(Page 3 of 6 pages for this article  <  1 2 3 4 5 >  Last »)

Filed under: GentryMedia Sister Sites •ProPhoto Coalition •ProVideo Coalition •Lighting •Production •Tips •Training •

LIGHTING STRATEGIES: Exploiting a Single Light Source

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Lighting •Tips •Training •

BOOK REVIEW: “How to Shoot Movies Without Shooting Yourself in the Foot”

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •

LIGHTING STRATEGIES: Placing the Fill Light for Faces

The following image isn’t a true representation of what I’m trying to show as the 3D modeling program I’m using doesn’t replicate the effects of light perfectly.

This is supposed to look like sun coming through a window on the right side of frame with a bounced fill on the left side. Adding a bounce fill opposite the main light source is a common way to fill situations like this, but it has its drawbacks. The main one is that in order to catch the most light the card must be almost directly opposite the primary light source, and that isn’t always optimal for lighting a face.

If I’m going for a dramatic look then having very little light in front of the face is okay. If not, there are a number of cosmetic issues that have to be addressed:

If light is coming only from the sides then the center of the face goes dark. There are dramatic situations that warrant this, but on other occasions it can look like a mistake.

As I’ve mentioned before, the one place that almost never gets light from this situation is the corner of the eye closest to the nose on the key side: a key from the side doesn’t quite reach this spot, and the fill light from the opposite side can’t reach it either. Worse, this spot is made to look darker than it is by being juxtaposed against the key side, which is brighter than the fill side. This isn’t a huge problem with this virtual gentleman but it is very common in real life.

Moving the fill card toward the camera, and more frontal to the actor, may solve this problem, but unless this fill source is very big and a fair distance away it can feel as if a bounce card is just outside of frame. It’s rare in real life for that to happen.

What I often try to do instead is fill from the key side. This means that the primary fill source (yes, you can have more than one) is on the same side of the lens axis as the key is. Here’s an example:

All the light in the scene is coming from the right side of the lens.

Here’s the key light only:

Here’s the fill light only:

And, once again, the combination:

I really like this look for a number of reasons. The first is that there’s a greater tonal range across the face than if the fill was placed on the opposite side of the key. Instead of bright on one side and dark on the other, the bright side transitions into a dark tone which then transitions further into a darker tone. I like the richness of the multiple tones. I also like that the fill shadow is falling the same direction as the key. Filling from the key side means all the primary shadows in the scene are being cast the same direction, and this has a certain aesthetic appeal. It’s also possible to hide one shadow within the other.

The need for some sort of fill on the opposite side of camera doesn’t always go away. In this case the 3D software I’m using doesn’t wrap the light around this gentleman’s face as much as it would in real life. The virtual fill light for this shot is placed at the top right corner of the virtual camera’s matte box, so the really dark area of the face should be receiving more light and the dark side of the face should be a little brighter. If that shadow is a bit too dark then it’s easy bring it up subtly with a distant bounce card.

“Wait!” you say. “You’re talking about adding an additional light in addition to a fill card. Isn’t that needlessly complex and a waste of time?” To which I say, “No. Anything that makes the subject look good fairly quickly is a good thing.” I usually have a key-side fill light, often a tungsten source in a medium Chimera or aimed through a diffusion frame covered with Lee 216 or 129, on standby for exactly this kind of thing. If an actor lands near a light source during the master shot, say next to a window or a table lamp, and the angle of light on their face isn’t optimal, placing the fill light on the same side of the lens and making it optimal can clean up a lot of badness. If I can’t put the key in the right place for the shot then I’ll put the fill in the right place instead.

It’s a very smoothing, flattering look, and even though it adds an additional shadow it often “hides” within the shadow of the key and is less noticeable. Also, we’re more able to accept multiple shadows that fall in the same direction as opposed to shadows that fall in opposite directions.

Here’s a top view of our key + key-side fill setup:

You’re used to seeing this in real life when it happens like this:

Which would look something like this:

See the double shadows to the left of the subject’s head? That looks very realistic to me. In fact it looks more realistic to me than if it were a single source, because light in rooms often comes from multiple windows and multiple light sources and a little sloppiness can make the image more real.

Also, if you think about it, you’re emulating the look of a large source by replacing this:

With this:

We’re basically placing a light at each end of where the diffusion frame would be. You’ll miss a lot of the gradations that happen by having a light source that fills the entire space between the two lights, but you gain the possibility of doing this:

It’s very difficult to cut a large source off a nearby background. It’s much easier to cut one of two smaller blended sources off the background, which will reduce the brightness of the background considerably. We’re always trying to control light on walls as they are rarely very interesting to look at. Their brightness and prominence may distract from what we should really be looking at: the actor.

There are variations of this technique that I use all the time, depending on the subject matter. For example:

In this case the fill light is from directly below, as if sunlight was hitting the floor directly in front of this person. This is a little stark so lets add a small fill card opposite the key:

To me this has the feel of someone standing near a window where sunlight is streaming in and bouncing off the floor.

Filling from the key side doesn’t solve every lighting problem, but it solves a lot of them. I find it much easier to light a set once and not have to completely relight closeups if I can simply smooth out the light that’s already there, even if the key is in an awkward place for that one shot. The key-side fill usually doesn’t draw attention to itself, it blends in with the lighting that’s in place, it’s fast and it makes nearly everyone look good. What’s not to love?

All the 3D examples in this article were rendered in Poser 9.

Art Adams is a DP who prides himself on his natural look. His website is at www.artadamsdp.com.

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •

LIGHTING STRATEGIES: What Makes Soft Lights Cast Soft Shadows?

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Production •Tips •

For You, a Panel Discussion

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Tips •Training •

LIGHTING STRATEGIES: Soft Light vs. Hard Light

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •

Pulse Width Modulation is NOT Your Friend

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •

LIGHTING STRATEGIES: Placing a Hard Key Light

Here’s typical “Rembrandt-style” hard light on a man. Men can tolerate side light more than women as men tend to be lit for ruggedness whereas women are more often lit for beauty.

Bringing the light around a bit is still quite flattering.

And here’s the dramatic extreme:

This is a great example of how hard light can reveal nose structure. Notice the shadow of the large bump just below the bridge of the nose. On a man this is rugged; on a woman, not so much. As with everything in lighting, sometimes this works and sometimes it doesn’t. Be aware and you’ll soon figure out what works and what doesn’t.

Here’s a technique that I learned when Stephen Burum, ASC spoke to my Advanced Cinematography class back in college:

Almost everyone’s face curves. The eye and mouth will be closer on one side of the face than the other. Generally people look better if that side of the face faces the camera and the key light is placed opposite the curve.

You’ll not always be able to block and light shots this way, but being aware of this may help you win the confidence of a difficult actor or actress.

Last but not least, here’s how I determine where other DPs place lights when I watch films and television:

I look in the actors’ eyes. In this case you can see the key light reflected high and to the right in one eye, and the virtual fill light I’ve placed immediately to the left of the camera, at lens height, in both eyes.. Judging light placement can be a bit tricky as the curvature of the eye adds distortion, but with practice you can learn to judge the placement, size and number of sources being used in a shot just by observing reflections in eyeballs.

In my next article we’ll look at soft keys. Until then… play around and see what kinds of faces work with different kinds of light. Every face is different. The trick is that you often can’t tailor your lighting to an individual, but you can learn how to fix common issues when they arise.

Art Adams is a DP who fails to flinch from facial photography. His website is at http://www.artadamsdp.com.

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •

The Simplest, Fastest Interview Lighting Setup—Ever.

Everything I used for corporate interview shoots had to fit on a cart, and uncut 4’x4’ bounce cards are unwieldy to pack and don’t go through doors easily. A local rental house came up with a trick to make them more portable; I’ve shamelessly stolen this trick and I now offer its theft to you:

Cut a 4’x4’ bounce card into two 2’x4’ pieces.

Lay them about a half inch apart on the ground.

Tape them together using 2” cloth tape.

The half inch gap allows you to fold them in half. Without that gap they’ll resist folding.

I worked with one client for about ten years shooting enormous numbers of talking head interviews, and I carried the following lighting/grip kit:

A four-light Arri kit, typically with a 1k open face unit, two 650w fresnels, and a 300w fresnel. The 1k open face was great for situation where I needed more punch out of my key light, such as when I was lighting an interview in a room lit by ambient skylight. (I only ever added half CTB to my lights for daylight correction as full CTB costs too much light and half correction to daylight allows the person to look a little warmer or the background to look a little cooler, depending on how you white balance.) Depending on the senstivity of the camera I could use either a 300w or 650w fresnel to light my bounce card.

Four of the previously-described collapsible 4’x4’ cards, made of foam core that is black on one side and white on the other. I could use the white side as a key bounce or fill, and I could use the black side for negative fill or to block large sources of unwanted ambient light, like windows.

Four C-stands.

Those basics allowed me to walk into almost any interview situation, sight unseen, and make it work. Block a window here, setup up my white card/black card setup there, turn on a light or two, and shoot.

Three more notes:

Sound people sometimes complain about the “slap” of sound bouncing back and forth between the two cards. This can be solved by used a 4’x4’ flag instead, or hanging a black sound blanket from a C-stand as negative fill, or covering a 4’x4’ card with black duvetine. Cloth absorbs sound, whereas foam core does not.

Some producers made fun of this setup and called it “the tunnel” or “the cave.” I responded by pointing out that it makes everyone look good on their schedule, and they couldn’t argue with that.

Make sure you fill the entire bounce card with light for greatest benefit. Spotting a lamp into the center of the card not only results in a smaller source with harder shadows but it’s also less efficient. Filling the entire 4’x4’ surface results in softer light, as the size of the source is much larger in relation to the subject, and reflects the most light possible.

Being fast is as important as making pretty images. This technique allows for both. Give it a try and see what you think.

Art Adams is a DP who loves lighting faces. His website is at www.artadamsdp.com.

Filed under: GentryMedia Sister Sites •ProVideo Coalition •

The Future of Technology is You

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •

Fill Light: The Underdog of Lighting

One interesting side effect of putting a large source near the lens axis is the glow that it casts in skin. Here’s a still from a green screen performance I shot shot for a PepsiMax viral:

(The actual video is here.) The budget for this project was very, very low and my entire lighting package consisted of a 1200w PAR, a set of Lowell lights, and six 4’x4’ black and white bounce cards. For this setup I set the camera (a Canon 5D) about 12’ away from the talent and set two 4x4’ bounce cards directly behind it, making a large 4’x’8 wide source. I then aimed every tungsten light I had into these cards, lighting them evenly. In this way I was able to light both the talent and the green screen beautifully, while casting few shadows on the green screen. I didn’t have the space or the lights to illuminate the green screen perfectly or separately (we shot this in the back room of a health spa) so I just lit down the lens axis with every tungsten light I had.

Top view.

Side view.

The glow in the center of their faces is beautiful. Just imagine what that would look like if you filled a shot that way: you’d have both the directionality of the key and softly glowing skin from the fill. Even if the key is temporarily blocked by a moving person or object the fill light takes over nicely.

PROS:

Soft light from near the lens axis looks natural and makes faces look their best.

CONS:

Fill light has to follow the camera around as it changes position.

Bounced fill from under the lens.

Here’s an example from a music I shot a while back. Here we see the key lighting the actress’s face (casting the nose shadow along the smile line), with the fill source lighting the shadows:

And here’s what happens when she tilts her face so that her hat cuts her key light off her face:

That beautiful hit in her eyes is a 2K bounced off a 2’ square white card directly under the lens. The edge of the card is almost touching the bottom of the lens. As the source is fairly small it did cast upward shadows, but they were so soft that they looked natural. Soft light from below, particularly from a big source underneath the lens (such as a 4’x8’ sheet of foam core) looks very much like ambient light that has struck the floor and bounced upward. The larger the source, though, the softer and dimmer the eye light, and sometimes a bright eye light is more important than softening a fill shadow. (Smaller, brighter sources cast harder shadows but create brighter eye lights. Large, soft sources cast indistinct shadows but are less visible in eyes.) Deep eye sockets can be hard to light, but a light placed under the lens will almost always reach into them.

NOTE:

I learn a lot about light placement in movies and TV shows by watching highlights in eyes during close ups. I can often tell the number of lights hitting the face, where they are in relation to the face (above, below, or level) and their approximate sizes. It takes a while to get a feel for this as the roundness of eyeballs tends to make large sources look smaller and the angle of lights can be exaggerated, but over time it’s possible to learn a lot about how other people are lighting by counting eye lights.

PROS:

Fill light is almost always seen in eyes so a catch light remains even if the key is blocked.

CONS:

Upward shadows, although under certain conditions this can look quite natural especially if the fill source is large.

Moving the fill light to the key side works wonders. One of the problems of putting the key opposite the fill is that it creates a competing shadow, falling opposite the key, that can be distracting; but placing the fill such that it casts a shadow in the same direction as the key is much less distracting. There are lots of natural situations where this might happen, such as a room where there are a large number of windows on one side of the room. Each window casts its own light, but as the light sources overlap none of the shadows is strong enough to be noticeable on its own as each window’s shadow is usually being filled by other windows.

For some reason our brains seem to accept multiple light sources that all come from one direction more easily than from opposing directions. (Roger Deakins, ASC, BSC exploits this technique often, as he has spoken many times about lighting large exterior sets by lining up large numbers of specular lights on a nearby hillside and not needing to diffuse them to make them feel like one source.)

One of the reasons this trick looks so good is that it creates multiple tones on the talent’s face. The traditional key-on-one-side-fill-on-the-other creates two tones: one side of the face is bright and one side is less bright. Filling from the key side results in a highlight that transitions through many tonalities as the key wraps into the fill, and the fill wraps around into deeper shadow.

This frame, from a music video I shot for the band the Tiger Lillies, is lit by two light sources and one bounce: there’s a large source to the far right, another source from just off the right side of the camera (filling from the key side), and a passive bounce on the left that is dim enough that it doesn’t create a bright highlight. (A “passive” bounce reflects light that is already in the scene, whereas an “active” bounce is a bounce card that is light with its own light.) You can see the two large sources in the actor’s frame right eye, and the fill-from-the-key-side source alone is visible in his frame left eye.

A diagram of the lighting setup. The half soft frost smooths the 12’x12’ bounce and makes the light appear more even across its surface from the subject’s perspective.

Another example using the exact same lighting setup. Note how the key light wraps around to transition into a darker tone created by the key-side fill, and then further around into darkness on the leftward facing surfaces. There are a lot more tones here than one would get from a traditional three point lighting setup or from simply putting the fill source over the lens axis.

It’s possible to fake the look of a much bigger source by using a soft fill from the key side. This can be handy when a large source would spill across the background, making it too bright, while a smaller source could be cut off the background. By defining the beginning and end of a large source it’s possible to emulate the look of a large key but with smaller, more controllable lights, put at the ends of where the larger source would be.

Filling from the key side works no matter where the light is coming from. If the key source is overhead, for example, then a fill light just above the lens axis will look very natural and “unlit” while creating a nice catch light in the talent’s eyes.

The trick with fill light is trying to find where to place it so that it does what you want without taking a lot of time to move it around. Time is always at a premium on set and it’s crucial to use fill light techniques to shorten setup time rather than increase it. In episodic television it used to be common to hang a space light over the set to create a base fill level, and then the key and fill lights would overpower it in the foreground while it gave some depth to the background. (Backgrounds are the hardest to light, because there’s always so much of it!)

Something I do a lot is to establish a key direction (soft light from a window or other source) and then build a big fill light that stays at the back of the set on the same side as the key. The farther the light is from the talent the deeper it will carry into the set and the more consistent in brightness it will be as actors move around.

Or, if the set is lit with a big enough source that I don’t need a specific fill light, I’ll only use one fill source on closeups to clean up faces. Ambient light can do nasty things to faces, creating dark eyes and accentuating wrinkles, so putting a fill light close to the lens just for closeups cleans a lot of that up.

I use a lot of negative fill, often because I don’t like what the ambient fill light in a room is doing to closeups. Maybe it’s too toppy, coming off the ceiling, or maybe it’s the wrong color due to a colored wall or carpeting. I’ll stand on the actor’s mark and look around to see what surfaces I see on the fill side that are radiating light (like a white wall) and then I’ll have the grips cover that with flags or duvetine. If the shot is wide then we may have to use larger solids farther back, but if I’m only worried about the closeup then we can use smaller flags closer and fuss less with the wide shot.

Sometimes the negative fill removes a little too much light and I’ll add a bounce card just to lift the shadows on the dark side of the face. I generally keep the bounce card as far in front of the subject as I can because, while a card alongside the talent’s face will catch more light, it may not reach into both eyes. If you want a harsher look, moving the fill light away from the lens axis can accomplish that.

The idea is to remove the ambient light that I don’t like and replace it with ambient light that I do like.

Sometimes it’s necessary to fill a deep set. This is occasionally easier than it sounds. If I’ve rigged a big soft source in one part of the set I can often add smaller fill sources from the same side of the lens axis to carry it deeper. These lights don’t have to be as powerful as the primary fill because the actors are probably still getting some light from the fill source, just not enough. If my fill reads T1.4 at the primary acting area, and T1.0 in the background, I only need to use a light that is half as bright as the primary fill to bring up that area. Here’s an example from the sitcom world:

Back when I was a camera assistant, this is generally how sitcoms were filled: there was a row of 4’x8’ pieces of foam core across the front of the set, high enough that the studio audience can see underneath, and hung in front of them were dozens of small, cheap open-face lights. This created a very soft beautiful fill that reached a fair distance into the set. The “A” people are nicely filled by this source.

The yellow lights boost the fill levels deeper into the set, where people “B” need a little help. These lights can be smaller units because they aren’t creating all the fill, they are just adding to it. Even though they are usually smaller in size the shadows they cast are usually washed out by the main fill source.

The blue “wing” lights are to wrap the foam core source around the front of actors who are standing at the far edge of the set. Person “C” would be very dark on the camera right side if that far right blue fill light wasn’t there.

(Note that I’m not showing key lights, which would typically be placed facing the opposite direction as back cross keys.)

Something to keep in mind is that distance solves a lot of problems. For example, it might be advantageous to fill the background of a shot simply by bouncing a light into the ceiling for the wide shot and then adding a fill light closer to the lens to wrap that light around faces for coverage. I’ve even filled with hard light at a distance, putting a tweenie directly over the lens and aiming it at someone in the distant background. If the light is directly over the lens it will cast a minimal shadow, and if the person you’re filling is far away from the camera and small in frame that shadow will never be noticeable.

Another interesting trick is coloring the fill light. This can be done in film cameras by pre- or post-flashing the film or using a device like the Lightflex or Panaflasher, both of which change the color of the shadows only. It can also be done simply by putting a colored gel on the fill light. This has the effect of lightly tinting the highlights and creating much more saturated shadows, and can be a great way to create mood. Just be careful not to use colors that compete with the set or wardrobe (filling a blue set with warm light) as the effect may be completely lost or distort other colors, and don’t dig yourself into a hole by finding that you now have to recreate that look for a sunlit day exterior. It is most definitely an option, though, and one I use fairly often. I’m shooting a spot in a darkened restaurant this weekend and I’m going to make the fill light warm and slightly red for a romantic and cozy feel. (I’ll probably use full CTO on the fill, and since HD cameras tend to see warm light as slightly greenish I’ll probably at 1/4 minus green to make the light a little more salmon in color and less yellow-green.)

