android.git.kernel.org. Am I looking in the wrong place?
On May 13, 10:47 am, SDS <sstrea...@gmail.com> wrote:
> Hi Ravi,
>
> Regarding the missing Content-Length header, this was on different
> streams. The test case you have has the Content-Length header (which
> is one reason I provided that test case specifically, to simplify the
> issue).
>
> Believe it or not, I'm having success with 128kbps MP3's to the device
> over 3G. Same goes with 192kbps and 256kbps. At 320kbps the problem
> becomes apparent again. What I find most strange about this is that
> the test case I sent you was at 192kbps but I'm still having problems
> with that particular track. The tracks I'm tsting with today are
> being transcoded on the fly by FFmpeg, so bitrate aside the encoder
> may be the notable difference there.
>
> Your thoughts?
>
> On May 12, 10:01 pm, rktb <yend...@pv.com> wrote:
>
>
>
> > Hi,
>
> > I was actually able to see the content length in the http response.
> > Here is a snippet from my log:
> > PVLOG:TID(0xe9bf8):Time=830:HttpParsingBasicObject::parseResponse()
> > file size = 5346201
>
> > But, yes, the auto-pause/auto-resume does indeed depend on whether or
> > not the content length is present in the header.
>
> > If you try to put the new OpenCORE code under your application, you
> > would need to write a new MIO [media input/output component] that
> > would dump the raw PCM data back to the java layer, which would then
> > use the AudioTrack API to render the data. But, I think that this
> > would be a bit involving.
>
> > I am still trying to think what other properties could be a factor.
> > Did you try:
> > - "smaller" content?
> > - lower bitrate content?
>
> > -Ravi
>
> > On May 12, 8:39 pm, SDS <sstrea...@gmail.com> wrote:
>
> > > Hi Ravi,
>
> > > Thank you kindly for your responses.
>
> > > I've tried MP3 streams without ID3v2, but I should point out a
> > > difference that may or may not be significant in the context of this
> > > defect. The HTTP headers for the MP3 streams were without a Content-
> > > Length header. In terms of the initial observations I reported, the
> > > intermittent behavior is identical but without the increasing buffer %
> > > callbacks. I'm assuming this is expected, but I'll let you decide.
>
> > > Regarding deployment of the latest code for OpenCORE with our
> > > application, so that we can take advantage of your fix, what would be
> > > involved from a development perspective, and how would you compare
> > > that level of effort with us rolling our own HTTP streaming / MP3
> > > decoding framework using native code?
>
> > > On May 12, 9:33 pm, rktb <yend...@pv.com> wrote:
>
> > > > One thing to try is to encode a clip without Id3v2, and see the
> > > > behavior. That would reduce the probability of "insufficient data" by
> > > > a bit.
>
> > > > -Ravi
>
> > > > On May 12, 5:02 pm, rktb <yend...@pv.com> wrote:
>
> > > > > Hi,
>
> > > > > I was able to reproduce the problem, and it is definitely timing
> > > > > dependent. For the eclair codebase, I found a problem in OpenCORE's
> > > > > mp3 parser node. That, however, has been fixed on the latest codebase
> > > > > available at kernel.org.
>
> > > > > Now, for applications that are being written for existing Android SDK
> > > > > releases, I don't have a way out yet.
>
> > > > > -Ravi
>
> > > > > On May 12, 12:47 pm, SDS <sstrea...@gmail.com> wrote:
>
> > > > > > I'm experiencing an intermittent and very frustrating issue with
> > > > > > MediaPlayer on Android when streaming MP3 over HTTP. I've done as
> > > > > > much research and debugging as I possibly can to try and find a
> > > > > > solution, but am coming up short. I can't find reference to this
> > > > > > issue anywhere, including b.android.com.
>
> > > > > > Essentially, what's happening is MediaPlayer will not come out of the
> > > > > > preparing state. The device is receiving audio data during this time,
> > > > > > as indicated by network sniffs and the MediaPlayer buffering callback
> > > > > > being invoked repeatedly with increasing percentage values, but
> > > > > > eventually the buffer appears to become full and stops receiving.
> > > > > > Because prepare() never returns, or more accurately when
> > > > > > prepareAsync() is used and the onPrepared callback is never invoked,
> > > > > > and neither the onError() nor onInfo() callbacks are invoked, the
> > > > > > application can neither call MediaPlayer.start() nor make any sort of
> > > > > > recovery.
>
> > > > > > Other notable observations:
>
> > > > > > - The problem is intermittent, meaning that subsequent attempts to
> > > > > > play the exact same MP3 stream from the exact same server may exhibit
> > > > > > this behavior or not
> > > > > > - The problem seems to be exasperated when the device is connected via
> > > > > > 3G versus wifi
> > > > > > - The problem occurs on different MP3's encoded by different software
> > > > > > - The problem occurs when streaming from different HTTP servers
> > > > > > - The problem cannot be reproduced on any of the Android SDK emulators
> > > > > > - The problem can be reproduced on the HTC Incredible, Verizon
> > > > > > Motorola DROID, and HTC Touch
> > > > > > - When MediaPlayer successfully returns from the preparing state,
> > > > > > there is an info message (what=1, extra=44) received. This info
> > > > > > message is not received during the problematic case.
>
> > > > > > Anyone ever experienced this, or even find a solution?
>
> > > > > > Thank you in advance.
>
> > > > > > --
> > > > > > You received this message because you are subscribed to the Google
> > > > > > Groups "Android Developers" group.
> > > > > > To post to this group, send email to android-developers@googlegroups.com
> > > > > > To unsubscribe from this group, send email to
> > > > > > android-developers+unsubscribe@googlegroups.com
> > > > > > For more options, visit this group athttp://groups.google.com/group/android-developers?hl=en
>
> > > > > --
> > > > > You received this message because you are subscribed to the Google
> > > > > Groups "Android Developers" group.
> > > > > To post to this group, send email to android-developers@googlegroups.com
> > > > > To unsubscribe from this group, send email to
> > > > > android-developers+unsubscribe@googlegroups.com
> > > > > For more options, visit this group athttp://groups.google.com/group/android-developers?hl=en
>
> > > > --
> > > > You received this message because you are subscribed to the Google
> > > > Groups "Android Developers" group.
> > > > To post to this group, send email to android-developers@googlegroups.com
> > > > To unsubscribe from this group, send email to
> > > > android-developers+unsubscribe@googlegroups.com
> > > > For more options, visit this group athttp://groups.google.com/group/android-developers?hl=en
>
> > > --
> > > You received this message because you are subscribed to the Google
> > > Groups "Android Developers" group.
> > > To post to this group, send email to android-developers@googlegroups.com
> > > To unsubscribe from this group, send email to
> > > android-developers+unsubscribe@googlegroups.com
> > > For more options, visit this group athttp://groups.google.com/group/android-developers?hl=en
>
> > --
> > You received this message because you are subscribed to the Google
> > Groups "Android Developers" group.
> > To post to this group, send email to android-developers@googlegroups.com
> > To unsubscribe from this group, send email to
> > android-developers+unsubscribe@googlegroups.com
> > For more options, visit this group athttp://groups.google.com/group/android-developers?hl=en
>
> --
> You received this message because you are subscribed to the Google
> Groups "Android Developers" group.
> To post to this group, send email to android-developers@googlegroups.com
> To unsubscribe from this group, send email to
> android-developers+unsubscribe@googlegroups.com
> For more options, visit this group athttp://groups.google.com/group/android-developers?hl=en
--
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from this group, send email to
android-developers+unsubscribe@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en
No comments:
Post a Comment