Kodi Community Forum
OpenELEC Testbuilds for RaspberryPi Part 2 - Printable Version

+- Kodi Community Forum (https://forum.kodi.tv)
+-- Forum: Support (https://forum.kodi.tv/forumdisplay.php?fid=33)
+--- Forum: General Support (https://forum.kodi.tv/forumdisplay.php?fid=111)
+---- Forum: Raspberry Pi (https://forum.kodi.tv/forumdisplay.php?fid=166)
+---- Thread: OpenELEC Testbuilds for RaspberryPi Part 2 (/showthread.php?tid=184866)



RE: OpenELEC Testbuilds for RaspberryPi Part 2 - popcornmix - 2014-03-20

(2014-03-20, 17:36)mikeb93 Wrote: Is there or will there be a solution to just stop playback and go back to the Home screen of xbmc when the TV shuts off? (cec-commands?) Or can this be achieved somehow? I'd really love a solution like this since I'm using my TV's sleep timer a lot...

It's a good question. Shutting down the Pi (with no easy way to restart it is almost never desired). I'll have a poke around...


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - mikeb93 - 2014-03-20

(2014-03-20, 17:46)popcornmix Wrote:
(2014-03-20, 17:36)mikeb93 Wrote: Is there or will there be a solution to just stop playback and go back to the Home screen of xbmc when the TV shuts off? (cec-commands?) Or can this be achieved somehow? I'd really love a solution like this since I'm using my TV's sleep timer a lot...

It's a good question. Shutting down the Pi (with no easy way to restart it is almost never desired). I'll have a poke around...

right now my pi gets its power from the tv usb port. But as you know this is not giving me enough power to overclock but does give me a solution to shutoff / turn on my pi dynamically with my tv.

commands like tvservice -o and tvservice -p don't actually stop playback...


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - Leopold - 2014-03-20

(2014-03-20, 17:51)mikeb93 Wrote: right now my pi gets its power from the tv usb port. But as you know this is not giving me enough power to overclock but does give me a solution to shutoff / turn on my pi dynamically with my tv.

If you have two USB ports next to each other on the TV you can use a Y cable to provide enough power to overclock. I believe each port usually provides 500mA so you would get a full 1A this way.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - mikeb93 - 2014-03-20

(2014-03-20, 18:04)Leopold Wrote:
(2014-03-20, 17:51)mikeb93 Wrote: right now my pi gets its power from the tv usb port. But as you know this is not giving me enough power to overclock but does give me a solution to shutoff / turn on my pi dynamically with my tv.

If you have two USB ports next to each other on the TV you can use a Y cable to provide enough power to overclock. I believe each port usually provides 500mA so you would get a full 1A this way.

unfortunately there's only one port. It's an 2010 Sony Bravia model
I am expecting a chromecast the next few days which I wanted to power through usb.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - tija - 2014-03-20

Just to understand all the actual development regarding multichannel sound.

If I use the actual Build from Milhouse should I keep 2.0 speaker configuration for multichannel passthrough sound?


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - FAMMAR - 2014-03-20

Maybe a quick solution? I am running the latest build but often I get Can't connect to network server


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - popcornmix - 2014-03-20

(2014-03-20, 18:24)tija Wrote: Just to understand all the actual development regarding multichannel sound.

If I use the actual Build from Milhouse should I keep 2.0 speaker configuration for multichannel passthrough sound?

I'm referring to multichannel PCM sound, requiring a HDMI connection to an AV receiver, and number of channels set to more than 2, e.g. 5.1 or 7.1.
This allows multichannel audio for non-passthrough formats (like AAC, PCM, flac, etc).
This case has only recently started to work with the Pi Sink.

If you are using spdif/toslink, then multichannel PCM is not supported (by your equipment) and number of channels should be set to 2.0 and passthrough enabled.
This case has worked for some time with the Pi Sink.

If you are in the first category, then you can set the number of channels to, e.g. 5.1. Leave passthrough enabled if supported.
If you then play a multichannel AAC/PCM/flac file you should get working multichannel audio.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - 606u - 2014-03-20

(2014-03-20, 00:08)MilhouseVH Wrote:
(2014-03-19, 23:15)606u Wrote:
(2014-03-19, 22:42)MilhouseVH Wrote: @606u: turn on debug log (wiki), specifically JSON logging, and I'm pretty sure the reason for increased load will become clear. Unfortunately some smartphone remotes are just not that smart...
Here it is.

Pretty much as suspected, the remote control app is spamming the client with JSON requests several times a second, most likely to maintain it's own state (what's playing, seek positiion etc.). Most (if not all of these) JSON requests are completely unnecessary, as the app could get the same information via JSON notifications without spamming the client.

