tag:blogger.com,1999:blog-5537325711190185140.post466483371627381690..comments2023-11-21T23:10:10.896+13:00Comments on Thundering Herd: Video seeking improvementsChris Pearcehttp://www.blogger.com/profile/13735147508549619230noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-5537325711190185140.post-74841643768333803552009-05-27T01:09:19.452+12:002009-05-27T01:09:19.452+12:00Kroc: you could have made that a lot quicker to re...Kroc: you could have made that a lot quicker to read by just posting the text on the web page :-)<br /><br />I believe Wikipedia has some technology like this, which also has the option of the Cortado Java applet. You should check in with them.Gervhttps://www.blogger.com/profile/06079277032027702609noreply@blogger.comtag:blogger.com,1999:blog-5537325711190185140.post-66171177619131841192009-05-27T00:09:23.171+12:002009-05-27T00:09:23.171+12:00Thanks for all the work you are doing. I’m looking...Thanks for all the work you are doing. I’m looking to solve the HTML5 video adoption issue, and have come up with this: http://camendesign.com/code/files/video_for_everybody/test.html I would really like to open a discussion with Mozilla regarding this.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-5537325711190185140.post-2423478688350118442009-05-24T19:30:58.945+12:002009-05-24T19:30:58.945+12:00Fantastic! This makes video seeking really smooth!...Fantastic! This makes video seeking really smooth!Alfredhttps://www.blogger.com/profile/14659465490974234666noreply@blogger.comtag:blogger.com,1999:blog-5537325711190185140.post-17071064062983755642009-05-23T07:01:45.995+12:002009-05-23T07:01:45.995+12:00Would it be a hack to add some sort of "x-time-ran...Would it be a hack to add some sort of "x-time-range" (with a "x-but-I-already-got-ranges" modifier) header such that the server could hand the client the pieces it needs to do the seek.<br /><br />For flash video you can augment the video with the time code to byte position correspondence for each key frame (see http://www.buraks.com/flvmdi/). Are there really no such option for ogg? And are there no meta-data or comment fields that could be hijacked for the purpose?Unknownhttps://www.blogger.com/profile/16298898435309335743noreply@blogger.comtag:blogger.com,1999:blog-5537325711190185140.post-54521979371715541892009-05-23T05:15:46.805+12:002009-05-23T05:15:46.805+12:00Colin: yes, it's arguably a bug in liboggplay. The...Colin: yes, it's arguably a bug in liboggplay. The 'more correct' behaviour would be to not output any frames until it sees the next keyframe.<br /><br />However, the desired behaviour is to output from the seeked-to frame, which requires seeking back to the previous keyframe - so Chris's approach is what you want to do regardless of what liboggplay does if you seek to a non-keyframe.Unknownhttps://www.blogger.com/profile/08200220127198890009noreply@blogger.comtag:blogger.com,1999:blog-5537325711190185140.post-90571264620106523042009-05-22T16:21:26.121+12:002009-05-22T16:21:26.121+12:00"The problem is that when liboggplay decodes the i..."The problem is that when liboggplay decodes the inter frame, it doesn't apply the inter frame to the frame that was actually prior to the inter frame, it applies the inter frame to some other frame (maybe the key frame from the previously playing segment?)."<br /><br />That sounds like a bug in liboggplay, no?Colin Barretthttps://www.blogger.com/profile/13209110183858813001noreply@blogger.com