HELP! Audio output is starving, playing silence

[Worf72]

Like other users, I'm having audio problems with VLC 1.0.0 on my PC: Intel i7 920, Asus P6T SE, 6 GB ram, Sound Blaster X-Fi, Windows XP 32 Service Pack 3.
The PC is running perfectly and I only have audio issues with VLC playing video files: .avi, .mpg, .flv, etc.
If I just play audio files with VLC (.mp3, .flac), it's ok.

VLC stops giving sound out every 5/7 or so seconds for a few tenths of a second.
(In fact the messages say: main debug: audio output is starving, playing silence)

The PC is a new build, so I just tried VLC 1.0.0 and 2 earlier beta versions.
(I already tried uninstalling and reinstalling, clearing cache and preferences, etc.)
I tried turning off/on "high quality audio resampling" and it doesn't affect the problem.
I installed the latest chipset and display adapter drivers after fresh OS install.

Here same debug messages:

playing example .mpg file #1:
main warning: output date isn't PTS date, requesting resampling (-41601)
main warning: audio drift is too big (-141079), clearing out
main warning: timing screwed, stopping resampling
main warning: mixer start isn't output start (-61172)
main debug: audio output is starving (174906), playing silence
main warning: output date isn't PTS date, requesting resampling (-42391)
main warning: buffer is 42397 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-42333)
main warning: timing screwed, stopping resampling
main warning: buffer is 79447 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-41000)
main warning: output date isn't PTS date, requesting resampling (-40073)
main warning: audio drift is too big (-141330), clearing out
main warning: timing screwed, stopping resampling
main warning: mixer start isn't output start (-56980)
main debug: audio output is starving (164909), playing silence
main warning: output date isn't PTS date, requesting resampling (-40959)
main warning: buffer is 40964 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-40580)
main warning: timing screwed, stopping resampling
main warning: buffer is 77010 in advance, triggering downsampling


playing example .mpg file #2:
main warning: backward_pts != dts (-66744)
main warning: backward_pts != dts (-66745)
main warning: backward_pts != dts (-66744)
main warning: output date isn't PTS date, requesting resampling (-44410)
main warning: buffer is 44229 in advance, triggering downsampling
main warning: backward_pts != dts (-66733)
main warning: backward_pts != dts (-66745)
main warning: output date isn't PTS date, requesting resampling (-40333)
main warning: timing screwed, stopping resampling
main warning: buffer is 82666 in advance, triggering downsampling
main warning: backward_pts != dts (-66744)
main warning: backward_pts != dts (-66744)
main warning: backward_pts != dts (-66745)
main warning: output date isn't PTS date, requesting resampling (-45964)
main warning: audio drift is too big (-126701), clearing out
main warning: timing screwed, stopping resampling
main warning: mixer start isn't output start (-51299)
main debug: audio output is starving (145904), playing silence
main warning: backward_pts != dts (-66744)
main warning: backward_pts != dts (-66744)
main warning: backward_pts != dts (-66745)


playing example .avi file #1:
main warning: buffer is 71477 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-41449)
main warning: audio drift is too big (-126085), clearing out
main warning: timing screwed, stopping resampling
main warning: mixer start isn't output start (-53170)
main debug: audio output is starving (154254), playing silence
main warning: output date isn't PTS date, requesting resampling (-42921)
main warning: buffer is 42536 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-40898)
main warning: timing screwed, stopping resampling
main warning: buffer is 72425 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-42030)
main warning: timing screwed, stopping resampling
main warning: audio drift is too big (-121666), clearing out
main warning: mixer start isn't output start (-60275)
main debug: audio output is starving (219631), playing silence
main warning: output date isn't PTS date, requesting resampling (-40819)
main warning: buffer is 40701 in advance, triggering downsampling
main warning: resampling stopped after 250000 usec (drift: 40420)
main warning: buffer is 44244 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-40921)
main warning: resampling stopped after 1876000 usec (drift: 70189)
main warning: buffer is 70008 in advance, triggering downsampling


I've already tried all the possible audio output modules and nothing changed except that if I select "Win32 waveOut extension output" the volume drops really very
low, the sound is no more interrupted by silence so often but crackles appear.