Code:
23:06:29  62.758083 T:2667574352   DEBUG: JSONRPC: Incoming request: {"method":"Player.GetProperties","id":-472574362,"jsonrpc":"2.0","params":{"playerid":1,"properties":["percentage","time","totaltime","partymode","position","canrepeat","canshuffle","repeat","shuffled","canseek"]}}
23:06:29  62.760704 T:2794067024   DEBUG: JSONRPC: Incoming request: {"method":"Player.GetItem","id":-613133993,"jsonrpc":"2.0","params":{"playerid":1,"properties":["album","artist","title","thumbnail","track","studio","showtitle","episode","season"]}}
23:06:29  62.881805 T:2794067024   DEBUG: GetMovieId (nfs://192.168.0.2/c/media/Series/Lilo & Stitch. The Series/s01e01 Richter (513).m4v), query = select idMovie from movie where idFile=467
23:06:29  62.891171 T:2794067024   DEBUG: GetEpisodeId (nfs://192.168.0.2/c/media/Series/Lilo & Stitch. The Series/s01e01 Richter (513).m4v), query = select idEpisode from episode where idFile=467
23:06:29  63.084656 T:2667574352   DEBUG: webserver: request received for /jsonrpc
23:06:29  63.085541 T:2667574352   DEBUG: JSONRPC: Incoming request: {"method":"XBMC.GetInfoLabels","id":456093242,"jsonrpc":"2.0","params":{"labels":["MusicPlayer.Codec","MusicPlayer.SampleRate","MusicPlayer.BitRate","VideoPlayer.VideoResolution","VideoPlayer.VideoAspect"]}}
23:06:30  63.725456 T:2794067024   DEBUG: webserver: request received for /jsonrpc
23:06:30  63.726749 T:2794067024   DEBUG: JSONRPC: Incoming request: {"method":"Player.GetActivePlayers","id":1121127974,"jsonrpc":"2.0","params":{}}

Note how there are two separate threads (T:2667574352 and T:2794067024) on the Pi both responding to JSON requests, suggesting that you have the same app running on two iOS devices and that when you start playing a movie on the XBMC client the app on each device goes into overdrive spamming the XBMC client with JSON requrests just to maintain it's own internal state. If you add this app to a third iOS device... well, you can see the problem, eventually you won't be able to play movies! ...

I've made a new log and verified with netstat:
Code:
# netstat -tn
Active Internet connections (w/o servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State      
tcp        0      0 192.168.0.108:22        192.168.0.106:51017     ESTABLISHED
tcp        0      0 192.168.0.108:80        192.168.0.102:54350     ESTABLISHED
tcp        0      0 192.168.0.108:756       192.168.0.2:2049        ESTABLISHED
tcp        0      0 192.168.0.108:80        192.168.0.102:54345     ESTABLISHED
tcp        0      0 ::ffff:192.168.0.108:9090 ::ffff:192.168.0.102:54346 ESTABLISHED

Multiple threads are servicing RPC requests (multiple libmicrohttpd threads visible in top), but there is only a single iOS client (192.168.0.102). Will take a look in the App Store for alternative remote control apps.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - Forage - 2014-03-20

(2014-03-20, 02:42)MilhouseVH Wrote:
  • FIQ FSM has been updated to resolve some issues, please re-enable and re-test. From popcornmix:
    Quote:There is an update to fiq_fsm (and rpi-3.13.y-next tree) which should fix the "FIQ reported NYET. Data may have been lost." issue.
    (Basicaly the fiq was getting disabled, and causing lost usb messages).

    This could help a range of problems with USB and fiq_fsm.
Unfortunately no success. The network connection still drops after a while.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - michbeck100 - 2014-03-20

(2014-03-20, 15:50)ijsbeer79 Wrote:
(2014-03-20, 15:10)Eoghlear Wrote:
(2014-02-10, 21:15)michbeck100 Wrote: I finally finished my XBMC add-on to install hyperion to Openelec systems. It's available at Github. Download the add-on as ZIP file from here . If there are any issues please pm me. I'll try to fix any issues ASAP and update the add-on if any hyperion updates are available or changes to the test builds break the installation.

Did anyone test this?

