Image steadiness test questions
Posted 02 March 2007 - 02:07 PM
Posted 02 March 2007 - 03:05 PM
I want to test a camera that I may buy. In the long run I hope to apply post-effects (composited elements, rotoscoping, etc.) to the footage, so I'm concerned about image steadiness. I've read here that the Arri S should produce solid footage, and I know the basics about shooting an image steadiness test: solid mount, sharp image of a test chart / lines, shoot twice underexposing one stop each time. But I don't know the important stuff - how to evaluate the results. So here are some questions. What's the best way to view the test results? Transferred to video? And do I need to be careful to use or avoid a certain type of transfer to make sure that it isn't influencing the results? Is there a way to quantify any "unsteadiness"? Like, is there a protocol for shooting a chart with an exact line spacing? Are there different kinds of "unsteadiness", and if so, are they indicators of different kinds of camera issues? Lots of questions - if you have answers or know of another source of info I'd appreciate it.
You seem to have a good handle on reg. testing. If your testing an Arri 16 you don't have to worry about the second pass being in the same perferation as the first pass as it is only one perf. per frame. Don't forget to off set the target (or camera) slightly in the vertical and horizontal planes on the second pass so you can detect any reg. problems. Vertical jitter would indicate a registration pin error (check for perf. damage) and
a horizontal error (sometimes called weave) would be a side rail problem. An image going slightly in and out of focus (more so in the central image area) would indicate a breathing problem due to incorrect pressure plate adjustment. Be sure your loops going in and out of the gate are the right size as this can effect the cameras registration. There is no need to transfer the film to any other medium just pop it into the projector and view. If the whole image that is both sets of lines are moving up and down relative to each other then it's just a projector error.