M Joel W

If the short is very effects-heavy, a dSLR may not be the best choice. The skew can make motion matching virtually impossible and extraordinarily time-consuming. Ask the director why he has chosen that camera, how it fits into the workflow he envisions, what specific aesthetic concerns were behind his camera choice (if any), and even if you do not agree with him then you will know what you'll have to compensate for if outfitting a dSLR. Maybe he wants deep focus and really long lenses? HDV would be nice for that. Or if it simply comes to giving up some of your fee to rent a better camera...ask yourself why you've chosen the project and what matters most to you.

At what ISOs and focal lengths were you shooting? The consensus among web tests seems to be that the t2i and 7D have the same image quality, but I too have noticed that the 7D seems to have better image quality and maybe a stop less noise, even with the same lenses. I'm scared to repeat that belief since it goes against the conventional wisdom. I've used them on the same set but never shot tests side-by-side. Also, what picture style were you using on each? Imo, neutral with highlight protection is the best for video, far better than standard (too much sharpening, resulting in halos and aliasing) or the custom ones online (which are generally too flat and result in bad tonality and weird saturation issues in the highlights, shadows, and skin tones while actually delivering no additional latitude). Oddly, I have found the kit lens to be rather good optically, but only toward the wide end. It's less impressive zoomed in, but at 18mm it's unexpectedly good.

Douglas' advice is more-or-less spot on. I would only add that the t2i (and presumably t3i and 60d) have significantly better LCDs than the 7d and pulling focus is a lot easier with them if still imprecise. The ratio concerns are real. You get eight meaningful stops, maybe, of dynamic range, and if you go over the clipping looks a bit digital (but still not that bad), if you let things go too far under and push them the noise is chunky, and if you light flat there's not much information bit-wise and it gets veiled in noise and compression artifacts. So you need to light to the medium. The dSLRs are nonetheless much more forgiving in this regard than most digital video except maybe the really high end stuff. They have approaching two stops more highlight detail than the hvx200, imo, but certainly way less than film. Andy, remember that the difference between a black object and white one in completely even light can still be as much as four or maybe even five stops through a spot meter. So if you're lighting that at a two-stop ratio you're already getting close to filling that 8-9 stop range. I mean, is the red really giving you 14 stops of dynamic range and a dSLR two stops? No. In terms of latitude, that's a factor of camera dynamic range and scene dynamic range, so there's no hard fast number, but a stop or so of latitude is certainly all the dSLRs have, I would agree. Not much room to fix exposure in post. My figures aren't too much more scientific than waving around a camera and light meter, sometimes together, sometimes independently of one another, but I know enough about ratios to be relatively confident in them, ballpark-wise. That said, if you just stick to using a two-stop ratio you may get way better results than approaching things how I do, though that wouldn't say much given my recent work. Heh.

If you're shooting a short for festival play or an indie feature or something, consider that while 16mm used to be seen as a cheaper alternative to 35mm, now that red and the 7d have monopolized the "cheaper than 35mm" market, 16mm is seen as something of an aesthetic choice and not a budgetary one. So in terms of promoting whatever it is you're shooting, you've got an advantage right there shooting film. It's going to look different and more expensive than video. It may also get more respect on set. The dSLRs have more like 8-9 stops of dynamic range, much better than most prosumer video (similar to the ex-1 and a bit less than the red, though, maybe?), and the resolution is poor but not dreadful (about 720p but good micro-contrast). Skew is really, really bad and if you want a handheld camera or a lot of camera moves with longer lenses just consider that a deal breaker. I went for it with lots of camera moves on a t2i and it's not so good. Aliasing is bad but generally manageable. The footage is very difficult to grade nicely. Skin tones and foliage (memory colors) are substantially better than with the red, worse than with film, but the lack of flexibility in post sort of offsets the nice colors out of the camera. In terms of aesthetics, of course you won't get any unbiased opinions here, but I think dSLRs produce nice video. A lot of very nice major ad campaigns are shot with them so if you're getting terrible results it's likely a matter of approaching the medium wrong. Yes, Rob's footage looks soft, low contrast, and sometimes out of focus, but it seems to be shot almost entirely wide open. And 35mm glass with be hazy, soft, low contrast, and more likely to go out of focus when shot wide open. I do, however, think the dSLR look, like the red look, will become increasingly associated with cheap content and that will hurt you in the long run. But really it depends on what you're shooting. For a short with no budget, a dSLR buys a lot of production value, especially when you consider how the light sensitivity and superior highlight handling (relative to other cheap video only) lets you skimp on lighting a bit compared with other inexpensive video. You can probably find a friend or local studio that has one so just try it out. Don't judge it by the first video you shoot. Video is free and instant so test the hell out of it. Start with neutral mode and with highlight priority on (what I use, ha) then try other settings or whatever. Shoot under lots of different light. See what works and what doesn't.

