[Windows] AudioEngine testers - Windows platform only - 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: [Windows] AudioEngine testers - Windows platform only (/showthread.php?tid=124600) |
RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-13, 12:08)PKOneTwo Wrote: @DDDamian: Please enable debugging and repost log - no it should definitely not be spamming the log like that but I need to see the debug-level lines to follow why it did that. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-13, 13:03)jpsdr Wrote: I've disabled in the BIOS the HD Via audio. Don't understand why there is trouble now and not before... I figured the onboard VIA was interfering with the nVidia device - glad disabling that fixed you up. I had a look thru your log for the one 2.0 flac that wasn't playing, and all log entries are good - nothing unexpected in them at all. Not sure if it's an issue with the file itself, which seems likely if similar ones play fine. You might want to use mkvmerge to demux/remux that file, alternately eac3to & more gui can use mkvmerge to do so. If you demux out the flac soundtrack and it plays okay as just an audio file try remux it back in and see what happens. This would indicate an issue with the way it was originally muxed, which is not unusual with some earlier versions of mkvmerge. For the MonInfo report - re-run MonInfo and browse through the reported devices until you see one that has some audio info in it - the device above was just the monitor, not the receiver. It will show as a seperate device. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-13, 17:15)loekf Wrote:(2012-05-12, 21:32)DDDamian Wrote: @loekf - there's the issue. WASAPI is just the render method - it is effectively passing thru right to the audio driver. Yes, Creative has not updated those drivers in ages, which makes me think they are unlikely to now for that card. I do think that your future plans should be based on HDMI, not on dedicated sound cards - you will need HDMI if you wish to use DTS-MA/TrueHD or 24/96 or higher audio and multichannel lpcm. DirectSound devices will always be reported as 2.0 channel, as this is the max for spdif pcm, and it won't affect passthrough like AC3. I do need to fix that for HDMI devices, but it's kinda secondary as with HDMI you should absolutely be using WASAPI anyways. DirectSound uses a mixer, which has benefits and disadvantages. Benefits include mixing (obviously) and resampling, but AE is capable of both on it's own. The big downsides are audio quality is worse than wasapi, and there is no support for HD audio. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-13, 21:13)TRaSH Wrote:(2012-05-12, 16:16)DDDamian Wrote: @TRaSH - i've seen that SQL error popping up a lot lately - perhaps time to try the full (non-portable) install, and go through your addons to see if any are causing the issue. The crash was an unrelated bug caused by a seperate threading project that I was able to track down - was fixed last night. JMarshall pushed some fixes last night addressing filepaths and the new SQL thumbnail caching which may have fixed the error you're getting - it's a new initiative by him to move the thumbs into the database, and he's been doing some bug-fixes on it. Awesome trash - glad to hear it. And yes, it will keep getting better As Martijn mentions, the commits to XBMC's master branch are the definitive changelogs although they cover everything, not just AE. Typically commits related to AE will have a [AE] prefix, but that's up to the dev doing the commit, and now that it's in master that can include any dev's work. We have a branch on gnif's and my repo's called AE_DSP which has some advanced work for the next revision of AE including DSPs. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-14, 13:52)Warstomper Wrote: I have beent trying to get AE(and Daniela's Patch and more recently using Nightlies) to work on my setup, but no matter what I do, DTS-HD does not seem to work properly. I have the following problems using AE: Hi Warstomper - just make sure debugging is enabled and play a single movie that you know works (perhaps with Daniela's patch) and I'll be happy to take a look. We've added quite a bit of debugging code to help us during this initial roll-out, so that's fine to start - if I need more info after reviewing I'll let you know. Cheers. (2012-05-14, 15:55)paolov74 Wrote: Hi DDDamian, With the exception of <transcodeto> and <allchannelstereo> the advancedsettings entries listed in post #2 of this thread are active. Transcoding is automatic now depending on other settinggs, and <allchannelstereo> is now a guisetting under System>Audio Output>Output Stereo to All Speakers (requires "receiver accepts multichannel lpcm"). I'll update the second post shortly. We're working on better logic and a seperate gui window for various settings to streamline it and provide easy access to upcoming DSP settings. RE: [Windows] AudioEngine testers - Windows platform only - TeKo - 2012-05-15 Seems to work with DTSHDMA and TrueHD but XBMC seems really slow (Movie posters in Coverflow take some time to load and other stuff) but maybe thats just a skin problem. Going to try different skins. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-14, 22:39)bambi73 Wrote: Hello all, Hi there - the log shows WASAPI is initializing correctly, but I don't see where you try to play something, so not sure what happens when you do. I would need a log where you actually attempt to play something. I would recommend with WASAPI. Thx. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-15, 18:22)TeKo Wrote: Seems to work with DTSHDMA and TrueHD but XBMC seems really slow (Movie posters in Coverflow take some time to load and other stuff) but maybe thats just a skin problem. Hi TeKo - a problem has been found in PAPlayer which is causing higher-than-normal cpu usage and will be addressed, but that's not the issue here. There has been work done as a seperate project to move thumbs into the databases, and that is resulting in slower thumbnail display. The dev handling that is looking at performance improvements in the new thumb-handling code (this includes covers). RE: [Windows] AudioEngine testers - Windows platform only - jpsdr - 2012-05-15 (2012-05-15, 17:50)DDDamian Wrote: I figured the onboard VIA was interfering with the nVidia device - glad disabling that fixed you up.I'm not so happy to be obliged of this workaround, as it worked perfectly fine before new AE... (2012-05-15, 17:50)DDDamian Wrote: For the MonInfo report - re-run MonInfo and browse through the reported devices until you see one that has some audio info in it - the device above was just the monitor, not the receiver.hmmm.... the problem is that only the [sample] have audio information. All my devices have only what i've provided to you... (2012-05-15, 17:50)DDDamian Wrote: It will show as a seperate device...... nope ! RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-15, 20:41)jpsdr Wrote:(2012-05-15, 17:50)DDDamian Wrote: I figured the onboard VIA was interfering with the nVidia device - glad disabling that fixed you up.I'm not so happy to be obliged of this workaround, as it worked perfectly fine before new AE... I'm sorry to hear that, but if you have another solution (mine worked) I'm all ears. You are not "obliged" to do anything if you think it inconvenient or have a better solution to making your multiple sound devices work together, including using Eden. Perhaps nightlies aren't for you. (2012-05-15, 20:41)jpsdr Wrote:(2012-05-15, 17:50)DDDamian Wrote: For the MonInfo report - re-run MonInfo and browse through the reported devices until you see one that has some audio info in it - the device above was just the monitor, not the receiver.hmmm.... the problem is that only the [sample] have audio information. All my devices have only what i've provided to you... If MonInfo doesn't detect it we won't either. The report you attached has no audio information reported. Check into EDID issues and resolve those - they are outside the scope of AE or XBMC. (2012-05-15, 20:41)jpsdr Wrote:(2012-05-15, 17:50)DDDamian Wrote: It will show as a seperate device...... nope ! See previous comment. RE: [Windows] AudioEngine testers - Windows platform only - jpsdr - 2012-05-15 (2012-05-15, 21:15)DDDamian Wrote: Check into EDID issues and resolve those. Wich means ? Sorry, i'm totaly lost here. Can you tell me what i have to do ? (Or a link wich will explain to me what i have to do ?). I've also trouble to understand what seems to be (at least for me) contradictory informations. - You said that if moninfo can't detect it, either you. But the device is present in Wasapi and Windows. - You said information show only 2 channels possibility, but i've configured 7.1 in both Windows and XBMC, and XBMC played correctly 5.1 files, either pathrough (DTS), or not (FLAC). Finaly, all of these made me totaly confused... RE: [Windows] AudioEngine testers - Windows platform only - bambi73 - 2012-05-15 (2012-05-15, 18:26)DDDamian Wrote:(2012-05-14, 22:39)bambi73 Wrote: today I compiled xbmc master and got no audio, completelly nothing, not even skin sounds. I mean it don't do any sound, not even "bubble sound" from skin. Of course I tried to play some video files, with same (bad) result. There is with debug log where I tried to play some playback. Thx EDIT: Ufff, please ignore both posts . My skin doesn't show scroll bar in settings and so I missed Audio output device option. Never thought that there is something under all these greyed out toggles Case solved, but maybe you should move this quite important option bit higher in list. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 @Bambi - i totally agree - we're looking at the best way to implement the gui settings. RE: [Windows] AudioEngine testers - Windows platform only - DDDamian - 2012-05-15 (2012-05-15, 21:40)jpsdr Wrote:(2012-05-15, 21:15)DDDamian Wrote: Check into EDID issues and resolve those. Most of what you're asking above, and all along, has to do with the detection of channels. You have an HDMI device which is only reporting two channels. If you are plugged into a TV that's fine, with a receiver it should report more. That's why I asked for the MonInfo report, which isn't reporting anything about your audio device at the end of the HDMI connection. It should report the device's audio capabilities, which is far different from "the device is present in Wasapi and Windows" Yes, it is present, and reports two channels. Passthrough is not dependent on the channel settings, it is "passed through" as is, encoded as it originally was, which is why it plays fine even though channels detected are only two. Typically this is due to EDID issues as mentioned. Frankly I don't care to be the google-bitch here (as some of the devs call it). Feel free to though. I've pointed you in the right direction several times here, fixed your issues, then got an "I don't like it", so you're on your own there. The AAC button does nothing at this point. As mentioned earlier "AAC will not play unless your device decodes it natively. If it's playing then obviously it does. RE: [Windows] AudioEngine testers - Windows platform only - jayhawk785 - 2012-05-15 DDDamian- Issue of xbmc recovering when the associated wasapi device is turned off seems to be fixed. However, my processor goes up to about 30% and stays there while it tries to reconnect hundreds of thousands of times. I'm not certain if others have mentioned it, but thought I'd post in here to get your take. I didnt see any mention of it in this thread since the end of april (well before merge). Code: 15:38:00 T:4212 ERROR: CAESinkWASAPI::AddPackets: GetBuffer failed due to AUDCLNT_E_DEVICE_INVALIDATED I can probably get a debug log if that will help--just figured you may have already seen this at some point and have an answer or hopefully a solution in the works thanks! |