• 1
  • 130
  • 131
  • 132(current)
  • 133
  • 134
  • 168
OpenELEC Testbuilds for RaspberryPi (Kodi 16.0)
(2015-11-07, 12:35)mule1 Wrote:
(2015-11-05, 21:24)mule1 Wrote: Is it usual that navigating (using skips) in a 3D movie using the latest testbuilds (started testing from build #1029) is yet nearly impossible?
In rare cases after a skip the movie continues to play. But most times Kodi only fills the buffer (shown in the GUI) but never continues to play the movie or the movie continues to play but only with 2-4 frames per second or if it continues there are heavy stutters. Stopping and trying to resume shows the same results. ISO seems to be even worse than MKV files. Navigating in non 3d movies is working as it should.

No one else having these issues? Is navigating (skipping forward, backward, fast forward, rewind) in 3d movies working flawless for you?

I will check this tomorrow as well. Perhaps also somebody else can check?
(2015-11-07, 13:53)popcornmix Wrote:
(2015-11-04, 00:07)popcornmix Wrote: The Advanced deinterlace at 1080i60 is expensive. It generates a lot of sdram traffic.
DVB USB dongles generate a lot of data, and depending on the driver, it may need to read prompty to avoid losing data.

The suspicion is that the heavy sdram traffic delays the USB driver occasionally and it loses a packet, causing corruption in the video.

Now there may be ways of working around this. Forcing the deinterlace to spread out its sdram accesses the right amount may solve this.
E.g. currently deinterlace might take 13ms every 16ms. If I were to make the deinterlace pause for 10us every 100us, it would still likely get it's job done,
and hopefully the USB driver won't be held up during the pauses.

Unfortunately it's not something I can test (I have encrypted cable at home, and too weak a terrestrial signal at both work and home).
I may be able to produce a test build that attempts this (perhaps with a config setting to tweak how frequent/long the pauses are).

@bagofcrap24 @polo_joe

I've added a test setting to see if we can fix the artefacts with advanced 1080i deinterlave and locally attached DVB dongle.
Use latest Milhouse build and add:
Code:
hvs_priority=0xN0000000

Replace the N with a hex digit from 0 to f (i.e. 0,1,2,3,4,5,6,7,8,9,a,b,c,d,e,f). That number determines how long to pause for, with 0 meaning no change in behaviour, to f means lots of pausing.
Start with 8, and if you still have artefacts then increase it, and if no artefacts then decrease it. Try to find the minimum setting that doesn't have artefacts.
Too high a setting will cause lagging of the video, as the deinterlace will take longer. Hopefully there will be a value that fixes artefacts and doesn't have lagging.

cannot test yet
haven't updated in about a week. Somewhere between 1029 & 1106 my DVB adapter no longer works on the Pi2.
It is seen in the TVH server however when I try to play any stream. I get a toast saying no signal.
I'll find the build where it stops working

[EDIT]
Last Working Build - #1101
upon updating to 1102 I get the popup toast saying no signal.
1101 all is well.
Using a "PCTV 292e Triplestick"

dmesg

TVHE log is full of this
Code:
2015-11-07 17:15:07.000 mpegts: 786MHz in DVB-T2 - tuning on Silicon Labs Si2168 : DVB-T #0

2015-11-07 17:15:07.001 subscription: 000B: "epggrab" subscribing to mux "786MHz", weight: 4, adapter: "Silicon Labs Si2168 : DVB-T #0", network: "DVB-T2", service: "Raw PID Subscription"

2015-11-07 17:15:07.002 subscription: 0005: No input source available for subscription "127.0.0.1 [ alex | Kodi Media Center ]" to channel "BBC NEWS HD"

2015-11-07 17:15:09.000 subscription: 0005: No input source available for subscription "127.0.0.1 [ alex | Kodi Media Center ]" to channel "BBC NEWS HD"

2015-11-07 17:15:11.000 subscription: 0005: No input source available for subscription "127.0.0.1 [ alex | Kodi Media Center ]" to channel "BBC NEWS HD"

2015-11-07 17:15:13.000 subscription: 0005: No input source available for subscription "127.0.0.1 [ alex | Kodi Media Center ]" to channel "BBC NEWS HD"
(2015-11-07, 19:04)bagofcrap24 Wrote: cannot test yet
haven't updated in about a week. Somewhere between 1029 & 1106 my DVB adapter no longer works on the Pi2.
It is seen in the TVH server however when I try to play any stream. I get a toast saying no signal.
I'll find the build where it stops working
Last Working Build - #1101
upon updating to 1102 I get the popup toast saying no signal.
1101 all is well.
Using a "PCTV 292e Triplestick"

I reported this at https://tvheadend.org/boards/5/topics/17...sage-18412 in thread with similar symptoms. #1102 and up use Linux 4.3 instead of 4.1.12 kernel. 4.3 or maybe even 4.2 appears to have broken a whole lot of DVB sticks. Your stick uses a different USB bridge (Empia instead of Cypress) but tuner and demodulator chips are same as in my stick (Si2168+Si2157).