playing example .ISO file #1 (DVD image) with "Win32 waveOut extension output"
main warning: PTS is out of range (-7265), dropping buffer
main warning: output PTS is out of range (62024), clearing out
main warning: input PTS is out of range (88735), trashing
main warning: input PTS is out of range (56735), trashing
main warning: input PTS is out of range (24735), trashing
main warning: PTS is out of range (-39265), dropping buffer
main warning: buffer is 64000 in advance, triggering downsampling
main warning: resampling stopped after 13784000 usec (drift: -770)
main debug: control type=1
main debug: control type=1
main warning: computed PTS is out of range (124602798), clearing out
main warning: output PTS is out of range (124646402), clearing out
main debug: audio output is starving (267265), playing silence
waveout warning: aout_OutputNextBuffer no buffer got next_date=94970030 ms, 0 frames to play,

As I said before, with other programs (GOM, KMplayer, BSplayer, MediaPlayer Classic) I don't have any problem with sound so is it a VLC bug?

I hope to receive feedback soon.

Thank you very much.

[VLC_help]

Does it help if you force vlc.exe affinity to single core?

[Worf72]

I tried setting affinity to just one cpu core, but it didn't change the sound problem!

[VLC_help]

And changing video output module doesn't matter to silence?

[Worf72]

Changing video output module doesn't matter whatsoever!
Today I upgraded to version 1.0.1, and the problem remains!

[slafsa]

Im having the same problem. Any solutions yet?

[tbessie]

I'm having the same problem: viewtopic.php?f=14&t=63200

How do you enable the logging where you're seeing this debugging info? I started vlc with --verbose=2 and pointed it at a log file, but see no log output.

UPDATE: OK, I enabled logging from the View-> menu, but I didn't see the "starving" messages when the sound dropouts happened. In fact, no message at all printed when it happened.

Curiouser and curiouser.

- Tim

[vaderflag]

I'm having the identical problem. Only in VLC across multiple video files (they work fine in WMP), on a Core i7 P6T, on-board RealTek High Definition Audio Driver.

main warning: output PTS is out of range (78302646), clearing out
main debug: audio output is starving (70667), playing silence
main warning: output date isn't PTS date, requesting resampling (-43667)
main warning: buffer is 43667 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-43125)
main warning: timing screwed, stopping resampling
main warning: buffer is 86147 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-42979)
main warning: audio drift is too big (-128584), clearing out
main warning: timing screwed, stopping resampling
main warning: mixer start isn't output start (-66728)
main debug: audio output is starving (179667), playing silence
main warning: output date isn't PTS date, requesting resampling (-41667)
main warning: buffer is 41500 in advance, triggering downsampling
main warning: output date isn't PTS date, requesting resampling (-42000)
main warning: timing screwed, stopping resampling
main warning: buffer is 82834 in advance, triggering downsampling

[Worf72]

It must be a VLC bug! I hope they can fix it soon because right now I'm forced to use other players...

[VLC_help]

There won't be any fix before we know what is causing the issue. Currently there isn't any way to replicate the issue, so it is hard to know what component behaves badly.

[cooper2k9]

Can confirm short 'silence' for a 1/10s or so. Only FLV, AVI, MP4 (maybe more). Not MP3 nor FLAC.

It makes VLC unusable... to me.

Dived down some versions (uninstall with deleting prefs, starting without modifications of prefs):
1.0.1: interrupted.
1.0.0: interrupted.
0.9.8a: interrupted.
0.8.6h: not interrupted.

To make it clear: It really happens whenever something like
main debug: audio output is starving (50023), playing silence
is in verbose-log (level 2).

This can be seen on pause/play in v0.8.6.h.

Occurrence is random, here everything from 1 to 4s.

It even does not help to switch ALL CPU features (I am also running a quite new CPU: AMD Phenom II 705e, 3 cores). It does not help to reduce system load. It does not help to assing VLC to a certain core.

Also tried to compare the audio-output modules on affected files. For MP4 and FLV these are appearing both:
coverter_float
scaletempo
bandlimited_resampler

It really seems a VLC problem, WMP nor Foobar2k are affected.

So it also seems not depending on some hardware (Realtek HD, onboard my MSI790GX-G65 mo/bo). BTW: XP, SP3.

Hope this helps to identify the issue.
cooper2k9.

