I'm attempting to draw text onto a video using either the avconv or ffmpeg commands. When specifying x= for where on the final video to place the text, the 'text_w' value is evaluating to 0, rather than the width of the rendered text as it should.

• I'm using Ubuntu 12.04
• I've got avconv version 0.8.3-4:0.8.3-0ubuntu0.12.04.1
• and ffmpeg version 0.8.3-4:0.8.3-0ubuntu0.12.04.1

Example command:

avconv -i test.mov -vf "drawtext=fontfile='/usr/share/fonts/truetype/ttf-dejavu/DejaVuSans.ttf':text='test text':x=text_w:y=50:fontsize=24:fontcolor=black" texted.mov

This command causes the text to be printed as if x were set to 0.

What I'd really like to be able to do is center the text horizontally using something like this:

avconv -i test.mov -vf "drawtext=fontfile='/usr/share/fonts/truetype/ttf-dejavu/DejaVuSans.ttf':text='test text':x=(main_w-text_w)/2:y=50:fontsize=24:fontcolor=black" texted.mov

Using ffmpeg for to attempt the same ends with the same result

ffmpeg -i test.mov -vf "drawtext=fontfile='/usr/share/fonts/truetype/ttf-dejavu/DejaVuSans.ttf':text='test text':x=(main_w-text_w)/2:y=50:fontsize=24:fontcolor=black" texted.mov