D.V.T.R. Editing Considerations for Multiplexed Audio vs. Separate Audio Edge Tracks
|Publisher:||Society of Motion Picture and Television Engineers|
|Publication Date:||1 February 1981|
|Conference Location:||San Francisco, CA, USA|
|Conference Date:||6 February 1981|
|Page(s):||72 - 79|
In a conventional helical scan video tape recorder, the video and audio signals are recorded and reproduced entirely independent of each other so that it is possible to picture-lock two video recorders and edit either picture or audio as desired. This statement may not be true of an all digital recorder wherein both the video and audio signals depend upon clocks in order to convey information. The T. V. signal, as we know it today, is a time structured system which needs an identifiable starting point and periodic updating in order to work. The digital representation of a T.V. signal while different in many respects will still be time structured. The audio signal, on the other hand has not been considered a time structured signal until the advent of the digital audio recorder. In the past a slight slippage of speed between two recorders might possibly be noticed as a change in pitch but in a digital world that same slippage of time base would cause a loss of clock synchronization with the intendent disastrous sound effects. To a degree it is possible to reclock two signals not running synchronously by using a process known as asynchronous buffering. The degree of slippage and the interval of time over which it occurs determine the amount of digital storage needed. The cost of such buffering and the problems created by the offset times usually preclude any buffering beyond that required for small instantaneous time variations like clock jitter or machine flutter.