maxotics

I'm confused by the question. The 720p is as good RAW as you'd get from a 5D3, it's just 720p, and you need to shoot crop mode to eliminate aliasing/moire. However, it isn't as sharp as 1080p, of any kind (unless that's scaled down to 720p). I'm not sure I'd use it in any situation where you really needed to depend on it, even as a crash-cam, because of the shutter bug. The camera just freezes up sometimes on ML (unless they fixed it; don't think they ever did). You can work around it, doesn't happen often, but again, an issue. If you can get one for $120 I'd get it because they can do so much. Probably the smallest camera you can get with mic in. As Mercer says, the camera with the Fujian CCTV lens ($30) is magical. The c-mount adapter is $10.

Well, if you try to program, or get someone to do it on the focus pixel remover, the problem is simple, you have a grid of 1280x720 pixels. You want to go to a list of locations and replace them with a pixel to the right say. So if you have a red focus pixel at 3,7, you want to get the next same-color pixel at 3,9 (3,8 would be green) It would be really EASY IF programming languages, or the computer, weren't built around an 8-bit (byte) mindset. So everything is in chunks of bytes, which in binary are like this 11111111 (which would be 256) The killer is that RAW video is stored as a frame of 14-bit values, like 11111111111111 (or 16,383). RAW video is NOT byte-based (like just about everything else in computers). The RAW file from ML is just consecutive 1280x720x14, or 12,902,400 bits per frame. In computers there are all kinds of "objects" where you can read/write bytes of data. There are NO easy objects for dealing with 14-bit chunks. so you have to read/write 14-bits on a bit level. (sometimes they are in reverse order) that is, you can't just load a file and say "Load 14-bit Chunks from this file into an array", you can EASILY do that with bytes, double bytes, etc. In short, binary math ain't for the Programming For Dummies crowd If you want to take your gluttony to a whole new level I have given you a fighting chance, I hope

You're killing me I almost bought another EOS-M yesterday! Then I remembered "almost every step along the way". However, it would be easier now because ML RAW Viewer is one piece of killer software. http://www.magiclantern.fm/forum/index.php?topic=9560.0 If only ML worked on the EOS-M 2 and/or 3, then there would be a future. Anyway, the guy who did ML Raw Viewer might put in the code to deal with the focus pixels IF there was enough demand. Or maybe we could find someone to port the code. (My serious programming days are over, just too old, when I try to flip the bozo-bit it ends up flipping me!). Between PDR and my thing, a good dev should be able to do it in a day or two. Again, need the demand. Mercer, you're probably all lone Anybody else out there?

It's a pain, and was a weakness of the software. If you can figure out which settings correspond to the mode you're shooting in, you'll be set. I'd try to get crop mode working first. Just shoot very small clips of a white wall and try out difference cameras/settings in PDR until the dots disappear.

Not sure what you're referring to, but if it's the difference between RAW (version 1) and MLV, then I would shoot RAW because the potential benefits of MLV, metadata, audio sync and compression, aren't going to give you much on the EOS-M. However, I haven't used ML in a while, so maybe there is something with MLV that gives you a benefit. Does MLV now compress RAW on the EOS-M? I would find it hard to believe the camera electronics would have any energy for it. If it doesn't, then yet, just shoot RAW.

