Michael Nash

Well, we're talking about a couple different scenarios here. You could transfer to 16:9 video, in which case you don't need any letterboxing unless you want a 1.85:1 or 2.35:1 aspect ratio within the 16:9 (1.85:1 is so close to 16:9 that you often don't need to mask any of the image). Either way, the 16:9 anamorphic video is automatically squeezed vertically and letterboxed by the DVD player when viewed on a 4:3 TV. And when transferring to 16:9 video, yes standard 16 has to be magnified more than Super16 to fill the width. The other scenario is transferring to 4:3 video, where you would have to add the letterboxing and possibly reposition the image. Telecine is usally the best place to do this since it gives you the cleanest image and the most flexibility. But it's also possible to add letterboxing and reposition the image vertically afterwards in video post. In a high end tape-to-tape online room you can get good results; with less expensive systems you'll lose a little resolution. And since you're just masking the 4:3 footage, you're not magnifying anything compared to the original 4:3 video original. It's just that Super16 would give you more negative area to fill up the letterboxed image. Most 16:9 TV's these days give you the option to frame multiple ways from multiple aspect ratio originals. TOO many options it seems, since often times it's done wrong. Sometimes you'll see someone playing the 4:3 version of a movie from DVD, on a 16:9 TV and telling the TV to blow up the image to fill the screen vertically. The whole thing comes out framed wrong and with a drop in resolution. So if you create a 16:9 anamorphic video master, a 16:9 TV will display the intended framing just fine (except for some possible edge cropping). If you create 4:3 video master, letterboxing will offer you some protection that the aspect ratio will be viewed correctly, but no guarantee. In any case, it would be good to shoot a framing chart for use in post, and try to shoot all your remaining material within that framing. Since you like the look of your existing footage when letterboxed, you should probably frame for "common center"; that is cropped equally top and bottom. Without proper viewfinder markings, just leave a little extra room past your subjects top and bottom when shooting.

Bob, compare the incident reading of your Minolta and your Sekonic in low level tungsten light (f2.8 at 500 ASA, 24fps). If it's like my 508C, your Sekonic will read about a half stop low compared to the Minolta (the Sekonic will say f2.0.5 while the Minolta will say f2.8). Switch the light to daylight balanced and the meters will agree; and at higher light levels they'll agree. This is a "known issue" with the Sekonic 508, which may or may not have been fixed with subsequent models. Also, check that the spot and incident readings agree -- usually they don't and need to be adjusted. Fortunately you can adjust them independently, and I was able to get mine to agree dead on.

TV broadcasts and badly tuned sets can alter the chroma of skin tones as well, for movies viewed on video.

That's about right.

Check the archives from previous years; this has been discussed many times.

It should be easy enough to test if the shutter stays open when you depress the plunger. Simply remove the lens or open the camera and pull back the pressure plate and watch what happens with the shutter when you trigger it. But I'm unclear about what you read about the exposure being 1/30 during stop motion. Is that from the K3 manual? That sounds to me like the shutter automatically exposes for 1/30 when triggered remotely. I doubt the K3 is designed with a true "capping" shutter, so you're probably better off playing it safe and keeping the lens covered bewteen exposures (maybe rig a tiny flap of duvetyne over the lens that you pull up during exposure, so you don't bump the camera too much). Does the K3 have reflex viewing (I seem to recall it does)? If so you'll need to cover the eyepiece during exposure, and possibly the rest of the time as well. There used to be a couple websites dedicated to the K3, so try a search and you might find more info. I remember doing stop motion on super8 years ago. The intermittent start/stop faked out the footage counter and I lost half my movie, because I ran out of film and didn't know it.

Great -- but the video camera doesn't even have to be linked to the film camera optics. It's just a reference for the balance between the projection and the light levels in the room. Then you meter and expose for the room, knowing the projection is within visible range. Phil suggested Polaroids for this same purpose, but you have no way of ensuring synch between the Polaroid camera and the projection. This is yet another use I've found for my cheap little digital camera (the same one I showed you a couple weeks ago, Phil). Even though you can't synch the camera externally, in preview mode with the LCD you can see light levels in the camera's native scan rate (probably 1/60). It's also got a dynamic range that's more like a film print, compared to a video camera.

