XO Wave
Editions (Why Buy?)
Songs 4 Bugs
Docs
  Installation
  XO Wave Tutorials
  Recording Basics
    OS X Maintenance
    Audio Performance
    Microphone Tips
    Stereo Mic'ing
    Compression
    EQ
    Time Code
    Dither
  Effects
  Windows
  Wizards
  Menus
  Reference
  Tech Notes
  README files
  FAQ
  Search Documentation
Contacting Us
  Support
  Mailing Lists
  Feedback
Links
Affiliates
format for printing
Google
xowave.com
documentation

Important Notice: XO Wave is now discontinued as we prepare to bring you the next generation Digital Audio Workstation called Xonami. This site remains available for anyone who purchased XO Wave in the past. However, please keep in mind that as discontinued software:

  • This site may not contain up-to-date information.
  • Technical support is discontinued, though we will do our best to continue to provide email support, especially to anyone who purchased recently.

XO Wave: All about Time Code

(c) grm_wnr

Timecode allows you to synchronize multiple distinct recordings, typically from different pieces of equipment. For example, you might use timecode to synchronize two analog tape decks so you can record more tracks. You might also use it to synchronize an analog tape deck with a VCR. Synchronizing audio and video is especially convenient because most professional video standards do not support more than two or four tracks of audio, which is hardly enough to do serious recording and editing.

According to industry lore, timecode was originally developed by the military in order to synchronize missile firings. The Society of Motion Picture and Television Engineers adapted the technology into the modern audio/video/film timecode standard that bears their name: SMPTE. What the SMPTE standard does is associate each video frame with a unique set of numbers which can be thought of as an "address". This "address" is represented as information indicating hours, minutes, seconds and frames. That is, each frame of a film corresponds to an exact SMPTE time written as HH:MM:SS:FF. For example, "03:23:14:05" means 3 hours, 23 minutes, 14 seconds and 5 frames.

Timecode information can be transmitted in a variety of ways. The most common method is to use an audio tone that can be recorded onto an analog tape track or carried over analog audio cables (and sounds like an annoying high pitched tone with frequent clicks). This type of timecode is called LTC (pronounced "LIT-see"), or Linear Time Code, because most traditional video formats carry analog audio information in a linear track on the edge of the video tape, whereas video is stored in diagonal bands through the middle of the tape.

Another way of transmitting timecode information is to "hide" it in the video signal by placing it in between frames of video. That is, it is placed in sections of the video signal which are not displayed on the screen. This time between frames is called the "vertical blanking interval" so this type of timecode is called Vertical Interval Timecode, or VITC (pronounced "VIT-see").

VITC timecode allows you to free up the audio tracks on video tape (although these are generally of poor quality) and it can allow you to read the timecode even when playback is paused. It also can handle offsets in the playback and record heads of video decks, which often delay the timecode. However, most LTC readers allow you to correct for the delay, so little is gained by using VITC. Furthermore, LTC has a historic place among musicians who use it to synchronize audio decks to other audio decks, and LTC equipment tends to be cheaper, so LTC is generally preferred.

Frame rates and tigers and bears, oh my!

The rate at which a video format advances through the frames is called the "frame rate". Unfortunately, this is where it gets complicated: different video and film formats use a variety of different frame rates. Some formats even count frames out of order by skipping over certain frames! Before you start to panic, let me assure you that it will all make (at least some) sense in a moment. But first, a little history to explain this confusing state of affairs.

When television was first popularized, the US decided to use a frame rate of 30 frames per second (fps). Each frame was separated into two "fields", each containing every other horizontal line of the image. Thus the "field rate" as it is sometimes called, is 60 Hz: exactly the frequency of US power, which was terribly convenient back in the day (before steady crystal clocks and clever synchronization circuitry). In Europe, where most power is 50Hz, they used 25 fps.