PINK DOT REMOVER: If you have a Mac you'd have to find the old Java pink dot remover. Don't know where the last one is. This might be it: https://github.com/Foorgol/PinkDotRemover It seems the EOS-M is a bit of a dead end since ML doesn't work on Digic 6's yet (EOS-M 3) and may never do. #1 MISCONCEPTION ON EOSHD: Some people here don't shoot people in story-telling formats, so often gloss over one of the primary benefits of RAW, which the EOS-M has, even in low resolution--high dynamic range and color depth. My experience is, even in a perfectly lit scene, if you shoot any H.264 of someone in a consumer camera, and then shoot that same scene in RAW, the RAW will have a more natural look (even though, in the end, you end up with 8-bit output like the H.264). Why this is I don't know. Perhaps someone can say. Anyway, the point is, it wasn't the resolution that got me away from the EOS-M. Even at 720, it looks beautiful to me--better, again, in a certain color feel than any 4K camera from Panasonic (I currently have the LX100). The problem was dealing with the pink dots, and worse, the "shutter bug". If I'm a filmmaker wanting to do documentaries, or I plan to shoot a lot, I'd go with Sony or Panasonic--each have their strengths. If I wanted to do a evocative piece with depressed people in the woods I could shoot RAW with the EOS-M. Anyway, shooting RAW on the Canon cameras is a huge pain, but NO ONE who has ever done it has complained about the results RAW, on any camera, gives a look that can only begin to be immitated in H.264 on professional cameras like Canon's C line. What about S-LOG, V-LOG. I'll probably get flamed. 8-bit is 8-bit. The log curves are useful for some lighting situation where you want to favor exposure (in highlights usually). It does NOT make 8-bit 14bit! What you gain in the highlights you would LOSE in the shadows. That can hurt skin tones and other things. End of rant If I only had $300 for one camera it would be the EOS-M. Remains, to me, the undisputed best utility still/video/RAW camera of all time! The next would be the Canon 50D (no pink dots and near 1080 RAW).

The chroma smoothing defeats the purpose of shooting RAW. Here's why. When you shoot RAW, you are getting a 14bit value for every 1280x720 pixel. Each of those pixels is either red, green or blue. When you "smooth" the red, green or blue pixels you're essentially truncating the 14bit values down to who knows what. The pink dots are removed at the expense of color depth. Since you don't know which pixel is a focus pixel, you have to degrade them all. The strategy behind the pink dot remover and the software I wrote, is to figure out the exact locations of every focus pixel. this can be down because they have a pattern that radiates out from the center of the sensor. Anyway, if the pixel is a red one, you go out and get the red value at the closes red pixel and use that value. It works very well and is very similar to the whole de-bayering process where you fill in missing color information. Unfortunately, which pixels are focus pixels changes with the crop mode you use. In crop mode they are closer together. when is full-sensor mode, so to speak, they are more spread out. If you have a Mac, find a copy of the Java pink dot remover and use that. You can use it on PC too. But if you have a PC, and shoot in crop mode, my software is very fast.

Yes, like TheRenaissanceMan says, you can do 1280 native focal length, but get moire. However, if you have shallow depth of field and are not shooting anything with hard edges you'll get that rich color look. The biggest problem are the focus pixels. You need to remove them or they create "pink dots". I wrote software for windows that does it well (but only in crop mode) https://bitbucket.org/maxotics/focuspixelfixer/downloads For other resolutions, you want the PinkDotRemover. The best combo, IMHO, is a 10-20mm type lens, then shoot crop mode, which will make it 40 to 80. Also, you'll need a 95MBS read/write card (even though it will max out at 40mbs). Here's a cool video I did using the EOS-M to show you what your sensor really sees in Bayer mode: https://vimeo.com/79857693 And here's one that show the basic quality, it's VERY filmmlike to me. Here's one with the kit-lens https://vimeo.com/76181035 And this one with a Sigma 10-20. Finally, here's a good one in H.264 with the EOS-M. I LOVED that camera. https://vimeo.com/75122636 I now have a BMPCC. But I LEARNED so much using the 50D (another great ML camera) and the EOS-M.

Since you can run Magic Lantern RAW on the 5D3 you get phenomenal video quality any day you care to, and as Gregormannschaft said, any upgrade to video quality will mean an upgrade in PC hardware (in fact, I am in the process of upgrading my PC equipment as I write). What you really don't have is something small with slow mo. So get the RX100 IV and wait. You may want an URSA mini in the future with EF mount (so you can keep using your Canon glass). My 2 cents!

Nice Job! I was reading a portrait photographer's book, back from film days, http://blog.kitfphoto.com/Zeltsman/chapter-12.html and his method was to get a basic fill-light going and then key from there, knowing he has a minimally exposed negative. I've been leaning that way myself. Anyway, those lights are PERFECT for that. You can either bounce them off cheap foam-core or put some tracing paper in front. Another design to look at is where you put the lights in a silver box and then curve a white surface above. The old way they used to make studio soft lights. Can't find an image easily on the Internet at moment. Either way, use some single light LEDs as Key and hairlights and you're good to go! And a tip for my U.S. brethren, after the November elections there is tons of foam-core thrown out by candidates, etc. The backsides are usually white and usable. So hit those dumpsters!

