Hi,
I'm building a video app for android with air and I cant stop netstream from streaming a video.
When a video is playing the user can choose to move to the next video but the current video still streaming even when i call ns.close();. On the desktop it works.
Is this a known bug? or am i doing something wrong...
The code is the standart netstream method, and when i want to load a new video:
ns.close();
ns.play("some url");
I need to solve this issue and the app is finished! please help!
was hoping this problem would be history in AIR 3.3.
if I am doing the self-contained AIR method properly, I am not seeing any improvement.
navigating between a number of videos, it hoses up on Android, while on the desktop, it's very sweet.
i should note that all my videos are streaming wirelessly on my network in *both* the case of Android and desktop. In neither case are the files local.
hope this helps. i'm probably not the most qualified to submit a bugbase report, but if need be..
Anyone can submit to the bugbase and those that have directly been impacted by a particular bug in a reproducable way are the most ideal to do it. Just jump on there, use the search and use a couple keywords like "android netstream" and see if anyone else has reported it already. If not, it would be a good idea for you to do it.
Hi,
This bug is confirmed by Adobe but there is no priority to fix it.
Vote for the following bug:
https://bugbase.adobe.com/index.cfm?event=bug&id=3190676
See notes for more details.
Philippe
Hi Zone3D. This is a known bug with AIR on Android. I just submitted the fix for it yesterday and it should be present in the next release of AIR 3.4. Take a look at this thread for the related discussion: http://forums.adobe.com/message/4563418#4563418. This is the bug that was fixed - https://bugbase.adobe.com/index.cfm?event=bug&id=3070987.
Thanks
Kalvyn
We just had a drop a couple days ago (http://labs.adobe.com/downloads/air3-4.html), so I think the next one will be in August sometime - maybe around mid-August? Not sure of the exact date though, sorry!
great results!!!! i have "Beta Build" "3.4.0.2410" now showing on the Android tablet and the videos are now streaming very sweet. better than sweet.. wicked.
note that I initially, erroneously downloaded "Beta Build" "3.4.0.3010" (the IOS version) as the regular SDK was not initially extracting in winzip. Don't make my mistake! Thanks!!!!!!!!!!!!!!
Althought the EULA changed about building apps with anything less than public releases, I'd only test with 3.4 and try to wait for at least the release candidate before you compile anything for a store. They're perfectly able to change something that may compromise your app before the public release, or worse, there's a new bug introduced you haven't found yet. The best bugs hide until a user does what you never think they'd do.
I had put my own version textfield but apparently Adobe adds a permanent one as well identifying it as a Beta. This helps reinforce not using it in the marketplace.
In the back of my mind I'm a bit curious why the fix was not in the later version that I would normally use when I develop for both Android and IOS which I have done in the past, ... rather I just hope it doesn't get missed somehow going to 3.4 public release.
note am using captive runtime always now to avoid most of those surprises but like you, do intend only to use the public releases.
Further feedback regards this fix. It certainly opens up video app development and allows for slick demos.
However, since videos are not freezing up I can now see clearly the buffering that's going on which seems to be another area that needs attention... Any hints/feedback regards this would be appreciated.
Specifically, I find a fair disconnect in behaviour between video playback on the desktop and on Android. This may be because of Android design but this should be noted since it might lead another change at Adobe's end.
The behavior is as follows (note that this includes my own created videos and videos obtained from the internet -- my videos work a little bit better and I know that one thing I usually do is CBR, constant bit rate which may help things) :
on the desktop===>: 100%, all wireless network video streams buffer flawlessly... roughly speaking, about 20 seconds of video every second, plus if you seek to an unloaded part, the video patiently waits until that part is loaded
on android===>: all wireless network video streams seem to have their own signature for buffering. Although many of the streams will buffer normal, maybe half buffer normally for 100 to 250 seconds of video data and then stop and wait for the playback to catch up, after which time they buffer slowly. This stop can therefore be 100+ seconds that no buffering is occuring. A few of the videos will even forget to continue buffering ( a netstream.buffer.empty is eventually fired and game over ) The behaviours are essentially repeatable for each video (I have my worst case video) although if you seek forward sometimes it will remember to resume buffering right away. Another symptom is that unlike the desktop, you cannot seek to an unloaded part, instead you generate a nestream.seek.invalidtime and the video simply resumes (I actually workaround this as I suspect it causes problems for me if I don't).
There are other discrepencies that I ask which is proper / better ? ===> on desktop a pause does not interrupt buffering, while on Android a pause stops the buffering....====> on desktop there seems to be a cache for the videos if you jump around between them while on Android everything is cleared out on the dispose()
as mentioned above, any hints/feedback would be appreciated. I'm trying to relate this to other outstanding forum and outside forum related issues and still thinking the fail part might be an Android/hardware realtionship but not sure and analyzing.
thanks.
Message was edited by: hiteck7 4:53pm Friday
North America
Europe, Middle East and Africa
Asia Pacific