Turn the page for lots more pictures, which a short description of what’s going on in each shot…

Filed under: Cameras •compression •Distribution •GentryMedia Sister Sites •HDSLR •ProPhoto Coalition •ProVideo Coalition •Lighting •Production •Software •Tips •

Blue Nile Shines Thanks to the Canon 5D and Apple Color

I did the final grade myself, as there wasn’t quite enough budget for a professional color grade, and I think it turned out pretty well. We had to create two versions, one in standard def for broadcast and an HD version for Seedwell’s website. Believe it or not, there’s quite a bit of standard def still out there: a number of systems, including Google TV (where this is running) will only accept standard def deliverables, and often they’ll insist on 4:3 and won’t accept letterbox! These systems are almost completely automated: you go to a website, pick the shows during which you want your spot to air, pick the number of times and dates that you want it to air, pay by credit card and upload a Quicktime file. The spot is inserted into standard def cable television streams and can start airing immediately.

The SD 4:3 version was fairly simple to grade. David had already reframed a number of the shots so I came in and did a little tweaking and moved on to color and power windows. When I got to the HD version, however, I had to back off on some of my vignettes as the highly compressed H.264 footage showed a lot of banding in certain shots, primarily the ones showing the blue-lit background bar. I don’t feel like I lost any of my look due to having to back off the grade, but it was interesting to see how much more I could get away with in SD than HD when it came to Canon 5D footage.

I should point out that the reason I was able to push the footage around as much as I did has much to do with Steve Shaw’s excellent Canon 5D gamma curves.

The client is thrilled, and the spot is airing now.

For the last few years it’s felt as if we’ve all been working around a “cult of the camera:” the camera became, for a short while, more important than the people using it. While different cameras have different strengths, and some cameras are definitely better than others, in the end the camera is one of the least important factors in achieving excellent results. It’s the people behind the camera that matter more than anything else, and in this case we had an exceptional crew and a fantastic creative team that turned out a product that had very little right to look as good as it did given the resources at our disposal.

Disclosure: I have consulted for Tiffen Filters and they gave me the filter mentioned in the article for free. I have no financial ties to Light Illusion and Steve Shaw other than that I bought their gamma curves and use them constantly.

Art Adams is a DP who makes just about any camera look better than it should. You can see the proof at www.artadamsdp.com.

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •

You’ve read my writing, now hear my talking

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •Web Video •

Anatomy of a Spot: T-Mobile

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Post Production •Production •Tips •Training •

DSC Labs Hawk Chart: The Simplest Color Chart That You Can’t Live Without

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Lighting •Production •Tips •Training •Visual Effects •

Arri Alexa and Rosco LitePads Come Through for OnLive’s First National Spot

The first shot added was of the front of the house, where a 1200w PAR raked the wood siding while the video projector lit the ceiling inside. We used a branch-o-lorus stacked in line with a nearby bush to break up the light a bit.

The closer light is a 1200w HMI PAR lighting the front of the house; the far light is a 4k HMI fresnel lighting the living room through floor-to-ceiling windows. They are covered to protect them from rain.

I love using multiple shadow sources because the interplay of the two patterns is much more interesting than a single pattern alone. The closer pattern provides some initial texture that isn’t terribly distinct, and then the openings in the second pattern act as apertures through which the image of the first pattern is projected. The result is a very interesting mix of soft and hard shadows that weave through each other, much like sunlight through dense trees on a sunny day: the interplay of shadows from leaves at different distances from the ground creates a mixture of hard and soft textures that are much more interesting than one layer of leaves alone.

The outside of the house looks really green because we lit green paint with greenish light. It looks really cool.

This setup was a lot of fun to light. The hard light raking across the far wall and the couch is from the 4k HMI fresnel outside the window on frame right. The flashes of light are from the video projector, rigged on a goal post directly over the TV and aimed through a small frame of either Hampshire Frost or Opal. The musical instrument on the left of frame is lit by a bounce card sitting just off frame left, and the actor is lit by both the 4K HMI outside as well as a Rosco Axium LitePad above and to his left. There’s an additional LitePad above and behind the actor but it’s not there for him; it’s meant to highlight the controller he’s holding.

Key grip Ernie Kunze rigs the slider on apple boxes for the TV scenes. The windows on the left are lit by the 4k HMI fresnel in the back yard, and the square of light lighting the windows in the back of the room is from the video projector rigged over the TV. Not bad for a projector that we picked up on a whim from Best Buy.

We used a 4’ slider to create a dramatic move over the couch and into the TV.

The controller, shot on an 85mm Ultra Prime at T2.8. The flash of red is from the video projector.

The controller needed a little extra care. The light coming through the window didn’t do it justice as hard light doesn’t illuminate black shiny surfaces well. Instead we lit it like a car, reflecting a light source in its upward-facing surfaces. Rosco had asked if I’d be willing to test out their new Axium series of LightPads, and shortly after I screamed “YES!” into the phone they shipped me a Gaffer’s Kit with a wide variety of LightPads of varying shapes and sizes.

The 1’x1’ was the perfect unit to swing over the product to give it a bit of glow. We added a 4’x4’ frame of Lee 216 diffusion to spread the highlight across the surface of the controller. If you think of the top of the controller as a mirror, we had to reflect something to reflect in that mirror in order to see it. We added the diffusion to make the reflection wider and wrap it around the controller’s curves.

The Rosco Axium LitePad on the right lights the controller through a frame of Lee 216 diffusion. The duvetine attached to the light and the back of the diffusion frame prevents spill light from bouncing off the back of the diffusion and washing out the background. The LitePad on the left, in the background, adds a little 3/4 back light to bring up the shadows on the actor’s fill side.

Rosco LitePads are uniquely designed in that they are the only LED product that doesn’t project light into the scene. Instead they glow softly in a fashion that makes me think of them as a flat Chinese lantern. They’re great for product shots, or hiding a light source in a scene without making it obvious that I’ve added an additional light, or for what I call “volumetric lighting”: since these “glowing” sources don’t project light objects have to pass near them to be lit—which helps the audience “feel” the size and shape of the space as the proximity of the actors to the light source changes. When I want to define a space by having actors pass through “glows” of soft light without drawing attention to those light sources, LitetPads, Kino Flos and China balls are my “glow” tools.

The base unit took a little more care. Eric wanted it to glow as it was turned on, so we had to light it two ways. In the “off” state it required the same kind of lighting as the controller, as it was made of the same black shiny material. In its “on” state we had to create a soft warm glow that dimmed up over the course of a second or two.

That’s me touching lights. Behind the TV is the LitePad that illuminates the back wall and creates a smooth highlight in the surface of the base unit. The narrow LitePad that I’m holding is meant to pop the front of the unit. In the very top of frame, directly above the lens, you can see the snoot hanging from the tweenie that is creating the “warm glow” effect. In the background you can see one side of the overhead speed rail rig that holds the video projector out of frame over the TV.

The base light was a narrow Rosco LightPad taped to the back of the TV. It reflected nicely in the top of the base unit and also added some light to the wall, which in turn was reflected in the shiny surface of the shelf holding the base unit. It was pretty late at night when we did this shot and I’m not sure how we could have done this as quickly and effectively without that one perfectly-shaped LitePad.

The front light was a very narrow LightPad strip that lit the flat front of the unit.

Products generally look a lot sexier when shot up close with a wide angle lens. We used a 20mm Ultra Prime and put the lens inches away from the product. (Ultra Primes can focus very close, typically around 1’ from the focal plane for the wider lenses.) Focus was placed between the front of the unit and the logo. I think we shot most of the interiors at T2.8 or so, but we lit this shot to T4 for increased depth of field.

The base unit with “glow” and added visual effects.

There are certain products that benefit from both hard light and soft light, and this base unit was one of those. The shiny black surfaces looked best with a large soft source reflected in them, but the logo has a matte non-shiny surface and required hard light. We lit the logo with a tweenie, rigged overhead with a snoot. Notice that the tweenie’s hard light has no other effect on the rest of the base unit other than to cast a shadow on the shelf. The base unit’s surface is like a mirror, and unless an object or light source is reflected in that mirror you won’t see the surface directly. As no part of that tweenie reflected in the product’s surface from the camera’s perspective, its effect was invisible except on the logo.

Me, director Eric Peltier and camera assistant Rod Williams finesse the base unit setup. You can see the “glow” tweenie at the top of the frame. The round white object directly below it, sitting on the edge of the shelf, is another custom-shaped LitePad that’s standing by (I don’t think we used it). The narrow LitePad lighting the front of the base unit is reflected in the TV.

The explosion was obviously done in post, although there are some physical effects present as well. Westernized Productions did all the post VFX work and a couple of their people came out with 2’x4’s and bounced the furniture up and down during the explosion sequence.

One interesting note: when we returned for the reshoot the team from Westernized asked me what I was doing that made their life so easy in post. They were working on another project that was also shot with an Alexa, and they were having a hell of a time compositing elements into that footage whereas they had zero trouble with my footage.

A few questions later I discovered that the other crew was rating their Alexa at EI 800, whereas I was rating mine at EI 400. The Alexa is very clean at 800 but there is still some noise, and compositors hate noise. My tests had shown that 400 gave me a much cleaner signal, and as I prefer a cleaner look I habitually rate the Alexa slower. It makes perfect sense that this would make VFX people happy: not only does less noise result in cleaner edges, allowing for better comps, but the comped elements don’t require added noise to make them match the background plates. (If you want to bring undue attention to a composite then add a very clean element to a slightly noisy background plate. The side-by-side comparison makes the differences very obvious.)

This shoot was great fun. It was also an honor to learn that the initial shoot went so well that the client escalated their expectations, paid for a re-shoot, and made a huge ad buy based on the quality of what Eric, my crew and I pulled off. The spot ran nationally in the U.S. on network and cable television throughout December 2010 and January 2011, and I’m told it still pops up from time to time on cable.

I can’t wait to show you the viral pieces we shot in May. They’re in post now, so hopefully I’ll have more to show in a couple of weeks.

Project: “OnLive.com” national spot

Production company: Rearden Studios

Writer/Director/Editor: Eric Peltier

DP: Art Adams

Camera assistants: Paul Marbury (day one), Rod Williams (day two)

Gaffer: Alan Steinheimer

Key Grip: Ernie Kunze

Lighting/grip from Lighting by Steinheimer

Camera and lenses from Chater Camera

Visual effects by Westernized Productions

Stills by Matt Polvorosa Kline, and used with his permission.

Disclosure: the LightPads mentioned in this article were provided at no charge by Roscoe for testing purposes.

Art Adams is a DP whose Xbox is very, very afraid of OnLive’s business model. His website is at www.artadamsdp.com.

Filed under: GentryMedia Sister Sites •ProVideo Coalition •NAB 2011 •Tips •Training •

The Secrets of the Chroma Du Monde, Explained Live (on tape) at NAB!

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Production •Tips •Training •

A Mix of Film and HD Doesn’t Scare Arri’s Alexa

Filed under: GentryMedia Sister Sites •ProVideo Coalition •

Where I’ll Be at NAB

Filed under: GentryMedia Sister Sites •ProVideo Coalition •Post Production •Production •Tips •Training •

The Secret Art of Slating: 25 Tips to Help You Slate Like a Pro

Filed under: Cameras •GentryMedia Sister Sites •Apps4Phones •ProVideo Coalition •Lighting •Mobile Devices •Pre-Production •Production •Software •Tips •

iPhone Apps: The Short List for the Average Cinematographer

Filed under: Cameras •GentryMedia Sister Sites •ProVideo Coalition •Production •Tips •Training •

Panasonic AF-100: The Good, The Bad, and The Ugly

The AF-100 has a lot of potential, and I’d definitely go with it when the alternative is an HDSLR. Its layout and controls are familiar and it’s designed to do what it does, so it’s much faster and easier to program, judge focus and expose than an HDSLR. It also doesn’t moire like an HDSLR will.

On the negative side, though, it’s still basically an HVX-200 with a larger sensor. It clips roughly the same way, which is not a good look in a large sensor camera, and it’s noisy too. The odd thing about the noise, though, is that it’s consistent throughout the exposure range: there’s no discernible change in noise between EI 200 and EI 800, and the exposure latitude doesn’t seem to be affected either. Typically you’d have more underexposure latitude and less overexposure headroom at EI 200, and the reverse at EI 800, but I couldn’t see any difference at all when I tested this camera during prep. At EI 1600 you can see that Panasonic is doing some noise reduction: Adam says the noise gets a little blocker, and there’s some odd motion blur that implies that they may be averaging noise across multiple frames in order to reduce it. Still, the camera is very usable at high EI’s and I’m not afraid to rate it at EI 800. In fact, for another project shot mostly under available light in a church, that’s exactly what I’m doing. (I rated the camera at EI 400 for this project.)

There are some differences between this camera and its predecessors when it comes to gamma settings. CineLike V has, in the past, been the all-around winner in the HVX-200 and HPX-170 as it rolls off fairly pleasantly at the upper and lower portions of the gamma curve and is pleasing on flesh tones. On the AF-100, though, CineLike V is very crunchy and does unpleasant things to flesh tones, causing them to chroma clip much too soon. As I mentioned before, I like to slightly overexpose flesh tones in certain situations and CineLike V doesn’t allow me to do that.

After a bit of testing I settled on “Low” gamma, which seems to push middle gray down the gamma curve and results in slightly crushed shadows but pleasantly stretched highlights. (Crushing highlights can make them look clipped sooner than they really are, so pulling gamma down opens up that highlight range and allows for a greater range of bright skin tones.) This is similar to what I do when working with a Sony EX1 or EX3: I use the CineLike 4 curve (which is a great all-around gamma curve) and then set overall gamma somewhere between -10 and -40, depending on the situation. This crushes the shadows a little but makes the highlights a lot smoother.

Speaking of highlights, I did a little testing on my own to determine whether I could return some detail to clipped highlights in post.

This picture is a still from the original footage, uncorrected. The shiny portions of the actress’s skin are getting into that compressed-highlight portion of the gamma curve where they start to look a bit electronic and unrealistic.

This is what happened when I went into Final Cut Pro and applied Tiffen DFX2’s Halo filter. Halo is my all-time favorite digital filter as it emulates the filter-in-the-telecine look: in days of old (this was popular in the late 1980s) there was a trend of putting a glass diffusion filter in the telecine path when transferring film negative, which had the effect of adding a glow to the shadows instead of a glow to the highlights. It also added a softening effect and desaturated the image as well. I’ve backed way off on that look, adjusting the filter so that most of the effect is desaturation with a touch of diffusion thrown in, and the results aren’t bad. I’ve probably gone a little too far in softening and desaturating, but you can see how the flesh tones look a little more filmic and pleasant without being overly saturated.

This next shot was a little more difficult:

Before…

What I wanted to do was to smooth and spread out that highlight so that the edges looked a little less electronic. I ran into a bit of a roadblock, though:

...and after.

This is interesting but it’s not there yet. I used the Halo filter again and combined it with a filter that’s not in Tiffen DFX2:  Skin Smoother. This filter places an adjustable mask over skin tone and then gives it an isolated blur. It’s a start but I’ll have to see this image in motion to see if it really works. (The Skin Smoother filter is part of an old filter package called 55mm, which evolved into the Tiffen DFX filter package.)

What jumps out at me is that I can see a weird cartoony effect in the skin tone highlights that looks to be an artifact of 8-bit compression. Yes, that’s right—the AF-100 records 8-bit AVCHD. As an experiment Jeff Regan brought a KiPro along and we recorded in ProRes as well, but unlike the Sony EX1 and EX3 cameras the HD-SDI output of the camera is NOT 10-bit.

Recording 10-bit out of an EX1/EX3 is a great solution when shooting green screens because the 8-bit internal long-GOP XDCAM codec does really horrible things to motion blur, making keying and rotoscoping very difficult. Recording the camera output to ProRes, however, results in a VERY clean signal that is great for visual effects work.

Not so with the AF-100. The HD-SDI output is 8-bit as well, and I question how good this camera will be for green screen work. We tend to need greater depth of field for green screen work anyway, so I’m always talking producers out of shooting green screen with an HDSLR in favor of an EX1/EX3 so we don’t have to worry about focus. (On larger shoots with Alexas and RED ONEs we tend to have the budget to light for a deeper stop.) It would be nice to have another option for low-budget green screen, but sadly the AF-100 is probably not it.

We’re going to compare the AVCHD and ProRes footage later on and see if there’s a noticeable difference between the two.

As you’ve read this far, you deserve a bit of a reward. The AF-100 did something really odd on this shoot, and while it hasn’t happened to me again I think you should know that it is possible. Watch this clip:

That’s the camera’s internal capping shutter spontaneously engaging.

Black balancing is recommended every time the camera EI is changed, probably because black balancing plays a part in noise reduction. You black balance by holding the white balance button down for a couple of seconds, at which point the camera caps itself internally and does its thing.

This problem seems directly related to having a heavy lens on the camera. As we were the first to use the AF-100 locally, and we were desperate to give it a try instead of using an HDSLR, Shooting Star Video provided a commonly available lens support setup with the full expectation that it would work properly with the AF-100 and the Alura.. Apparently the lens support bracket allowed the lens to sag slightly on occasion. When that happened the capping shutter engaged, sometimes for only a moment but occasionally it stayed closed until we smacked the lens or adjusted the lens support bracket. (Smacking tended to happen during takes, of which only two were ruined.)

I’ve since used this same camera with Shooting Star’s new bracket specifically made to support this lens and it’s worked flawlessly. The current theory is that the lens shifted in the mount in such a way that it pressed against a couple of contacts and triggered the internal capping shutter, but we don’t know for sure what was going on. Just know that a heavy lens like the Alura must be properly supported (and that’s a good idea on any camera).

If this happens to you with an Alura please feel free to smack the lens. It can take it, and it feels good after the surprise of suddenly seeing black through the viewfinder. Then re-seat the lens and check the support bracket.

The great reveal: words scratched into a hood, to be added in post. At magic hour, under low contrast illumination, the AF-100 is beautiful.

Shooting with the last of the day’s light.

I have to admit that under controlled lighting this camera looks great.

Lower end cameras that exhibit a lot of contrast are like slide, or transparency, film: it doesn’t handle extreme contrast well at all, but under controlled circumstances it can yield very pretty pictures indeed. This is true of all the small camcorders, but it is especially true of the AF-100. The limited contrast, the horrible clipping and the inability to record 10-bit from the HD-SDI spigot hold this camera back from being, by far, the best in its class. But given the choice between this camera and an HDSLR—and if small size and form factor aren’t a concern—I’ll pick this camera every time. The combination of reduced depth of field, video camera controls and focus and exposure feedback win over the HDSLR any time.