When color video was introduced, some people feared that a subtle problem might cause interference between the video and audio carrier signals. To solve this, the color television standard dropped the frame rate from 30 fps to 29.97 fps. (interestingly, some have argued -- although far too late for it to do any good -- that this change was completely unnecessary). The new standard, called NTSC (named after the committee that created it: National Television Systems Committee) is the analog video format used for television and video in much of the world.

The result of this change in frame rates was that SMPTE timecode slowly drifted out of sync with reality ("wall-clock time"). This matters a lot to people setting up broadcast TV schedules, so a new SMPTE timecode type, called "drop-frame timecode", which skipped over certain frames, was created to more closely match real time.

The trick is to skip over certain frames so that timecode and "wall-clock time" are close together, allowing timecode to "catch up" with reality. Although you won't need it on a day-to-day basis, you can impress other audio engineers at cocktail parties if you know exactly how it works, so here's the rule:

Every frame that is of the form: XX:XX:00:00 or XX:XX:00:01 is dropped unless it is on a ten minute interval, i.e.: XX:X0:XX:XX. That is, the first two frames of every minute, except every tenth minute, are dropped.

To distinguish between the two slightly different timecode standards, drop timecode is delimited with ';' instead of ':' between fields. Many timecode readers (both hardware and software) show "sub-frames", which are simply hundredths of a frame, as well. So non-drop timecode looks like:

HH:MM:SS:FF.sf

and drop-frame timecode is of the form:

HH;MM;SS;FF.sf

Because all video in Europe uses the either the PAL (Phase Alteration Line) or SECAM (Séquentiel couleur à mémoire, French for "sequential color with memory") standard, both of which run at 25 fps, and never had any need for drop-frames, European technicians often don't understand drop-frame rates, and as far as the European AV community is concerned, this is just one more silly thing about Americans. European audio software often gives more timecode options than actually exist, because, understandably, they just can't make heads or tails of all this.

Timecode formats are commonly written as the frame rate followed by ND for non-drop or DF for drop-frame. 29.97DF is often abbreviated as 29. It is a common mistake to forget to include the DF after 29, but don't make that mistake since some production houses actually use 29ND!

One other frame-rate worth mentioning is the frame rate of film: 24 FPS. Film, like PAL or SECAM video, is never drop-frame, so you won't have to worry about that confusion. In fact, generally speaking, you won't have to worry about this frame-rate at all because, for editing purposes, film is usually converted to video in a process called telecine (Pronounced "tele-SIN-ee" or "tele-SIN-a" or "TEL-uh-SIN-a") which also changes the frame rate.

Which timecode should you use?

You may think you don't need to worry about all this too much, since you just bought an advanced timecode reader that has an "auto" feature that automatically figures out what kind of timecode it's reading and you can just use that, right? Unfortunately, it can be very difficult for such devices to accurately detect different types -- especially when the source is noisy, damaged or otherwise imperfect. Even with good quality tapes and equipment it can be hard to tell the difference between 30ND and 29ND, so you should always check with the source of the timecode to verify its format. You don't want to get to the end of editing a two-hour movie only to find out you're out of sync by 7.2 seconds! In case you can't get in touch with the editing house that sent you your tape, or you are trying to generate the right timecode type yourself, here are some good rules of thumb:

  • Never use drop-frame with anything but 29.97 fps!
  • In the US, and other NTSC countries, color video uses 29.97 fps. Long productions must use drop frame, but some short productions such as commercials and even music videos use non-drop.
  • In the US, and other NTSC countries, black and white video uses 30ND, although black and white production, when it is done, usually uses color equipment and, therefore, 29DF.
  • In PAL and SECAM countries, which includes all of Europe, and Australia, most of Asia, and Africa, and some of Latin America, both color and black & white use 25 fps (non-drop).
  • Pure audio work frequently uses 30ND, or 25ND.
  • Film usually runs at 24 fps.
  • European video run at 25 fps.
  • Never use drop-frame with anything but 29.97 fps!

Happy Sync'ing!

-- Bjorn Roche


Legal & Copyright This page was last modified January 2008.
up
© XO Audio 2005-2008.
All rights reserved.