Throughout the years, we get plenty of questions from filmmakers all over the world. This month we intend to recap three of the questions and our thoughts on them.
Q: I’m a new filmmaker who just shot his first film. Some of my footage was underexposed and, when I brightened it enough to be useable, there was a lot of grain. However, the rest of my footage looked really great. What do I do?
A: Well, you first look to see if you can either reshoot the footage that turned out badly or edit your film so you don’t have to use the mis-lit footage at all. If the footage is necessary and can’t be reshot, then you have to make the good footage you have look just as bad as the bad footage. This is a hard thing to do, but absolutely necessary. This is because disparity in video quality always looks awful. It’s far better (and far less noticeable) to have uniformly mediocre visual quality that is consistent rather than footage that looks good in one shot and bad in another shot. The bad footage will always look horrible when intercut with good footage.
Q: I’m looking at a couple of basic video cameras to start filmmaking with. One is a really nice high-def camera that records to hard drive, but it doesn’t have a mic jack, just the onboard mic. The other is an SD camera that records to tape that has a mic jack. I’m really leaning toward the HD camera. What would you suggest?
A: Although we as filmmakers tend to be visual people, audio is always more important than video. Movies like The Blair Witch Project and the Bourne Ultimatum had nauseating camerawork, yet they were successful due to the fact that they had very good audio. Beautiful looking films with awful audio never even see the light of day for distribution. As such, get the SD camera with the mic input. Just be aware that, on more basic cameras, there tends to be auto-gain to the audio levels. This means that the camera will often try to enhance the audio volume if no one is speaking, causing the audio to be too loud when the recording starts again. As such, try to get a camera that has manual audio controls and no auto-gain if at all possible! (For more information on recording audio on your first projects, check out our introduction to audio article here.)
Q: I’ve heard that if you’re going to shoot greenscreen footage, you’ve got to have an HDV camera. Is that true?
A: No, it’s not. In point of fact, you can shoot quite competent greenscreen footage with an SD camera, if you’re creative, light the shots properly, and you understand the camera’s limitations.
The limitations of low budget cameras for greenscreen work do not have anything to do with the resolution of the camera but with how many pixels of color are recorded. This is because the human eye is far more sensitive to light than it is to color, so most low-budget cameras compress their images by throwing out a lot of the color information. While the human eye isn’t aware of this loss in the recording, when you try to go through and do color-based processing (which greenscreen keying is) you realize that there’s a lot of missing information that must be compensated for after the fact.
In a standard definition camera, the ratio of light to color recorded is 4:1:1. This means that for every four pixels recorded, four pixels of light information is recorded, but only one pixel of color information is recorded. On HDV cameras (and PAL SD cameras) the ratio is 4:2:0. While this may seem to suggest that two pixels of color are recorded for every four pixels of light, in reality it has nearly the same ratio as SD cameras. This is because, while two out of four pixels of color are recorded in the first line, NO color pixels are recorded in the second line. As such, the same number of color pixels are being recorded every two lines as in the SD camera. (Now, HDV does have higher resolution so the actual pixels are smaller which can help your key “appear” a little cleaner to the naked eye.)
With that said, you really start seeing improvements in color recording in true HD cameras like the HVX200, which records at 4:2:2. Cameras like this do in fact record two color pixels for every four light pixels and, because it’s HD, it has more pixels than the SD camera. (To give you an idea, the big cameras that are commonly used in Hollywood films for greenscreen work usually record at a 4:4:4 ratio, which means they use such low-compression that they retain as much color information as light information.)
With that said, even though each low-budget camera has its limitations, you can in fact get some good keys with nearly any style of camera. I’ve seen an amazingly impressive trailer for one greenscreen film called Dirty Trowsers (that looks a lot like Sin City) that was shot with the standard definition Panasonic DVX100B. To find out more about how you can key well with low-budget cameras, be sure to check out the review Ryan Graham did of our new book, GreenScreen Made Easy, which specifically goes in-depth on these subjects.
Hopefully these questions and answers have served to help shine some light on these things for other readers who have wondered about them, as well. We’ll start doing more of these questions and answers sections in future issues.
God
Bless,
Jeremy Hanke
Editor
Microfilmmaker Magazine