Like other users, I'm having audio problems with VLC 1.0.0 on my PC: Intel i7 920, Asus P6T SE, 6 GB ram, Sound Blaster X-Fi, Windows XP 32 Service Pack 3.
The PC is running perfectly and I only have audio issues with VLC playing video files: .avi, .mpg, .flv, etc.
If I just play audio files with VLC (.mp3, .flac), it's ok.

VLC stops giving sound out every 5/7 or so seconds for a few tenths of a second.
(In fact the messages say: main debug: audio output is starving, playing silence)

The PC is a new build, so I just tried VLC 1.0.0 and 2 earlier beta versions.
(I already tried uninstalling and reinstalling, clearing cache and preferences, etc.)
I tried turning off/on "high quality audio resampling" and it doesn't affect the problem.
I installed the latest chipset and display adapter drivers after fresh OS install.

As I said before, with other programs (GOM, KMplayer, BSplayer, MediaPlayer Classic) I don't have any problem with sound so is it a VLC bug?

Thank you very much.

[Worf72]

cooper2k9,
you did a great analysis.
I hope the VLC developers can fix the bug soon.

[overlayed]

Here is one of many feeds that's been giving me the audio grief.
mms://media2.bloomberg.com/btv_US200.asf

At the moment, if I want the audio to play, I'd have to re-initiate the audio after the video starts - first disable the audio track, then re-enable the audio. The trick works every time, but having to do it all the time is starting to get to me.

The problem wasn't there until v1.0.0, and it' remains after the v1.0.1 update. I always delete the cache and preferences at each updates, so that should eliminate possibility of corrupted preferences.

Could someone please followup on whether my problem with this particular video feed is reproducible on other systems?

[VLC_help]

overlayed: I can confirm your problem. You get errors like

Code: Select all

main debug: b_seek=0 th*avg=6505 skip=0 main warning: output date isn't PTS date, requesting resampling (40504) main warning: buffer is 79076 late, triggering upsampling main debug: b_seek=0 th*avg=6505 skip=0

to your message log?

[overlayed]

With --verbose=2, the short answer is yes.
The long answer is - I spent some time matching the output of --verbose=2 to --verbose=1, and this is what I found ...

First the --verbose=1 output:

Code: Select all

access_mms info: selecting stream[0x1] audio (21 kb/s) access_mms info: selecting stream[0x2] video (181 kb/s) access_mms info: connection successful access_mms error: no data received access_mms warning: cannot fill buffer access_mms warning: cannot receive header (1/10) asf warning: unknown asf object (not loaded) asf warning: unknown asf object (not loaded) access_mms warning: unimplemented query in control main warning: late picture skipped (1429001 > -9) main warning: late picture skipped (1362001 > -9) main warning: late picture skipped (1295001 > -9) main warning: late picture skipped (1562001 > -9) main warning: late picture skipped (1495001 > -9) main warning: late picture skipped (1231001 > -6) main warning: late picture skipped (1166001 > -6) main warning: late picture skipped (1112001 > -254) main warning: late picture skipped (1047001 > -254) main warning: late picture skipped (993001 > -190)

At this point the video started with no audio output. The "late picture skipped" warnings happen at random, and I'm guessing it has more to do with video delays than the audio.

I then proceed to disable the audio track and re-enable it.

Code: Select all

scaletempo warning: bad input or output format scaletempo warning: input and output formats are not similar main warning: no audio filter module matching "scaletempo" could be loaded

Now there are both video and audio.
Finally, I press the stop button and the rest comes

Code: Select all

main warning: received buffer in the future main warning: can't get output picture main warning: can't get output picture asf warning: unknown asf object 0x26f18b5d-0x4584-0x47ec-0x9f5f0e651f0452c9 asf warning: unknown asf object 0x1806d474-0xcadf-0x4509-0xa4ba9aabcb96aae8 access_mms error: failed to send command

Equipped with that, I now try to match the --verbose=2 messages to --verbose=1, the first place I looked was before "access_mms warning: unimplemented query in control"

