Kodi Community Forum
AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Printable Version

+- Kodi Community Forum (https://forum.kodi.tv)
+-- Forum: Development (https://forum.kodi.tv/forumdisplay.php?fid=32)
+--- Forum: Kodi Application (https://forum.kodi.tv/forumdisplay.php?fid=93)
+--- Thread: AudioEngine branch - DO NOT REQUEST BINARY BUILDS (/showthread.php?tid=78289)



RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-11

@cdhesse
Its not a bug sorry, this is ALSA failing to open the HDMI output device, you need to get that fixed before any HD content will work.


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-11

Progress update:

* Wrote a little buffer class to replace manual buffer wrangling with with bounds checking, should help prevent buffer overflows in future and help identify them when they do occur, it have already identified a few problems that have been corrected.
* PAP is now playing things properly again, of special note is streams with infinite length (streaming radio) or non-indexed audio files (captured audio).
* Don't know if I mentioned this one, but I added S16 -> Float dithering using both normal and SSE intrinsics, this should improve the dynamic range of S16 audio considerably (so I am told).
* Pass-through seems to be fully functional under Linux, need more testing to confirm this though.
* Changing the output device now applies the setting instantly again, this was broken in an earlier merge.
* PAP now continues onto the next song properly on streams with invalid total times
* Merged working OSX code (someone needs to test this)
* Disabled pulse audio by default (until it gets fixed)
* Merged WASAPI fixes from DDDamian.


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-12

Progress update (wow, this is new, there is usually a 1/2 dozen or so posts between my updates):

* AE has been updated to master (windows and OSX projects broken, awaiting fixes)
* Mute now toggles correctly
* Volume level is saved and restored correctly
* Rare deadlock issue when PAPlayer stop button clicked has been fixed

We are getting VERY close now people, the merge with master is one of the last large leaps before AE is merged into master! Its been a long journey but we are finally at the end. XBMC developers are now actively testing AE as we nut down the remaining bugs.

IOS/OSX is working in AE now also (except the broken build files from the merge ofc), which is awesome, there are no major blockers preventing AE from merging into master any more.

This doesn't mean that AE is finished, it just means that AE will become part of the nightlies and it will force all the devs to deal with AE, exposing bugs and allowing us to get it rock solid for the next XBMC release.


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Berggren - 2012-04-12

(2012-04-12, 12:52)gnif Wrote: Progress update (wow, this is new, there is usually a 1/2 dozen or so posts between my updates):

We are getting VERY close now people, the merge with master is one of the last large leaps before AE is merged into master! Its been a long journey but we are finally at the end. XBMC developers are now actively testing AE as we nut down the remaining bugs.

This doesn't mean that AE is finished, it just means that AE will become part of the nightlies and it will force all the devs to deal with AE, exposing bugs and allowing us to get it rock solid for the next XBMC release.
Good work Gnif and all the others behind the scene!
Looking forward in the release and the joy it will bring to the whole XBMC scene!

Again, Congrats from me Big Grin




RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - erhnam - 2012-04-12

(2012-04-12, 12:52)gnif Wrote: Progress update (wow, this is new, there is usually a 1/2 dozen or so posts between my updates):

* AE has been updated to master (windows and OSX projects broken, awaiting fixes)
* Mute now toggles correctly
* Volume level is saved and restored correctly
* Rare deadlock issue when PAPlayer stop button clicked has been fixed

We are getting VERY close now people, the merge with master is one of the last large leaps before AE is merged into master! Its been a long journey but we are finally at the end. XBMC developers are now actively testing AE as we nut down the remaining bugs.

IOS/OSX is working in AE now also (except the broken build files from the merge ofc), which is awesome, there are no major blockers preventing AE from merging into master any more.

This doesn't mean that AE is finished, it just means that AE will become part of the nightlies and it will force all the devs to deal with AE, exposing bugs and allowing us to get it rock solid for the next XBMC release.

Nice work! Maybe we can update the first post of this topic to summarize the high level changes? new formats, separate AE library, windows, osx and linux improvements, etc? What do you think?




RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-12

(2012-04-12, 14:12)erhnam Wrote: Nice work! Maybe we can update the first post of this topic to summarize the high level changes? new formats, separate AE library, windows, osx and linux improvements, etc? What do you think?

Heh, the list of fixes and improvements is so long I cant remember 1/2 of them until I use it and go 'oh yeah! that works properly now too'. But if you wish to start compiling the features go for it, you started this thread and as such I believe you can edit your post to do so Smile.