And I have been picking this camera regularly over HDSLRs. As I mentioned earlier, I’ve been shooting a charity project in a church and getting wonderful results under available and slightly augmented lighting conditions, and I’ve got a corporate talking head shoot later this week where this camera is the perfect choice for the space we’re shooting in and the budget we’re working with.

Every tool has its purpose, but I wish this tool was just a little more versatile. Then again, if it was it would quickly cut into Varicam sales… and we can’t have that. Or Panasonic can’t. (I wouldn’t mind a bit.)

But one has to wonder if a future software upgrade that made clipped highlights more pleasing wouldn’t add a lot of life to this product.

The Shooting Star Video list of reasons to choose the AF-100 over an HDSLR:

(1) Built-in ND’s (a first for a single chip camera)

(2) Time code/external time code (not available in HDSLRs)

(3) XLR audio with manual control, phantom power, headphones, speaker, uncompressed audio recording

(4) Waveform/vectorscope, zebras

(5) False color focus assist

(6) HD-SDI, HDMI and composite video monitoring

(7) Wide variety of lens mounts, allowing for the use of a wide variety of lenses (including PL mount) without modifying the camera

(8) Half the rolling shutter skew of an HDSLR

(9) No aliasing/moire worth noting

(10) The ability to undercrank & overcrank at 1080/60p over 24p or 30p

(11) Extensive menu control of camera parameters (not offered in HDSLRs)

UPDATE: I just finished using the same AF-100/Alura combo from Shooting Star Video today and I carried it all over the place—upstairs, downstairs, indoors and out—for a small charity project I’m helping director Ian with. I had no problems with the internal capping shutter at all.

Disclosure: The Tiffen DFX2 filter package referenced in this article is a review copy sent to me by Tiffen at no charge.

Art Adams is a pretty sharp guy, though his background is often out of focus. His website is at www.artadamsdp.com.

Filed under: Business •Production •Tips •Training •

Career Advice for the Young DP

Filed under: Cameras •Production •Tips •Training •

Phantom Adventures: 1000fps on a Budget

The point of this setup was to show off the laptop, which uses Rambus technology in some fascinating way. (I don’t remember exactly how, but it has something to do with RAM.)

And here’s what it looked like behind the scenes:

High-speed cinematography creates the illusion of objects defying gravity, and the quilt did a marvelous job of making an otherwise dull product shot at 24fps look startlingly beautiful at 1000fps.

We tried something a little different with this setup: we aimed two Maxibrutes into an 8’x8’ silver/gold checkerboard and aimed that through a 12’x12’ frame of full grid. This rig, positioned just off the left edge of frame, washed the set with warm beautiful evening light. We also added some soft warm light through the curtains, in contrast to the slightly hotter and less warm light from the previous setup.

Medium shot of mom.

Mom’s medium shot on the Phantom monitor. Jay has moved his trigger point about 1/5th of the way down buffer time line.

Mom is backlit from over the top of the set, probably with the nine-light Maxibrute, and this light really accentuates the flow of the quilt. The back light brings out texture, and as this shot is all about flow and texture it was important to rake some light across the fabric to enhance the feeling of flow.

The laptop shot was fairly straightforward: the best way to light shiny surfaces is to reflect something in them, so we put a piece of showcard on a stand and reflected it in the laptop’s shiny surface from just outside the frame.

This shot is interesting in another way. Can you see the blue light on mom’s right arm, or on the quilt in mom’s medium shot above? The laptop screen wasn’t on for this shot as the display images were going to be added later, and even if it was on it wouldn’t have been bright enough to interact with mom at all. Part of selling the illusion that the monitor is on is to show that interaction, so we put a 1k Tota light gelled with CTB on a wood tray over the keyboard to create the subtle glow of the screen. (Subtle glow… from a 1000w positioned three feet away??? Welcome to 1000fps.)

This next shot was a bit more difficult. In addition to the silver/gold bounce through grid on frame right we had to bring in a 12’x12’ silk on frame left to light the floor and make it look “normal.” (Otherwise it was a little too obvious where our key light was coming from.) We also tried to light the laptop keyboard to make it pop, so we hung a Firestarter par about 5’ overhead to really punch the keys. At some point I looked down and noticed a crater in the keyboard… we’d melted it! Fortunately director Jono Schaferkotter was able to stage the quilt action to cover up the crater and we got the shot.

I remember holding my hand under the Firestarter, just above the keyboard, and thinking “That isn’t so hot.” Ten seconds later it nearly caught fire! Lesson learned…

Turn the page for an evening out on the town…

Filed under: Cameras •Production •Tips •Training •

Alexa ISO Settings: The Least You Need to Know

Filed under: Cameras •Production •Tips •Training •

Arri Alexa’s Dynamic Range: It’s All in How You Use It

Behold! I bring you ISO 400:

This is where things start to get interesting.

I haven’t changed the T-stop on the lens or the lighting on the chart. Notice on the waveform that the notched chip, which used to be middle gray, is now one stop darker than middle gray (at -1). Chip #6, in the image of the chart above, is now middle gray, a distinction originally held by chip #7. Changing the ISO from 800 to 400 costs one stop of overexposure latitude and shifts all the steps down one toward black.

Making all the stops darker means there should be one more stop in the shadows, but it’s way down there just past -8 and it’s barely visible. What I do see is that noise decreases noticeably:

Noise shows up in a waveform trace as a fuzzy line, and by comparing these two images of the darkest part of the frame we can see that the toe of the ISO 800 trace is fuzzier than the toe of the ISO 400 trace. Not only that, but the black level itself is a couple of units lower. Switching to ISO 400 cost us a stop of overexposure latitude but gave us cleaner and crisper blacks.

Let’s look at the Rec 709 results:

This is a bit more interesting. What I’ve noticed is that step +5 appears in roughly the same position in relation to white clip at both ISO 800 and ISO 400:

Although there’s another stop of overexposure latitude under ISO/EI 800, the distance of stop +5 from clipping is about the same. In fact, when I go back and look at the ISO 800 Rec 709 chart, I can’t tell the difference between stops +6 and +7—and I can tell the difference between +5 and +6 at both ISO 800 and ISO 400. What this seems to imply is that, at ISO 800 or 400 in Rec 709 mode, stop +5 is the brightest stop that will hold significant detail. At ISO 800, stops +6 and +7 are so close together that they will most likely appear the same, and at ISO 400 stop +6 is the clipping point.

It appears that ISO 400 might offer cleaner blacks and less noise, with hardly any compromise in the highlights, when shooting Rec 709. That’s very interesting to keep in mind. But, at the same time, it’s worth noting that maximum white in LogC is down to about 90% (or, in extended mode, just under 100%). As the ISO decreases our “bit bucket” is getting smaller. Will that make a difference in the grade? Most likely not… but it’s interesting to note.

Let’s proceed down the ISO scale to 200:

We’ve now lost another stop of overexposure latitude, and our notched reference is down two stops from where it started at ISO 800. More on that in a moment. First, let’s look at noise levels:

Once again, lowering the ISO gives us crisper and cleaner blacks, as indicated by the ISO 200 trace looking less fuzzy than the ISO 400 trace and the black level sinking a bit lower.

Once again, the “bit bucket” is getting smaller, as maximum white only reaches about 82% on the waveform (85% or so in extended mode). Does it matter that LogC doesn’t allow the brightest tones to reach white in this mode? Not really. The white level will be set in post, requiring that the entire range be stretched out a bit, but 10-bit 4:2:2 or 4:4:4 material should be able to handle that.

Let’s look at Rec 709:

Whereas stop +5 had been somewhat consistent in distance from white clip and separation from stop +4 between ISO 800 and ISO 400, stop +5 is now just plain white, with no detail beyond it. Let’s compare highlight steps between ISO 400 and ISO 200:

In ISO 200, step +4 is about where step +5 was before. It’s interesting to see that the chart shows very clear separation between stops +4 and +5, whereas other ISO’s under Rec 709 crushed the last couple of highlight steps together to the point where stops beyond +5 were almost meaningless. The good news is that highlights and brighter tones will have more contrast and appear snappier as the steps above middle gray are pulled farther apart, resulting in more contrast.

ISO 200 seems like a perfectly viable option in situations where low noise and less than five stops of overexposure latitude are acceptable. One situation that screams out for this configuration is green screen.

Let’s take a look at two vertical charts before we move on to our last ISO test. Here’s ISO 400:

And here’s ISO 200:

We can see by eye how the steps disappear off the dark part of the chart as the ISO decreases.

Let’s move on to ISO 1600, on the last page, for a final surprise…

Filed under: Cameras •GentryMedia Sister Sites •ProPhoto Coalition •Production •

Canon 5D: How much dynamic range does it have, really?

Filed under: Cameras •Lighting •Post Production •Production •Tips •Training •

My Love Affair with Alexa

I watch the monitor as director Ian McCamey directs by radio, which is what you have to do when working with long lenses.

Welcome back to The Embarcadero. We’ll return to Treasure Island in a bit. Meanwhile, the following shot happened just a bit after sunset:

85mm Super Speed, T1.3, 180 degree shutter.

Same shot, LogC.

Look at all that shadow detail in the LogC image!

The following shot occurs later on in the edited montage but was shot around the same time:

85mm Super Speed, T1.3, 180 degree shutter.

Same shot, LogC.

The only difference is that I kept tilting up to reveal the bridge in the latter shot to reveal some shapes that will be enhanced later. The lighting consists of sky “glow” combined with nearby street lights, some of which can be seen in the bottom right corner.

These shots will have symbolic significance in the final project. Right now I’ll just point out that the sky read T0.7 1/2 reflected at the time and we had to darken it and remove some of the blue in the grade to make it match the other shots. A digital grad was added to the top of the frame over the bridge.

We didn’t do anything really spectacular here other than to wedge a high-hat between the raised walkway and a lens case in order to get the lens where Ian wanted it—directly over the left edge of the walkway, to create a strong vertical line that runs into the distance.

Here’s a shot that we did shortly before Colin’s opening shot on the previous page:

35mm Super Speed, T1.3, 180 degree shutter. All natural lighting.

The same shot, LogC.

As I look through most of these shots I think to myself, “Well, not much to write about here. We placed the camera, set a T-stop, and shot. Nothing special.” And maybe that’s what is so special: we did remarkably little to get these shots. We moved quickly, sure—we had to complete all these shots, along with some additional shots not featured in this montage, in a few hours in high traffic areas without legal authority to do so—but we were able to get good results with a minimum of lighting/grip work.

That’s not to say that we should immediately eliminate our lighting and grip departments. Lighting people is relatively easy (once you know how to light faces, which is an art form and craft in itself) but they are only part of the equation; we always need a background to shoot them against, and backgrounds are often quite big. In a situation like this, where the background is the Bay Bridge, Treasure Island and The Port of Oakland, we can’t light it—but the Alexa gives us the ability to use the existing light to illuminate the deep background. We’ll still need to light the talent for mood and beauty, and we’ll need to light the immediate background for mood, but distant backgrounds that are self-illuminating can now add an additional layer of depth to a night exterior image.

Besides, there’s a difference between “illumination” and “lighting”: “illumination” means that there’s enough light for an exposure, but says nothing about the look; “lighting” creates moods and tells stories, and often doesn’t happen by accident but instead by careful planning and execution.

I do hope that this kind of lighting (or un-lighting) doesn’t become standard. Shooting at T1.3 is not a nice thing to do to camera assistants. I had an excellent one in Paul Marbury, but I’d hate to make him, or any camera assistant, pull focus wide open all the time. One should have mercy on one’s minions. I prefer not to shoot wider than T2.8 when possible as that gives focus pullers a fighting chance, and most lenses work better when stopped down a bit. These Super Speeds would have looked a lot sharper if we’d elected to shoot at a T2 instead of T1.3. (A good rule of thumb is that all lenses tend to look their best stopped about 2 1/3 stops down from wide open.)

The nice thing about shooting at EI 800 with the Alexa is that it only takes 12fc to get to that T-stop. So while you can shoot at T1.3 with (gulp!) 3fc of light, at that point you’re doing more damage control than anything else: it’s not about adding light at that point, although it’s probably a good idea; it’s more about getting rid of stray sources that do ugly things to your actors. You never know when a car will turn the corner a half mile away and completely change the modeling on your leading lady’s face.

How lucky: a meter reading near Colin’s face says T1.4, and I have a lens that will do that!

Preparing to shoot Colin walking in front of the bridge. While Ian directs from his monitor and Paul gets focus marks, I take a light reading off the fog cover that’s lit by the Port of Oakland lights on the other side of the bay. It was quite lucky for us that we had that brightly-lit foggy background; otherwise the bridge would have appeared as a dark blob in most of the shots.

The following shot was the one that gave us fits as we had to wait for trains and hope not to be blocked by a crowd of pedestrians:

200mm Nikkor at T2, 270 degree shutter. The dark parts of the train read T1.05, and the background highlights read T1.4 1/2.

The same shot in LogC.

The light on Colin’s face is—yup, you guessed it—a LightPanel Micro held by gaffer Simon Sommerfeld off frame left. Colin started his walk about 25’ away from Simon and exited frame with about 10’ to go. I never thought an LED light would make such a difference over a long distance, but it did.

Shooting the train shot from atop a (not so grassy) knoll.

The following shot was a test of the Alexa’s ergonomics:

85mm Super Speed, T1.3, 180 degree shutter, pulling my own focus. Colin’s face was T1.4 when directly in front of a streetlight.

Same shot, LogC.

Look how much detail there is in the LogC image! I didn’t bother metering that as it was what it was, and I never thought I’d see so much shadow detail. This is the dead of night, with the light coming predominantly from the row of streetlights on the left of frame.

The Alexa’s ergonomics were fine: the camera has a nice little notch for handheld use and it’s surprisingly well balanced for a camera that looks like a large brick. A few more sit-ups per day and this shot would have been a bit steadier, but overall it did what Ian wanted.

The following shot was a LOT of fun:

200mm Nikkor, T2, 180 degree shutter.

Same shot, LogC.

The main light, as always, was a LightPanel Micro, held this time by production assistant Jim Feeley. Gaffer Simon Sommerfeld was sixty feet behind and to the left of Colin holding a flashlight I’d bought for $10 earlier that day. During the shot he swept it through the grass, to great effect.

Just for fun I took a spot reading of the grass in its lit and unlit states. My notes say that unlit read T0.5 and lit read T0.7. It’s hard to believe we were able to pull such detail out of the murk of underexposure.

And now, back to Treasure Island:

200mm Nikkor, T2, 270 degree shutter.

The same, LogC.

While setting up for a vista of San Francisco Ian noticed Paul cleaning water droplets from the lens and checking his work with a flashlight. He thought it was cool, so we shot it.

While it looks like we may have used a grad or a power window at the top of frame to darken the sky, we didn’t. That’s fog. Welcome to San Francisco in August.

200mm Nikkor, T2, 270 degree shutter.

The same, LogC.

Camera assistant Paul Marbury was grabbing a focus mark on the bridge in preparation for a shot of Colin when I heard him exclaim something along the lines of “Holy crap!” We turned around and saw this ship moving under the Bay Bridge, beautifully backlit by the lights from AT&T Park. Ian’s one comment: “I hope we’re rolling.” We were. Paul hit the roll button very, very quickly. (We punched in on this shot in post as someone didn’t know we were rolling and stepped into frame left.)

And that’s about it for our adventure. Ian spent a couple of hours the next morning doing a rough cut and we graded it immediately before projecting it for the first time. Ian is the in process of re-editing and adding some visual effects, at which point I look forward to showing you what this was really meant to be.

I asked Ian for a comment about workflow, as he cut the piece and has a background as a visual effects editor at The Orphanage. He had this to say:

“Aside from producing amazing low-light images that feel genuinely ‘filmic,’ the post pipeline was a dream. I typically cut my own stuff, so being able to start cutting with the files that came directly from the camera without any transcoding was fantastic. And having access to the LogC content for grading—it’s just a simple workflow. Finally someone in the camera world got it right!” -Ian McCamey, director/editor

ARRI Alexa San Francisco Demo Footage Shoot

A Digital Cinema Society co-production

Director: Ian McCamey

DP: Art Adams

Gaffer: Simon Sommerfeld, DCS Northern Calfornia President

Camera Assistant: Paul Marbury

Production Assistant: Jim Feeley

Behind-the-Scenes Stills: Adam Wilt

Talent: Colin Stuart

Production Support: Leigh Blicher, Videofax

Prototype Alexa Camera courtesy of ARRI and Michael Bravin

Join the ARRI Alexa community at ARRI Digital

Grading by Shane Mario Ruggieri

Scratch system by Lucas Wilson/Assimilate

Shot entirely on location in San Francisco without anyone’s knowledge or permission

Art Adams is a director of photography who is not afraid of the dark. His website is at www.artadams.net.

Filed under: Cameras •Tips •Training •

Next Stop: The Last Stop! RED MX Latitude Tests

Filed under: Cameras •Production •Tips •

IR Cheat Sheet, Updated to Version 2

Filed under: Cameras •Production •Tips •Training •

Arri Alexa and Far Red: A Problem That’s Already Been Solved

Filed under: Cameras •Production •Tips •

RED MX IR Tests: The New Sensor is Similar to the Old Sensor

Filed under: Cameras •Post Production •Production •Tips •Training •

Step into the Matrix: What I Learned from Examining RED’s Build 30 Color Science

The big question is this: if the blue photosite sees both blue and green light, but can’t tell the difference on its own, how does the processor know to remove green’s influence from the blue channel? It can’t know just from looking at the blue photosites, because a photosite can only detect light and not color, so the processor must be looking elsewhere. The green channel seems the obvious place to look for information about green in the image, therefore RedColor must be taking information from the green channel in order to remove green from the blue channel.

I set up the following node structure in Apple Color’s Color FX room: the idea was to split the camera’s RGB signal into its individual red, green and blue components, subtract some of the green channel’s signal from the blue channel, and reassemble red, green and blue into an RGB signal again:

Nothing is done to red or green, but blue is modified by blending a negative value of the green signal into it (in this case the “Blend” value was -.7). Since blue photosites react to both blue and green, but the green photosites reliably react only to green, we should be left with a richer blue if we subtract the green signal from the blue signal.

This becomes a bit more clear if we look at a waveform again. The first waveform shows Camera RGB before the node tree, and the second shows Camera RGB after the node tree:

And here’s the difference in the chart’s appearance:

Another side effect of subtracting green from the blue channel is that blue becomes much more highly saturated, which is something I see now in RedColor.

Do I have any hard evidence that this is what RED is doing? No. But I’ve certainly found a viable way of fixing the blue channel in a way that looks remarkably consistent with RedColor.

Don’t expect my results to match RedColor as I’m not doing anywhere near the kind of finessing that RED, or any other manufacturer, does with their proprietary color science. I’m just demonstrating one method of cleaning up a specific anomaly, albeit a fairly dramatic one.

Lets look at some vectorscope patterns. These aren’t perfectly white balanced (I used camera presets in order to avoid tweaking the image data unnecessarily) but we’re just looking at the overall pattern so that shouldn’t matter.