More glass is one factor, but not to the extent that more glass means more light loss (with modern coatings you lose very little light and contrast even with complex designs); it's mostly a factor of more glass costing more money and resulting in an enormous lens. The front element has to be as large as the maximum aperture and so a 240mm f1.4 lens would need a 171mm-wide front element, which is HUGE and very expensive and would likely result in tons of flare. Image circle is another factor. I shoot some large format and lens designs list coverage in degrees (so the longer a lens the more coverage) and the word is that the widest lenses of a given design are also the sharpest over their (more limited) image circle. So it gets really hard designing fast, wide, sharp lenses for bigger sensors. I have a 135mm f5.6 plasmat that weighs half a pound (and is disgustingly amazing); its 240mm f5.6 brother weighs more than four times as much and seems even bigger than that. Equally important is the fact that you need some depth of field to shoot. Imagine pulling focus at ultra-telephoto t1.4 on super35. No one could do it! I assume it would be possible to make a t1.4 zoom for super35, but it would be incredibly expensive and probably have rather poor image quality (flare, spherical aberration, etc.) and be almost unusable in terms of size. You'll notice it's the same with stills. For 35mm still cameras you usually get primes in the f1.4 range; for 6x7 and 6x6 cameras f2.8 is about as fast as it gets; for 4x5 and 8x10 f5.6 is fast (f2.8 exists and by most accounts is amazing), for ULF Schneider makes f11-f22 lenses. The limitation seems to be related to the size of the front element and depth of field. Despite normal aperture values of f16-f64, focusing in 4x5 is an art and a science. This is mostly second hand and maybe full of fuzzy science. I have a book on lens design but am scared to read it. It's full of real science. I'm just repeating what I've been told and have found from experience.

You've probably made up your mind judging by the date on which you posted, but I would think twice before building one of those home depot things. All they are is a counterweight with no gimbal and you might as well just tie a weight to the bottom of your tripod and operate from the center of gravity for even less money. The result will be equally poor. I bought a glidecam 2000 about six years ago (for my long gone dvx) and have used it with the dvx, 7d, t2i, hv30, etc. Also used the glidecam 4000 with the vest, which might be more appropriate for an xl2. These are usable to the extent that they'll actually help stabilize your shots a bit way when used correctly (they have both counterweight and gimbal), but they are a pain to balance and the slightest error balancing will result in wonky footage. Unless you build up a lot of intertia and get perfect balance the look is still more smooth handheld than steadicam, but they are good for the money. None of these cheap options will compare with a real steadicam rig and experienced operator. Not even close. And they'll all tire you out really fast with a camera as large as the xl2. That said, I do find the glidecam useful and it's inexpensive. It's just not great. Raimi was doing a trick shot at a low frame rate. Pretty effective, and the guy is a genius with the camera, but that technique is not really versatile.

Nikon's old lenses are really great so I think you will be pleased. If you go that route keep the kit lens. It's surprisingly good and has IS. f3.5 is close enough to f2.8 that you can get away with it even if your scene is lit to f2.8 and 18mm is a focal length I find I use a lot. I can intercut between it and nikon lenses and the difference in color rendering/contrast is so trivial. Also recognize that you can't really do on-camera zooms effectively without a constant aperture zoom...so if that's important to you go for the zoom. I used to recommend buying from keh.com and getting BGN lenses, but their prices have risen so shop around. Even eBay or craigslist can be okay and some local camera stores have good selections; old nikkors are durable and plentiful. If you shoot stills I think you will find the t2i to be a very nice camera for that, too. Liveview makes manual focus primes usable for stills--and the screen on the t2i is great.