I originally missed the part where you said you were shooting in B&W. But shooting color neg and printing in B&W IS an option in that case. But let me get this straight -- you're shooting someone sitting on a real moving train at night. What exactly do you hope to see? It sounds to me like someone sitting in a chair, perhaps a little of the train interior, and MAYBE some moving lights out the window. That sounds like the perfect opportunity for poor-man's process. All you need is a seat that looks like it could be from a train (prop rental), a piece of plexiglass for a window, and some out of focus highlights moving in the BG (all this framed fairly tight). Then light it any way you want. Or, frame out the window and add some strobing light onto the actor's face from the window side. In the edit add some train sound effects and open the scene with an exterior shot of a moving train. But if you really need to shoot on a real train, you can also try those battery powered fluorescent tubes, the kind made to be put under cabinets. Get a few of them and place them wherever you like; one on the seatback in front of the actor and maybe one directly overhead. They're small and lightweight and easy to stash in a backpack. Or, you could have your actor hold real still and undercrank the camera to get a little more exposure. You have to hold the camera very still (probably a tripod) and lights outside will move faster, but sometimes trains go slow when they're near a stop.

It's also going to be really difficult to put up any large overheads or frames between trees. It's not like you're going to fly a huge silk over the TOPS of the trees, so anything you put up has to fit conveniently between trunks and branches, yet still manage to cover your action and keep the stands hidden. Or, you could try to tie off the silks to the trees with ladders. Sounds like slow going to me. You might be better off shooting your wide- and medium-shots around sunset when you have even, diffused skylight; then you can pick up closeups during the day with smaller overhead silks for the actors and flags for the background. Or, try to pick a location that's in the shadow of a hill or very dense trees so that you have shady light for a longer portion of the day. A really dense fog from an outdoor fogger can give you more continuous light, if you have access to such a machine and fog suits your story. Look at the DVD extras on "Insomnia" for this.

This also brings up a good question: how does Kodak define "latitude"? I've always understood that to mean how much you can over- or under-expose an image and recover a decent looking image. That's different than the "dynamic range," which is how much extreme bright or dark detail the film can capture. There is a HUGE difference between these two concepts -- some films will capture detail down to five stops under, yet I wouldn't dare tell someone the film has "five stops of underexposure latitude" and have them think they can underexpose five stops and still recover an image! By the same token, a fast grainy film may look like crap if you underexpose it, meaning it has NO latitude for underexposure. Yet it may still capture detail several stops under key. Have I got the terms wrong? Or am I being overly nit-picky? To me it seems important to draw this distinction, even if almost everyone says "latitude" when they're really talking about dynamic range.

If that doesn't work, try doing a manual white balance on the screen with just the raw light of the projector, before you thread up the film.

That's a bit of a moving target, unfortunately. It's variable with the native gamma of the film stock and the gamma of the telecine setup. You really kind of have to narrow it down to a particular stock on a particular telecine at a particular facility with a particular colorist who knows the look you want. Easy, right? ;-) I've seen stuff that was three stops over start to burn on older telecines, and stuff that's five stops over hold every bit of detail under 100 IRE on others. In general the newer and better the telecine, the more control over highlights you have. Most Kodak Vision stocks have a gently sloping shoulder curve and will hold highlights best in telecine. Older EXR stocks have a straighter gamma curve and will clip more quickly. I'm less familiar with Fuji.

There's a really good (and lengthy) segment about that on the new recut "Alien" DVD.

You could also consider compositing the projected image, if the shot is a lock-off. Then you can film at any light level you want (just do me a favor and insist to your editor that the black levels of the projected image aren't darker than the screen it's projected on -- drives me nuts whenever I see that!) A trick I've learned for balancing exposure between live action and a screen or monitor image is to simply look at the video tap image (or a similar on-set video camera). Basically, if it looks reasonable on the tap it won't look any WORSE on film. Of course you still need to synch the projector and camera. I've shot Betacam footage of projected 16mm with a double-bladed projector, and had it come out just fine. Of course the light levels were very low.