This is Camera RGB, under tungsten light, with no correction.

This is the same Camera RGB signal processed through the node tree, with green subtracted from blue.

This is the same signal processed via RedColor, with no additional tweaks.

Clearly my node tree correction doesn’t make Camera RGB look anywhere near as good as RedColor; I don’t know enough about color science to pull that off. What I notice when comparing these vectorscope images is that my nodal tree correction skews the colors along the blue-yellow axis:

By the way, if there was one color you’d expect to be affected dramatically by this subtraction process, what would it be? My guess is that cyan would suffer the most as it contains equal amounts of green and blue. In my head, I see cyan as leaning toward blue: if cyan is supposed to be a 50/50 mix of blue and green, but the blue photosites add a bit of green so the green/blue mix ends up being 50/70 (because green sees only green, but blue sees blue plus some green), subtracting green from blue should result in a skew towards blue. And that’s exactly what we see on the vectorscope: cyan, in RedColor, trends toward blue.

Cyan is one of the hardest colors to reproduce for any camera so I have to wonder if this “blue filter passes some green” issue isn’t fairly common. In the early days of color television it was difficult to make a red phosphor that glowed as brightly as the green and blue phosphors, so rather than dim the green and blue phosphors to match to match the dimness of the red phosphors the solution was to add green to the red phosphor to help it glow brighter.

This process worked but it resulted in orange-ish reds that plagued color television for quite a long time. I have to wonder if there’s a similar issue in creating blue dye filters for photosites: silicon is more sensitive to long wavelengths of light (infrared and red) and least sensitive to short wavelengths (blue) so perhaps manufacturers intentionally use greenish-blue filters on blue photosites to boost the otherwise weak blue signal.

Initially I thought this was the end of the story: I figured out how RED removed green contamination from the blue channel—huzzah!—but I was still confused by the fact that the vectorscope that resulted from my bit of Color FX digital magic still didn’t look anything like RedColor. That’s when it occurred to me that I’d only seen the tip of the iceberg, and the only way to figure out how all the RED’s colors interact was to (gulp!) try to build my own color matrix.

So that’s what I did, and I learned a ton. More on the next page…

Filed under: Cameras •Lighting •Production •Tips •Training •Visual Effects •

Lighting Against The Wind: The Making of a Mime Music Video

The lighting was fairly simple but took a while to set up as the technical crew consisted of myself and Ted Allen, my intern. Originally I wanted to use a Leiko to create a spotlight effect, but the Leiko we found at the studio didn’t have an iris and I couldn’t make it project a pool of light that was the right size. Instead I used a 1k Baby on full spot, with a couple of wire doubles in it to balance it with the background.

I have a healthy respect for the old masters of hard light as it can be very difficult to use, especially in color. Black and white lends itself to hard light because it’s automatically an abstract medium; our eyes expect color images to look a bit more realistic, and although there are many hard light sources in the everyday world they often blend together into a low contrast ambiance. A busy street at night is a great example of this concept: street lights, shop lights and headlights create an environment that feels like night, and contains a lot of sharp shadows, but often doesn’t contain rich, deep blacks or areas of severe contrast.

It’s difficult to light faces well using hard light. The higher the contrast on the face, the more precisely the lights must be placed. It’s very easy to make people look bad with hard light. Contrast is a key player, which means fill light plays a major role. In these images from Indiana Jones and the Last Crusade, note that while the key light is not placed “perfectly” the reduced contrast created by the fill light smooths the rough spots on faces:

The first three Indiana Jones movies were shot by the late Douglas Slocombe, BSC, who was the last reigning master of classical hard lighting. Whenever I want to study hard light I pull one of these films off my DVD shelf.

Classical hard light modeling illuminates the face from above and to one side with the intent of casting the nose shadow along the “smile line,” an invisible line that connects the corner of the nose to the corner of the mouth.

Classical hard lighting, with the nose shadow extending down the smile line.

I wasn’t terribly worried about lighting Nanishka with hard light, as her face is beautiful and can take light from nearly any angle, but I still wanted a properly placed fill to help her out during those moments when her head was turned at an odd angle to the key. It’s possible to compensate for an unfortunately-placed key light by placing the fill light in a pleasing position; often the placement of the fill light is more important than the placing of the key.

To get the look I wanted I knew the fill had to come from a large-ish source placed as close to the lens as possible. To that end I placed a 2’x3’ bounce card directly below the lens and hit it with a floor-mounted Baby. The combination of source size and placement near the lens gave her skin a luminous quality while also reaching under the brim of her hat and creating large eye lights.

Soft light from below does wonders for eyes and hat shadows.

Fill from below looks very real and natural, as if the light sources at the top of the frame are creating ambience by bouncing of the floor. Bouncing off the floor, however, has a very different look, and the steeper the bounce angle the more the talent will look “underlit.” Sometimes this is good, but when lighting for beauty instead of mood it’s better to fill with a large source near the lens rather than near the talent.

The effect of the fill source alone: Nanishka on the right has ducked out of her key light.

Occasionally a fill source placed beneath the lens will cast shadows upward onto walls and other background objects, but if the fill source is big enough these shadows will be very soft, natural and realistic. Sunlight striking a floor through a window will result in exactly these kinds of shadows in a room, and while we often don’t notice them consciously the subconscious picks right up on them.

The exposure was driven by the background: two Babies lit the black curtains and a 200w Inky on an overhead C-stand arm lit the background bench from above. This Inky really determined the F-stop as we weren’t able to rig a Baby on the C-stand arm due to weight. Exposing for the background bench meant pushing the ISO and opening the shutter to eliminate perceived noise, but after seeing most of that noise disappear in post I’ll keep the shutter closed down and increase the ISO in the future.

Two more Inkies served as backlights, one for each mime position. A 4’x4 tube Kino Flo with Lee 122 “Fern Green” lit the green cards that covered the bench (so that Nanishka could be inserted later, in her third on-screen role) and another Kino Flo hidden behind the bench cast a glow onto the curtains from below.

The foreground lighting setup.

I had Ted follow Nanishka with the “follow spot” Baby during her performance, and all the other lighting stayed the same to ensure consistency between the different performance passes.

Ted Allen plays follow spot operator when he’s not working as camera assistant, electrician or grip.

While our little video appeared to be the most professional entry in the Pink Martini contest the band apparently preferred a more rough and home-grown approach. We didn’t even make the top five. Although we’re all disappointed that our hard work didn’t result in a decisive win, place or show, we’re quite happy to have created a cute little art piece that has a special place on our reels.

This is what happens when you let a DP touch his own lights. The Leiko mounted to the post is not working, nor are any of the other lights in the shot beyond the Baby at the top of frame, the Baby aimed into the bounce card below the lens, and the green Kino Flo behind Nanishka.

You can read the director’s account of the project here.

All behind-the-scenes photographs were created by, and are property of, the extremely talented Brooke Duthie, and used with his permission.

Stills from “Indiana Jones and the Last Crusade” are used for education purposes only and are the property of Paramount Pictures.

Music: Pink Martini

Director/Editor: Ian McCamey

Producers: Ian McCamey & Nanishka Camberos

Director of Photography: Art Adams

Dancers/Mimes: Nanishka Camberos

Story and Choreography: Ian McCamey & Nanishka Camberos

Choreography Consultant: Suzette Slaughter

Makeup/Wardrobe: Maria O’Reilly

Camera assistant/electrician: Ted Allen

Production Assistant: Whitney Kahl

Studio: The Producer’s Loft, San Francisco (Special thanks to Vic Ferrer, Owner/Benefactor, and Benoit Lacasse, Stage Manager Extraordinaire)

Art Adams is a director of photography who lights like a mime: silently and with practiced skill. His website is at www.artadams.net.

Filed under:

A Cine Gear Find: TechScout Touch

Filed under: Cameras •Production •

GearNex: The Next Generation of Gear Head

Filed under: Cameras •Editing •Lighting •Post Production •Production •Training •Visual Effects •

VFX Tell the Story in California State Fair Spots

Gaffer Alan Steinheimer examines the lighting while key grip Joseph Scott removes the green screen from the set.

I was quite excited about using the new RED color science, but I was disappointed to see that, once again, a RED software change meant dealing with an entirely new camera.

There are two ways to determine a camera’s EI: the film method, and the video method.

The film method is based on determining where a film stock picks up enough density to create an optimal image. The point where the density becomes acceptable determines the bottom of the straight line portion of the gamma curve. (I’m not sure I’ve got that totally right, but one thing I can count on when writing for the web is that someone will correct me in comments.)

The video method looks at where middle gray falls: if one lights a gray card and then exposes the gray card for 18% gray (usually 42-45 units on a waveform monitor) the resulting combination of shutter speed and aperture will suggest the EI of the camera at its current settings. In this case the EI of a WYSIWYG camera has been predetermined by the manufacturer.

RED uses the film method in determining its EI of 320, basing that number on a “sweet spot” that offers the best compromise between highlight detail and noise. In my early tests I used the video method, looking only at RAW mode and placing an evenly exposed 18% gray card at dead center on the histogram, and came up with an EI of 160. I rated the camera at 320 on my very first RED shoot and was quite unhappy with the amount of noise I saw, and on every other shoot since I’d rated the camera at 160 and enjoyed a silky-smooth and noise-free look. I lost a stop of highlight latitude but I almost never missed it.

In the past I’d manipulated the output settings (changing View>Exposure to -. 4) to darken RedSpace’s gamma to match RAW’s gamma, so that if I toggled between one and the other the only difference I saw was that the image became desaturated in RAW view. The client saw a WYSIWYG image at my preferred EI, which if uncorrected would have looked a stop too bright. I relied almost entirely on my meter to set the stop, and verified each exposure by briefly toggling into RAW.

This time, when I fired up the camera running Build 30, I discovered that RAW’s gamma no longer tracked the same way. The only two options were RAW and REDColor (fine, no problem with that) but now RAW’s gamma was nearly exactly the same as RedColor’s! Toggling into RAW didn’t darken the image anymore; instead, both RAW and RedColor showed an effective EI of 320!

It was common knowledge that RED was applying some sort of gamma correction to RAW before Build 30, but now RAW changed so completely that (1) I clearly wasn’t seeing “raw”, and (2) I couldn’t rate the camera at 160 anymore as I couldn’t verify my exposures by looking at the image in RAW mode. Previously, if something looked too hot in RedColor and EI 160, I could toggle into RAW and reassure myself that none of the channels was clipping. For the first time in the two years since my first RED shoot I was forced to rate the camera at 320.

The good news is that the RED’s shadow noise has been much reduced, and while I saw some patches of noise on the monitor during the shoot it almost completely went away when we viewed the footage in post on a broadcast monitor. Any noise you see in the final spots is due to the addition of artificial film grain in After Effects.

As for REDColor… I’m a big fan. It’s a bit too saturated for my taste but it worked well for a spot aimed at kids. I know that on future shoots I can desaturate the image in the View menu without sacrificing underlying color data. What most impressed me was how the new color science handled tungsten light under a daylight white balance.

Tungsten light has a lot of red in it, which is what makes it appear warm, but it also has a lot of green. If you look at a vectorscope you’ll see that orange and green are adjacent, and it’s very easy for a saturated orange to “tip over” into green. I’ve seen this happen on both film stocks and video cameras when shooting saturated warm elements like firelight or raw tungsten light under a daylight white balance. REDColor handled the warm tungsten crate light wonderfully, rendering it as a very natural and pleasant warmth.

I should mention that I shot this entire project on daylight preset white balance, and we didn’t do much to the color during the grade. We tweaked the contrast a bit but the color is as the camera saw it. Very saturated objects of known color, like skin tone and grass, can be tough to shoot because odd color shifts are readily apparent. Desaturated images are much easier to photograph as odd color shifts are less obvious and more easily forgiven. This spot called for punchy colors and Build 30’s color science delivered exactly what we needed. (I owe a big “Thank you! ” to RED’s Graeme Natress for his constant efforts to perfect RED’s colorimetry.)

Well, that’s it. Suffice to say that I love my job, I love going to work, and I love solving artistic and technical challenges like the ones we saw here. When working with a talented director and a great crew it’s easy to deliver more bang for the client than the budget should allow.

Best boy electric Ernie Kunze models Poppy’s wardrobe.

Director: Ian McCamey

Producer/AD: Tom Ruge

Production Company: Rough House

Agency: Runyon, Saltzman & Einhorn

DP: Art Adams

Gaffer: Alan Steinheimer

Best Boy Electric: Ernie Kunze

Key Grip: Joseph Scott

Best Boy Grip: Cliff Henry

Camera Assistant: John Gazdik

DIT: Jay Farrington

Camera equipment by Chater Camera

Post by Rough House

Editor: Marc Cebrian

Compositor: Kit Klangsin

RED ONE, 4K HD, 23.98p (plus 4K 16:9 23.98p and 3K 30p)

EI 320, Software Build 30

Zeiss Super Speeds (18mm, 35mm, 50mm)

No filters for interiors; Schneider Tru Cut IR 750 and Tru-Pol for exteriors

Art Adams is a DP who has both a barn to shoot in and a mother who can sew. His website is at www.artadams.net.

Filed under: Cameras •Lighting •Tips •Training •

Everyone Looks Sexy at 1000fps

Filed under: Cameras •Lighting •Production •Visual Effects •

The Tiger Lillies Finally Set Sail

Filed under: Cameras •Production •Tips •Training •

Random Tips from a Professional Camera Operator

Filed under: Cameras •Production •Tips •

Two New Sharp-Looking Charts from DSC Labs

Filed under: Cameras •Lighting •Production •Tips •Training •

The Making of an Epic Media Project

THE BRITISH BANKERS CLUB, MENLO PARK

This was a fairly simple lighting setup:

The flag at top left keeps hard light from the HMI’s lens from spilling into the shot. Before rolling I always stand in the middle of the set, on the talent’s mark, and look around for light sources that shouldn’t be there. HMI lenses frequently leak hard light through barn doors, and occasionally a shiny barn door will act like a mirror, reflecting light from the lens into the shot.

Our talent consisted almost entirely of company executives with busy schedules, so the trick was always to get the lighting in place before they arrived and then change it as little as possible while shooting, in order to maximize our coverage options. If most of the lighting stays the same, and all that’s left is moving the camera around and adding a little fill for closeups, it’s possible to get a lot done in a very short amount of time.

In this case most of our lighting comes from a 4’x8’ bounce card, sitting horizontally on the floor, to sidelight the upstairs lounge of this restaurant/bar. Placed about 6’-8’ away from the talent, the 4’x8’ source wrapped nicely around their faces. No fill was necessary.

This entire setup was lit with HMI’s. I think we have an 800w Joker bouncing into the 4’x8’ card, with another HMI PAR bouncing into a 4’x4’ card around the corner to light the background. (You can see its effects on the pillar behind the blond gentleman.) There’s also a PAR off to the right augmenting the background ambience and bringing the back wall up a bit in exposure.

As this setup was on the second floor of a building that had no elevator, and given that I had a grip/electric crew of two for the day with no dolly grip, I pitched Jono on the idea of using a small 4’ slider for our dolly moves. By mounting it between two apple boxes I was able to do some very long, slow and smooth moves in a cramped space without forcing my small crew to carry a heavy dolly upstairs. It worked out very, very well.

Sliders are designed for short, super smooth moves on small tabletop shots, but they work well in cases where the budget or the location decide against the use of a normal dolly. I’m not sure who made this slider but this page gives you a better idea of what they look like.

Originally I exposed this shot at EI 160, but with the camera set to EI 320. I’d discovered a while back that setting the camera to EI 160 yielded an image that looked too dark on the monitor, and while the RED is not a WYSIWYG camera most clients tend to treat it that way. By leaving the meta data EI set to 320 and overexposing the image, the monitor image looked okay although the highlights were a little toasty and were meant to be fixed in post.

Here’s the original exposure:

And here’s the color graded shot:

(This still is a little darker than it is in the video. I think there was some sort of color or gamma shift when converting YUV video into an RGB still.)

Later on in the shoot I figured out how to work around this exposure issue, with a little help from my friends. More on that shortly.

RAMBUS CUBICLE FARM, PALO ALTO

Cubicles can be a bit dull, so I decided to try something new in lighting this one: we laid Kino Flo tubes along the back edge of the cubicle wall and left them in the shot. The traditional way to light a cubicle is to poke lights over the walls, but that felt a bit boring to me. This project was not served by simple illumination. It had to look cool as well.

I love leaving lights in the shot when it’s appropriate. We save a lot of time by not having to hide them anywhere, and they give the shot a cutting edge look. You’d never be able to work in a cubicle lit this way, but it looks great onscreen.

The light in the back hallway is one of ours, and we created little pools of light in distant cubicles that aren’t seen in this particular take. I don’t think we’re using any fill other than some cool 5600k light on the outside wall with the nameplate.

I shot this scene with a Schneider 1/2 CTB filter, to correct the raw tungsten light halfway to daylight for better skin tone and color rendition. By splitting the difference between tungsten and daylight, and using both kinds of light in the shot, I’ve found I can create a pleasing warm/cool contrast very quickly.

We did some closeups of one of these two gentleman that can be seen in the behind-the-scenes video but that aren’t in this edit. I believe his closeup will be used in a series of spots that are currently being cut from this project’s raw footage.

Our next meeting is in the conference room, just down the hall…

Filed under: Cameras •Lighting •Post Production •Production •Tips •

Low-Budget PSA’s, Shot on RED, Prove that Budget is Not a Barrier to Excellence

Here’s the first shot of “Lineup”:

And here’s the first lighting setup:

I exaggerated an old moonlight trick to create a stylistic industrial look: some of the prettiest and most interesting blues have a lot of green in them, and a common recipe for night lighting calls for 1/2 CTB and 1/4 plus green gels for a cool-but-not-pure-blue feel. Another variation is to use 1/2 CTB and the old arc correction gel White Flame Green, which creates a silvery moonlight that can be stunningly beautiful.

In this case I wanted a not-so-subtle mercury vapor lighting look, so we swapped all the Kino Flo tubes to daylight (5600k) and put 1/2 plus green on the unit itself, along with some diffusion (probably Lee 250). By setting the white balance at 4300k (not 3200k—ignore the white balance temp in the diagram) I not only compensated for my Schneider 1/2 CTB filter (which converts 3200k light to 4300k) but the 5600k light goes a little cool and the 3000k light goes a little warm. I’m white balanced exactly between them.

If the woman were to look up she’d see the Kino Flo directly over her head—which means that her face would get little front light from it. The cop in the background, on the other hand, would see the unit stretching out in front of him. That’s why he is lit very softly, with a lot of blue-green light wrapping around his face, while her face is mostly lit from inside the room.

Using lighting in this way helps to define a 3D space in a 2D medium and is a very easy way to create interest. Even though we can’t see it there’s clearly a light over them: she’s in front of it in space, and he’s behind it.

Lighting the foreground warm and the background cool also creates a classic warm/cool color contrast, which is a very interesting look. Warm colors tend to attract our attention and make objects seem closer, while cool colors tend to reduce interest and seem farther away.