I've actually found my bad lenses perform better on the t2i than on older, lower res cameras (in still mode), and that in video mode a really sharp lens results in aliasing more than it does a great image. Buying compact primes or the like seems like a waste to me for such a technically poor camera as the t2i, except to the extent they will be arguably more "future proof" compared with worse lenses. I also disagree with Hal to the extent that most of the automatic features are worthless in the t2i's video mode: autofocus is mostly useless; no one uses autoexposure, really; and beyond that it's simply a matter of rotating a physical aperture ring versus manipulating aperture via a dial. For stills the difference between lenses with adapters and EF-mount lenses is tremendous (and you will want a good adapter if you go that route; the cheap ones get stuck on lenses and don't have infinity focus calibrated correctly and that's a killer if you're using wide angle primes with floating elements or want to pull focus to infinity adroitly), but for video, using an adapter is fine. Otherwise, I agree entirely. The Canon zoom is a great choice if you can afford it. But I would add an 85mm lens for CUs and some inserts. I also like the Tamron 17-50mm IS for half the price, but it has poor bokeh and a frustratingly small focus throw (I believe so too does the Canon). IS is absolutely crucial for handheld work, especially for 50mm and above, when skew gets horrible, so don't get a zoom or telephoto lens without IS if you can help it, as it will limit what you can shoot a bit. For $1000, a Tamron 17-50mm IS and 85mm f1.8 Canon (be careful with skew on that one; it lacks IS) makes a strong kit. In this case the kit lens is redundant. The Canon 17-55mm IS zoom is superior if you can afford it. For cheaper, the kit lens (mostly useful as an 18mm f3.5 IS prime), a 28mm f2.8 nikkor, a 50mm f1.8 nikkor, and an 85mm f2.8 nikkor (plus adapter) would make a good kit for much less money. I actually prefer mf primes for their long focus throw but for a documentary-like situation I'd rather have the fast zoom. Check keh.com or ebay for used mf nikkor primes. Also invest in ND filters, an ND grad, a polarizer, and whatever else you like. The Cokin P system is nice. What matters MOST is covering a good range of focal lengths with acceptable (f2.8 or faster, imo) speed. I use 17mm-85mm for 90% of shots, maybe, and occasionally up to 200mm, maybe. A nature videographer in Africa would go much tighter. Terry Gilliam would go much wider. Above all else, get the focal lengths you know you'll use.

In theory it could be helpful, but in movie mode the camera hunts for focus really, really slowly and the noise of the autofocus motor is loud.

I own a t2i and have shot quite a bit on the 7d. If you're planning on renting out a kit and freelancing it might be worth getting the 7d simply for purposes of advertising/appearing more "professional." The t2i is about half the size and does not get taken very seriously due to its price and form factor. This can be an advantage if you're using it as a crash cam or with a tiny steadicam type rig. But if you have to cheap out and buy a t2i that does not say much about your level of professionalism and investment in your craft. The conventional wisdom seems to be that the image is exactly the same. I'm not sure, but there's certainly not a big difference. Image quality is so close it shouldn't play in to your decision. The 7d has a few real advantages: 1/3 stop ISO settings (which, surprisingly, isn't really that big a deal since you can set most lenses within the 1/3 of a stop, though it's very nice for maintaining depth of field within a scene and not having to relight as much) and vastly more flexible white balance settings. The t2i has no good settings between 3200K and 5600K except "white fluorescent," which is okay with cool white fluorescent lights and unacceptably red with incandescent lights. So if you like shooting at different white balance settings between daylight and tungsten or baking in slightly different color, this is a pretty big deal. The t2i has tons of daylight settings, though, for whatever reason. The 7d can also go to like 2700K or so, which is more neutral with most incandescent lights; 3200K is a bit warm. The 7d also supports external monitoring at useful resolutions and the interface is trivially better. The build quality is of course vastly superior, which may matter if you plan to shoot in the rain or extreme heat. The t2i has a few advantages of its own: autofocus during video capture (which is 99.9% useless) and a significantly sharper, larger, and brighter screen. I can't judge focus worth anything on the 7d's LCD without zooming in, and that's not available during capture, and that makes pulling focus really hard. The t2i's screen is still much lower res than its capture (720X480 vs somewhere around 720p), but I can spot aliasing and focus errors much more easily with it. This is a moot point if you set up external monitoring correctly with the 7d. As always, it boils down to preference.