Code: Select all

asf debug: found object guid: 0xd2d0a440-0xe307-0x11d2-0x97f000a0c95ea850 size:162 asf debug: read "extended content description object" asf debug: - 'WMFSDKVersion' = '9.00.00.3265' asf debug: - 'WMFSDKNeeded' = '0.0.0.0000' asf debug: - 'IsVBR' = 'false' asf debug: found object guid: 0x86d15240-0x311d-0x11d0-0xa3a400a0c90348f6 size:214 asf debug: read "codec list object" reserved_guid:0x86d15241-0x311d-0x11d0-0xa3a400a0c90348f6 codec_entries_count:2 asf debug: - codec[0] audio name:"Windows Media Audio 9 Voice" description:" 20 kbps, 22 kHz, mono" information_length:2 asf debug: - codec[1] video name:"Windows Media Video 9" description:"" information_length:4 asf debug: found object guid: 0x75b22636-0x668e-0x11cf-0xa6d900aa0062ce6c size:50 asf debug: read "data object" file_id:0xe88c20f2-0xf44a-0x44be-0x8019be83f36c0575 total data packet:0 reserved:257 asf debug: + 'Unknown' GUID 0x0-0x0-0x0-0x0000000000000000 size:0pos:0 asf debug: + 'Header' GUID 0x75b22630-0x668e-0x11cf-0xa6d900aa0062ce6c size:2654pos:0 asf debug: | + 'Stream Bitrate Properties' GUID 0x7bf875ce-0x468d-0x11d1-0x8d82006097c9a2b2 size:38pos:30 asf debug: | + 'File Properties' GUID 0x8cabdca1-0xa947-0x11cf-0x8ee400c00c205365 size:104pos:68 asf debug: | + 'Header Extension' GUID 0x5fbf03b5-0xa92e-0x11cf-0x8ee300c00c205365 size:1823pos:172 asf debug: | | + 'Language List' GUID 0x7c4346a9-0xefe0-0x4bfc-0xb229393ede415c85 size:39pos:218 asf debug: | | + 'Extended Stream Properties' GUID 0x14e6a5cb-0xc672-0x4332-0x8399a96952065b5a size:88pos:257 asf debug: | | + 'Extended Stream Properties' GUID 0x14e6a5cb-0xc672-0x4332-0x8399a96952065b5a size:110pos:345 asf debug: | | + 'Unknown' GUID 0x26f18b5d-0x4584-0x47ec-0x9f5f0e651f0452c9 size:26pos:455 asf debug: | | + 'Metadata' GUID 0xc5f8cbea-0x5baf-0x4877-0x8467aa8c44fa4cca size:224pos:481 asf debug: | | + 'Padding' GUID 0x1806d474-0xcadf-0x4509-0xa4ba9aabcb96aae8 size:1290pos:705 asf debug: | + 'Stream Properties' GUID 0xb7dc0791-0xa9b7-0x11cf-0x8ee600c00c205365 size:150pos:1995 asf debug: | + 'Stream Properties' GUID 0xb7dc0791-0xa9b7-0x11cf-0x8ee600c00c205365 size:133pos:2145 asf debug: | + 'Extended content description' GUID 0xd2d0a440-0xe307-0x11d2-0x97f000a0c95ea850 size:162pos:2278 asf debug: | + 'Codec List' GUID 0x86d15240-0x311d-0x11d0-0xa3a400a0c90348f6 size:214pos:2440 asf debug: + 'Data' GUID 0x75b22636-0x668e-0x11cf-0xa6d900aa0062ce6c size:50pos:2654 asf debug: found 2 streams main debug: selecting program id=0 asf debug: added new audio stream(codec:0xa,ID:1) asf debug: added new video stream(ID:2) main debug: b_seek=0 th*avg=13520 skip=0 main debug: using demux module "asf" main debug: TIMER module_need() : 8.000 ms - Total 8.000 ms / 1 intvls (Avg 8.000 ms) main debug: looking for a subtitle file in C:\Program Files\VideoLAN\VLC\ main debug: looking for decoder module: 36 candidates qt4 debug: Updating the geometry qt4 debug: Updating the geometry qt4 debug: Updating the geometry qt4 debug: Updating the geometry qt4 debug: Updating the geometry qt4 debug: Updating the geometry dmo debug: DMO codec for wmas may work with dll=wmspdmod.dll dmo debug: found DMO: WMSpeech Decoder DMO dmo debug: DMO input type set dmo debug: DMO output type set dmo debug: GetOutputSizeInfo(): bytes 16384, align 1 main debug: using decoder module "dmo" main debug: TIMER module_need() : 110.000 ms - Total 110.000 ms / 1 intvls (Avg 110.000 ms) main debug: thread (decoder) created at priority 2 (../.././src/input/decoder.c:315) main debug: thread started dmo debug: ProcessOutput(): failed main debug: looking for decoder module: 36 candidates avcodec debug: libavcodec initialized (interface 0x342000) avcodec debug: using direct rendering avcodec debug: ffmpeg codec (Windows Media Video 3) started main debug: using decoder module "avcodec" main debug: TIMER module_need() : 22.000 ms - Total 22.000 ms / 1 intvls (Avg 22.000 ms) main debug: thread (decoder) created at priority 0 (../.././src/input/decoder.c:315) main debug: thread started access_mms warning: unimplemented query in control main debug: `mms://media2.bloomberg.com/btv_US200.asf' successfully opened main debug: b_seek=0 th*avg=10010 skip=0 main debug: b_seek=0 th*avg=8840 skip=0 main debug: b_seek=0 th*avg=8255 skip=0 main debug: b_seek=0 th*avg=7900 skip=0 main debug: b_seek=0 th*avg=7670 skip=0 main debug: b_seek=0 th*avg=7375 skip=0 main debug: b_seek=0 th*avg=7280 skip=0 main debug: b_seek=0 th*avg=7200 skip=0 main debug: b_seek=0 th*avg=7135 skip=0