"Are they just a brand, whose intent is to leverage as much margin as possible out of gullible rich customers or are they a serious imaging firm for pros?" Leica has a way of looking at photography, a philosophy, that is about the best optics and fastest/smallest/most precise deign where focus, aperture and shutter speed are quickly accessible for pure "manual" photography. Yes, mostly rich people can afford that philosophy, but it's unfair to say gullible because most photographers respect Leica's philosophy and what the camera can do. Is their Monochrome $9K because it's a luxury, or because the market is small? Have they ever been a serious imaging firm for professionals? Even when I was a kid, Leica was beyond many professional photographers. It makes sense for Panasonic to work with them. However, most Panasonic and Olympus glass is MFT. Lack of lenses still bedevils the e-mount (the Batis 85mm is still not shipping in quantity). Not sure Panasonic can come out with other lenses or confuse its current users. What's interesting about this camera is that is shows how confused/uncertain Leica is about where the high end market is moving. Although I'm the first to argue that full-frame gives a DOF look you can't get in MFT, I'd also argue that if Panasonic can deliver a GH5 with RAW it would be a camera to be reckoned with. All these log formats are fool's gold in my opinion. You can't make 8-bit anything but 8-bit. Guess what I'm saying is I doubt Panasonic will ever come out with a full-frame camera under the Panasonic brand. I think what we're looking at is that camera under the Leica name where it would stay.

That's the way I thought 30 years ago. After years and years or watching movies and TV and reading books I've flipped the other way. Just as there is a technical side to depth of field, or resolution say, there is a technical side to story-telling, how much is focused on one character, what is shown with dialogue. For example, a filmmaker could be sold on the low dynamic range of a RAW camera. But if he used it on a comedy it would probably fall flat. Comedy is always sharp and bright. Or it could be the other way around, a filmmaker looking to tell the story of a Syrian refugee for example, and shoot with bright camera that can't get a thoughtful mood. I know many people will read what I'm saying and think 'what about this exception, or that exception'. It's not about rules. It's about what works, and has been working for centuries, though the technology and cultural styles may change. Creativity, to me, isn't this "personal" thing that is unique to an individual, it is personal to that person in his/her PLACE IN TIME. An artist serves their generation. They seldom come up with anything new. Of course, when you're young you feel omnipotent, powerful, creative, immortal. I can only say from my experience that young artists who try too many things that haven't worked in the past end up with the same result. So it is important that one understand the basic technical aspects of story telling before they choose camera A or B. There's a big difference between shooting people in a comedy, a drama, a documentary, or random scenes of buildings and fountains. The technology of the camera should fit the technique of the story.

Glad to hear you're getting back to your core genius! There continues to be a huge need out there for people who want to do cool things with consumer cameras. I posted a video on how to run Magic Lantern on an EOS-M two years ago and it has 10,000 views. It got 43 last week. A quick NEX7 video I did has 15,000 views. I bought a Canon 50D a few months ago, and run ML on it again, the same camera that brought me to EOSHD in the first place. For the money, it is STILL a very powerful creative tool, which I'm sure you'll agree. Lately, I've been playing around with the LX100. The 4K photo mode is very interesting, not a lot of work has been done on that, or on other cameras. Davinici Resolve, I've given up on it because it's just too complicated. Gimbals are another complex gadget that will need guidance as they get cheaper and better. Or the Panasonic G6. I finally bought one and Andy Lee is right, a perfect camera in its way, timeless. And yes, the BMPCC is a very creative camera but many people don't understand why. In short, the original reason for EOSHD is as relevant as ever! I've learned a lot here at EOSHD and I look forward to the old Andrew