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - HeresJohnny - 2012-04-12

I would expect congratulations are in order! Thanks for your work, gnif!


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - cdhesse - 2012-04-12

(2012-04-10, 21:03)cdhesse Wrote: Hi gnif,

DTS-HD (or anything else passthrough) still doesn't work for me.

I've tried HDMI, and custom hw:0,9 as well as hdmi:NVidia for passthrough but always get this error. The applicable part of the log looks to me to be:

14:37:40 T:140249841436416 INFO: CAESinkALSA::Initialize - Attempting to open device hdmi,AES0=0x06,AES1=0x82,AES2=0x00,AES3=0x0e
14:37:40 T:140249841436416 ERROR: CAESinkALSA::Initialize - snd_pcm_open_lconf(-2) - hdmi,AES0=0x06,AES1=0x82,AES2=0x00,AES3=0x0e
14:37:40 T:140249841436416 ERROR: CAESinkOSS::Initialize - Failed to get supported formats, assuming AFMT_S16_NE

Full log can be found at http://pastebin.com/ekSTZy5R

I hope this is a bug and not something I'm doing wrong...

I understand you are telling me I have a support/setup issue, but I've done lots of testing since this post, and if it is a support/setup issue it is one that completely evades me.

Bottom line, I couldn't get it working, so I did a complete install of XBMCBuntu (Eden). The audio setup for passthrough was hw:CARD=NVidia,DEV=9 and DTS and DD worked fine after the fresh install.

I then checked out and compiled AE and rebooted. With the same audio setup for passthrough, the snd_pcm_open_lconf(-2) error shows up again (the same device string works for normal audio in AE, just not passthrough). I'm confused by this, but minimum I'm thinking that potentially there is something different with how AE tries to open the device which others would see to when going to AE from Eden? For this reason, I'm asking again in the development thread, to see if there is something gnif thinks is worth looking at more closely. If need be I can reinstall fresh again, play a file which works in Eden, then upgrade/compile AE again and show it failing to provide both logs. Would this be something desired?

In addition - speaker-test works up to 192000 6 channels with device hw:CARD=NVidia,DEV=9 too


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-12

@cdhesse - Sure, I am interested if it was working as you say, please provide full logs for both Eden and AE builds via pastebin.com


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Philmatic - 2012-04-12

(2012-04-12, 12:52)gnif Wrote: We are getting VERY close now people, the merge with master is one of the last large leaps before AE is merged into master! Its been a long journey but we are finally at the end. XBMC developers are now actively testing AE as we nut down the remaining bugs.

You sir, are a gentleman and a scholar! Excellent work!


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - grote002 - 2012-04-12

Hey, i pulled the latest version from git and i get the following error while running configure, the compilation from 2 days ago didnt have this problem

I am running ubuntu 10.10

Code:
checking for LIBBLURAY... yes
checking for libbluray soname... libbluray.so.0
checking libbluray version... ./configure: line 369: test: if: integer expression expected
configure: error: you
./configure: line 266: return: if: numeric argument required
./configure: line 276: exit: if: numeric argument required

What am i missing to compile this?



RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - cdhesse - 2012-04-12

(2012-04-12, 18:42)gnif Wrote: @cdhesse - Sure, I am interested if it was working as you say, please provide full logs for both Eden and AE builds via pastebin.com

Hi Gnif...

Here is the fresh XBMCBuntu/Eden log where passthrough works: (DTS)
http://pastebin.com/LnLQ30TZ
Interestingly - it says that it fails and reverts to "default"?


Here is the AE log where it doesn't:
http://pastebin.com/JxyTQizz

In the AE log I also played a 5.1 FLAC just to show it working in non-passthrough mode...


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-13

@grote002 - this is unrelated to AE, seems like you need to perform a distclean on the source as the configure that has been generated for libblueray is invalid.


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - nirmv - 2012-04-13

I get a compilation error (commit 0b490f8...)

PeripheralCecAdapter.cpp: In member function ‘virtual bool PERIPHERALS::CPeripheralCecAdapterUpdateThread::SetInitialConfiguration()’:
PeripheralCecAdapter.cpp:1191:16: error: ‘class CSettings’ has no member named ‘m_nVolumeLevel’


Thanks


RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Fneufneu - 2012-04-13

it's m_fVolumeLevel