You can install just the firmware (just start.elf and fixup.dat nothing else) and then try the fix. I'll be curious if this gives the ARM core enough breathing space for USB to work normally.

Can you zap HD channels with MMAL enabled and OMXplayer disabled? I am using TVheadend and with this setting I have to stop playback and select a channel fresh, else it will buffer to death and then hang.
Katastrophentourist
So run build 1101 and replace the firmware with that from 1106 then perform tests?
And by zap you mean to change from one channel to the next without stopping via the EPG?
Just using channel up or down.
I think I've seen that behavior before where it hangs and you have to stop then chose a new channel
@MarkT: the kernel patches that fixed 4.2 are already included in 4.3, no doubt more patches are needed - if you spot anything new please let us know.
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
(2015-11-07, 20:15)bagofcrap24 Wrote: So run build 1101 and replace the firmware with that from 1106 then perform tests?

Yes, using a working buid with firmware (start.elf/fixup.dat) from latest build should be fine for testing.
(2015-11-07, 20:33)popcornmix Wrote:
(2015-11-07, 20:15)bagofcrap24 Wrote: So run build 1101 and replace the firmware with that from 1106 then perform tests?

Yes, using a working buid with firmware (start.elf/fixup.dat) from latest build should be fine for testing.
Even with 0xf there is still some artefacting with mmal. Though it is much less often.

Even at 0xf I'm not seeing any video lag

Edit
I'm lying. It's heavily effected by what you watch.
Pretty much anything with panning shots.
E.g. nature/sports it will cause artefacts every few seconds.
(2015-11-05, 00:49)Milhouse Wrote:
(2015-11-05, 00:35)SB9x Wrote: i have noticed, when i put a tar update into the update folder, the rasperrys network performance is very poor under openelec.

Most probably your SD card write performance, assuming you are wired (with WiFi all bets are off).

You can test your network performance with iperf (install from the OpenELEC Unofficial Repository) - this is the best way to test your true network bandwidth as it eliminates all other variables.

If your network performance is particularly slow, or even dropping packets, then this could account for playback problems. Are you wired or WiFi? Test with wired, as WiFi problems are likely to be down to your router, dongle, neighbours etc. and not something relevant to this thread.

i have now found out my problem. the first problem was the sdcard and the second problem was, the wired connection between openelec and the second raspberry with samba, had a fault utp network cable and now i using a new one (sftp) and now - it works.
(2015-11-07, 20:57)bagofcrap24 Wrote: Even with 0xf there is still some artefacting with mmal. Though it is much less often.

Yes, by channel zap I ment change channel up or down. I get excessive "Buffering...xx%" messages and then a freeze with MMAL. Not in OMXPlayer.

If you run a "journalctl -f" log tail on the RPi2, do you see continuity error lines such as these?

Code:
Nov 07 20:48:21 RPI2 tvheadend[673]: TS: Kabel Deutschland/114MHz/arte HD: H264 @ #6661 Continuity counter error (total 1)

Followed by video artifacts briefly after a new message pops up? How many per hour do you get? I used to use a Terratec H5 Rev.2 with said Empia USB chip with a Micronas DRX-K tuner. Its isosynchronous USB endpoints still give me nightmares. If possible, try a USB dongle with Cypress chip. Those use bulk endpoints and I think (just imho) those are better tolerated by an RPi2 under SDRAM bandwidth starvation. Using a Cypress stick cut those errors down by a factor 50 or more in my setup. From 3-5 video dropouts per minute with frequent pitch problems due to resampling catching up, to now maybe one every 20 minutes. I am also using dwc_otg.fiq_fsm_mask=0xF as kernel command line on the FAT32 partition. For testing, use a powered USB hub and make sure you do not ever (even during boot) get that colored square top right in the display (low voltage indication).