To be honest, I'm not entirely sure what to take away from this, but it seems "WMSpeech Decoder DMO" failed to process the output in the beginning, and that's why the audio output failed.

Moving forward ... I then take a look at what happened at the re-initiation.

Disabling audio track:

Code: Select all

main debug: control type=17 main debug: thread times: real 0m15.687500s, kernel 0m0.000000s, user 0m0.000000s main debug: removing module "dmo" main debug: killing decoder fourcc `wmas', 0 PES in FIFO qt4 debug: Updating the geometry

Re-enabling audio track:

Code: Select all

access_mms debug: recv command start_sequence:0x4a000001 command_id:0xb00bface length:32 len8:4 sequence 0x00000006 len8_II:2 dir_comm:0x0004001b main debug: control type=17 main debug: looking for decoder module: 36 candidates dmo debug: DMO codec for wmas may work with dll=wmspdmod.dll dmo debug: found DMO: WMSpeech Decoder DMO dmo debug: DMO input type set dmo debug: DMO output type set dmo debug: GetOutputSizeInfo(): bytes 16384, align 1 main debug: using decoder module "dmo" main debug: TIMER module_need() : 10.000 ms - Total 10.000 ms / 1 intvls (Avg 10.000 ms) main debug: thread (decoder) created at priority 2 (../.././src/input/decoder.c:315) main debug: thread started main debug: creating aout main debug: looking for audio output module: 3 candidates aout_directx debug: OpenAudio qt4 debug: Updating the geometry qt4 debug: Updating the geometry aout_directx debug: found device: Primary Sound Driver aout_directx debug: found device: Conexant HD Audio output aout_directx debug: device supports 2 channels (DEFAULT!) aout_directx debug: device supports 1 channel aout_directx debug: Windows says your SpeakerConfig is stereo aout_directx debug: creating DirectSoundThread main debug: thread (DirectSound Notification Thread) created at priority 15 (../../.././modules/audio_output/directx.c:427) main debug: using audio output module "aout_directx" main debug: TIMER module_need() : 142.000 ms - Total 142.000 ms / 1 intvls (Avg 142.000 ms) main debug: output 's16l' 22050 Hz Stereo/Mono frame=1 samples/4 bytes main debug: mixer 'fl32' 22050 Hz Stereo/Mono frame=1 samples/8 bytes main debug: filter(s) 'fl32'->'s16l' 22050 Hz->22050 Hz Stereo/Mono->Stereo/Mono main debug: thread started aout_directx debug: DirectSoundThread ready main debug: looking for audio filter module: 24 candidates main debug: using audio filter module "converter_float" main debug: TIMER module_need() : 26.000 ms - Total 26.000 ms / 1 intvls (Avg 26.000 ms) main debug: found a filter for the whole conversion main debug: looking for audio mixer module: 3 candidates main debug: using audio mixer module "float32_mixer" main debug: TIMER module_need() : 28.000 ms - Total 28.000 ms / 1 intvls (Avg 28.000 ms) main debug: input 's16l' 22050 Hz Mono frame=1 samples/2 bytes main debug: looking for audio filter module: 1 candidate scaletempo warning: bad input or output format scaletempo warning: input and output formats are not similar main warning: no audio filter module matching "scaletempo" could be loaded main debug: TIMER module_need() : 2.000 ms - Total 2.000 ms / 1 intvls (Avg 2.000 ms) main debug: looking for audio filter module: 1 candidate scaletempo debug: format: 22050 rate, 1 nch, 4 bps, fl32 scaletempo debug: params: 30 stride, 0.200 overlap, 14 search scaletempo debug: 1.000 scale, 661.000 stride_in, 661 stride_out, 529 standing, 132 overlap, 308 search, 1101 queue, fl32 mode main debug: using audio filter module "scaletempo" main debug: TIMER module_need() : 1.000 ms - Total 1.000 ms / 1 intvls (Avg 1.000 ms) main debug: filter(s) 's16l'->'fl32' 22050 Hz->22050 Hz Mono->Mono main debug: looking for audio filter module: 24 candidates main debug: using audio filter module "converter_float" main debug: TIMER module_need() : 3.000 ms - Total 3.000 ms / 1 intvls (Avg 3.000 ms) main debug: found a filter for the whole conversion main debug: filter(s) 'fl32'->'fl32' 22050 Hz->22050 Hz Mono->Stereo/Mono main debug: looking for audio filter module: 24 candidates main debug: using audio filter module "trivial_channel_mixer" main debug: TIMER module_need() : 0.000 ms - Total 0.000 ms / 1 intvls (Avg 0.000 ms) main debug: found a filter for the whole conversion main debug: filter(s) 'fl32'->'fl32' 24255 Hz->22050 Hz Stereo/Mono->Stereo/Mono main debug: looking for audio filter module: 24 candidates main debug: using audio filter module "bandlimited_resampler" main debug: TIMER module_need() : 1.000 ms - Total 1.000 ms / 1 intvls (Avg 1.000 ms) main debug: found a filter for the whole conversion

Conclusion? My guess is that there's probably nothing wrong with "WMSpeech Decoder DMO", otherwise it won't work at later stages. May be at the first initiation something else is missing and caused ProcessInput() not to accept, and that is why nothing is coming out.

Just doing some guesswork.

[net7runner]

I have the exact same problem

HOWEVER

...switching to OpenGL output fixes things.

Note that this is an issue ONLY when playing video - when playing an mp3 or something, there is never any hitching.

ATI Radeon 4870
Latest drivers
i7 920

I strongly suspect that there's something wonky with the ATI drivers (or they're trying to do some crazy video correction that's screwing things up).

[net7runner]

...a few things to add:
-Enabling/disabling audio doesn't fix things
-Even switching to OpenGL, DVD playback still hitches. DivX/XVid/etc. does not.

[VLC_help]

DVD playback still hitches

even when VIDEO_TS is played from hard drive?

overlayed: yep, DMO decoders don't always work correctly in VLC. But I would guess that this is something else.

[net7runner]

even when VIDEO_TS is played from hard drive?

Yep. Exact same warning messages as others have posted.

Summary:
-DirectX: starves on pretty much any content I can find to throw at it
-OpenGL: only seems to starve on DVD content (whether played from a disk or from a decrypted .VOB file)

[Worf72]

The same for me, but you gave me the idea of changing from "default video output" to "OpenGL" and that's resolved my "playing silence" problem!

Thank you!!!

