Closed Bug 846571 Opened 11 years ago Closed 6 years ago

Embedded youtube videos don't use Gaia Video app, inline playback is slow and stutters

Categories

(Firefox OS Graveyard :: Gaia::Video, defect, P3)

ARM
Gonk (Firefox OS)

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: Harald, Unassigned)

References

()

Details

(Keywords: perf, Whiteboard: [c= p= u= s=])

After having bug 834965 fixed I tested a few videos. They now play inline whereby I had expected them to open the Video app, like videos do on the youtube website.

The inline playback in my tests was slow and stutters badly. I guess they use the Webm software codec. Playing the same videos in the Video app was "buttery".

This affects all apps and websites that embed youtube videos (there are a lot).
We are getting html5 video (ie webm):

I/GeckoDump(  614): Loading http://www.youtube.com/get_video_info?html5=1&video_id=eRsGyueVLvQ&cpn=x0BXFHEEQlMJ-cdt&eurl=http%3A%2F%2Ftestno.de%2Fyt.html&ps=native&el=embedded&hl=en_US&sts=136200532652&width=320&height=215
I/GeckoDump(  614): Loading http://r16---sn-o097znez.c.youtube.com/videoplayback?expire=1362121998&cp=U0hVR1FNT19OUkNONV9JR1NKOmY2XzhHSVFtVWlR&ipbits=8&fexp=923501%2C901700%2C904307%2C916612%2C920704%2C912806%2C902000%2C922403%2C922405%2C929901%2C913605%2C925006%2C906938%2C931202%2C908529%2C920201%2C930101%2C906834%2C926403%2C901451&sparams=cp%2Cid%2Cip%2Cipbits%2Citag%2Cratebypass%2Csource%2Cupn%2Cexpire&newshard=yes&ratebypass=yes&id=791b06cae7952ef4&upn=9kH7yop6Xsw&ms=au&mt=1362097893&mv=u&source=youtube&itag=43&sver=3&key=yt1&ip=63.245.220.224&cpn=x0BXFHEEQlMJ-cdt&signature=BE511B10A0E130D5FB32ADE91A618865A67BCC9E.7199F7A2012D2D9CBEFB2FA971708C948A926A06
E/GeckoConsole(  614): Content JS LOG at app://system.gaiamobile.org/js/identity.js:25 in onMozChromeEvent: mozChromeEvent audio-channel-changed
E/GeckoConsole(  614): Content JS LOG at app://system.gaiamobile.org/js/identity.js:25 in onMozChromeEvent: mozChromeEvent visible-audio-channel-changed
I/GeckoDump(  614): Loading http://i3.ytimg.com/vi/R6MlUcmOul8/mqdefault.jpg
I/GeckoDump(  614): Loading http://i2.ytimg.com/vi/IN6w6GnN-Ic/mqdefault.jpg
I/GeckoDump(  614): Loading http://s.youtube.com/s?html5=1&ns=yt&ps=native&el=embedded&hl=en_US&cr=US&cpn=MZdemMR4t0Yzb6Sx&docid=eRsGyueVLvQ&referrer&plid=AATW0jexSOK3848d&fmt=43&fs=0&screenw=320&screenh=480&w=320&h=215&vw=640&vh=272&playerw=320&playerh=215&vid=-Wgw5T5ojB8yNW8z6rhLDdRSOD4zIqE1C&rt=2.647&mos=0&volume=100&eurl=http%3A%2F%2Ftestno.de%2Fyt.html&playback=1&len=888.522&fexp=923501%2C901700%2C904307%2C916612%2C920704%2C912806%2C902000%2C922403%2C922405%2C929901%2C913605%2C925006%2C906938%2C931202%2C908529%2C920201%2C930101%2C906834%2C926403%2C901451&hmempf=5&hmemdf=5&hmempresented=2&hmempainted=0&st=0.054&et=0.054
I/GeckoDump(  614): Loading http://s.youtube.com/s?html5=1&ns=yt&ps=native&el=embedded&hl=en_US&cr=US&cpn=MZdemMR4t0Yzb6Sx&docid=eRsGyueVLvQ&referrer&plid=AATW0jexSOK3848d&fmt=43&fs=0&screenw=320&screenh=480&w=320&h=215&vw=640&vh=272&playerw=320&playerh=215&vid=-Wgw5T5ojB8yNW8z6rhLDdRSOD4zIqE1C&rt=12.674&mos=0&volume=100&eurl=http%3A%2F%2Ftestno.de%2Fyt.html&vtmp=1&len=888.522&fexp=923501%2C901700%2C904307%2C916612%2C920704%2C912806%2C902000%2C922403%2C922405%2C929901%2C913605%2C925006%2C906938%2C931202%2C908529%2C920201%2C930101%2C906834%2C926403%2C901451&hmempf=93&hmemdf=93&hmempresented=93&hmempainted=0&st=0.054&et=4.626

But we don't use the hw decoder for webm afaik, hence the poor performance.
Keywords: perf
Whiteboard: [c= p= u= s=]
Will the media team take a look at this and see if it is still a problem?
Status: NEW → UNCONFIRMED
Ever confirmed: false
Flags: needinfo?(hkoka)
Flags: needinfo?(sotaro.ikeda.g)
Flags: needinfo?(hkoka)
(In reply to Dave Huseby [:huseby] from comment #2)
> Will the media team take a look at this and see if it is still a problem?

Confirmed mpeg4 video playback. No problem about inline playback.
Flags: needinfo?(sotaro.ikeda.g)
Setting this back to NEW. UNCONFIRMED refers to the filer status, not the bug status.

Sotaro, can you clarify whether we still don't use the HW decoder for Webm, per comment 1? If we don't, would that still cause the performance issues in question?
Status: UNCONFIRMED → NEW
Ever confirmed: true
Flags: needinfo?(sotaro.ikeda.g)
Priority: -- → P3
Current b2g try to use HW decoder for webm playback if the phone's gonk is since JB. It is fixed by Bug 986381.
Flags: needinfo?(sotaro.ikeda.g)
Does that mean we can go ahead and close this out as obsolete?
Yes. We can close this bug.
Firefox OS is not being worked on
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.