Let me see if I can shed some light on this.

Time-code character format (TCF) is a precise way of representing
time-code, with additional information describing time base
derivation. It takes the form of


where HH is hours, MM is minutes, SS is seconds, FF is frames and
XXXX is an optional sample count. "i" represents separators each of
which conveys its own meaning. The hours, minutes and seconds follow
the ascending progression of a 24 hour clock starting with 0 hours, 0
minutes and 0 seconds to 23 hours, 59 minutes and 59 seconds. The
frames are numbered from 1 to n where n is indicated by the frame
count and time base indicator (the first "i"). The second separator
"i" codes film framing information, the third separator indicates
video field and time-code type [PAL, NTSC non-drop frame, NTSC drop
frame]. The separator between frames and sample count indicate the
sampling frequency. The sample count itself is the integer sample
count beyond the video frame count.

Some examples:


This 1 hour, 20 minutes, 20 seconds and 10 frames, the frame count is
30 with an unknown time base, film framing is not applicable, the
video field is field 1, NTSC non-drop frame, the sampling frequency
is 48000 with a sample count of 1000.


This is zero hours, 10 minutes, 17 seconds and 1 frame, the frame
count is 30 with a time base of 1.001, film framing is one to one
NTSC, and the video field is field 2, NTSC non-drop frame.

The standard has a series of tables that detail the meaning of each separator.

I am not sure that you need to have the individual SMPTE values since
TCF can handle all of them (and more) in its coding scheme.



>Based on David's input, I'm willing to add a TCF
>value to the list of time codes, but I'm reluctant at
>this point to remove the individual SMPTE values.
>I'm also concerned that a single value of "TCF" may
>be inadequate, and that AES31-3 may specify some
>variation in TCF (just as SMPTE has its variety
>of flavors), and so we may need more than one
>"TCF" value.  But without a copy of the standard
>in front of me, that's hard to judge.  I'll add
>a TCF value for the moment, leave in the SMPTE
>values, and I'll try to order up a copy of the
>AES standard to review to see if we should put
>additional, more extensive support for TCF
>time code values in.  If David could supply some
>brief (sentence or two) descriptive language
>regarding TCF for me to include in the documentation
>section of the schema, I'd appreciate it.
|  David Ackerman               |  Telephone: 617-495-2794         |
|  Audio Preservation Engineer  |  Fax: 617-496-4636               |
|  Archive World Music          |  e-mail: [log in to unmask] |
|  Harvard University           |                                  |
|  Cambridge, MA 02138          |  PGP key 0xE928B52F              |