[Carnacky]

Same or quite same problem here viewtopic.php?f=14&t=63478, but for me it doesn't help changing to openGL...any help? i isolated the problem and i think it's a video problem...strange uh? i think so 'couse i've tried to put the video output on ASCII mode and i had no problem with audio
For me i've solved trying to put video output on Windows GDI...any other gives me same problem. With windows gdi this is the messages:

main debug: Buffering 0%
main debug: Buffering 8%
mpeg_audio debug: MPGA channels:2 samplerate:48000 bitrate:128
main debug: Buffering 16%
main debug: Buffering 25%
main debug: creating aout
main debug: Buffering 33%
main debug: Buffering 41%
main debug: Buffering 50%
main debug: Buffering 58%
main debug: Buffering 66%
main debug: Buffering 75%
main debug: looking for audio output module: 3 candidates
main debug: Buffering 83%
main debug: Buffering 91%
main debug: Buffering 100%
main debug: Stream buffering done (325 ms in 2 ms)
main debug: no usable vout present, spawning one
main debug: looking for text renderer module: 2 candidates
aout_directx debug: OpenAudio
main debug: thread (fontlist builder) created at priority 0 (../../.././modules/misc/freetype.c:473)
freetype debug: using fontsize: 2
main debug: using text renderer module "freetype"
main debug: TIMER module_need() : 16.000 ms - Total 16.000 ms / 1 intvls (Avg 16.000 ms)
main debug: thread started
main debug: looking for video filter2 module: 20 candidates
freetype debug: Building font database...
freetype debug: Finished building font database.
freetype debug: Took 1000 microseconds
main debug: thread ended
swscale debug: 32x32 chroma: YUVA -> 16x16 chroma: YUVA with scaling using Bicubic (good quality)
main debug: using video filter2 module "swscale"
main debug: TIMER module_need() : 22.000 ms - Total 22.000 ms / 1 intvls (Avg 22.000 ms)
main debug: looking for video filter2 module: 20 candidates
yuvp debug: YUVP to YUVA converter
main debug: using video filter2 module "yuvp"
main debug: TIMER module_need() : 75.000 ms - Total 75.000 ms / 1 intvls (Avg 75.000 ms)
main debug: window size: 720x416
main debug: looking for video output module: 7 candidates
wingdi debug: creating Vout EventThread
main debug: thread (Vout Events Thread) created at priority 0 (../../../.././modules/video_output/msw/wingdi.c:247)
main debug: thread started
wingdi debug: DirectXCreateWindow
main debug: looking for hwnd module: 3 candidates
qt4 debug: requesting video...
qt4 debug: Video was requested -1, -1
aout_directx debug: found device: Driver audio principale
aout_directx debug: found device: Realtek HD Audio output
aout_directx debug: found device: ATI HD Audio rear output
qt4 debug: Updating the geometry
qt4 debug: Updating the geometry
qt4 debug: Updating the geometry
qt4 debug: Updating the geometry
qt4 debug: New caching: 100
qt4 debug: New caching: 100
aout_directx debug: device supports 2 channels (DEFAULT!)
aout_directx debug: device supports 1 channel
aout_directx debug: Windows says your SpeakerConfig is stereo
aout_directx debug: creating DirectSoundThread
main debug: thread (DirectSound Notification Thread) created at priority 15 (../../.././modules/audio_output/directx.c:427)
main debug: using audio output module "aout_directx"
main debug: TIMER module_need() : 180.000 ms - Total 180.000 ms / 1 intvls (Avg 180.000 ms)
main debug: output 's16l' 48000 Hz Stereo frame=1 samples/4 bytes
main debug: mixer 'fl32' 48000 Hz Stereo frame=1 samples/8 bytes
main debug: filter(s) 'fl32'->'s16l' 48000 Hz->48000 Hz Stereo->Stereo
main debug: thread started
main debug: looking for audio filter module: 24 candidates
aout_directx debug: DirectSoundThread ready
qt4 debug: Video is resizing to: 720 416
main debug: using hwnd module "qt4"
main debug: TIMER module_need() : 103.000 ms - Total 103.000 ms / 1 intvls (Avg 103.000 ms)
main debug: using audio filter module "converter_float"
main debug: TIMER module_need() : 61.000 ms - Total 61.000 ms / 1 intvls (Avg 61.000 ms)
main debug: found a filter for the whole conversion
wingdi debug: created video sub-window