The lighting effect is most dramatic when the woman steps up to the window in complete shadow, at which point the inside lights come on. (In this edit that light change is hidden by the fade up at the beginning of the spot.) I had a 4’x8’ bounce card rigged inside the room, laying on the desk inside with two lights aimed at it. (It’s the same room used for the last shot of the previous spot, “Perp,” and we’re just looking into the window that the previous character was looking out of.) The 650w fresnel spotted into the card is there to provide a little glow to the inside of the room and provide a hint of light to the woman’s face as she enters frame, while the 2k (operated by a film student I’ve been mentoring, Ted Allen) provides most of the light when she lands on her mark. The subtlety of the 650w was lost in the edit, which happens quite often; the important thing, though, is to try to add those little touches when possible because we never really know how the edit is going to turn out until it’s done. It’s better to add a nice little touch and not see it used than to miss an opportunity and then stare at that gaping creative hole every time you see the finished product. (Such touches, though, are easily sacrificed if the schedule doesn’t allow for them. Fine brushstrokes are nice, but broad brushstrokes are most important. The faster you can paint with the broad strokes, the more time you’ll have for the fine strokes.)

There’s a piece of black foam core leaning against the wall on the other side of the window to make the reflections more noticeable. We also enhanced the window reflection in post by selecting it with a rectangular window in Color and increasing the exposure inside it.

Here’s a slightly different angle:

You can see the subtle intermixing of colors on the actress’s face. I really like that look. The cop’s face in the background is a nice contrast. You can also see the sharpening we added in Color to enhance our gritty look. (This was shot on a 135mm Zeiss Ultra Prime, probably at T2.)

Here’s what the setup looked like:

The nicest wide shot included some plasterboard wall on the side of the window toward the camera, so we put Duvetine up against the side of the window to make the wall go black. You may notice that the tilt angle of the camera alines with the actors’ heads. (The piece of paper on the wall is a lighting diagram left over from my Tiger Lilies shoot on the same stage.)

Our other setup showed the actual lineup, which included a couple of actors from the previous spot plus a crew member. Ian originally sent me this still to illustrate the look he wanted:

I took one look and knew exactly what to do:

Soft light from below can be very flattering, but in this case we over-sharpened the image and cranked up the contrast to make it look a bit sinister. Here’s a behind-the-scenes still:

My film student, Ted, is on the left. (He was drafted at the last minute because he fit a certain age demographic we didn’t have cast.) And here’s a color-graded screen grab:

As you can see, Ian and I like vignetting our images. (The vignette looks a lot more subtle on a video monitor than it does here.) Don’t let anyone tell you that you can never light “flat” by putting a big soft source near the camera. Depending on the size of the source, what side of the lens axis it’s coming from, and how far off-angle to the lens it is, it’s possible to create a large number of very interesting, rich and subtle looks.

The trick with a lighting setup like this is to light the bounce cards as evenly as possible so there’s one shadow, not two. If the lights are too spotty or too close to the cards they’ll create two hot spots, one at either end of the card, which results in double shadows. There’s a little bit of a double shadow in the midst of this shot but it’s not horrible.

Ideally we’d have evenly lit the cards with a row of Lowell Tota-lights placed in a row on the ground, but as we didn’t have any we punted with the 2k’s. It worked fine. It’s important to understand what’s important and what’s not important in a lighting setup, and in this case we were able to get 90% of what was necessary quickly with two lights instead of five or six.

We did the dolly move on a doorway dolly, which was not the best method possible but it was what our budget allowed. We really needed track to make the move useably smooth, but as we didn’t have any we used the unsteadiness of the move to simulate the woman’s POV as she looks down the lineup.

One last note: we shot this spot in 16:9 but Ian cropped it to 2.4:1 “scope” letterbox during the edit. Normally I don’t like having my work reframed in post, but I have to admit that it looks great. There are exceptions to every rule, and the rule of “don’t touch my footage!” sometimes gets in the way of someone else’s creativity making the piece better. I don’t encourage this, but I also take credit when it works.

That’s about it. Very simple overall, although it took a while for me to learn to light that way. Lighting simply pays huge dividends: fast and pretty are popular modalities both in cinematography and in dating, and both can result in a lot of callbacks—albeit for different reasons.

All original photos are copyright 2009 their respective owners, who in this case are Tim Blackmore/Meets the Eye and Luke Seerveld.

Art Adams is a DP who is fast and pretty, but in a very different way from how he was 20 years ago. His website is at www.artadams.net.

Filed under: Cameras •Lighting •Pre-Production •Tips •Training •

A Tale of Forbidden Love, Shot on RED

This sequence was shot on day one at a location halfway across San Francisco from the first day’s location.

This was the project’s very first shot, and it’s a shot I really enjoy from an operating standpoint. I really enjoy how the animal control officer leans out from behind the cop, but I love the primarily vertical move that follows the cop as he approaches the door. During rehearsal I found a frame that worked for the end of the shot, but during the move I didn’t frame to accommodate the animal control officer at all. I framed the cop so that he would land in the right place for the end composition, and then I let the animal control officer land in his proper place without moving the camera.

By framing for the final composition early, as the cop advances, and leaving the frame unbalanced in anticipation of the other actor stepping into his final mark, I created a much stronger shot than if I’d moved the camera around and tried to keep everyone in the frame. Creating an unbalanced composition creates tension, and letting the actors settle into their proper positions over time, without constantly trying to find a balanced frame, gradually releases that tension. I thought this subtle touch helped create more tension in an already emotionally tense scene.

Compositions don’t happen only in space; they also happen over time. I think this is a great example of a composition whose every frame may not work as a still but that pays off over the course of the shot.

Half of camera operating is knowing when NOT to move the camera.

The only lighting in this shot is a bounce card taped to a window just off frame right. You can see it light the cops cheek and give his face a touch more modeling as he approaches the door. I love soft light sources that interact with actors and objects as they move through a space. It creates a greater sense of space and depth in a 2D image.

During the location scout I discovered that the sun would set directly down the entryway of the house (thanks to Suunto and Sun Path). We decided to start shooting at around 2pm and do all of our east-facing shots while the sun was too high for front light but was perfectly positioned for backlight. All of our shots facing out of the house (the cops, looking out the house door, the woman on the street) were shot between 2-4pm. When the sun was low enough to provide flattering front light on faces we turned around, at about 4pm, and shot the rest (the wide shot of the house, the lead actor waving at the woman, the lead actor in the doorway with the dog, etc.). We shot our way out of the house early in the afternoon and then worked our way back in as the sun went down.

We really had to race, though, because while the sun set in such a way that it lit the front door of the house until it was gone, the rest of the street fell into shade very quickly.

For example,

This was among our last shots of the day, and you can see the sun is nearly gone. It’s hitting the actor but not the dog. A rectangular Power Window around the dog allowed us to brighten the dog enough that the lack of sunlight striking him isn’t really noticeable.

This was our second shot of the project. The hot streak of light on the dog is pretty cool, and it’s a complete accident. It’s reflected off an outside window. I added a little bit of fill here by bouncing a 1200w HMI PAR into the wall and ceiling over the camera. The idea was to lift the shadows just enough to see what was going on, but not enough to feel lit.

If you look at the top of the actor’s arm you get the sense that the light was coming from above. Ideally the light would have come from below the lens, as that’s where reflected sunlight would have normally come from, but we didn’t have the technology to get that much light into that small a space. The 1200W PAR was placed around the hallway’s 90-degree turn into the living room as there wasn’t enough room in the hallway for myself, the camera, the camera assistant and the talent.

The direction of the fill light isn’t that critical here, though, because the dimmer a light is the less you can discern its direction and quality. That’s not to say that you can be sloppy, but fill lights or underexposed key lights can be cheated more easily because they create less contrast than bright sources do. Contrasty scenes require more precise light placement than low contrast or dark scenes.

The little bit of light reflecting onto the underside of the actors face, and giving him a little bit of facial modeling, is from a 2’x3’ bounce card laid on the ground just outside the door, in the sunlight.

While we had sun for most of this shot, the fog was starting to roll in and the exposure changed a bit during this take. Our colorist added a ramp to even out the worst of it, and he did a great job. This shot was done entirely with natural light, and worked out so well because we calculated the sun’s path in preproduction and opted to take advantage of it.

This shot, too, was all natural light. For the first take I tried to light up a little of the hallway, just to open the shadows a little, and it didn’t work at all so I turned the light out. The shadows on the actor’s face are natural, a result of the number of power lines crisscrossing the average San Francisco street.

We did this shot at 2k, 120fps just as we were losing the sun off the house. The colorist created another ramp that brightened the dog as he entered the shade. If I were shooting with a baked-in camera I’d try to do a stop pull, but on the RED it’s better to let the exposure go and track it in post, where it can be finessed. This assumes that the shadow portion of the shot isn’t horribly underexposed, which could result in a change of noise quality as the shadows are boosted. I pushed the exposure as far to the right as possible in order to ensure good shadow density.

When I first watched the 2k footage in REDCine I was worried about how soft it looked. I’m happy to say that Spy Post’s processing of this footage eliminated most of that softness. The difference in sharpness between this shot and the others was minimal in the telecine suite, and that’s where it looks the best it ever will.

This sequence was the last east-facing shot of the day. We shot everything facing outward, toward the cops and moving out of the house into the street, and after these shots we turned around for our big wide shot and worked our way back in.

There’s no fill on this other than sunlight bouncing off the row of houses lining the street—and that ambient fill looks quite good.

We rolled on the dog for several minutes and just let it do what it wanted to do. That’s the advantage of shooting HD: you can roll and roll and roll and not worry about missing that perfect moment.

This is our wide reveal shot. The shaving cream and toilet paper are real; the spray paint was added in post. The producer was tempted to create actual graffiti but elected not to as it was his uncle’s house and he didn’t want the neighbors to gossip.

We’re starting to play “beat the sun” in this shot. You can see it’s already creeping up the sidewalk and we haven’t started getting our reverse angles yet. Fortunately none of the reverses required much, if any, lighting—something we discovered, and planned for, on the scout. You can work really quickly if the lighting is in place and all you have to do is move the camera around.

There’s a small bounce card just off frame left, but that’s it for lighting. The sun is in a great position, and while it’s not quite getting into the actor’s eyes his face is modeled very nicely.

I love this frame. It’s dynamic and it carries him all the way from the front door to his end position without a single framing adjustment.

I was truly fortunate to have a location that offered such beautiful natural light, along with a producer and director who were happy to plan our schedule around my lighting needs. As a result we moved very, very quickly, and completed this sequence in about six hours. That might be considered a bit slow normally but as my entire camera/grip/electric crew consisted of a camera assistant and two PA’s that were intermittently available I think we did remarkably well.

This was a very satisfying experience, not just because the cast and crew were a lot of fun but because we did some nice work very simply and quickly. I used to love assisting and operating on episodic television shows because it never got boring: I was always moving, always thinking, always trying to figure out how to squeeze a little more out of the shot. At the end of the day it was fun to look back and think about how much shooting we’d done in such a short time, and how nice it looked overall. That’s how I felt at the end of this shoot: proud that we’d done such nice work so quickly and with so little.

Thanks are due to my camera assistant Jamie Metzger, second day gaffer Chris Galdes, and especially to the director and producer who I won’t name yet as they don’t want this short film popping up in Google searches while they’re submitting it to film festivals. (They’re listed in the credits.) Thanks especially to colorist Cary Burens of Spy Post for doing a great job grading this project.

Art Adams is a DP who works doggedly to improve his reel. His web site is at www.artadams.net.

Filed under: Cameras •Lighting •Production •Tips •

Anatomy of a Spot: Commonwealth Club

Filed under: Cameras •Lighting •Production •Tips •Training •Visual Effects •

World’s Only “Death Oompah” Band Gets Virtual Reality Music Video

There’s so much to shoot on a project like this that it’s crucial to design lighting setups that can be used for a lot of different things. The hanging and swinging Vista Beam worked well for the end shot of the video, which is a night exterior lit by “tungsten” lamps (although all the additional lighting was daylight balanced) and for the ship’s interior. It was a fairly simple setup, and for a while the goal-posted Vista Beam lived alongside our brighter, bigger daylight setup, allowing us to switch between them as needed.

The vast majority of what we shot, though, was lit by the big broad daylight setup, and once set up it never changed. We just moved the camera around as fast as we could and focused on shots and performances instead of lighting.

Setting up for a high angle on the ship’s captain, who’s desperate to shoot an albatross—possibly over a bad debt or excessively high phone bill. [TB]

For example, we had a lot of shots of the ship’s captain searching for, and eventually shooting, an albatross. First, the search:

The ship is a model that Mark shot in his studio using a home-built motion control rig. The lighting isn’t a perfect match, but Mark says that the entire image will be a lot softer by the time he’s done with it. Here’s an example of the look he’s going for:

This is a temporary comp, where the captain was shot with the RED and a stock “eye” was dropped into the end of the telescope. We re-shot this on our last shoot day (day three of three) with the telescope sweeping from side to side, along with a shot of the actual actor’s eye.

The captain aims at the albatross, before…

and after:

The captain takes a break after a hard day of hunting albatross and beating his crew:

And a rough comp:

The green screen on the floor was a scrap, as we didn’t want to mess up the pristine and heavily-discounted green screen in the background.

What you don’t see in the shot above is the crew floating just under the surface of the water. Rather than stay aboard and tolerate their abusive captain, they offed themselves. (I felt like that the one time I worked on a reality TV show.) They were shot separately, in a small pond built in a corner of the stage, by the stills crew, so they’ll appear perfectly motionless under a layer of animated water.

Left to right: Director Mark Holthusen, me, and the ship’s captain. Adam Wilt is hidden behind my girth. Blond hair by Alan Steinheimer. [TB]

The captain lines up for a shot, first against green:

[TB]

...and then for real:

Mark and his crew had a great method for figuring out whether an angle would work: I’d line up a shot and they’d shoot a still of it off the director’s monitor, a 17” Panasonic LCD. Then, using Adobe Photoshop, they’d comp it into a still of the background plate. I seem to have a latent talent for determining the correct camera angle from rough story boards: Mark would hand me a storyboard page, I’d place the camera, his gang would check my shot against the background plate, and nearly every time I was dead on.

Although I used an O’Connor 2060 for the high shots off our scissor lift, I used a Gearnex gear head for everything else.

I really enjoyed working with it, although I don’t think it has yet reached perfection. There should be a bit more of a difference between the two gears (one is twice as fast as the other) and the head can bind a bit when a heavy camera is tilted all the way down or up. Also, the size of the cradle is so small that it pays to make the camera as short as possible—in the case of the RED, that’s aided by mounting the battery and hard drive off the side of the camera instead of off the back.

For the most part, though, it was a joy to use, and I operated a number of shots with it, including closeups. There is definitely something unique about the look and feel of a gear head, and I’m thrilled that someone has finally found a way to make such a head that is affordable to the HD crowd. I’ve had very few chances to use one over the course of my HD career, and I miss them terribly.

And last but not least, let’s talk a bit about the RED ONE. I really do enjoy using the camera, in spite of its many shortcomings, but it is still a major cause of headache and heartache. While ultimately not that big a deal, as we didn’t lose any footage, we had a number of hard drive faults that prevented us from rolling on a number of occasions. It only took a few seconds to clear the fault, typically by hitting the record button a couple of times until the camera rolled again, but it certainly does make one’s heart sink to see that big red warning message pop up in the viewfinder. Fortunately it’s fairly fast to play back the last clip in-camera to ensure that it actually exists.

There seemed to be no reason for it. The fault never happened when the camera was moving. Strange.

Then there was the saga of the RED 18-85 zoom. This is, in theory, a great zoom: it encompasses the entire range of a normal prime lens set, and it opens up to a T2.9, which is quite handy when shooting with a slow camera like the RED. Unfortunately accurate focus marks are an as-yet unrealized upgrade option: when horizontal, the markings on the RED 18-85 lens that we used were fairly accurate, but they drifted severely when the camera was tilted at a severe angle. For example, while shooting off the scissor lift at a 45-degree down angle and with the zoom fully wide at 18mm, eye focusing on a subject 20’ away yielded a distance of 50’ on the lens. Under such conditions one can’t zoom in, focus and zoom out as the focus doesn’t track properly; one has to pick the proper focal length and use the RED’s focus-checking digital zoom to verify sharpness.

We shot a fair amount of high-angle footage from a scissor lift, and the zoom was a lifesaver as it saved us a lot of scissor lift drive time, but the lack of focus integrity made me nervous.

Focus, focus, wherefore art thou, focus? [TB]

Adam Wilt’s employer owns three RED cameras and two RED 18-85 zooms. Adam told me a story about how one of the zooms, the one we used, was fairly accurate out of the box—at least when fairly level—but the other was completely off. Focus marks weren’t even close to accurate. He sent it back to RED for adjustment, and then called a while later to see how it was doing.

“We don’t find anything wrong with it,” said RED.

“But the focus marks don’t line up at all!” said Adam. “How can this be okay?”

“It’s within our specs,” said RED.

If RED isn’t able to repair it Adam may have to dedicate one of his company’s RED cameras specifically to that lens, mis-adjusting the back focus in order to make it usable. But I guess one can’t complain too loudly. The lens costs 30% of what a real lens would, and you get what you pay for. It is usable, but it’s not camera assistant friendly. For what we needed, it was fine: I needed to find the proper camera height and angle first and the proper frame size second, and having a zoom for those high-angle shots sped us up considerably. Fortunately we didn’t have to follow focus on any of those shots.

It’ll be interesting to look at some still frames down the road and see how the RED zoom compares to our earthbound lenses, a set of Zeiss Ultra Primes.

Left to right: Adam Wilt, me, producer Jason Santos. [TB]

I can’t close this article without saying some nice things about Adam Wilt. As well as being a friend, and my local guru of all things electronic, he served as my camera assistant for this project as a condition of using Meet’s the Eye’s RED package. I was a camera assistant for about five years and I consider myself to be somewhat “old school” in my expectations of camera assistants, having been whipped into shape (literally, in some situations) by several well-seasoned IA veterans who didn’t let me get away with anything, and I knew we were going to have some tough shots to do. Adam surprised me on every level. Not only was he superbly detail oriented, he was also calm and cool and after a while didn’t blink when we threw a tough shot at him with very little rehearsal. He tackled some very tough focus pulls and did remarkably well. I’m glad he chose to stick in there with me rather than kill himself and float, hauntingly, next to the dolly for the duration of the shoot.

Someone shot behind-the-scenes footage that appeared recently on Youtube. The music video coverage starts at about 2:15.

NEW: Meets the Eye has posted more behind-the-scenes photos.

