0, by merging pixels, get's miraculously 4:4:4, is wrong.      You will never reverse the 4:2:0 to 4:4:4.   However, unless you are working in a set like this:   ... you can do with 4:2:0. Because a green screen will have only green pixels in 4:2:0. It's the outlines that cause problems. The Keylight method to blur the edges and mix them with purple is no longer state of the art, as mentioned above.

How is that? A scaling without without visible aliasing (without just skipping pixels) means interpolation, averaging values. How can this increase the accuracy of the individual color information?

  Though recording vertically is a good advice, your explanation is faulty. By condensing the color information you get worse color accuracy. This is all about spacial resolution.

  Not if it made your $600 50D into a fully valid 10-bit 422 camera. Which it doesn't. If it did, the news hadn't been chinese whispers. People just don't grasp what ML and BM did for us ordinary Joes.    I guess now that people start and develop a conscience for much better quality, it will be hard to sell old fashioned camcorders, no matter how many megapixels. I hope so. Bit depth is not considered redundant data anymore. But perhaps we are only a minority, too small to ignite a change of perspective. The dogs bark and the caravan moves on.

Hodgetts says exactly what I think about 4k. As far as Atomos is concerned, they talk about the Samurai, HD-SDI!

      If the Ninja exists for years and sells well, how can it be useless? Demand determines supply. Marketing creates demand.

  You record an 8-bit 4:2:0 signal "as" ProRes or DNxHD 10-bit 4:2:2. You have an intraframe codec for editing. You can as well take the AVCHD and transcode it with 5D2RGB.   Once your hardware and software can deal with AVCHD, the output quality can theoretically be even better (ProRes is a copy, nothing can top original data). In reality, there is no visible difference. Don't believe it? Test for yourself.    Atomos never cleared the confusion, they saw all the DSLR aficionados as lucky chance. So they fanned the hype further. It was Atomos at Photokina who announced which of the new VDSLRs had clean HDMI out. Some had no full resolution, some had artifacts, some produced psf like the GH2, but of no damned importance.   RichST wrote:     How do they conjure True HD by this? Not by combining pixels alone. It's the lenses' resolution that only sees one 'superpixel', because that's it's limit. A very sharp photographic lens, calculated for 30 megapixels and above is actually worse. How do they conjure 4k by this (C500)? We haven't reached HonestHD yet, what do we want with false 4k? 

  First paragraph: It's not a 4k camera, nor is it likely to output video beyond 1080p ever. HDMI outputs 10-bit indeed, but it's 8-bit color depth and 2-bit of control signals (or so).    Second paragraph: There is planned obsolescence going on, and if Sony really tried hard to make better cameras, they probably could. On the other hand, I don't believe it's easy to build a perfect camera for a budget. It's either that the big companies are not in touch with the VDSLR users, that they don't understand what we are asking for (easiest thing in the world, I was told by an engineer, to shit raw signals, spares you a lot of work for the processor and the codec implementation). Or it is that we are too few. That for a majority of Sonys consumer customers this camera is just the right alternative to an NEX, because they even hate to be bothered with lenses. 

