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 - 2012-04-22 Updates: * Optimized numerous things in the pass through code path * Fixed an issue with the merged apple code incorrectly running on other platforms * Found a long outstanding bug in CSoftAEStream::GetDelay (this is huge! it may fix many sync issues) * Streams can now start in a paused auto start state, the stream resumes once buffers are filled. @einhänder - I had a look at your logs, your running an old build, your issue is completely unrelated. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - einhänder - 2012-04-22 (2012-04-22, 18:20)gnif Wrote: @einhänder - I had a look at your logs, your running an old build, your issue is completely unrelated. Hm, that's strange because according to the logs I built the branch on 20.04.2012. I used that version as a base for my report. Also I copied the git commit (cb7d1bee09f50dca60a9aee22290f80adc1c14b5) from the logs which match the version I pulled on that day. EDIT: However right now I'm about to build a version based on you commits from today. Let's see if it changes anything. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-22 @einhänder - perhaps you did not have debug enabled, your log is missing all the new Enumerate entries at startup. Just found a bug also when skipping tracks that are in different formats, I will look into it tomorrow, bed time for me, 3am here RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - einhänder - 2012-04-22 Ok, I'll double check the debug logging, although I did set Code: <loglevel hide="true">1</loglevel> in advancedsettings. Is there any additional debug logging facility for AE branch? I'm going to post new logs as soon as I will have sorted out the logging problem. Good night. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Wanilton - 2012-04-22 Hello Gnif, too late for sleep.. I Try compile your recent version to windows, but this is broke in vcproject and vcproject.filters (missing includes), do you try fix this for us, please, when you have a time, of course. AEEncoderFFmpeg.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\encoders\AEEncoderFFmpeg.h(23): fatal error C1083: Cannot open include file: 'Interfaces/AEEncoder.h': No such file or directory 1>c:\xbmc-hd\xbmc\cores\audioengine\Sinks/AESinkWASAPI.h(27): fatal error C1083: Cannot open include file: 'AEDeviceInfo.h': No such file or directory SoftAE.cpp SoftAESound.cpp 1>..\..\xbmc\cores\AudioEngine\Engines\SoftAESound.cpp(22): fatal error C1083: Cannot open include file: 'Interfaces/AESound.h': No such file or directory SoftAEStream.cpp 1>..\..\xbmc\cores\AudioEngine\Engines\SoftAEStream.cpp(27): fatal error C1083: Cannot open include file: 'AEFactory.h': No such file or directory AEPPAnimationFade.cpp 1>..\..\xbmc\cores\AudioEngine\PostProc\AEPPAnimationFade.cpp(23): fatal error C1083: Cannot open include file: 'AEFactory.h': No such file or directory AESinkDirectSound.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\sinks\AESinkDirectSound.h(23): fatal error C1083: Cannot open include file: 'Interfaces/AESink.h': No such file or directory AESinkNULL.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\engines\SoftAE.h(32): fatal error C1083: Cannot open include file: 'Interfaces/ThreadedAE.h': No such file or directory AESinkProfiler.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\sinks\AESinkNULL.h(25): fatal error C1083: Cannot open include file: 'Interfaces/AESink.h': No such file or directory 1>c:\xbmc-hd\xbmc\cores\audioengine\sinks\AESinkProfiler.h(25): fatal error C1083: Cannot open include file: 'Interfaces/AESink.h': No such file or directory AESinkWASAPI.cpp AEBitstreamPacker.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\sinks\../Interfaces/AESink.h(25): fatal error C1083: Cannot open include file: 'AEAudioFormat.h': No such file or directory AEBuffer.cpp AEChannelInfo.cpp AEConvert.cpp AEDeviceInfo.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\utils\AEDeviceInfo.h(25): fatal error C1083: Cannot open include file: 'AEAudioFormat.h': No such file or directory AEPackIEC61937.cpp AERemap.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\utils\AERemap.h(23): fatal error C1083: Cannot open include file: 'AEAudioFormat.h': No such file or directory AEStreamInfo.cpp AEUtil.cpp AEWAVLoader.cpp 1>c:\xbmc-hd\xbmc\cores\audioengine\utils\AEWAVLoader.h(24): fatal error C1083: Cannot open include file: 'AEAudioFormat.h': No such file or directory DVDAudioCodecPassthrough.cpp CrystalHD.cpp Thanks, your job is great, congratulations from Brazil. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - grote002 - 2012-04-22 @gnif I tried the latest version and Dolby True HD seems to be working correctly now, no problems with the sound anymore, but DTS HD still doesnt work RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Maxoo - 2012-04-23 @einhänder : add debug when you compile. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-23 @grote002 - Can you please provide a full log for DTS HD, and if possible a sample file also. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - einhänder - 2012-04-23 (2012-04-23, 11:40)Maxoo Wrote: @einhänder : add debug when you compile. As far as I remember the debug flag is necessary to get additional debugging symbols relevant for meaningful backtraces (gdb/core dump)? Nevertheless I'm going to build it with "debug" enabled to see if there will be a difference. However I thought it would be a good idea to move "~/.xbmc" and to start with a fresh profile in order to see if there is something involved in writing the enumeration info to the log. I set up my audio settings as usual and surprisingly everything works like it should with that new profile - all issues were gone. But there is a "but". As soon as I completely resembled my former settings it went back to the reported problems. Yet I wasn't able to track down which setting enforces the faulty behaviour. Could anyone give me an example how the enumeration information in the log looks like please? So I could compare if these details are finally included. EDIT: False alarm, yesterday evening it worked like described above, today it does not. At least it is not alway reproduceable. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - gnif - 2012-04-23 @einhänder - fincheresque posted a sample a few posts back RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - fincheresque - 2012-04-23 I did? Oh well, I'm going to re-compile tonight and test again. I had the issue with a build built on Saturday where if I restarted lightdm, I would get no sound on the initial run of XBMC, but if I exited XBMC, and went back in, sound was fine. Unfortunately I didn't have debugging turned on, and my wife was looking forward to starting Season 4 of SoA...so I couldn't go back and try and duplicate the issue with debugging turned on at that particular moment (real life...always interferring). I'll try again tonight and hopefully get some information the important folks can use. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - einhänder - 2012-04-24 @gnif Now I've got XBMC to print the needed information. There was no need to rebuild the bin, just using the parameter --debug helped. Here is a log from my standard profile right after a fresh start: http://paste.pocoo.org/show/586242/ If you want me to re-post the logs from my previous report showing what happens during different situations then just let me know please. RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - fincheresque - 2012-04-24 @gnif I still can't get pastebin to accept my pastes. I don't know what the issue is... I presume it's me because nobody else seems to be complaining about it. Multiple PC's and browsers here just hang when posting. In the mean time, I've noticed that in the log there's an extra line that's there when audio fails: Code: INFO: CAESinkALSA::EnumerateDevicesEx - Unable to open hdmi:CARD=NVidia,DEV=0 for capability detection So it seems as though something is stealing the card, but again this wasn't an issue in builds prior to the large hold on development leading up to Eden's release. Stupid question, anything I should be --enabling or --disabling with the ./configure prior to compile? Here's the audio enumeration code: Code: 19:14:53 T:3019335440 DEBUG: CAESinkALSA::EnumerateDevicesEx - Skipping device hw:0,2 as it does not have PCM playback ability RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Maxoo - 2012-04-24 If pastebin doesn't work, use http://www.anonpaste.tk/ RE: AudioEngine branch - DO NOT REQUEST BINARY BUILDS - Memphiz - 2012-04-24 little offtopic, but i can't stand to clear these things up: @einhänder - i guess activating debugging in XBMC system settings would do the trick aswell @fincheresque - sounds like a mtu problem. You could try to reduce the MTU below 1500 (e.x. 1400) on the network adapter you use for pastebin. |