I slipped on my Vibram KSOs yesterday and went out to run a few camera tests for an upcoming project.
Specifically, I was testing a new camera rig configuration and evaluating the stability of video shot while running.
The camera rig is built up from Gini parts, providing a wide and stable grip with the camera at the center. This configuration made the camera very easy to handle while running. The articulating handles are essential as they allow me to adjust the grips so that my wrists are in a neutral position. In this simple configuration, the side handles could also be adjusted to change the angle of the camera without having to change the position of my arms.
- Panasonic GH2
- Canon FD 28mm/2.8 (using a micro 4/3 to FD lens adapter)
- Rode VideoMic Pro
- DIY variable ND filter
Though I didn’t use any of the recorded audio, I used the Rode VideoMic Pro and “Deadcat” windscreen to evaluate the quality of the audio as I will want environment sound for the actual project.
I was surprised by how stable the footage was of my running and walking around. I made no effort at all to move smoothly as the purpose of this test was to determine just how bad the video could get. It does make sense though, because most of the potentially shaky movement was vertical, the effects of which are greatly minimized because the camera is pointing down at the ground.
Nonetheless, I am heartened to know that the footage can indeed look quite smooth.
Unfortunately, the GH2’s codec completely falls apart with this type of footage. It just can’t handle the fast moving, constantly changing textures (grass, sidewalk, etc.). I will have to test my hacked GH1 to see if the increased bitrate alleviates the issues with image quality. I could hack the GH2 to shoot with a higher bitrate, but I don’t want to void my warranty just yet.