mozilla :: #media

17 Apr 2017
02:38bwualwu: Is bug 1348879 duplicated with bug 1350947?
02:39firebothttps://bugzil.la/1348879 DUPLICATE Playbutton not visible when opening a video from givemesport.com in a new tab
02:39firebothttps://bugzil.la/1350947 NEW, alwu Postpone loading NPAPI Flash content until the tab is resumed
02:43bwukentuckyfriedtakahe: After double confirming with kaku, we do shutdown decoder if the video is moved out of viewport. What is the problem you see in www.playstation.com/en-us/games/horizon-zero-dawn-ps4?
02:45alwubwu : yes, see comment 3
02:46bwuI was in a triage meeting and just want to make sure. cool. I am going to change the duplicated bug number.
03:01kentuckyfriedtakahebwu: how quickly do we shut them down?
03:13bwukentuckyfriedtakahe: we reply on the visible event from dom. IIRC, if that video is moved out 3 times of viewport size, the decoder will be shutdowned. vice versa.
03:16kentuckyfriedtakahebwu: perhaps we should talk to kats about a visible state
03:17bwuvisible state?
03:18kentuckyfriedtakahebwu: when we're unable to keep up, we should prefer maintaining the videos that are currently on the screen
03:25bwuI think the current mechanism is fine. It is a tradeoff. If we shutdown videos quickly, user may feel bad if she or he scroll back quickly.
03:27bwuWith current mechanism, we resume videos in advance if the video enters 3 times of viewport size.
03:39bwuwhich is the upside of current mechanism.
04:00Caspy7ok, so, it looks like the most recent driver for a Radeon may need blacklisted or at least that AMD may need contacted (?) https://www.reddit.com/r/firefox/comments/65elkv/fullscreen_in_youtube_causes_system_wide/ https://www.reddit.com/r/firefox/comments/65ra4r/fullscreen_with_html5_on_twitchyoutube_is_laggy/
04:13bwualwu: https://bugzilla.mozilla.org/show_bug.cgi?id=1348909#c2 at that time, the nightly is still 55, right?
04:14firebotBug 1348909 DUPLICATE, nobody@mozilla.org Focusing the Web Skype tab while being called does not output the call audio
04:15alwubwu : ah, yes
04:15alwubwu : but I think you can ask them retest all tests after fixing Flash and pinned tab issue
04:18bwuThat's good idea!
06:37bwualwu: bug 1348326 is not a regression, right?
06:37firebothttps://bugzil.la/1348326 REOPENED, alwu@mozilla.com BBC video/audio fails to load after focusing the videos tab
06:38alwubwu : I don't sure what's your definition about regression here, it was caused by block autoplay feature
06:40bwualwu: patch
06:40alwubwu : it was caused by block autoplay feature
18:03Caspy7on our minimum system requirements page https://www.mozilla.org/en-US/firefox/52.0.2/system-requirements/
18:04Caspy7under the heading "For optimal functionality, we recommend the following libraries or packages:"
18:04Caspy7Shouldn't we list PulseAudio and a version?
21:15padenotCaspy7, yes we should
21:22Caspy7padenot: do I leave this to you? Is there a next step? A bug?
21:24padenotCaspy7, dunno, I'll try to find someone tomorrow, I might now the right person, or a person that will know the right person
21:24padenotCaspy7, it's 11:23pm here, everybody is asleep
21:24Caspy7ok
21:24padenot(and the person is french)
21:25padenot(otherwise it makes no sense)
21:25Caspy7ah :)
21:54Caspy7are we making bug 1224973 apply to videos in the current tab that are not in the viewport?
21:54firebothttps://bugzil.la/1224973 FIXED, dglastonbury@mozilla.com Reduced decoder usage for background videos
21:54Caspy7if not, does a bug need filed?
21:55Caspy7the reason I'm bringing this up is because a user in #firefox is complaining of a page giving very poor performance and apparently it has multiple videos on it playing simultaneously (like 7?)
21:55Caspy7so if we're at a point we can cover this, it'd be awesome
23:24decoderpadenot: gerald: can any of you possibly help me with a question regarding media/libstagefright/system/core/include/utils/TypeHelpers.h ?
23:24geralddecoder: I can try.
23:25decodergerald: i was wondering about this line:
23:25gerald(Though I've dealt with stagefright, but not that particular file)
23:25decoderhttp://searchfox.org/mozilla-central/source/media/libstagefright/system/core/include/utils/TypeHelpers.h#134
23:25decoderin the tests, this and the following functions are repeatedly called with n=0
23:26decoderthen they overflow
23:26decoderis this intended? it looks very weird to me
23:26geraldSure sounds weird! Let me have a look...
23:26gerald(but sadly, not totally unexpected from stagefright)
23:26decoderi guess because it's n-- and not --n the n is first evaluated for the while loop, then decremented
23:27decoderso the loop exits right away i guess?
23:31geralddecoder: Yes, the loop exits right away, so the overflowed number is not actually used.
23:37decodergerald: i see. i will blacklist it somehow
23:37decoderthx
18 Apr 2017
No messages
   
Last message: 36 days and 18 hours ago