This is a huge long shot, but was the footage by any chance shot with HMIs on magnetic ballasts? Shooting with a 180º shutter on such lights will produce almost exactly the phenomenon you've mentioned.

It might be the power. And if you're using thin gauge wires, particularly long ones, you can expect some big drops in color temperature. I've never heard of tungsten bulbs changing color temperature over time, though. I've also noticed tungsten lights seem very warm relative to, say 3200K kinoflos, so it may just be that.

LACC looks pretty great, I can't believe I hadn't heard about it until now. Don't think it will be my MFA destination since they don't seem to offer one and what I'm looking for personally is pretty specific, but the moral of your message resonates. Is it open admission? Can you take classes out of order? I see they have view camera classes (which are hard to find, everyone's going digital) and I eventually want to take some of those, maybe a few years down the line.

Thanks all. Going to get every last one of those books and keep all this in mind as I visit different schools.

Double post! Delete.

There's a fat chunk of irony in your post and it's this: Michael Bay's favorite directors are the Coens. So I'm sure he'd be glad to hear you enjoyed his movies (and you're right to the extent that he's a great director and story issues in both aside, Transformers 2 was brilliant next to GI Joe), but I don't think he'd be glad to hear you trash the Coens. True Grit wasn't their best work, but it was a great piece of filmmaking from our era's best American directors. Michael Bay comes from directing commercials and he's VERY good at a few things. He can deliver fast exposition better than anyone else, sometimes elegantly (though the image), sometimes inelegantly (through wild lines or what have you). He can set up the stakes of a scene very fast and very coherently. He also has an AMAZING eye for action and a great eye period. His style of direction (hyperkinetic frontal primarily objective action/comedy?) is unique and inimitable, however it's been very influential. I think Bay is to some very small extent the Spielberg of our jaded, degenerate generation. I mean that as a huge compliment. Still, you're making a serious mistake thinking you're the only audience member. True Grit did great business and a lot of people, me included, were really entertained by it. It's also a pretty unique film for the Coens (a Spielberg story with a Coens twist) and it was shot quite well. There are metrics, as objective as you're going to get in art analysis, by which Transformers 2 is exceptional. There are others by which True Grit is. Beyond that it's a matter of subjective opinion, or some would argue it's all opinion with occasional big words to dress it up.

I haven't read most of those books. I'll pick them up. Most of the books I've read on cinematography have either been hopelessly basic or hopelessly specific. I get that it's hard to find the happy medium but like "here's a car interior, how do you we light it for these moods;" "here's the corner of a white-walled room with the camera pointing straight at it and nowhere to rig stuff; how do we deal with it;" "here's ext. forest night; how do we light it for these moods;" etc. with simple diagrams would be amazing. I haven't found much like that. ASC articles are wonderful but I can't apply those lessons to such small videos as my own. I have also found books on directing to either be too theoretical or too specific, written as though by movie fans and not working directors. As for schools, well, the application deadline for next fall has passed and I think those are the only two schools I may get into, having sent in applications to a few places. They were also among my top choices going into all this. I don't want to spend as much money as NYU costs. Don't have it. Can't justify it. Not going to spend years building up my reel until I can get in. Here is my directing and lighting work. You can see I still have a ways to go and am all over the place in terms of aptitude, from ghastly beginner at directing to having some skills lighting.

I want to end up in LA. I believe FSU and Austin are the only schools at which I might be admitted. I don't have a strong enough background to get into NYU/USC, etc. and cannot afford that kind of debt, either. What should I do? Both schools claim they move the vast majority of their graduates to LA, so maybe I can still work it out? Also, what textbooks did you read while at NYU/AFI? I've read most of the common ones and found few helpful.

I applied as a DP and will probably end up working as one. But if I'm not mistaken, getting proper coverage and maintaining a consistent eye line are responsibilities of both directors and DPs. Certainly blocking is the director's job but the DP needs to understand what function the blocking serves. It's been tough for me not being able to understand how coverage will edit together, as that informs everything from screen direction to the order in which you shoot. So whether or not I ever work as a director, I certainly want to learn the skills needed for the job. I wouldn't want to direct unless I knew how to DP and vice versa.