Why would you? Hyperion could be installled to OE already. See theire wiki: https://github.com/tvdzwan/hyperion/wiki/Installation-on-RPi-with-OpenELEC

Because you can get easily updates for the Hyperion package once I finally get to update the plugin. And some users might not be so familiar with installing Hyperion from the console. At the moment it was just for testing purposes. But I will update the plugin to support auto updates.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - MrNice - 2014-03-20

Gotham build: #0320b

Audio test with PAPlayer (default)
Flac/16/44.1/2CH => OK
+++++++++++++
MCH:
6_Channel_ID.wav
Wave
PCM
Little
Signed
00001000-0000-0100-8000-00AA00389B71
5s 836ms
Constant
4 234 Kbps
6 channels
Front: L C R, Back: L R, LFE
44.1 KHz
16 bits
+++++++++++++++
ForUntoUsSurround88.flac
FLAC
Free Lossless Audio Codec
30s 104ms
Variable
5 837 Kbps
6 channels
Front: L C R, Back: L R, LFE
88.2 KHz
24 bits
+++++++++++++++
Both have drops every few seconds
FL, C, FR, BL, BR are OK
But LFE seems not mapped (I don't have subwoofer but I should have low frequencies in the front R and L (large band speakers))

OMXPlayer plays fine.

Youy can find these file here

Skin issue:
Every time I open System>system>video output, I get the message "Save resolution. Would you like to keep this change"

Question:
With the 2 players I have a drop at the very beginning of the file and can't listen the half first second.
Is the best way to add in advancedsettings.xml?
Code:
<advancedsettings>
  <audio>
    <streamsilence>1</streamsilence>
  </audio>
</advancedsettings>



RE: OpenELEC Testbuilds for RaspberryPi Part 2 - popcornmix - 2014-03-20

(2014-03-20, 21:54)MrNice Wrote: Gotham build: #0320b
Both have drops every few seconds

Do the drops go away when running full screen (tab on keyboard)?


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - 606u - 2014-03-20

(2014-03-20, 15:34)popcornmix Wrote:
(2014-03-19, 22:11)606u Wrote: Problems from this older post still stands with both -- PAPlayer and DVDPlayer. I've recorded a short clip to demonstrate the issue (the volume level is a bit low, tho): 24-bit, 44 kHz 5.1 sounds pretty bad: music is heard from front left channel all the time, center channel occasionally, the rest are mostly mute. Audio breaks every now and then, when garbled audio starts, moving cursor (in songs list) up and down fixes it for a short time (while cursor is sliding up or down), then it breaks again.

The audio break up (a repeated rumbling noise) should be fixed in latest Milhouse build.
If you still believe channels are mapped wrongly, then can you upload a sample file and describe your audio settings.
Confirming that multi-channel playback audio break up is fixed in OpenELEC-RPi.arm-Milhouse-20140320013122-r17977-ge57e1bc.

However, at least at my end channels are still mapped wrongly. However I don't claim audio files are correct and in Audacity I don't see indication which channel is which. If I interpret what I see and what I hear correctly, here is what I think current mapping is:
  • channel 1 => front left,
  • channel 2 => center,
  • channel 3 => front right,
  • channel 4 => rear left,
  • channel 5 => rear right,
  • channel 6 => assuming LFE, as I don't have one.
Except that in my files channels 1 and 2 are front, 3 is center, 4 is LFE, 5 and 6 are rear, which explains the result. You can find sample files at Dropbox link I've sent you previously, PM me if you don't have the link anymore.

Update: All the files (ALAC-encoded 24-44, 24-88,2 and 24-96 at 5.1) plays back correctly in full screen with no info ribbon; 24-96 is on the edge, tho, with average CPU idle below 10%.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - 606u - 2014-03-20

(2014-03-20, 21:54)MrNice Wrote: Every time I open System>system>video output, I get the message "Save resolution. Would you like to keep this change"
...
With the 2 players I have a drop at the very beginning of the file and can't listen the half first second.
I have the same two problems, except I got the message "Save resolution..." twice: here is a short clip; message also appears when video settings are opened right after a restart (while Pi is restarting TV and receiver are switched on and TV's active input is the receiver).

Update: Initial interval where no sound is heard could possibly be related to the receiver: the time it takes to detect the format of input signal and set-up the appropriate decoder, hence it is audible on 1st track. Pause seems to also appear when audio format is changed, like PCM => DD or between different bitrates and frequencies of PCM. If that is so, there is little that can be done.


RE: OpenELEC Testbuilds for RaspberryPi Part 2 - MrNice - 2014-03-20

@ 606u
You can download test files here
go to Multi-Channel Examples (WAVE_FORMAT_EXTENSIBLE)
To check channel position
files are 3 and 9 Mo

(2014-03-20, 22:00)popcornmix Wrote:
(2014-03-20, 21:54)MrNice Wrote: Gotham build: #0320b
Both have drops every few seconds

Do the drops go away when running full screen (tab on keyboard)?

Not at all
With full screen I can see the cursor going from left to right. It freeze and sound drops.