Song: “Living Hell”

Album: Rime of the Ancient Mariner

Group: The Tiger Lillies

Director: Mark Holthusen

Producer: Jason Santos

DP: Art Adams

Camera Assistant: Adam Wilt

Gaffers: Alan Steinheimer, Ernie Kunze

Key Grips: Kyle Rudolph, Mike Best

Production Designer: Claire Mack

Props and Wardrobe: Georgie Perrins

Makeup: Aurora Bergere

VFX lineup: Michal Horevaj

Studio: Meets the Eye Productions, San Carlos, CA

Studio support: Tim Blackmore

Special thanks to Meets the Eye executive producer Marshal Spight and Lighting by Steinheimer for their help in making this production possible.

Photos marked [TB] are copyright 2009 by Tim Blackmore/Meets the Eye Productions. Photos marked [AW] are copyright 2009 by Adam Wilt. Thanks are due to both of them for allowing me to use their work in this article.

Art Adams is a DP who snacks on albatross at theatrical events. His web site is at www.artadams.net.

Filed under: Production •Tips •Training •

So You Want to Work with Cameras

Filed under: CS4 •Tips •Training •Web Video •

ADHD Guide to Flash Video for the Web

I’m at the point where I can test my movie. I select “Control > Test” and the following window appears:

There’s one more step before we can export files. Go to “WIndow > Component Inspector” and click on the video window on the stage:

It’s very important that the “source” tag reflect a local file. It should contain ONLY the file name of the video file without showing any path. If there’s any path information in front of the file name, remove it.

All looks well. It’s time to export my final Flash files. First, I save the Flash project to the SAME DIRECTORY that contains my encoded F4V video file. Select “File > Publish Settings” and this page appears:

In this case the two elements that I’m looking for, the .swf Flash movie file and the HTML file, are both selected. Select “Publish” from the bottom left and the following files appear in the empty directory to which you’ve saved your Flash project.

“Rachel Williams 60p ProRes Final.f4v” is the video asset.

“Rachel Williams.fla” is the Flash project that I created above. All the following files, that were generated by the “Publish” process, will be created in the same directory as this file. This is the ONLY file you won’t upload to your web server.

“Rachel Williams.html” is a web page that contains instructions for embedding the Flash movie. I can pull the embed code from this web page and place it in another page, or I can simply open this HTML page from a link—possibly in a smaller window floating on top of the link page. Gray Box is a neat tool for that. This page is optional.

“Rachel Williams.swf” is the Flash movie that contains the stage and my video file.

“SkinUnderAll.swf” is the player skin.

VERY IMPORTANT: In order for the “Rachel Williams.swf” Flash movie to play properly, IT MUST BE IN THE SAME DIRECTORY as the video file (“Rachel Williams 60p ProRes Final.f4v”) and the player skin (“SkinUnderAll.swf”). Remember the step where we used the Component Window to remove any path names from the video file name? That simplifies matters by allowing the main .swf Flash file to easily find the video asset file and the player skin .swf as long as they are all in the same directory.

If you use the “Rachel Williams.html” page, it too should be in the same directory as the others. The only file you wouldn’t upload to your web server would be the .fla Flash project file.

Flash allows direct embedding of video into an .swf file, reducing the number of files that must be placed in the same server directory to the final .swf file and the player skin .swf file. The advantage of keeping the actual video asset (.F4V file) separate is that it allows for progressive downloading. This means the Flash player will automatically buffer the download and start playing the video as soon as it has enough of it loaded. Otherwise, embedding the full video in the .swf requires the entire .swf file to be downloaded before playback can begin.

Here’s the “Rachel Williams.html” web page in action.

And here’s an article about the video project used in this demonstration.

Art Adams is a DP who is too cheap to hire a web designer. His web site is at www.artadams.net.

Filed under: Cameras •Lighting •Production •Tips •Training •

Rough Guide to Flicker-Free HD Shooting

Filed under: Cameras •Tips •Training •

What You Need to Know about Green, Blue and RED

Silicon sensors can’t, at the moment, detect the color of light. They can only detect when light is present. That’s why sensors use colored filters: when the camera detects light at a photosite, it checks its memory to see what kind of filter covers that photosite and then logs the light as belonging to that color. If light strikes a photosite and the sensor knows that photosite has a blue filter over it, it logs the color of the light as blue.

But sometimes the red, blue and green filters aren’t perfect and they pass more than one color. That’s what I believe is happening in this case: the blue filter lets mostly blue light pass to the photosite, but it also lets a little bit of green through as well. Under daylight, which has a lot of blue in it, the green is easily overwhelmed by blue light and has no effect; but under tungsten light there’s not enough blue to overwhelm the green, so the small amount of green light adds to the photosite’s exposure. That means the RED sees a little blue in everything that contains green.

This has broad implications for the colorimetry of the RED under tungsten light. You can see how many colors are affected in the drawing below:

As any color that contains green is affected, almost two third’s of the RED’s color response under tungsten light is distorted by this phenomenon. As most of us learned in film school, the best way to make bright colors look desaturated is to add some blue—and that certainly appears to be the case here. This is a strong argument to use daylight sources or blue filtration when shooting with the RED ONE.

The good news is that RED’s white balance really is meta data. Nothing is “baked in” to the look, and that’s the RED’s greatest strength.

How RED is eliminating blue noise in Build 20 is still a mystery. They claim they aren’t using traditional noise reduction algorithms, so the only means left that I’m aware of is to desaturate the blue channel at the level where blue noise occurs. Under Build 20 the RED is still noisy, but the noise isn’t speckled with blue anymore. It’s more subtle, and in a way more pleasing.

The RED ONE is not the greatest camera ever made, not by a long shot; but it is by far the best $17,500 camera in existence, and you certainly get a lot of value for your money. Just be aware that it’s not currently possible to make a 4k single-sensor camera for that price without taking a few shortcuts. By understanding a camera’s flaws (and every camera has flaws) we can learn how to take advantage of their strengths and make them look their best—which makes us look good as well.

Art Adams is a DP who hates feeling blue. His website is at www.artadams.net.

P.S. At Adam Wilt’s request, below is the waveform for a Sony EX3 shooting the same chart under tungsten light:

Filed under: Cameras •Production •Training •

RED Color Comparison

Filed under: Cameras •Production •Tips •

Tiffen Goes Into Production on T1 Far Red Filter

Filed under: Cameras •Production •Tips •Training •

The Vertical Bar Graph Raw Exposure Meter

Filed under: Cameras •Post Production •Production •Tips •Training •

RED’s Blue Noise: Where It Went, and Other Color Anomalies You Should Know About

In this article I found that RED Build 20, under tungsten light, sees green in the blue channel, and I thought this was a defect in the sensor’s spectrometry: maybe the blue filtration on the photosites isn’t very good and lets some green light sneak through when there’s not enough blue light to overwhelm it. Now I’m changing my mind.

My theory is that the blue channel’s sensitivity to green light in 3200k mode is intentional and based in software, not hardware. In order to test my theory I used a tool I don’t dig into very often: the Final Cut Pro Channel Mixer plugin.

This plugin is an interesting beast: it reassigns color information from one color channel to another, in effect making a color channel sensitive to a different color. For example, it’s possible to tell the blue channel to recognize blue not just in colors that contain blue, but in colors that contain green as well. The amount of blue that the blue channel sees in green colors is adjustable, from a little to a lot.

By making the blue channel a little sensitive to green you eliminate the ability to obtain really saturated greens, because everything that’s green is going to have a little bit of blue mixed into it. But it also guarantees that if there’s very little blue light in tungsten light, the blue channel will always have some exposure because tungsten light has a lot of green in it. In effect you’re tapping into the green channel’s healthy exposure to beef up the blue channel’s unhealthy exposure.

While I can’t prove that this is what RED is doing, I can show that it’s possible. The proof is in the pudding, so turn the page for dessert…

Filed under: Cameras •Production •

CameraMania!

THE CANON 5D

The Canon 5D was built as a response to a request by Associated Press and Reuters, who were trying to cut costs by having one person shoot both stills and video of an event. They mandated that the 5D’s movie mode be fully automatic, as they didn’t want to scare still photographers off of using it, and they dictated a frame rate of 30 fps for web streaming.

Once it became clear that the motion picture community wanted to use this new tool for professional production work, Canon released an update that offered a manual exposure mode. There are third party hacks that offer zebras and other exposure tools, but I’ve not had the opportunity to try them yet.

Like RED, there are cries amongst some in the production community that this is the only camera you’ll ever need to create phenomenal money-making projects. Well… maybe. While it’s possible to create art with virtually any tool, professional production tools require the virtues of reliability, predictability and telemetry: cameras have to work, they have to work the same way all the time, and they have to tell you what they’re doing. That last part is where the 5D falls short.

WHAT IT’S GOOD FOR

The 5D creates excellent HD images for a small, cheap camera. The price, more than any other factor, is what’s driving 5D-mania right now: for US$2,600 you can buy a camera body that will deliver full 1920x1080 HD using a single 35mm-sized sensor. And the sensor is actually slightly bigger than a 35mm film frame, which means that motion picture lenses would vignette if they were mounted on this camera.

It’s very sensitive under low light, making it ideal for natural light cinematography. It’s inexpensive, making it a great crash camera. It fits into an established line of accessories, making it cost effective for those who already own Canon gear.

While the camera doesn’t offer any critical exposure telemetry (zebras, histograms) in movie mode, the LCD is accurate enough to judge coarse exposure and contrast. It’s necessary to roll through the F-stops in order to see if a highlight is clipping, but what you see is darned close to what you’ll get.

If you need a small HD camera with 35mm depth of field that doesn’t attract a lot of attention, this may be the camera for you.

WHAT IT’S NOT GOOD FOR

Following focus. Canon DSLR lenses are designed for action photography, which relies heavily on autofocus. Autofocus is not so good for motion picture photography, and it’s not enabled in movie mode anyway. You can focus before rolling, but once the camera is recording it will not try to adjust focus for you.

And you can’t adjust it very easily yourself. The manual focus ring on a Canon autofocus lens is not attached mechanically to anything. It relays data to a servo, making it nearly impossible to retain physical focus marks. Also, the lens focus markings themselves are small and very close together, making it impossible to reliably hit a mark twice.

The same is somewhat true of manual focus still lenses. They were never designed to be focused by tape measure, so the markings are rough estimates and not reliable at all. The markings are also packed so close together that being off by less than a millimeter can result in a completely soft shot.

It’s nearly impossible to see critical focus on the camera’s low-res LCD screen. There is a zoom function that does show image detail pixel for pixel, but you can’t use that while rolling. And besides, by the time you see that the shot is out of focus it’s too late.

Zacuto has solved a lot of the ergonomic problems of handholding the 5D for long periods of time, but their sunshade solution for the LCD is far from perfect. It is mounted using Velcro and tends to slip, and it’s hard to tell whether the black frame line that you’re looking at is the edge of the LCD or the inside of the sunshade that is now blocking part of the image. The sunshade magnification is also too high, which is great for judging focus but not so good for being able to see the entire frame at once. You have to scan the frame constantly to see all of it, and it’s easy to miss the tiny record indicator at the top right edge of the frame.

The camera records 8-bit variable bit rate H.264. H.264 is an excellent codec, but 8-bit color reduces how much the data can be massaged in post before banding occurs. I found that this camera’s footage requires some color correction because its eight stops of latitude forced me to protect highlights more than I wanted to. It’s necessary to use UDMA CF cards to ensure that the camera can write data fast enough to avoid severe bit rate-constrained compression artifacts. The camera also likes being held level, as too many diagonal lines may cause additional compression artifacts.

The camera shoots at 30 fps, not 29.97. Make sure your sound person knows this if you’re shooting double-system sound.

Before using a matte box and filters with this camera be sure you’ve acquired or manufactured donuts for every lens you plan to use. Still lenses are much smaller than film lenses and most of the pre-made cine donuts don’t work satisfactorily, resulting in a lot of wasted time and paper tape.

SUMMARY

This camera makes great images, but imaging power alone does not a great camera make. A truly usable production camera must conform somewhat to the process of filmmaking, and this camera isn’t quite there yet. Which is not a surprise, as it was never really meant for professional filmmaking.

For the price it can’t be beat, and the pictures speak for themselves. But this isn’t really a follow-critical-focus set-critical-exposure great-exposure-latitude camera. Within its limits it works extremely, but don’t expect to shoot your next full feature on it. At least not easily.

Art Adams is a DP who believes that the person behind the camera makes more difference than the camera itself. His website is at www.artadams.net.

Filed under: Cameras •Tips •Training •

RED Build 20 Torture Tests

Daylight Underexposure

Once again, here’s the baseline exposure under 5600k light. In this example we’re going to pay close attention to the vectorscope below.

It’s hard to see in this still frame, and harder to see using Final Cut Pro’s built-in scopes, but both the colors and white look pretty good here. White, the spot at the center, is a little on the cool side (it bulges toward the blue vector) but otherwise is a nice tight ball that retained this shape consistently throughout the overexposure tests.

It’s interesting to note that steps in the “legs” of the blue channel are fatter than than either the red or green channel, which means that blue is picking up more noise than the other channels.

The vectorscope dots are suddenly a bit fuzzy. The spot of white in the center is a little larger and has changed shape, possibly indicating noise, and the dots of color surrounding it are a little larger as well. It’s a surprisingly noticeable change for a half stop decrease in exposure.

The “steps” in the blue channel “legs” are getting a little thicker and less distinct, particularly on the bottom. The lower steps in the red channel look a little thicker as well. This thickness is consistent with those channels becoming noisier.

At this exposure both white and the colors look less distinct as they accumulate noise.

The steps in the red and blue channels are both noticeably thicker due to noise.

All the dots on the vectorscope are fuzzier and larger than before, indicating a large build-up of noise.

The “steps” in the red and blue channels below 18% gray are very thick now, and green is starting to catch up. The black chip in the blue channel (the blob between the “feet” of the blue channel) is much blurrier than in previous exposures, and red isn’t far behind.

Based on what I see here, EI 320 really does seem to be the sweet spot of exposure. At that rating, tones five stops under 18% gray hold up well and sit just at the top of the noise floor. Any further overall decrease in exposure, such as would happen if the camera were rated at 500 or 640, results in a rapid increase of noise in the blacks.

Rating the camera slower than EI 320 also results in the white chip at 2 stops over 18% gray becoming noticeably noisier.

The noise that we see here is exaggerated as I zoomed into this chart from the larger frame to isolate the color and brightness values. There may be slightly less apparent noise when viewing the full frame, although if you look at the full frame movie of the test (see last page) you’ll see that what we’re seeing on the enlarged chart is a pretty solid indicator of what’s happening overall.

We switch over to tungsten light on the next page…

Filed under: Cameras •Post Production •Production •Tips •

Busy Person’s Guide to Color Correcting the Canon 5D

I’m being pulled backwards on a wheelchair, shooting handheld. I stopped way down on this one, probably to F8, just to make sure I didn’t run into focus issues as the distance between the wheelchair and tricycle changed. Not that the dolly grip wasn’t excellent. (It was the director. He did okay. Don’t tell him, he’ll get all proud.)

This shot is okay on its own, but it could be a little better. I think I underexposed a little bit to protect highlights that snuck in over the course of the shot. I found myself constantly coddling the highlights on this camera. Someday they’ll be all grown up and out on their own, and then what? I should have toughened them up a bit.

I boosted the highlights a little to make the sun on Cary’s legs “pop” a little more. Silly, spoiled highlights.

I do love a good vignette. In this case the Vignette module in Looks adds a little interest by reducing the evenness of the sunlit ground and bringing the eye to the center of the frame.

The Soft Edges module adds a little more interest. It’s not exactly the Lensbaby effect, but it has a nice rough feel to it.

As this spot was all about color I enhanced the red of the tricycle and made it the same vector as Cary’s shirt. You haven’t seen his shirt yet, but you will in the next shot. It’s a nice rich Communist red, a nice tribute to the united workers of the world.

And the final touch, the Tiffen DFX Bronze Glimmerglass filter—a soft kiss of effervescent autumnal coloration added to an otherwise sultry summer day.

——————————————————————————————————-

I love lighting simply, and this was about the simplest lighting setup ever. I just put a 4x4 bounce card on the far side of the water fountain, where it leaned against the wall and bounced sunlight back into Cary’s face. Fast and simple, which is the way you have to work around four-year-olds. Not that he was much trouble at all, and in fact he was so well-behaved for the most part that I swore the director must have been slipping him cough syrup between takes.

Cary’s shirt in this shot became a color reference for the rest of the spot. I made the red in his shirt as saturated as possible while still remaining broadcast “legal” and matched the vector his shirt created on the vectorscope in every other shot in the spot. I’m going to call it Cultural Revolution Red.

This is the Leica 35mm, around F4.

As you can see, the “bones” of the shot are there. I underexposed a little to hold the highlight on his shirt. The light on his face is very nice but I want to make the highlights a little brighter and the shadows a little darker. I’m always protecting the highlights, which is a problem as they’ll never grow up properly that way.

Lift/Offsett/Gain works differently from Lift/Gamma/Gain. When you adjust lift, gamma or gain, you’re adjusting either the blacks, the mid-tones or the highlights, but you’re doing so in a way that causes them all to interact. In this case I wanted to boost the gain and the gamma to make the highlights on Cary’s face a bit brighter, but without lifting the blacks—and I couldn’t make that happen with Lift/Gamma/Gain. I’d raise the gamma, which pulls the blacks up, and then lower the blacks back to where I wanted them, pulling the gamma back down.

I’d always wondered what Offset/Gamma/Gain was good for. It turns out that Offset simply adds or subtracts values from the entire range. If you lower or raise Offset, every level in the image raises or lowers equally. In this case I was able to increase gain and boost gamma fairly severely, stretching them both away from black without raising black as well.

If you compare this image to the one above it, you’ll see that the shadow side of Cary’s face is nearly the same but the highlights on his face are brighter and richer. That’s the look I was going for, and I couldn’t get it any other way.

Our old friend Edge Softness blurs the sides of the frame and directs attention in toward the fountain and Cary’s face. It’s subtle, but it works.

The Shoulder module is a dangerous tool to use when color correcting off an LCD monitor. I use two calibrated Apple Cinema Displays, and what I’ve learned is that these displays reproduce highlights vastly better than a Rec 709 monitor can. In the past I’ve boosted highlights to a level where they look good on my display, although the waveform says they’re off the chart, and when I burn those projects to DVD and watch them on a Rec 709 monitor the highlights are gone.

The Shoulder module is like a knee control. It puts a heavy slope on the exposure curve in the highlight range, bending out-of-control highlights back to within “legal” limits. I use this tool judiciously as it’s very easy to boost the highlights out of range and then bring them back with Shoulder in a fashion that looks fine on my LCD display but makes them unwatchably bright on a broadcast monitor.

In this case I used just a touch of Shoulder to reign in the overexposed shirt highlights. I brought them them down until they were just touching 100% on the waveform.

These particular highlights are obviously terribly spoiled. Over time they wanted to throw an impromptu party and there was nothing I could do but deal with them in post. Keep reading, you’ll see.

Cary’s shirt became the standard red reference that I used for every subsequent shot. I made note of what vector it fell on in the rotational axis of the vectorscope (which displays hue) and the lengthened the vector (increase saturation) until the red was just short of the red box on the vectorscope. I then took a little bit of saturation out of the highlights as the red shirt was starting to turn blue. (More on that later.)