Um, how about available light? You can shoot just about anything with 500 ASA these days, even slower if it's daytime. A light will just attract attention. 7218 pushed one stop would get you an exposure at very low light levels. I shoot a ton of ENG Betacam at roughly 500-640 ASA, and hardly ever need a light for exposure -- just fill for aesthetics, if needed. I recently shot a story on the exhibit at LA's Natural History Museum (LA: light/motion/dreams -- great exhibit, go see it if you're here). Very low light levels, shooting with a D-30 at 500 ASA and +6db gain (one extra stop of exposure), I was shooting anywhere between f1.8 and f4 depending on the shot. Looked great.

The DVX-100 is a very "in demand" camera these days, so it will have a value and use beyond your documentary. The DVX is a relatively small "handheld" camera, which means you can't really shoulder mount it. This is generally a plus since it's lightweight and easy to put just about anywhere to get the shot you need, but if you're following people around in a real-life documentary situation you're ALWAYS holding the camera up in front of you. The DV-500 is an ENG-styled camera that's designed to sit on your shoulder, and it's light enough that you can walk around that way all day and not really feel it. It's also light enough that you can wave it around for different angles if you want.

It depends on the show, network, and so on. Here in the US, Betacam SP (analogue, uncompressed) is probably still the most widely used format. Digibeta (4:2:2) is used a bit, but probably even more as a post format. A couple of the network news affilates here in LA are using DVCPRO50, which has the same compression as digibeta. I've heard smaller news markets are using Beta SX (compression?) and DVCPRO (25MB/4:1:1). DVCAM (25MB/4:1:1) gets used every now and again, but isn't as common next to Betacam SP.

You don't have to take an all-or-nothing approach to your education. Don't be a slave to your meter or abandon it completely just yet. It sounds like you need to learn the dynamic range and "look" of the film stock you're using, and a light meter is a good tool for that UNTIL you can learn to trust your eye. Give yourself a chance to develop that skill with some practice. You don't have to spend big $$ on film for testing. You can shoot exposure tests with a 35mm still camera, on motion picture stock provided by RGB labs in Hollywood. They'll then print them on slides for you. Figure that the "average" motion picture negative gives you a ballpark of + or - 4 stops before it goes completely black or white (give or take, depending on a lot of little variables). That means that if an actor walks into an area that's 2 stops below exposure, his face will be about "half way" down into the shadow; dark but still visible. Same thing for overexposure. Try this out and learn what it looks like, then you'll be less afraid to use contrast in your lighting.

It really depends on the look and quality of the 16mm footage. A 25P (PAL) camera will better match the motion of film transferred to video (I'm assuming you're editing in PAL/25fps), but a better quality camera like a 2/3" chip Beta SX will probably give you better resolution, contrast and color control than the DVX-100. If the film was shot well with good lenses and a high-quality transfer, then you're going to need every bit of resolution and dynamic range from the video camera you can get. If the film footage is at all soft due to cheap lenses and a mediocre transfer, then the DVX-100 footage in 25P might match well. Is there a progressive scan Beta SX camera available? If not, you'd have to de-interlace the footage to get the motion to match that of film. That usually drops the resolution somewhat, so you'd be back to the look of the DVX-100. It's really something you'd have to test. But I don't understand your question about the Avid Symphony -- are you concerned about the image quality matching, or are you referring to EDL's and timecode?

You mention the price of film but nothing about the price of camera rental. Do you have a deal with someone for a free camera package for the duration of your project? If so, great. But if not -- don't underestimte the price of a camera, lenses, and accessories over the length of time it takes to shoot a feature film. If you're inexperienced with film and want to upgrade from consumer DV, you might consider shooting your project on an intermediate format like the Panasonic SDX-900 camera in 24P. It will give you a look that's vastly superior to pro-sumer equipment (although not exactly like film), yet the footage and sound will be on video and easier and more familiar for you to edit. Rental price of this camera will add up as well, but your footage costs will be far below that of shooting film. In the interlaced video world there are cameras like the Sony DSR-500 that record an excellent image onto DVCAM tape, which you can firewire straight into your computer from a DVCAM deck. Of course if you want to shoot film and can afford it, go for it! But if price is a factor and you want to produce a film of better quality than what consumer DV offers, then professional video might be the way to go.