As you mentioned, lack of competition put that camera firmly in the rich-person dilettante market. Difficult for even a working photographer to spend $3,000 for a one-trick pony. But oh what a trick! Also, the Ricoh GR has amazing stills quality, APS-C sized sensor (pitiful video) and the only retractable prime lens I know about. I take that camera almost everywhere. If fits in almost every pocket and is near silent. The RX1, though it's certainly tiny compared to a D800!, is not much smaller than an A7 with a prime. Unless you need that absolutely smallest camera, then you're better off in the A7 line IMHO. Again, if one lusts after full frame IQ, you can get pretty close with the Ricoh GR. Here's a shot that shows one of the great benefits of the GR. I was able to set it to TAv. I knew I needed a fast enough shutter speed to catch her, and wanted f4 of DOF. The camera set the ISO from that. The image is a bit noisy, but a whole lot less than my cell phone or MFT camera would have been.

The New York Times didn't mention the Orthodox Jewish connection either. That's a 3-pointer for you ahsanford

It seemed VERY wise to not include the word because the first reaction of many was to boycott B&H. When readers learned the company is owned by Orthadox Jews they rightfully wondered why Al Jazeera of all organizations would not mention it. Right or wrong, is this not realistic to expect? I'm pretty sure any other journalistic enterprise would have mentioned it because, again, B&H considers themselves Orthadox Jews first and merchants second. If they didn't, they'd be open Saturday. Right? As for the sweatshop, the article does not compare B&H to any other companies. There are no supporting statistics that these issues are unique to B&H. It's all about context, as others have said. Again, I have no problem with Al Jazeera. I do have a problem with weak journalism.

Unfortunately, most settlements are not instigated by the victims, but are brought on by lawyers. I'm not saying that B&H didn't abuse workers, only that all companies, when they reach a certain size, are targeted by lawyers who make the real money. It's the American way. I have a fair amount of experience in Wall Street settlements and what the public sees as some large number, is actually quite small, a cost of doing business. In other words, even if B&H were no better or worse than any other company, it would still end up paying $5.7 million to settle SOMETHING. Generally, various law firms attack companies and join together to get something. Not pretty! I'd like to say the government is un-biased, but it too, is run by politicians looking to score points with voters (who never look at the details). We all accept that this happens, I believe, because we understand the best we can do is keep a fire under everyone's butt. If B&H was really bad they would get fined out of business. If one doesn't buy what I'm saying then I ask, have you never gotten a ticket (for your car) that really had little to do with safety and a lot to do with shaking you down? Ahhh life!

Shallow or controllable DOF IS quality to me and I bet every cinematographer who ever pointed a camera at anyone One of the interesting aspects of photography is many inventions are created by lone wolfs. So Go Sanveer!!!!! However, you might take some classes in optics, or read some books, before you start taking your phone apart, not that I follow my own advice here Why don't you get a Raspberry PI and the camera module? That's how I'd try out your ideas. https://www.raspberrypi.org/products/camera-module/

Anyone who has spent even 30 seconds in B&H picks up that it's run by Orthodox Jews. To mention that nowhere in the story, whether the New York Times, or Aljazeera makes one question the reporter's competence, if not their eye-sight. Because of the work-ethic of Orthodox Jews, (like closing on Saturday) I can't see how it is not relevant. Also, to add to my previous post. There are probably hundreds of electronic stores in NYC alone. B&H is not successful because of cheap prices only. It's a very well-run store.

B&H and Adorama, and many other stores, are run by Orthodox Jews. It sounds like the warehouse workers have grim work, but so do most Jewish workers in these stores. They are bused from their communities in Brooklyn and live very spartan lives, which is why these stores succeed (lower prices). You read about "tribes" in the Middle East but don't need to travel there to see one in action. Go to B&H. Then go to Williamsburg, Crown Heights or Borough Park. You won't see a lot of money flowing about. Yet they gather round and protect their own. I find these communities creepy. Then I remind myself Park Avenue is just as creepy in its own way too. There are no easy answers here, like boycotting a store. The truth is, wherever you live, if you weren't born into money getting work is very difficult--especially if you're an immigrant. In that story the only job many of those workers could get was at B&H. Again, believe it or not, B&H doesn't treat its own Jewish workers much better. The difference is that the Jewish employee believes in something more than money. Or maybe they're too scared of raising prices (to pay workers more) and losing their customers to Mormons.