Last but not least, the Bronze Glimmerglass adds a snifter of happiness to the proceedings.

More about the ups and downs of coloring on the next page…

Filed under: Cameras •Tips •Training •

Abel Cine offers New Mattebox System for Canon 5D Mark II

Filed under: Cameras •Production •Tips •

My First Shoot with the Canon 5D

Camera assistant Satsuki Murashige demonstrates proper slating technique for young trainee Cary Sommerfeld

(all behind-the-scenes photos courtesy of Adam Wilt)

The spot concept revolved around a four-year-old boy and his father playing in a park. Director Sommerfeld cast his four-year-old son, Cary, in the lead role, and Cary performed beautifully with only occasional lapses in emotional maturity. We found ourselves working at the limits of what the camera could do, which was both freeing and incredibly frustrating at the same time. Speed and efficiency were of the essence, and the 5D and its handheld rig were not always as cooperative as I would have liked. At the same time, though, the quality of the image and the reduced depth of field suited our purposes perfectly.

My biggest complaint about the 5D was the fact that, in movie mode, the camera has a standby timer that toggles the camera into still mode in order to save battery power. When movie mode is selected, the camera’s shutter is held open and the LCD display remains on, drawing a fair bit of power. The standby timer seems to sense just when you’re about to roll, and with a “click!” it closes the shutter and turns off the LCD display. A quick poke at the movie mode button turns the camera back on, but the delay can be a little startling. I’m hoping that Canon can harness the predictive power of this particular circuit and use it to roll the camera instead of shutting it down. (I’ve since learned that there is a menu item that disables this timer indefinitely.)

The Zacuto Sniper DSLR rig

The rig that Zacuto provided us for testing worked brilliantly in all but one regard: its first-generation eyepiece magnifying loupe, or Z-Finder. This is a conical-shaped hood that attaches to the LCD viewfinder by velcro, shielding the display from sunlight and allowing for critical examination of focus on the 5D’s low-res LCD screen. This loupe employed just a little too much magnification, which meant the eye had to scan the frame in order to see the entire image and made it occasionally difficult to see the recording indicator, which is a red dot at the top right of the screen. This resulted in some unexpected behind-the-scenes footage early in the day.

The other issue I experienced is that the first-generation Z-Finder shifted on the velcro, and as there are no viewfinder markings overlaying the image (such as action safe or a crosshair) it was occasionally difficult to discern whether the black edge I was looking at was the edge of the LCD or the edge of the loupe itself. As a result some of my early test shots were off level as I was framing the image based on a frame edge that turned out to be the inside edge of the loupe and not the actual frame edge.

Zacuto has a new Z-Finder that attaches to a hard frame, which is in turn mounted to the camera using strong double-sided tape.

The new generation Zacuto Z-Finder

Otherwise, the Zacuto rifle-style rig worked very well for this camera. It didn’t take all of the weight off of my forearms but it did help considerably by shifting some of it to the front of my shoulder. Also, holding the rig tight against my shoulder made it very steady and easy to operate.

A bit about focus and donuts, on the next page…

Filed under: Training •

NorCal Digital Cinema Society Meeting Tonight, 6pm

Filed under: Cameras •Production •

Facebook Mobile spot, Shot on RED

Filed under: Cameras •Production •Tips •Training •

Black Glimmerglass and Hollywood Black Magic Filter Tests

Filed under: Cameras •Production •Tips •Training •

Diffusion Confusion

FIX IT IN POST?

Both Zupka and Tiffen are in agreement that there’s no true replacement for using a filter in front of the lens.

“If you’re doing it up front, in the camera, you’re affecting the image in a 14-bit color space,” says Zupka, referring to the color depth at which many HD imagers capture data. “On tape you’re down to 10-bit, or 8-bit. The higher the bit depth, the larger the palette of colors to choose from. Also, any time you manipulate the red, green or blue gains in a camera to create a look you’re susceptible to adding noise from pushing a color channel too far.

“Also, filters like polarizers and diffusion interact with the lighting and texture of the object you’re shooting. You can’t recreate that in post.”

“I like to quote Hank Harrison,” says Tiffen. “He said, ‘When you put a filter on a camera you are dealing with an infinite number of luminance levels and variety. You lose that once you go to any recording medium. It’s ALWAYS different when you try to affect a recording instead of the real thing. There’s nothing that replaces the real world as a source of information.’”

Having said that, Tiffen was a major proponent of the software package that is now known as Tiffen DFX. “Post filtering is an increasingly valuable tool for arriving at the look we want. As with everything else—lighting, gels, cameras, lenses, set design, and costumes, to name a few—there are many variables that offer control points. Post filtering is another relatively new addition to the arsenal of effects. The idea is to use the camera, with its digital and optical capabilities, to get as close to the end result as is reasonable. Then post options can be used to bring the final image to successful fruition.”

When I suggested that Tiffen DFX effects were similar to, but not a match for, the Tiffen filters they claim to emulate, he said that didn’t matter. “There isn’t a practical difference because the people who see the final result aren’t doing comparisons between the actual filter and the emulation. As long as you’re consistently using one or the other, there is no need for concern. It’s all a matter of practical differences and circumstances.”

He offers a word of warning, however. “Fixing it in post is more and more frequent. You have to know that there’s only so much range you can adjust in post, due to the flaws in the recording medium. You’re better off getting it as close as possible in camera if you can. Shooting it clean and fixing it later only works if you know exactly what you can do later.

“Filter use comes down to visualization,” says Tiffen. “If you know what look you want, use the filter that will give it to you.”

On page 4: diffusion in the digital age…

Filed under: Cameras •Pre-Production •Tips •

Using the ND .90 Filter as a Preproduction Tool

Filed under: Cameras •Production •

Far red on the EX1/EX3/F35: “It’s a feature, not a bug!”

Filed under:

Look for me at CineGear

Filed under: Cameras •Pre-Production •Production •

The Last Far Red Filter You’ll Ever Need?

Here’s the Sony F35 with the same setup. The problem here is lack of consistency: I have to work with different fabrics at different rental houses because I haven’t settled on my own standards yet. This test session, though, showed me how to set up a consistent black fabric test chart. More on that below. But first, here’s the F35 without, and then with, the Tiffen no-ND far red prototype filter:

If you saw no difference, ha! Ha, I say! Because there is virtually no difference. While there may be a minute amount of far red contamination in the target fabric, this particular fabric is not a good far red test subject for the F35. Obviously the F35’s far red filtration is better than the EX1/EX3 (you get what you pay for!) although there are fabrics that will cause the F35 to react more strongly.

Just for fun we did something you’d almost never, ever do, which is add a bunch of ND while shooting under tungsten light. In the following stills we tested both a Formatt ND .90 and Schneider ND .90 to see how they enhanced far red, and also to see whether the Tiffen prototype worked equally well under both circumstances:

There are two interesting things to note here:

(1) The ND filters are slightly different colors, which is no surprise: it’s extremely hard to make perfectly neutral ND filters, and each manufacturer’s dyes shift in slightly different ways. In this case Schneider’s filter looks a little more red, while Formatt’s filter looks a little more yellow.

(2) The Tiffen prototype no-ND far red filter cuts far red when added to an ND .90, which is pretty surprising as it’s really only supposed to be used on its own, with no additional ND. That’s a strong filter!

Just for fun we decided to really stress the F35, adding both the Schneider and Formatt ND .90 filters together to create a double-whammy of far red:

You’d never do this, but it’s fun to see what happens. It’s also amazing to see that the Tiffen prototype filter cleans up nearly all the far red, even at an effective ND 1.8! I’m starting to wonder if this filter is stronger than it needs to be for its intended use. I’m also really impressed that Tiffen found a way to snip off an offending portion of the spectrum without using a hot mirror, avoiding all the issues that a hot mirror creates (ghosting, vignetting, etc.).

An interesting side note is that we discovered that we could see far red shifts in fabrics simply by looking at them under tungsten light through an ND .90, which completely surprised us. (This should also work in daylight, although I haven’t tested this yet.) The overall color shifts a bit warm, because ND filters are rarely neutral, but the far red contamination jumps right out at you with no subtlety at all. In fact, looking through an ND .90 by eye resembled the combination ND .90 + ND .90 chart showed above.

This could turn out to be a great preproduction trick when shooting with the Sony F35, EX1, EX3, and Panavision Genesis cameras. It also means that I can walk the aisles of a fabric store looking for problem fabrics with which to build a standardized test chart. I’ll probably do that next week, and then we can start looking at which materials are the worst offenders.

Before then, look for me at CineGear!

Art Adams is a DP who often sees red. His web site is at www.artadams.net.

Filed under: Cameras •Post Production •Production •Visual Effects •

Porsche Spot VFX Breakdown

Filed under: Cameras •Production •Tips •

Why Does Blue Hurt So Much?

Filed under: Cameras •Production •Tips •Training •

Tiffen tests new no-ND IR filter for EX1/EX3/F35

I shot the DSC chart full frame, both clean and with the Tiffen prototype filter. Here’s the clean version:

Here’s the version shot through the Tiffen prototype and white balanced:

I can’t see any difference by eye, but that means nothing. The eye is extraordinarily easy to fool. We really need to compare how these charts look on a vectorscope and see if there are any color shifts. Here’s the clean chart:

I artificially colored this screen grab for a reason. We’ll get to that in a moment. Here’s the chart with the Tiffen prototype in place and white balanced:

Now, if we take both those images and overlap them:

The images have been colored in exact opposite colors, with the clean chart made red and the filtered chart colored cyan. If the Tiffen prototype was cutting significantly into the red spectrum I’d expect to see the top of the cyan pattern pulling in toward center. I don’t see that here. There’s a whisper of a mismatch but I don’t see anything to worry about. The filter seems to cut exactly what it needs to and no more.

This was the first test prototype. Tiffen says they will be sending me another version to test soon. Their goal is to release this new filter in a month or so. If you own a Sony F35, EX1 or EX3 it’ll be well worth looking at.

Art Adams is a DP who tries to write clearly about technology, with very little filtration. His web site is at www.artadams.net.

Filed under: Cameras •Production •Tips •Training •

Optically White Balance the RED with Schneider CTB Filters

Filed under:

Whatever Happened to Ira Tiffen?

Filed under: Cameras •Hardware •Production •

Hands on with the Gearnex Gear Head

Filed under: Cameras •Post Production •Production •Tips •Visual Effects •

Tiffen DFX 2 Digital Filters

Filed under: Lighting •Production •Tips •

Rosco Shows New IR Filter (and other goodies) at NAB

Filed under: Cameras •Production •Tips •

Gear Head for the RED

Filed under: Cameras •Production •Visual Effects •

Porsche F35 Spec Spot Revealed

Filed under: Cameras •Production •Tips •

IR Filter Cheat Sheet

Filed under: Lighting •Production •Tips •Training •

Mysteries of Color and Light

Knowing how color works in light can be quite handy, especially when you want to add saturation to an object. A great example is blue/green screen photography: there’s a reason that super blue and super green Kino Flos exist, and that’s because the dyes that color blue and green screens aren’t perfect reflectors of just blue or green. They reflect mostly blue or green but they reflect some of the rest of the spectrum as well, which washes out their primary color. The solution is to light a screen that reflects mostly blue or green light with PURE blue or green light. A mostly blue or mostly green screen lit with pure blue or green light will reflect only pure blue or green.

Tip: to super saturate a green screen when using tungsten lights, use Lee 122 “Fern Green” gels. (Thanks to DP Geoff Boyle for that tip.) To super saturate a blue screen under a tungsten environment, light it with uncorrected HMIs or daylight Kino Flos. (This applies only to screens that are separate from the talent. Do not use these tricks on screens where the talent is standing on, or touching, the screen.

You can pop an object’s color just by adding light containing a bit more of that color. For example, if you’re lighting something that’s made of warm wood and you want to make it warmer, add 1/4 CTO to your lights and see what happens. You’ll be surprised how much adding a hint of an object’s color to its light will punch up its saturation.

On to the last page, where we talk more in depth about additive and subtractive color…

Filed under: Business •Web Video •

The Quicktime Conundrum, Part 2: Solved by our Readers

Filed under: Business •Web Video •

The Quicktime Conundrum

The X.264 codec is known to be a little buggy. I like to encode the videos on my web site as 720x405 Quicktimes, as this seems to be a nice size for both viewing and for keeping file sizes down. I came to the 720x405 size by starting with the standard def video spec 720x480 and dividing the width (720) by 1.78 (or 16/9). The result is 404.49, and for quite a while I’ve been happily plugging 720x405 into Compressor and getting excellent results.

These stills are from some regional spots I shot a couple of years ago for Kaiser Permanente:

This is how the original 720x480 video frame looked when exported from Final Cut Pro. NTSC uses non-square pixels, so when this video clip is shown on an NTSC monitor the pixels will be stretched horizontally to an aspect ratio of 1.21:1 to yield a 16:9 frame.

Computer displays utilize square pixels for better resolution, so in order to make a non-square pixel NTSC image look proper on a computer we have to do a bit of resizing. This image was squeezed vertically to yield a proper 16:9 image at 720x405, although it could easily have been stretched the other direction to yield 853x480.

Unfortunately my 720x405 settings caused the X.264 encoder to crash. It worked fine at 720x480, and I encoded a 4:3 file at 640x480 without difficulty, so what was the problem? There had to be something about the 720x405 aspect ratio that X.264 didn’t like. But what could that be? I pulled out my calculator and, once again, divided 16 by 9, and divided that into 720. The answer was still 404.49. But: I’d been rounding up to 405; what if I rounded down to 404?

It worked perfectly. It seems that the X.264 encoder only works on video that it considers to be precisely 16:9 or 4:3, and if your calculations give you a frame size in one dimension that can be rounded up or rounded down to the nearest pixel then one value will probably work while the other won’t.

The last step was de-interlacing my NTSC footage. Compressor did a bang-up job of that, but the unwanted gamma shift Compressor added ruled it out. I didn’t quite trust X.264’s de-interlace option, so I dug around in my collection of FCP filters until I came across G Smart De-Interlace, part of Graeme Nattress’s File Effects plugin package. It worked like a charm.

Here’s a frame from some standard def footage that shows clear interlaced artifacts (horizontal lines) around the gentleman’s outline, particularly around his shirt and on his forehead.. On a TV set this kind of motion artifact isn’t an issue, but it’s very clear on computer playback.

This is the same frame cleaned up using the G Smart Interlace plugin in Final Cut Pro. It does an excellent job.

Perhaps someday Apple will deign to fix this “feature” of Quicktime, but don’t hold your breath. Meanwhile, this workaround is a little bit of a pain but it does work—and Quicktime H264, as a showreel format, is well worth it.

I’m on to Adobe Encore now. Encore will allow me to lay out a showreel DVD and then also export it as a fully functional Flash file that emulates a physical DVD, menus and all. More on that when it happens.

Filed under: Cameras •Lighting •Production •

Anatomy of a Porsche Spec Spot

After our first rehearsal it became clear that we needed to hide a light in the car itself. Alan Hereford offered a battery powered LED light that he’d picked up at Walmart as an experiment. It was the perfect color to emulate the cool blue/purple light one sees in high-end car dashboards. It was a small ring light, and we hung it off C-47’s attached to the heating vents on the dashboard. It ran off four AA batteries.

’

Alan Hereford’s LED ring light adds a high-tech glow to the interior of the Porsche. The stop sign in the background is lit by the 1k open-face hidden behind the building and will be removed in post. Notice that the two street lights to the left of the stop sign show vertical smears: that’s an artifact of the F35’s striped-sensor technology and reminds me of a vertical anamorphic flare.

The first shot was a close-up of three people being dragged to the car. We shot the two women separately because they didn’t overlap (they would be separated into different elements later), and then shot the guy separately. We did a number of passes: first they just sat there, then they stood up and sat down abruptly, and then they pushed themselves away from the car. All these elements would be sped up, blurred, and occasionally reversed in order to create one shot where they were dragged to the car as if by a magnet.

We ran into a little bit of a problem in post: as we were shooting the guy’s plate the one overhead building light that we couldn’t reach shut down, apparently due to a headlight hitting a daylight sensor. For a short time we were afraid we’d have to shoot all the women’s plates again in order to match the new lighting. Fortunately the building light re-ignited within a minute or two but took a while coming up to speed, and as we were pressed for time we shot as soon as the color of the light looked stable. In post we found out it wasn’t quite up to speed yet, and there was a slight color shift that was visible on the white hood of the car. If it wasn’t for our white car we never would have noticed the difference. That plate will require a small amount of color correction to make it match the others.

This is why, when doing multiple passes on a scene for VFX work, it’s good to control every light in and around the shot—assuming, of course, that you have a budget that can buy you the manpower and tools to do so.

After that we shot a master of the car pulling up and repeated a number of elements of our three main talent being dragged to the car. And after that we lowered the camera and shot another angle of the same actions again.

Hidden in plain sight: That black object you see tucked against the wall in the bottom left of frame is one of our Kino Flos. No one will ever know what it is.

Last but not least we shot the end shot, or the punch line of the spot.

We started setting up in daylight, around 5pm, and started shooting around 7:30. We were finished by 11:30, thanks to a little help from my friends and co-workers.

Berkeley police drove past us five times while we were shooting, and I was fairly sure we were going to get busted for shooting on the sidewalk without a permit or aiming bright lights into the street. They never bothered us. On my way home I drove around the block and discovered that there was an actual club (as opposed to our fake club) on the next street over, and that was what the cops were watching. It was only through great force of will that I drove around, and not over, the hundred or more people who had spontaneously decided that 8th Street was their living room.

We recorded all the footage via HD-SDI to a Panasonic HPM-110 P2 deck, using the AVC-Intra codec. The spot is currently in post and I hope to be able to show you the finished piece in a week’s time.

While I shot a couple of behind-the-scenes photos for this article, the majority were shot by Rod Williams. Thanks, Rod!

Filed under: Cameras •Production •Tips •Training •

EX1/EX3 IR Filter Shoot-Out

SECOND UP: FORMATT

In previous tests I’ve discovered that Formatt seems to have one flavor of IR cut in their filters. For this test we only tried the Formatt stand-alone Hot Mirror:

The Formatt Hot Mirror seems to cut about as much as the Schneider ND 1.2/Tru-Cut 720 combo filter: there’s some improvement, but not nearly enough to make a huge difference.

At this point the clues were starting to add up, but I was still too focused on my “bad Hot Mirror” theory to see them. The Tiffen Hot Mirror IR tests on the next page led me farther astray.

Filed under: CS4 •Post Production •Production •Training •

Fake Tilt-Shift Cinematography

Filed under: Cameras •Production •Tips •Training •

F35 IR Filter Shoot-Out

FIRST UP: SCHNEIDER OPTICS HOT MIRRORS

I’ve been informed by Sony engineers that the F35 has a 700nm cut filter built onto the sensor, which would mean that the red contamination we’re seeing is far red, not IR. Far red is at the edge of the visible spectrum and is a color of light that our eyes can’t see very well, if at all, but it is not yet pure infrared heat. The far red “leak” typically occurs between where ND filters start passing non-visible light (around 680nm) and where the IR filter on the sensor cuts (700nm).