I've been lighting for a few years and my progress has slowed to virtually nothing. Same material, same approach, similar results. I also want to learn to block/stage/cover a scene. Am I barking up the wrong tree? I know a job isn't guaranteed, but I think my limited abilities are what's keeping me from finding more work and are certainly what's making my work less fulfilling for me. I get what you're saying about the issue with alums in other cities, though.

Not an option for me personally since I can't afford the better schools in LA, nor could I get into them, chances are. If I do get into Austin and FSU, would you recommend not attending based on location? If neither school will land me a job in film I've still got time to look elsewhere. UCLA is too competitive (and expensive) and I can't think of any other really strong state schools in LA. This may just be my lack of knowledge on the subject. As for AFI, I sent my reel (which is pretty mediocre) to faculty and they said it was worth applying (before I decided I didn't want a 100% cinematography program), so while you do need a reel and some experience form what I understand, I don't think it needs to be amazing.

Definitely true; I'm going for an MFA next fall and, as money is a big concern to me, likely won't be attending an LA program. But the truth is, if you know you want to be a DP and you know you want the best education, AFI is it. Personally, I'm deciding between FSU (if I get in) and UT Austin to learn both cinematography and directing. Very different programs, but both are state schools and more affordable than their glossier brethren. If you're not tied to living in LA and money is an issue for you, both might be worth looking into. I've also heard good things about SCAD and Full Sail, but, not having the money to afford them, didn't even consider them as options.

AFI. Elswit ALSO studied at USC and AFI if I'm not mistaken. For cinematography, AFI is the best school there is. Bar none.

I was accepted for the fall for an MFA and am visiting in a month. I'll let you know my impressions then. I'm not sure what to expect, but I've heard good things about the school and the culture in Austin. Austin requires you do some documentary work and take classes outside your concentration, not sure that's what you want or not. I don't think the program is tops for cinematography specifically (since, well, AFI pretty unequivocally is), but I've got some directing and theory-related tendencies so I don't want a hardcore cinematography conservatory as much as you might and the program seems like a good balance between hands-on stuff and theory, at least for me. AFI is generally for people who've already studied or worked in film for quite some time, too, and want to hone one very specific part of their craft. I don't know your level of experience; if you're already very experienced with film and have the money AFI may be worth checking out. I very highly recommend checking out FSU's film school. They're very intense, relatively affordable, and have no theory or documentary focus. Very genre-centric, relative to other places. They seem to be the fastest track to a comprehensive education in directing or lighting, and they pay for film and have great gear and the faculty seems strong and no-BS. Tons of structure and conservatory style. Honestly it's all what you make of it, so visit and apply all over.

You're falling back on the same pseudo-technical talk, though. Yes, sampling theory says you need n*2 pixels to accurately sample n lines (or n/2 line pairs), but in reality no one is sampling things that perfectly and camera manufacturers, even the most conservative ones, allow for a little moire, because the lens will knock the resolution down to the point where aliasing is not a significant issue more often than not. The system needs to sample at n*2 pixels to get n "mathematically perfect" pixels, but the sensor itself can sample more aggressively and the lens will knock it down to something acceptable. In stills mode, the 7D pushes 3100 lines per image width with aliasing (2500 without) according to published tests using a sharp lens. In theory it should only resolve about 5184 * .7 bayer efficiency /2 (nyquist sampling) or 1800 lines. But in practice it somehow does better, largely because a pixel does not have to be mathematically perfect to be worth anything visually. We don't sample perfectly according to theory in the real world. If we had to, a 2k film scan would resolve 500 line pairs or have tons of visible aliasing. But it resolves more than that with very little aliasing. And then microcontrast enters the picture; the integral of the area under film's mtf curve, the most widely accepted measure of perceived sharpness, is likely no better (and probably worse) than digital's even if the resolution doesn't extend out quite as far. My own tests on still film bear this out. My sharpest 25 megapixel scans of velvia have detail nearly as fine as a 25 megapixel digital camera's in high contrast areas (and no aliasing). But image quality is subjectively comparable to maybe 5 megapixels, at the very best, because the stuff is so grainy and detail gets fuzzy fast. That said, "true 3.2k" out of the red does sound totally impossible. That's substantially better than the 7d in stills mode despite a more aggressive OLPF and a significantly lower resolution chip. But who knows? Sampling theory is just that: theory.