Sanveer, not sure you're listening to the answers given by others above Sensor size issues are similar to the saying in automobiles, "there is no replacement for displacement". Certainly a turbo-charged 4 cylinder car match a large engine at high speeds, but not at low ones. Similarly, in good light and wide DOF cell phone cameras can hold their own. In LOW light, or when shallow DOF is needed, nothing works as well as a full-frame sensor. A lot of it is physics. It is NOT because the companies aren't trying, or are purposely crippling the cameras (for example, why would Apple care if it hurts Nikon, Canon, Sony, etc?). Also, zoom lenses HAVE trade-offs. Look at a RAW file BEFORE IN CAMERA ADJUSTMENTS and you will see this. Even prime lenses have their issues with DxO makes software to deal with. As you point out, most lenses don't resolve their sensors. Think about that! You're wondering why cell phones aren't as good as DSLR but recognize that DSLRs still aren't as good as they should be! I agree!

To follow on Cantsin's points. All zoom cameras make lens distortion computations. They use computations to adjust the mirrors in deep-space telescopes to remove atmospheric blur. My gut feeling is 80% of most photographers don't understand how a bayer sensor works and why it has inherent color problems that reduce true "color" resolution to 33% of the sensor's stated resolution. I don't know how you guys feel, but all my experiments in photography/video surprise me in how poorly camera equipment works if you take the hype literally. Even when I have great light and expensive equipment I'm surprised at how limited the image data is; that is, how much is lost in the shadows, or whites. As for high ISO noise, sure, full frames like the SONY A7S are amazing compared to other cameras BUT, again, don't live up to what I want, or expect, if digital photography was perfected. In short, digital photography is far, far from perfect. For example, little exposure mistakes make a big difference. Really, if you step back, you wouldn't think so. I can understand non-photographers thinking a camera with 16 lenses is stupid (like a car with 16 wheels), but people on this forum? A big mistake that light.co is making, IMHO option, is going for the mass market. What they need to do is get first adopters on board, like Andrew or people on this forum. They're trying to skip a step. Lytro did the same thing. I hope this forum thread, and the one on DPreview, motivates Light.co to focus on one benefit of their technology and effectively educate photographers about what it is, and how it can help them. I have a BMPCC. I hardly use it (though I'd love to). It only really does one thing but that one thing is something no other camera can do at that size and price. That's why I bought it. I believe that's why Blackmagic is successful (yes, I know they're not perfect). Why would I want the "Light cameras"? After two videos and reading their site, I still have no idea what exactly the camera does that others don't.

It's a great idea, but they botched the explanation of it. There are two problems they're solving. NOISE: By using multiple sensors they are working with a higher sample of readings. A full-frame DSLR uses larger sensels to sample more light. Keep in mind bayer sensors have inherently color problems because each pixel only samples one channel. More pixels gives a cleaner image. FOCUS: By using multiple shots at different depths of field, and advanced stitching algorithms, they're able to composite an image that shows a shallow depth of field; however, these shots will not be as good as a DSLR because only the optics can perfectly get the focus/blur at each length. However, combined with the lower noise, they will be much better than the algos on single sensor cell phones. My guess is the first market for these cameras are event photographers. Discreet cameras, lower noise and subject differentiation.

Right on the money, Oliver! Superb! Way to keep at it until you have it just right.

There's a thing called brightness, luminance, dynamic range, what have you. That bring perception to about 16 million "colors", still within 8 bits. The sensor outputs 14 bits of information. It's a long, complicated answer to why you would want 14 bits. The best way to understand is by doing. Take a high dynamic range shot with your DSLR in both RAW and JPG. Then open in photoshop, whatever, and work to improve the highlights or shadows. You will soon see what can be done with a RAW, 14-bit image, vs an 8 bit image. As for the s-logs. They are for specialized situations one should want to avoid in the first place. There's no free lunch in data-bit land.