I have done greenscreen compositing with my DVX100, with my XH A1 and nothing since, just 'tests'. Both must be considered inferior by any means compared with modern cameras - and modern software.   Back then it was Ultimatte and/or Primatte, plugged in After Effects. Today I would highly recommend to try the very easy Keying-Filter within FCP X, should you happen to have access to it. To achieve visually perfect results, you have to reveal the 'advanced' options, but once you allow the new background's colors to counteract spill (tool: 'Light Wrap') and 4:2:0 edges (instead of complementary tint and much edge feathering), you will love it.   Being in no way an expert for chroma keying, I still think the spill issue doesn't get much better with 4:4:4. Reduce spill by seperately lighting the greenscreen (as recommended above or in every greenscreen tutorial), that's more than half the battle.    exotics wrote:     Arguable. The larger the green background, the more spill, almost unavoidable. A very rich broadcaster in my country built a green hell of a news studio, and it looks terrible. I followed the advice of Maschwitz in his DV Rebel: Make the green background as small as possible, make it a green towel, light it as little as possible (but as evenly as possible, outside, with an overcast sky, you will get the least spill). Let it just barely cover your foreground object, cut away everything else with garbage masks. Even allow an occasional hand or foot to stick out, a few frames of roto won't hurt you, but the key will be better.   Of course, this is not an option for a 'broadcaster' or someone who needs to isolate an ensemble of dancers.   Testing greenscreen techniques is cheap and easy. Multi-talented cats move and have whiskers. Let your cat play with a laser pointer in front of a green background. Then comp it into your slightly tilt-shifted kitchen and let yourself be hunted. The incredible shrinking EOSHD user.

  Great fan of making ofs, fan of the DV Rebel's Guide, great fan of the traditional make-believe approach that the great masters of the silent movie era began, great fan of meticulous storyboards and so forth, I hope this old clip might inspire you: https://www.youtube.com/watch?v=WRS9cpOMYv0

  Of course, if you have the third DSLR or EVIL or so, you take all the industry's buzzwords with a grain of salt - and spit them out. And I'd tell traditional, pre-DSLR videographers that they should compare the price as well. Had this been a 600 € camera for hobbyists ... Had the quality of the official demos really convinced us ...   I sometimes stop for a moment and see my life as that of a compulsive collector of electronic waste, with very little creative output, a pathetic wannabe. Though there is truth in this, there also are devices that make me go out and actually do something. The one most precious to me is my small, smart, versatile and quality-wise still impressing GH2 (have the G6 since recently, waiting for BMPCC). Does it have moire? Very rarely. Does it have good resolution? You bet! Does it cost 1200 €? A used one with a kit lens 14-40 and an SLR magic 12mm or 25mm. That's why this Sony is nothing to write home about.

  This will avoid moire and other artifacts and probably will increase the overall resolution when filming test charts, but there is a serious error in reasoning nonetheless. If the glass is calculated for a 5k sensor, it's not perfect for any other resolution. More so, if it is a giant zoom lens. In the recent years, there were lengthy discussions on various lenses for DSLR videography and their effect on the final image, and the choices for video were mostly not the same as for stills. This approach takes the (1) inadequate sensor and combines the (2) cameras' strategies to downsample the resolution with an (3) optimized glass, found by trial and error.    The whole point of preferring VDSLRs over conventional camcorders or even FS700 or C500 (Shane Hurlbut: "videoish look") is the images' analog look-and-feel (not to be confused with an artificial 'cinematic look'!!!). As I see it, this look is achieved best by having interchangeable lenses whose characteristic bokeh predominates digital dodging.   Having less moire becomes the sole feature of the new Sony, and that doesn't suffice. What does 'intelligent downsampling from a higher baseline' mean in english? It means smudging pixels. The official clips are not exactly inspiring, and frankly, I think I can't stand another romantic landscape with muzak, intended clearly to leave me trembling in awe.       It's not as if we hadn't seen any ultra high definition / 8-bit / tiny sensor clips from consumer cams so far. The JVC something combines four 1080 signals to a threethousand something pixel image. Saw the official clips. Saw a live-signal at a trade show. Razor sharp image, no striking artifacts, but beyond doubt the ugliest mess of a pixel pulp ever. If everything is sharp in a giant image, it looks completely two-dimensional. Suitable for filming your fish tank.   When will people abandon the idea that higher resolution (alone) will improve *quality*?   EDIT: There is a new camcorder from Sony, coming in a few weeks for ~ 6000 €, the Z100. It does real 4k @ 600mbps, has 10-bit 422, 50p and 60p: http://www.youtube.com/watch?v=hrHoniHNhko   Certainly this is the *quality* I asked for. But it has a diminutive sensor. Comments?