wingdi debug: Vout EventThread running
wingdi debug: disabling screen saver
main debug: looking for audio mixer module: 3 candidates
main debug: using video output module "wingdi"
main debug: TIMER module_need() : 125.000 ms - Total 125.000 ms / 1 intvls (Avg 125.000 ms)
main debug: Deinterlacing available
qt4 debug: Updating the geometry
main debug: using audio mixer module "float32_mixer"
main debug: TIMER module_need() : 5.000 ms - Total 5.000 ms / 1 intvls (Avg 5.000 ms)
wingdi debug: GDI depth is 32
main debug: input 'mpga' 48000 Hz Stereo frame=1152 samples/969 bytes
main debug: got 1 direct buffer(s)
main debug: pic render sz 720x416, of (0,0), vsz 720x416, 4cc I420, ar 186923:108000, sar 1:1, msk r0x0 g0x0 b0x0
main debug: pic in sz 720x416, of (0,0), vsz 720x416, 4cc I420, ar 186923:108000, sar 1:1, msk r0x0 g0x0 b0x0
main debug: pic out sz 720x416, of (0,0), vsz 720x416, 4cc RV32, ar 186923:108000, sar 1:1, msk r0xff0000 g0xff00 b0xff
main debug: looking for video filter2 module: 20 candidates
swscale debug: 720x416 chroma: I420 -> 720x416 chroma: RV32 with scaling using Bicubic (good quality)
main debug: using video filter2 module "swscale"
main debug: TIMER module_need() : 1.000 ms - Total 1.000 ms / 1 intvls (Avg 1.000 ms)
main debug: indirect render, mapping render pictures 0-15 to system pictures 1-16
main debug: looking for audio filter module: 1 candidate
main debug: End of video preroll
main debug: Received first picture
freetype debug: using fontsize: 26
scaletempo warning: bad input or output format
main warning: no audio filter module matching "scaletempo" could be loaded
main debug: TIMER module_need() : 5.000 ms - Total 5.000 ms / 1 intvls (Avg 5.000 ms)
main debug: looking for audio filter module: 1 candidate
scaletempo debug: format: 48000 rate, 2 nch, 4 bps, fl32
scaletempo debug: params: 30 stride, 0.200 overlap, 14 search
scaletempo debug: 1.000 scale, 1440.000 stride_in, 1440 stride_out, 1152 standing, 288 overlap, 672 search, 2400 queue, fl32 mode
main debug: using audio filter module "scaletempo"
main debug: TIMER module_need() : 1.000 ms - Total 1.000 ms / 1 intvls (Avg 1.000 ms)
main debug: filter(s) 'mpga'->'fl32' 48000 Hz->48000 Hz Stereo->Stereo
main debug: looking for audio filter module: 24 candidates
main debug: using audio filter module "mpgatofixed32"
main debug: TIMER module_need() : 1.000 ms - Total 1.000 ms / 1 intvls (Avg 1.000 ms)
main debug: found a filter for the whole conversion
main debug: filter(s) 'fl32'->'fl32' 52800 Hz->48000 Hz Stereo->Stereo
main debug: looking for audio filter module: 24 candidates
main debug: using audio filter module "bandlimited_resampler"
main debug: TIMER module_need() : 0.000 ms - Total 0.000 ms / 1 intvls (Avg 0.000 ms)
main debug: found a filter for the whole conversion
main debug: End of audio preroll
main debug: looking for video blending module: 1 candidate
blend debug: chroma: YUVA -> RV32
main debug: using video blending module "blend"
main debug: TIMER module_need() : 6.000 ms - Total 6.000 ms / 1 intvls (Avg 6.000 ms)
main debug: Post-processing available
main debug: Decoder buffering done in 290 ms
main warning: PTS is out of range (-10000), dropping buffer
main warning: PTS is out of range (-33000), dropping buffer
mpgatofixed32 debug: libmad error: bad main_data_begin pointer
qt4 debug: Qt: Entering Fullscreen
main debug: control type=1

For information, i have a realtek integrated sound card and an ati radeon 4870 as a video card...

[VLC_help]

Maybe this has something to do with audio thread priority. Video thread has much higher priority and because of this, it might choke audio in some cases.

[Carnacky]

I tried giving more priority to vlc process by the specific check in the options, but no effect. How can we solve it? only with windows gdi?
thank you

[VLC_help]

I didn't mean vlc.exe priority, but
main debug: thread (DirectSound Notification Thread) created at priority 15 (../../.././modules/audio_output/directx.c:427) the one given to audio playback thread.

[Carnacky]

I understand. How can I solve it?
thanks