@Milhouse Props to you for being totally above the kernel patching game with regard to video hardware. I'll keep my eyes open.
Katastrophentourist
I always use a powered hub for any accessories to the pi.
Never seen the rainbow voltage indicator on my pi but I have on a friends, I use a proper block PSU (adjustable) and get bang on 5.0v at the pi.
Sadly I don't have any other tuners to hand to experiment with so have to run tvh on my x86 Ubuntu machine to bypass the issue for now.
I don't have time at the minute to check the journalctl but I'll verify tomorrow for you.
As for the channel zapping, it works fine for me even with mmal (this used to be an issue for me in the past but then one day it just started working). I don't have the mpeg2 license on this pi so can't test SD channels but worked fine switching between HD.
New OpenELEC Jarvis build #1107: RPi / RPi2
(Supercedes previous build)

Code:
# uname -a
Linux rpi512 4.3.0 #1 Sat Nov 7 21:03:08 GMT 2015 armv6l GNU/Linux

# vcgencmd version
Nov  6 2015 21:55:08
Copyright (c) 2012 Broadcom
version 40eee6c6d44edb1db4302217927a66a8dffd9374 (clean) (release)

# lsb_release
OpenELEC (Milhouse) - Version: devel-20151107210218-#1107-ge1c98c8 [Build #1107]

# vcdbg log msg 2>&1 | grep DTOK
001693.176: Kernel trailer DTOK property says yes

# Kernel device tree status: Enabled

Based on tip of OpenELEC master (e1c98c83, changelog) and tip of XBMC master (8e4ecd8e, changelog) with the following modifications: Build Highlights:
  1. Added PulseAudio support for Bluetooth and network audio - thanks @lrusak

    To enable Bluetooth Audio:
    • System > Settings > Audio: Select "PULSE:" audio output device
    • OpenELEC Settings add-on: Enable PulseAudio service
    • OpenELEC Settings add-on: Pair Bluetooth dongle with Bluetooth speakers/headphones
    From now on, whenever the Bluetooth speakers/headphones are connected audio will play over Bluetooth (otherwise HDMI).

    Known issues:
    • Kodi will crash if the Bluetooth device is disconnected during playback - stop playback before disconnecting
    • Bluetooth audio is delayed slightly, and will required an audio offset of about +100ms to restore sync
Build Details:
  1. XBMC:
    • [EPG,PVR] Fix deadlock caused by Noitifyobservers call while holding instance's mutex (PR:8355, 1 commit, 1 file changed)
    • FIX: [droid] regression on media keys with CEC (PR:8364, 1 commit, 3 files changed)
    • FIX: [droid] some intent filters MUST be registered in code (PR:8365, 1 commit, 2 files changed)
  2. pvr.stalker:
    • bug squashing (PR:30, 4 commits, 6 files changed)
  3. pvr.vdr.vnsi:
    • More code cleanup and many memory leaks fixed (PR:34, 3 commits, 11 files changed)
  4. service.openelec.settings/pulseaudio (lrusak):
    • New commits in this build:
  5. Additional commits/pull requests/changes not yet merged upstream:
    • Added: [env] patch: Build pulseaudio for all projects
    • Added: [env] PR:4427: pulseaudio and kodi
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
IQaudio PiDigiAMP+ Broken

I use RPI2 with touchscreen and the PiDigiAMP+ as a music player. On the latest builds I only get a series of clicks coming from the speakers. The last working build for me is #1031.

I have tested with 1102, 1104, 1106 and 1107 all show same symptoms. Series of clicks from speakers and playback time display runs very fast, approx. 2 to 3 times real time.

aplay -l

**** List of PLAYBACK Hardware Devices ****
card 0: IQaudIODAC [IQaudIODAC], device 0: IQaudIO DAC HiFi pcm512x-hifi-0 []
Subdevices: 1/1
Subdevice #0: subdevice #0

I don't know if it makes any difference but I haven't licensed any of the codecs as I only use it for audio playback. Fault apparent when RPi boots, start up sounds are only heard as a series of clicks, playing from HDD or internet radio app only produce clicks and fast running time display.

On another note - it would be great to control the play back volume using the touchscreen interface?
Like S3nox I seem to be sseing the exact same issue but I have RPI-2 with a HiFiBerry Digi+ (runnig over SPDIF to my amp).
Everything works on 6.0.0, but in both dev builds 1106 and 1107 I have the S3nox issue, clicking noises and the player runs the counter
at about x5 speed, and no sound from movies. ( I have not tried earlier ones).