We’re about to find out whether that’s true or not. Schneider Optics kindly provided me with three filters to test: a 680nm cut filter, a 720nm+ND 1.2 combination filter, and a 750nm cut filter. If there really is a built-in 700nm cut filter on the chip then we should see some results from the 680nm filter, but not the others.

First, here’s the 680nm filter:

This is the Schneider Tru-Cut IR 680 filter, in combination with a Schneider ND .9 and ND 1.2. It has a very strong dichroic coating, which gives it a cyan cast and costs between a half stop and a full stop of light. As we’ve seen, an overall color cast isn’t an issue at all: what’s important is what’s left once we white balance through it.

The Tru-Cut 680 does an excellent job of removing all far red or IR contamination from this shot.

BUT…

Here’s the flat field test, with the Tru-Cut 680 on a 16mm Ultra Prime:

Dichroic coatings tend to show up most when viewed at an angle other than perpendicular, and in this case the lens angle of view is too extreme around the edges of the filter. Let’s look at a 24mm lens:

Better, but there’s still an obvious cyan cast to the edges of the frame. Let’s see if the 32mm works:

At this point it appears safe to say that the cyan edges have given way to an overall cyan cast that can be easily white balanced away. Let’s take one more look, this time at a 50mm lens:

This image looks virtually the same as the 32mm, which means that 32mm is the widest lens one could safely use with the Tru-Cut 680 to avoid color vignetting.

The next filter in the Schneider series is the Tru-Cut IR 720 + ND 1.2 combo:

It seems that Sony’s engineers were correct. The 720 filter shows virtually no effect. The final Schneider filter is the Tru-Cut 750:

This filter has no effect at all, and when we compared the two filters we discovered that the 720 had some effect but not nearly enough. This can be explained by the sensor’s IR filter having a gradual roll-off effect that starts at 700nm, instead of a hard cut of everything above 700nm.

Just for the record, here are the flat field tests on the Tru-Cut 720 and 750 filters. First, the 720 as viewed through a 16mm lens:

And then the 750, viewed through the same lens:

We saw no evidence of a color vignette, only an overall color shift that is easily removed by white balancing.

While these filters don’t work so well with the F35 they do have some positive results on the RED.

Let’s move on to Formatt’s offerings…

Filed under: Cameras •Production •Tips •Training •

RED Hot Mirror Shoot-Out

FIRST UP: SCHNEIDER OPTICS HOT MIRRORS

Schneider Optics kindly got the ball rolling by loaning me three different filters with three different degrees of far red/IR cut: The Tru-Cut 680, Tru-Cut 720, and the Tru-Cut 750. First up is the Tru-Cut IR 680, in combination with a standard Schneider ND 1.2 filter:

The Tru-Cut 680 is probably the lowest cut filter that we tested, and as you can see it does an excellent job of cutting IR/far red. The black jacket is rich black, with no obvious color shift, and the red kimono is still nice and red. The problem, though, is that the dichroic coating on the filter is very heavy. Dichroic IR filters typically have two sides: the side that faces out appears to be salmon-colored when viewed at an angle, and the side that faces the lens appears cyan when viewed the same way. On wide angle lenses, such as a 16mm lens, the angle of view around the edges becomes so extreme in relation to the dichroic coating that the cyan hue becomes visible:

In my Sony F35 tests (to be published shortly) we detected vignetting on lenses wider than 50mm when using this filter, although it was most noticeable on the 16mm and 24mm. So while the Tru-Cut 680 is a great filter for cutting IR, it’s not so good on wide angle lenses.

Next up is the Schneider Tru-Cut 720 + ND 1.2 combination filter:

This filter cuts most of the IR from the image after color correction. Before color correction, though, with the blacks lifted up a bit, we can see a little maroon-colored IR infiltration:

For practical purposes the Tru-Cut 720 filter removes all the IR necessary to recover good blacks in post. The flat field test of this filter shows it has a slight warmth to it:

This tint is easily removed in post. The edges of the frame show no obvious color fringing beyond the natural brightness vignette of the 16mm UltraPrime, so this would be a good choice for an all-around IR ND filter. (We only received this filter as a Hot Mirror/ND combination, so we didn’t test a Tru-Cut 720 clear filter—if it exists. Before this test I’d not heard of the Tru-Cut 720.)

The final Schneider filter tested is the Schneider Tru-Cut IR 750, which has been a staple of local rental houses for the last six months. We stacked it with a Schneider ND 1.2 and viewed the results:

After color correction the filter seems to be doing a fine job, although a look at the frame prior to color correction shows that there is some IR/far red contamination in the jacket:

When I mentioned this to one rental house tech, all he had to say was “Every RED rental we’ve had for six months has gone out with one of these filters, and we’ve had no complaints.”

While the Tru-Cut 750 doesn’t do a perfect job, it seems to do its job well enough. Here is the fiat field test on the 16mm lens:

While this JPEG shows some slight color vignetting that appears to be a side effect of JPEG compression. The actual DPX file shows no color vignetting at all.

Schneider was among one of the first companies to offer Hot Mirror filters for the RED, and their offerings hold up well. While the 720 and 750 filters don’t do a 100% perfect job of cutting all IR/far red, they do contain it well enough to effectively eliminate contamination after post color correction.

The camera assistant who helped me with this test, Rod Williams, noted that Schneider Optics coats their filters with a slippery finish, making it very easy to remove fingerprints and other smudges.

We move on to Tiffen IR ND filters on the next page…

Filed under: Cameras •Post Production •Production •Tips •Training •

The Not-So-Technical Guide to S-Log and Log Gamma Curves

THE MAGIC OF S-LOG REVEALED

Here, in all its glory, is how S-Log works:

(Courtesy of Digital Praxis)

The gamma curves that we’ll look at on the next page are based on the bottom graph, and it’s useful to notice a couple of things about it.

The steeper the curve the finer the brightness steps, and the flatter the curve the bigger the steps. The steepest part of the curve is between 0% and 70% on a waveform monitor, which is where the most perceptual steps of brightness are; after 70% the curve becomes very flat and is meant to hold highlight values whose steps can be wider because, while our eyes appreciate seeing highlight detail, we aren’t very good at perceiving fine luminance steps in it.

Brightness data is stored in 10-bit steps, or 1024 values for each channel. The internal code values for different shades of brightness, and their equivalents on a waveform monitor, look like this:

It’s handy to note that, if you move the decimal point of the code values one step to the left, they are very close to their equivalent waveform value. For example, middle gray’s code of 394 becomes 39.4, which is very close to a waveform value of 37.7%.

One of my first questions to all the people interviewed for this article was: “With only 370+ steps from black to gray, and another 250+ or so from middle gray to white, and with another nearly 400 steps dedicated to highlights, are there enough steps to prevent banding when they are expanded and manipulated in post?” The answer, universally, was “Yes: you’ve been watching this for years. This is based on the Cineon spec for digitizing film, and you’ve been fine with it so far.”

There are three reasons why this curve won’t play well on a regular Rec 709 display:

(1) S-Log lifts the blacks considerably, which tends to desaturate the colors

(2) The brightest highlights are heavily compressed

(3) It’s a high dynamic range image whose gamma is not Rec 709 compliant, which desaturates and flattens the image

There’s much talk about using LUT’s and such for viewing an S-Log-encoded image correctly during grading, but that’s not really correct or necessary. Unless your footage is destined for another kind of media, like film, imposing a LUT isn’t necessary.

For TV deliverables you can get an idea of what’s happening on set by configuring one of the F35’s monitor ports to overlay Rec 709 gamma on top of the S-Log image. It’s not going to be a perfect rendition but it’ll give you an idea of what’s going on. When configured this way the F35 output will display a curvy Greek “gamma” symbol in the corner, along with the numbers “709”, to remind you that Rec 709 is being applied to that output. (Your DIT can watch the uncorrected S-Log signal on another monitor.)

While you might be tempted to paint this image to make it prettier you shouldn’t touch the paintbox at all. Painting a Rec 709 rendering of an S-Log image will result in a reshoot, and while the rest of the crew will thank you for another day’s work, you won’t be there.

Another viewing option is to use one of the Sony BVM-L LCD displays, which offer a built-in S-Log-to-Rec 709 decoding function.

To sum up, S-Log very efficiently puts 12 stops of dynamic range into a five stop bucket by remapping brightness data in a way that makes sense to our eyes, and then throws away the data between the perceptual steps that our eyes can’t see. It’s a very good form of visually lossless compression.

S-Log isn’t the end of your options, as I’ll prove on page four…

Filed under:

No pain? NO GAIN!

Filed under: *VIDEO* •Cameras •Production •Training •

No 85 Necessary on the F35!

Filed under:

Correction: Schneider Optics Sings the CTB’s

Filed under: *VIDEO* •Cameras •Post Production •Production •Training •

Schneider Optics Sings the Blues

Filed under: *VIDEO* •Cameras •Production •Training •

Return of IR

Filed under: Cameras •Production •Training •

The Sony F35 and ND Filters

Filed under: Cameras •Training •

The Not-So-Technical Guide to the Sony F35

CRAMMING 12 STOPS INTO A FIVE STOP BUCKET

Anyone who is familiar with the menu system of a Sony F900R already knows the vast majority of the menus in the F35. Several of the primary gamma curves are roughly the same. We’ll go into those in a moment. But first, let’s talk about why we need aggressive gamma curves.

The original HD spec is based on the old NTSC television standard of five stops of dynamic range, encompassing 0-100% on waveform monitor. That’s it. The Rec 709 HD standard only allows for about 2.5 stops of overexposure latitude, between approximately 45% (or 18% “middle” gray) up to 109%, as seen on a waveform monitor.

Our happy misfortunate is that sensors have become much, much better and can now see well beyond the five stops of latitude provided for in Rec 709, but all of our engineering and monitoring tools and data pipelines (HD-SDI) are designed around the Rec 709 legacy of having a 0%-109% “bit bucket”. At a demonstration of the Sony F23 held by Videofax in 2007, local video engineer Fred Meyers described how he came up with the concept of the bit bucket while doing tests for the film Speed Racer. “The trick,” he said, “is not only using the biggest bucket possible, but making sure you fill it all the way.” We’ll talk about this a bit more when we start comparing curves.

How do we cram a total of 12+ stops of dynamic range into a five stop bit bucket?

Those of you who said “bigger hammers” are banished to the grip truck. The correct is answer is “aggressive gamma curves.”

The following graphs are intended to communicate general ideas and are not exact measurements of anything known to man.

The Rec 709 gamma is a very simple one, and although it has a curve to it it’s easier to envision as a straight line:

At around two to two-and-a-half stops over middle gray, exposure hits a hard ceiling and clips. This is completely unlike film, where at some point one stop of exposure change does not result in a one stop difference on film—and the highlights slowly and gradually lose detail until they disappear into featureless white.

Knee circuits help somewhat but never really do the trick. They try to create an artificial slope that’s shallower than the Rec 709 curve but doesn’t roll off gently the way film does:

Knee circuits are notorious for causing color distortions in highlights, which is why we never use them to reign in highlights on flesh tone: the affected areas tend to turn a metallic green color, which does no one any favors.

A curve is fundamentally different. While a knee circuit is just trying to make blown-out highlights look a little prettier by bringing the exposure in the highlights down in an attempt to make detail visible, curves are actually grabbing extended dynamic range information available on the sensor and remapping it to fit into the confines of the Rec 709 spec:

Instead of starting at a knee point and trying to force detail out of bright, desaturated highlights, a curve gently draws information from beyond Rec 709’s normal cutoff down into the 0-109% bit bucket. Sony describes how much information the camera is recording beyond Rec 709’s usual boundaries by displaying a percentage number on the side of the camera. As best I can tell the percentages break down in terms of doubles and halves—the way most everything in photography works:

100% is what would normally have been the broadcast clip point for Rec 709, about two stops over middle gray.

200% is another stop beyond that point that the sensor can see but couldn’t fit into the Rec 709 gamma space without using a curve.

400% is another stop beyond that point.

650% is a normal “limit” to how much information can be pulled off the chip, and represents about a half stop increase beyond 400%, or an additional 2.5 stops of information that normally wouldn’t fit in Rec 709—for a total of 4.5 stops of exposure latitude above middle gray. (In some modes Sony claims a maximum of 800%, or an additional half stop beyond 650%.)

Depending on the curve used and the gain applied the percentage number will increase or decrease, giving you an idea of how much dynamic range the camera offers in that particular operating mode.

What do these curves actually look like? You’ll see on the next page…

Filed under:

Even Rembrandt Sang the Blues

Filed under: Lighting •

Lighting Advice for Budding DPs

Filed under: Cameras •Editing •Lighting •Post Production •Production •Training •

My First Shoot with the Sony F35

Filed under: Cameras •Lighting •Pre-Production •Production •

The RED Outdoors

Filed under: Cameras •Lighting •Post Production •Production •Training •

Lighting Simply for the RED

This shot was lit very simply, by bouncing a 575w PAR into the hardwood floor just to the right of the shot. Light from below feels like ambient light to me, and the feel of this shot is that sunlight is striking the floor just outside of frame.

Here’s the original shot, brought in from RedCine with minimal correction:

Here’s the shot with some basic color correction, setting the pedestal, gamma and gain using Magic Bullet Looks:

The plant looks a bit down to me. At the time it looked okay but now… it needs a little pop. Initially I tried using Looks’ spot exposure adjustment, creating a round feathered pool of increased exposure around the plant, but that just didn’t look right. Instead I decided to use a “negative ND grad.”

An ND grad knocks down the exposure in half of the frame and then blends back to normal exposure in the other half. If you look at an ND grad the top, or one side, will look dark and the bottom, or opposite side, will appear clear. The dark feathers to clear in the middle of the filter to hide the area of transition. Magic Bullet Looks has a digital ND grad that emulates this affect as long as nothing in the affected area is clipped. (Clipped highlights turn gray when darkened and look very much like dark clipped highlights—or, in film terms, “a mistake”.) In this case, though, I wanted to do the opposite: I wanted to increase exposure in one part of the frame and have a transition to normal exposure in the rest. So I set the digital grad in Magic Bullet Looks so that it darkened the left side of the frame… and then I entered a negative value for the grad. That lightened up the area I’d just darkened:

This worked very nicely. The effect is completely hidden.

One other trick: the beam in the ceiling in the background on the far right was really bright due to reflected sunlight from outside, so I used a polarizing filter to reduce the shine and make it less bright.

If you haven’t eaten yet you’ll love this next setup…

Filed under: Cameras •Post Production •Production •Training •

The RED Camera and Green Screen

Meet Adam Loften. He was our assistant editor/data wrangler on the shoot, and he was kind enough to stand in for our little test. The light is all tungsten and read around 2900k on my Minolta Color Meter 2.

You may notice some dimples and such in the green background. That’s because we saved a bit of money and went with a 30’x40’ fabric green screen over a cove in order to save money on paint.

Here’s the green channel:

It’s very clean. My previous tests on RED Build 15 (this test was shot on Build 16) showed that green was consistently the most solid color channel under both daylight and tungsten light. On 18% gray card tests shot under daylight the RED’s histogram showed green in the center with the red and blue channels almost perfectly aligned under it. Under tungsten tests green stayed in the center but red advanced to the right of the histogram (overexposed compared to green) and blue receded to the left (underexposed compared to green) with little overlap of color.

Knowing this, it seemed pretty safe to expect a good key based on green lit with tungsten light. I’ve also noticed, as have others, that the RED is overly sensitive to green and magenta color shifts in light. (In pure light terms this means the RED is sensitive to too much green and too little green. There is no magenta color in the visual color spectrum: magenta is simply an absence of green in light.) When not shooting full body green screen we often super-saturate the green with Super-Green Kino Flo tubes or by using green gel (“plus green” or Lee Fern Green). Modern software keyers look not just at the brightness level of the screen but also at how far the screen’s hue is from the next nearest color. The farther away it is from the next nearest color the better the key. Full body green screens don’t allow for the use of colored light, as it will contaminate the subject, so we’re forced to use only white light. In this case, having a camera that’s a bit oversensitive to green turned out to be a good thing.

Here’s the red channel:

As you’d expect, there’s not much red in green screen. (When looking at color separations, light tones show where there’s more of the separated color, and dark tones show where that color is absent. In this case, Adam’s skin and the slate have a lot of red, but his shirt and the screen don’t.) Zooming in shows a little bit of a rough edge between his shirt and the screen but it’s not too bad. We’re not really keying based on red but noise in an underexposed channel can result in a noisy image and, possibly, a compromised key.

Here’s the blue channel:

Gulp.

There’s a fair bit of noise in there. The lightness of the screen tells me there’s blue in the screen, which isn’t unexpected: if you look at any green screen on a vectorscope you’ll see that it’s not pure green but is a little shifted towards the blue vector. What’s surprising is how noisy it is. Even though the screen is light, telling me there’s blue exposure in there, it’s very noisy, as is Adam’s shirt. How can a green screen that obviously has some blue in it look noisier than the relatively clean red channel? There’s a lot less red in a green screen than there is blue!

Here’s my theory:

Tungsten light has a lot of red in it, so the red channel has a healthy exposure even though there’s not much red in the green screen, because what little red is there is healthily exposed. The 3200k viewing preset in the camera is probably holding back a lot of red that is actually present in the raw image. So while the screen looks dark in the red channel it’s actually getting enough red exposure that it’s not very noisy.

The blue channel probably looks bright because it is being boosted due to RedAlert trying to mix three color channels to create a white balance of 3200k. Red is reduced and blue is increased to match the level of the green channel. But the blue channel boosting in RedAlert is only increasing the brightness in this image. It may increase the visibility of some of the noise but it’s not increasing noise overall, because noise is a sensor artifact and we’re way beyond the sensor’s influence now.

Silicon sensors tend to be least sensitive to blue, which is why the blue channel tends to be the noisiest. Many cameras, such as the F900 or the Varicam, are balanced for tungsten light by boosting the blue gain in relation to the others. Most of the time this isn’t noticeable until the camera’s overall gain is boosted, and then the blue noise quickly becomes apparent. When shooting outdoors, an 85 filter simply balances the light for the tungsten-jiggered sensor.

The RED has no such filtering and gain manipulation. It records a “RAW” signal (at least in theory—let’s not get into what really constitutes RAW these days!) supposedly with no manipulation of the red, green and blue gains. (There is probably some, particularly in the highlights.) The result is a camera that does very well in daylight because there is a LOT of blue in daylight, and that increase in blue makes up for the lack of blue sensitivity in the sensor. But under tungsten light the red channel gets more exposure because of the increased warmth of the light and the blue channel is a bit starved.

Let’s see what difference a filter makes…

(Page 3 of 6 pages for this article  <  1 2 3 4 5 >  Last »)

                   

 

You must be registered to comment. This is an effort to reduce spam. Please REGISTER HERE.