I've made up my own using the specs from Samuelson's Manual. You can email me what frame you're looking for and I'll see if I've got it already. It's also a cinch to draw up your own in any draw program if you've got the specs and the time to sit down and do it.

As in "Blair Witch Project" Neal Fredericks? Welcome to the forum! Moviecam is much preferable -- the option to go mag on top or mag on the back helps in tight spaces, beyond the BL configuration. The BL is always a tank, the Compact doesn't have to be if you strip it down. Handheld is more comfortable with the the Moviecam, although not the lightest weight camera for that. And of course the optics rock. Very operator and assistant friendly. BL's sometimes rent cheaper, and aren't bad cameras. Just not as flexible as the Compact. The only bad thing I've ever seen happen with the Compact was on a shoot in the desert where sand got in the gate. The gate design in that camera is two separate pieces of metal, which allowed a little sand to get stuck in the crevice and scratch the film. Of course this isn't really a design flaw, as desert shooting is always a risk and sand in ANY camera is a bad thing!

You can also try to drown out or overpower the green influence of the fluorescents by adding more daylight with HMI's. If there are pockets of deep background you can turn off those practicals and bounce a daylight balanced light into the ceiling instead. Or simply cover the offending practical light with minusgreen gel, usually 1/8 or 1/4. As long as you don't see the fixture itself in frame, it's a quick fix. But you might also consider shooting with a filmstock that's less sensitive to mixed color temps. Daylight balanced stocks like Kodak or Fuji 250D will minimize (but not eliminate) the color difference. Fuji 500D has a fourth color layer that's supposed to minimize fluorescent green even more. I've only seen the demo and never tried it myself, but it seems to work reasonably well for this.

I shoot tons of Betacam and personally I have almost no use for 100 IRE zebras. If it's blown out, I can easily see that in the viewfinder. The knee function sometimes restores a little color to burnt out highlights that you can't see in the VF, but the signal is still up there at 100 IRE. Personally I like zebras somewhere in the middle like 70 IRE as a reference for exposure. It is true that Caucasian skin comes up about 70 IRE on the key side when properly exposed, but I don't live and die by that number. Exposure of any shot is determined by the levels throughout the whole frame and not just the skin tones alone. And of course in dramatic lighting, you can put as much highlight and shadow as you want anywhere you want. These days when shooting ENG (uncontrolled lighting and no monitor), I rely more and more on the viewfinder for exposure. The zebras are still there, but since skin tones are moving in and out of shadow the 70 IRE rule of thumb goes out the window. If the viewfinder is in good working order and calibrated for contrast and brightness to color bars, you can begin to trust what the image is telling you. Basically, if it looks good in the VF it won't look any WORSE in color and on a professional monitor. Sometimes you get stuck with an old camera with a crappy viewfinder, but I try to familiarize myself with the quirks of whatever camera I'm using before I roll anyway. Regarding Marty's situation, it sounds like errors in the engineering might be responsible for the levels being different. But another thing I've noticed in the past few years as video cameras have included more and more internal signal processing, is that the gamma, black stretch, and knee of cameras can be wildly off from each other; far more so than in the old days when all you could adjust was white balance. IRE levels and exposure can vary between cameras, just as it would between say reversal film and low-contrast negative. The old rules of thumb for video don't always apply anymore.

Another trick is to flag the wall or ground where the shadow falls, so that your eye goes to the light on the subject instead of to the shadow. Typically any kind of sidelight will include a "bottomer" to flag or at least soften the light on the ground. This is especially important on sidewalk night exteriors, where the concrete is relatively bright when lit, compared to the dark background. You usally don't miss the little bit of side or edgelight when it's below the knees, and it's generally better to lose a little definition on the feet than it is to have distracting mulitple shadows. But use your judgement -- there are all kinds of "rules" out there that they teach you in school, but it's okay to break them AFTER you learn them. I'll tell you, when my lighting and DP skills really took off was when I decide to throw out all the rules they taught me in school, and start lighting things the way I wanted to! I'm not suggesting your ignore what you're taught, just don't be afraid to go BEYOND the rules if you need to.