Cracking the Video Time Code

Pat Bailey
July 2005

» Next Image Previous Image «

Many devices use time code of one sort or another to synchronize with other devices and with the real world of atomic clock time. For our purposes, time code is the standardized system that video cameras use to assign a digital time signature to each frame of video that's shot. Essentially, there are two kinds of time code: drop frame and non-drop frame. One of the important things to know about these time codes is that they are simply a way of counting video frames. It is a misconception that drop-frame time code actually drops any frames of video.

Instead, it merely drops frame numbers. Think of drop-frame time code as pushing back the second-hand of a clock to make synchronization adjustments. Pushing back the hand has no effect in the real world. When the system drops frame numbers, no changes are ever made to the video frames themselves. The only difference between drop and non-drop frame time code is in how they count frames.

The Society of Motion Picture and Television Engineers (SMPTE) created drop and non-drop frame time code. While it might seem logical to count in whole-number frames per second, technical issues with the NTSC video signal (the North American video signal standard) mean that 30 frames per second won't work. The NTSC signal, developed in 1953, uses 29.97 frames-per-second video signals in order to be backward compatible with black and white televisions (among other reasons). The signal starts out like a black and white signal, but then uses three frames to deliver the color information.

The result of the NTSC video signal is that time code frame-counts gradually drift away from time as it is measured in the real world. For instance, when we use non-drop frame time code (and we usually do), you'll notice that the time code duration reads slightly longer than real time: 3.5 seconds or 108 frames longer for every hour of video. By contrast, drop-frame time code remains accurate with real time by dropping (that is, not counting) all of the video frames, specifically skipping 108 numbers every hour.

Another kind of time code is longitudinal time code (LTC). This time code system converts information into a bit-stream that the camera records on audio track two on compatible machines. LTC is either recorded when shooting or dubbed onto your original tape prior to editing. Some of LTC's disadvantages include the elimination of one audio track and the fact that the time code is not visible when the tape is paused or slowly jogged during playback.

Vertical Interval time code (VITC) eliminates these problems with LTC. VITC (pronounced vit-see) is recorded as part of the off-screen video signal, but, like LTC, VITC is bit-stream information. The VITC bit-stream exists in the video signal's raster-line, however, instead of one of the audio tracks. VITC reads accurately when the tape is paused, but must be recorded during shooting or onto a copy of the original tape.

Since drop-frame and non-drop frame time codes are just two methods of counting video frames, using one or the other is sometimes a matter of the editor's preference. In some instances, however, you'll make this decision based on your client's needs or your editing system limitations. If you work with Mini DV time code, you'll use drop-frame time code, but other systems, such as DVCAM and DVCPRO, offer the choice of using drop-frame or non-drop frame time code. Remember, it is just a system of counting frames, so the image is unaffected by all of this accounting.

For personal and non-commercial purposes, using either time code will be sufficient. The decision really begins to matter primarily when you are working on commercial and professional broadcast projects. Professional broadcast video must use drop-frame time code, because of the need for timing precision.

• Digg This!
• del.icio.us
• Technorati
• StumbleUpon
• Reddit