Roku Developer Program

Developers and content creators—a complete solution for growing an audience directly.
cancel
Showing results for 
Search instead for 
Did you mean: 
renojim
Level 8

isPlaybackPosition inconsitency for HLS v3.x vs. v5.x

I've never seen this mentioned before, but an HLS stream with a starting index other than 1 (or is it 0?) produces different indices for isPlaybackPosition messages depending whether the box is running v3.x of the firmware vs. v5.x.

For example, if I have the following playlist:
#EXTM3U
#EXT-X-VERSION:2
#EXT-X-TARGETDURATION:10
#EXT-X-MEDIA-SEQUENCE:69025
#EXTINF:10,
segment_69025.ts
#EXTINF:10,
segment_69026.ts
#EXTINF:10,
segment_69027.ts
#EXTINF:10,
segment_69028.ts

After 10 seconds, the first playback position reported under v5.x of the firmware is 10, but under v3.x of the firmware the offset of 69025*10 seconds is added to the position giving a playback position of 690260. I'm not sure which is correct, but it would be nice if they were consistent.

Likewise, setting the playback position with the PlayStart attribute works differently under the two different firmware versions, but it is consistent with the playback position reported. The problem occurs if you start such a stream on a v3.x box and try to resume on a v5.x box (or vice versa).

-JT

edit: changed Seek() to PlayStart'; I haven't looked at how Seek() acts.
0 Kudos