Tascam HD-P2 and Sony HDR-FX1 success stories?

[A. J. deLange]

Yes, that's the stickler. It is a real bummer that the TAO LTC Export is no longer made (http://www.spcomms.com/ltcexport/). I suppose there are some of these still out there and if one is lucky he might find one (none on Ebay at the moment). You might try contacting Sweetpea to see if they could find one for you. The pity is that I don't believe there is much to tranlating the LANC code to SMPTE. I expect it's just level translation and perhaps modulating a baseband code onto a 1kHz carrier or something of the sort. Tying it to the video (which the Export did) would be a little trickier but a good electronics designer could probably reproduce the Export in fairly short order.

The Avit product I was referring to is http://www.avitresearch.co.uk/lancmtc.htm. Send them £75 and it's yours. The problem is that you now need a box to convert MTC to LTC and, as you say, these are relatively expensive, are rack mount or half rack width and most run on AC mains. I've done what you want to do with the MOTU Digital Time Piece. It only draws 7 watts so it's possible to run it in the field from a small battery and inverter for a fairly long time. Not exactly what you had in mind I'm sure.

[A. J. deLange]

Originally Posted by Scott Pierce

But does the HD-P2 timestamp the .bwf files with LTC? Then when in FCP5, then what?

I now have the answers to these. Yes, the HD-P2 does timestamp the files with time code. If you open one of the files and locate the 'bext' chunk and navigate your way through it eventually you will come to a date time group in ASCII format e.g. 2006-04-0220:07:40. This is the time at which the recorder opened the file and is simply a reading of the system clock. Eight bytes of hex data immediately follow this DTG e.g. ...07:40ABCDEFGHIJKLMNOP. Rearrange these bytes as OP MN KL IJ GH EF CD AB. Convert this to decimal by multiplying the decimal equivalent of the hex digit O by 16^15. Add P times 16^14, M times 16^13 and so on. This is the value of the starting timecode in samples. Divide by the sampling rate in samples per second. That's the time code in seconds. Convert to hours minutes and seconds. Take any remaining fractional seconds and multiply by the frame rate to get the frame number. Remember that frame numbering starts with 2 on the minute unless it's a multiple of 10. You can now import the clip into FCP, go to Modify>Timecode and enter the LTC you just calculated. This cut will now display time code correctly whatever you do with it within FCP.

Alternatively, and much easier, you can run the .wav file through Sebsky tools which will create a .mov file with a timecode track. This can be imported into FCP and will display proper time code.

I've been experimenting with an HD-P2 with an XL-H1 and find the Sebsky tool path workable. Sound stays locked for at least 10 minutes (longest cut I have tested) but there is a 4 frame offset between the audio and video.