And audio would be out-of-sync by the way.
Posts by Da Flex
-
-
- activate debugging at GUI (HUD will appear on top)
- reproduce the error (wait for reconnect)
- login by SSH and run pastekodi
- post the resulting URL of pastekodi
-
- activate debugging at GUI (HUD will appear on top)
- login by SSH and run pastekodi
- post the resulting URL of pastekodi
-
Read here about the new analog audio behavior: Click!
-
So you just have to edit the UUID's for successful booting. Problem solved?
-
Because LE 10 Beta is working, no further suggestions from my side.
-
We need more info. Activate debugging at GUI, reboot, login by SSH and post the resulting link of the pastekodi command.
-
Do you overclock?
-
Yeah, could be driver confusion due to downgrade. Do a fresh install.
-
I am having the same issue with rpi 3 and kodi 19. When I turn off the TV and come back to it the next day, I can't get kodi to wake up using any keys on any remotes. I have to turn it off and then on again.
When the TV is switching off, the RPi stops A/V output by default. When switching on the TV, there must be miscommunication between your RPi and TV, so A/V will not switch on.
Go into input device settings, and select CEC adapter. Change the CEC settings to ignore TV switch-off signal. So RPi's A/V stays active all the time, and should still run when you switch the TV on again.
-
Yes, /flash/config.txt is the correct place for it.
Even on LE 10 Beta?
-
You need an additional audio input device for this task.
It's possible to route input to output on this device.
Problem: I don't think you can route audio from an external input device to another external output device (your USB soundcard).
-
Just to make this complete: HEVC (aka H265) is suggested by dev CvH as the best long-term solution for encoding. If you want to play your video library on multiple platforms, go for that codec. (RPi 4 can hardware-decode it.)
At second place comes H264. (RPi 4 can hardware-decode it.)
MPEG-2 is third place. (RPi 4 can only software-decode it. RPi 3 can hardware-decode it by license purchase.)
VC-1 is working on powerful hardware (x86 PC's), but can overcharge weaker hardware, like RPi's, due to missing multi-threading. (RPi 4 can only software-decode it. RPi 3 can hardware-decode it by license purchase.)
-
You talk about music videos, and I assume the audio data rate is higher at those. A single RPi core can't handle the decoding of such data rate anymore. -> re-encode or overclock
-
After conversation with dev's, I can say that the ability of ffmpeg's multi-threading depends on the codec. It's currently not available for VC-1. Parallelizing VC-1 decoding is a heavy task. So my suggestion is to re-encode videos by MPEG-2 (software decoding) or H264/265 (hardware decoding).
-
Yes, that sync error is visible at the log. The rest of the log looks good, so I'm thinking it's a decoding bug (WMA pro / DTS). A dev should have a look at this, I have no idea about a hot fix.
Code
Display More... 2021-04-07 14:20:09.319 T:942 INFO <general>: CDVDAudioCodecFFmpeg::Open() Successful opened audio decoder wmapro 2021-04-07 14:20:09.319 T:942 INFO <general>: Creating audio thread 2021-04-07 14:20:09.319 T:945 DEBUG <general>: Thread VideoPlayerAudio start, auto delete: false 2021-04-07 14:20:09.319 T:945 INFO <general>: running thread: CVideoPlayerAudio::Process() 2021-04-07 14:20:09.319 T:945 DEBUG <general>: CVideoPlayerAudio - CDVDMsg::GENERAL_PAUSE: 1 2021-04-07 14:20:09.319 T:942 DEBUG <general>: Prepare - Start position set to last stopped position: 482000 2021-04-07 14:20:09.320 T:811 DEBUG <general>: EGL Debugging: Error: EGL_BAD_SURFACE Command: eglSwapBuffers Type: EGL_DEBUG_MSG_ERROR_KHR Message: dri2_swap_buffers 2021-04-07 14:20:09.322 T:811 DEBUG <general>: CDRMAtomic::FlipPage - Execute modeset at next commit 2021-04-07 14:20:09.323 T:942 DEBUG <general>: SeekTime - unknown position after seek 2021-04-07 14:20:09.323 T:942 DEBUG <general>: CVideoPlayer::FlushBuffers - flushing buffers 2021-04-07 14:20:09.323 T:945 DEBUG <general>: CDVDAudio::Pause - pausing audio stream 2021-04-07 14:20:09.323 T:944 DEBUG <general>: CDVDVideoCodecDRMPRIME::Reset - flush buffers 2021-04-07 14:20:09.345 T:945 DEBUG <general>: CVideoPlayerAudio - CDVDMsg::GENERAL_SYNCHRONIZE 2021-04-07 14:20:09.345 T:944 DEBUG <general>: CVideoPlayerVideo - CDVDMsg::GENERAL_SYNCHRONIZE 2021-04-07 14:20:09.345 T:942 DEBUG <general>: CVideoPlayer::SetCaching - caching state 2 2021-04-07 14:20:09.346 T:945 DEBUG <general>: CDVDAudio::Pause - pausing audio stream 2021-04-07 14:20:09.346 T:942 DEBUG <general>: CDVDClock::SetSpeedAdjust - adjusted:0.000000 2021-04-07 14:20:09.346 T:942 DEBUG <general>: Prepare - starting demuxer from: 482000 2021-04-07 14:20:09.352 T:811 DEBUG <general>: CWinSystemGbmGLESContext::PresentRender - Sending display reset to all clients 2021-04-07 14:20:09.352 T:811 INFO <general>: VideoPlayer: OnResetDisplay received 2021-04-07 14:20:09.352 T:945 DEBUG <general>: CVideoPlayerAudio - CDVDMsg::GENERAL_PAUSE: 0 2021-04-07 14:20:09.352 T:944 DEBUG <general>: CVideoPlayerVideo - CDVDMsg::GENERAL_PAUSE: 0 2021-04-07 14:20:09.359 T:811 DEBUG <general>: ffmpeg[(nil)X]: [image2] Custom AVIOContext makes no sense and will be ignored with AVFMT_NOFILE format. 2021-04-07 14:20:09.396 T:942 DEBUG <general>: CVideoPlayer::HandleMessages - player 2 reported state: 0 2021-04-07 14:20:09.396 T:826 DEBUG <general>: OnAVChange: CApplication::OnAVChange 2021-04-07 14:20:09.396 T:942 DEBUG <general>: CVideoPlayer::HandleMessages - player 1 reported state: 0 2021-04-07 14:20:09.399 T:942 DEBUG <general>: CheckPlayerInit - dropping packet type:1 dts:479317000.000000 to get to start point at 482000000.000000 2021-04-07 14:20:09.401 T:942 DEBUG <general>: CheckPlayerInit - dropping packet type:2 dts:481397000.000000 to get to start point at 482000000.000000 2021-04-07 14:20:09.401 T:944 DEBUG <general>: CVideoPlayerVideo - Stillframe left, switching to normal playback 2021-04-07 14:20:09.403 T:942 DEBUG <general>: CheckPlayerInit - dropping packet type:2 dts:481439000.000000 to get to start point at 482000000.000000 2021-04-07 14:20:09.405 T:942 DEBUG <general>: CheckPlayerInit - dropping packet type:2 dts:481480000.000000 to get to start point at 482000000.000000 ... -
Ah, now I can see. I should censor that way of posting a version number.
-
The user never mentioned the LE version. For LE 9.2.6 it's the config.txt way, for LE 10 Beta it's kernel options.
