theora - Mar 2009 - ffmpeg2theora: Garbled AC3 Audio

I recorded a television show using my HDHomeRun. The resulting file
contains MPEG-2 video and AC3 audio. Specifically, ffmpeg2theora
shows:

Stream #0.0[0x1e0]: Video: mpeg2video, yuv420p, 1280x720 [PAR 1:1 DAR
16:9], 38810 kb/s, 59.92 tbr, 90k tbn, 119.88 tbc
Stream #0.1[0x80]: Audio: ac3, 48000 Hz, 5.1, s16, 448 kb/s
Pixel Aspect Ratio: 1.00/1   Frame Aspect Ratio: nan/1

When I feed it to ffmpeg2theora using the -a switch for quality-based
encoding for the audio it sounds fine but as soon as I replace the -a
switch to specify a bitrate (specifically -A 128 -c 2 -H 44100) the
audio is garbled and sounds horrible.

This isn't new to 0.24 but with the alpha release of Theora 1.1
("Thusnelda:) I wanted to look into it Theora again.

Just wanted to follow up with more information that I learned: I  
recorded a different program where the audio was still ac3 but in  
stereo (not 5.1) and the -A 128 -c 2 -H 44100 flags did work correctly.

This was the audio stream, as reported by ffmpeg2theora: Stream  
#0.1[0x80]: Audio: ac3,  48000 Hz,  stereo,  s16,  192 kb/s

So it appears to be an issue with it being 5.1, not with the ac3 audio  
as I originally suspected.

Oh, and I can do the 5.1 to stereo conversion in FFmpeg proper, so I  
downloaded the source for for FFmpeg 0.5 and FFmpeg2theora but I'm not  
sure what FFmpeg2theora is doing differently.