Triode;669885 Wrote:
> Sounds good - the main change is in how stream endings are handled - the
> code should now treats errors and end of list again streams
> differently...
>
> If you find any errors then output of /var/log/messages would be
> interesting.Mon 21 November
> Otherwise - interested if stream is played to the end as with the
> proxied streaming before moving on. (I think it is, but interested in
> additional test resource here!)
I have played out half a dozen streams under both proxied and direct
streaming scenarios, and found no audible differences at the end of the
streams.
Foolishly I attempted a quick millisecond accuracy check, by playing
out over the headphone socket and digitally recording the result in my
computer. That failed because the Radio doesn't believe that my
computer's line in socket is a headphone.
However the log might give a better clue:
I noticed that the proxied streaming method appears to deliver a very
slightly shorter stream (about 30ms over a stream of 3 minutes
duration). I couldn't detect whether this miniscule difference occurred
at the beginning or the end of the stream. The stream I used was Bells
on Sunday - broadcast today at 00:45.
Proxied streaming - first run - 7,918,155 samples
Code:
--------------------
Nov 21 12:56:13 squeezeplay: INFO audio.decode - decode_start_handler:279
init decoder aac
Nov 21 12:56:13 squeezeplay: INFO audio.decode - Playback.lua:447 connect
192.168.1.70:9000 GET /stream.mp3?player=00:04:20:29:a4:29 HTTP/1.0^M
<snip>
Nov 21 12:59:13 squeezeplay: playback_callback:346 Audio underrun: used 437
frames, requested 441 frames. elapsed samples 7918155
--------------------
Proxied streaming - second run - 7,918,155 samples
Code:
--------------------
Nov 21 13:06:50 squeezeplay: INFO audio.decode - decode_start_handler:279
init decoder aac
Nov 21 13:06:50 squeezeplay: INFO audio.decode - Playback.lua:447 connect
192.168.1.70:9000 GET /stream.mp3?player=00:04:20:29:a4:29 HTTP/1.0^M
<snip>
Nov 21 13:09:51 squeezeplay: playback_callback:346 Audio underrun: used 437
frames, requested 441 frames. elapsed samples 7918155
--------------------
Direct streaming - First run - 7,919,478 samples
Code:
--------------------
Nov 21 12:59:49 squeezeplay: INFO audio.decode - decode_start_handler:279
init decoder aac
Nov 21 12:59:49 squeezeplay: INFO audio.decode - Playback.lua:447 connect
95.140.226.102:1935 nil
<snip>
Nov 21 13:02:48 squeezeplay: INFO audio.decode - messageType4:263 message
type 4 - user control message event 1: EOF - exiting
Nov 21 13:02:48 squeezeplay: INFO audio.decode - readL:771 end of stream
Nov 21 13:02:50 squeezeplay: playback_callback:346 Audio underrun: used 138
frames, requested 441 frames. elapsed samples 7919478
--------------------
Direct streaming - Second run - 7,919,478 samples
Code:
--------------------
Nov 21 13:03:28 squeezeplay: INFO audio.decode - decode_start_handler:279
init decoder aac
Nov 21 13:03:29 squeezeplay: INFO audio.decode - Playback.lua:447 connect
95.140.226.102:1935 nil
<snip>
Nov 21 13:06:27 squeezeplay: INFO audio.decode - messageType4:263 message
type 4 - user control message event 1: EOF - exiting
Nov 21 13:06:27 squeezeplay: INFO audio.decode - readL:771 end of stream
Nov 21 13:06:29 squeezeplay: playback_callback:346 Audio underrun: used 138
frames, requested 441 frames. elapsed samples 7919478
--------------------
I think that's enough investigation for now. I've heard enough
Grandsire Triples for one day.
--
mrw
------------------------------------------------------------------------
mrw's Profile: http://forums.slimdevices.com/member.php?userid=38299
View this thread: http://forums.slimdevices.com/showthread.php?t=53229
_______________________________________________
plugins mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/plugins