see http://xbmclogs.com/pne9xlx8e

I get lots of this in the log file:
15:41:21 14.908775 T:1963274240 NOTICE: Running database version Addons20
15:41:23 16.770435 T:1963274240 NOTICE: start dvd mediatype detection
15:41:23 16.822298 T:1963274240 WARNING: CSkinInfo: failed to load skin settings
15:41:24 17.909468 T:1963274240 WARNING: Previous line repeats 1 times.
15:41:24 17.909632 T:1963274240 WARNING: JSONRPC: Could not parse type "Setting.Details.SettingList"
15:41:24 18.032183 T:1963274240 NOTICE: ActiveAE DSP - starting
15:41:24 18.071171 T:1963274240 NOTICE: initialize done
15:41:24 18.071386 T:1963274240 NOTICE: Running the application...
15:41:25 18.572470 T:1963274240 NOTICE: starting zeroconf publishing
15:41:25 18.575850 T:1963274240 NOTICE: WebServer: Started the webserver
15:41:25 18.576258 T:1963274240 NOTICE: starting upnp client
15:41:25 18.593479 T:1636881312 NOTICE: ES: Starting UDP Event server on 0.0.0.0:9777
15:41:25 18.593771 T:1636881312 NOTICE: UDP: Listening on port 9777
15:41:25 19.042219 T:1740632992 NOTICE: ## OpenELEC Addon ## 0.6.8
15:41:25 19.050814 T:1740632992 WARNING: CSkinInfo: failed to load skin settings
15:41:26 20.098246 T:1963274240 WARNING: Previous line repeats 1 times.
15:41:26 20.098425 T:1963274240 ERROR: Control 1500 in window 13001 has been asked to focus, but it can't
15:41:59 53.026920 T:1504760736 WARNING: CSkinInfo: failed to load skin settings
15:42:02 56.095501 T:1963274240 WARNING: CreateLoader - unsupported protocol(sources) in sources://music.tbn
15:42:32 85.662323 T:1963274240 NOTICE: CDVDFactoryInputStream::CreateInputStream - <f1> /var/media/Media-USB1/TEST/test-1-mp3-320.mp3 <ssif>
15:42:49 102.591774 T:1963274240 WARNING: CreateLoader - unsupported protocol(sources) in sources://music.tbn
15:42:52 106.255547 T:1636881312 NOTICE: ES: Client from 127.0.0.1 timed out
15:43:40 153.883499 T:1875899296 ERROR: CAESinkALSA - snd_pcm_writei(-32) Broken pipe - trying to recover
15:44:00 173.674927 T:1963274240 ERROR: Previous line repeats 14 times.
15:44:00 173.675095 T:1963274240 WARNING: CreateLoader - unsupported protocol(sources) in sources://music.tbn
15:44:02 175.849640 T:1875899296 ERROR: CAESinkALSA - snd_pcm_writei(-32) Broken pipe - trying to recover
15:44:06 180.093369 T:1963274240 ERROR: Previous line repeats 4 times.
15:44:06 180.093552 T:1963274240 NOTICE: CDVDFactoryInputStream::CreateInputStream - <f1> /var/media/Media-USB1/TEST/test-1-mp3-320.mp3 <ssif>


The: "ERROR: CAESinkALSA - snd_pcm_writei(-32) Broken pipe - trying to recover" is the line sthat then keeps croping up.

Any thoughts?
(2015-11-08, 20:19)javaboyuk Wrote: Like S3nox I seem to be sseing the exact same issue but I have RPI-2 with a HiFiBerry Digi+ (runnig over SPDIF to my amp).
Currently some important patches to the bcm2835 DMA driver are missing in the 4.3 kernel tree so I2S soundcards won't work.

These patches are included in a rather huge rework PR, https://github.com/raspberrypi/linux/pull/1178 which is currently being tested. Could take a while until all issues are resolved, but after that PR gets merged I2S soundcards should start working again.

so long,

Hias
New OpenELEC Jarvis build #1108: RPi / RPi2
(Supercedes previous build)

Code:
# uname -a
Linux rpi512 4.3.0 #1 Sun Nov 8 21:03:15 GMT 2015 armv6l GNU/Linux

# vcgencmd version
Nov  6 2015 21:55:08
Copyright (c) 2012 Broadcom
version 40eee6c6d44edb1db4302217927a66a8dffd9374 (clean) (release)

# lsb_release
OpenELEC (Milhouse) - Version: devel-20151108210225-#1108-g25c23fe [Build #1108]

# vcdbg log msg 2>&1 | grep DTOK
001690.459: Kernel trailer DTOK property says yes

# Kernel device tree status: Enabled

Based on tip of OpenELEC master (25c23fef, changelog) and tip of XBMC master (6c008a3b, changelog) with the following modifications: Build Highlights:
  1. Cosmetics
Build Details:
  1. OpenELEC:
    • Cosmetic: Fix typo in filename (PR:4430, 1 commit, 1 file changed)
  2. XBMC:
    • Fix JSON output of album and artist mood to be array as in spec (PR:8358, 1 commit, 2 files changed)
    • [jenkins] buildsteps update (PR:8366, 3 commits, 27 files changed)
Texture Cache Maintenance Utility: Preload your texture cache for optimal UI performance. Remotely manage media libraries. Purge unused artwork to free up space. Find missing media. Configurable QA check to highlight metadata issues. Aid in diagnosis of library and cache related problems.
  • 1
  • 130
  • 131
  • 132(current)
  • 133
  • 134
  • 168

Logout Mark Read Team Forum Stats Members Help
OpenELEC Testbuilds for RaspberryPi (Kodi 16.0)10