Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: WASAPI output component version 3.0 final (Read 163357 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

WASAPI output component version 3.0 final

Reply #100
Sounds like a driver bug. Old 2.1 version of the component tried to use floating point output in 32-bit mode. This was changed for v3.0 and it now reverts to PCM mode if float mode fails. Your drivers report it as supported, yet fail horribly with it.

WASAPI output component version 3.0 final

Reply #101
For the benefit of any one having glitching/pausing problems with WASAPI Event Mode, I've managed to fix my issues by adjusting advanced settings within Foobar.  I'm playing back FLAC's and MP3's from my external 320GB hard drive, using Foobar2000 v 1.3.1 and WASAPI 3.2.3 on Windows 7 (64 bit).  My specific issue was the playback of MP3's regularly pausing and/or stuttering when using Event Mode.  The issue did not occur with FLAC's, or when using Push Mode.  I'm on a work laptop that is locked down so hard I can't change anything in the OS, even down to my task bar auto hide or turning the date/time display on or off.  I've found Event Mode drastically clearer and more open sounding compared to Push Mode, so I've tried just about everything possible to get it working glitch free.

The main setting that fixed the glitching was Advanced>Playback>Thread priority>Thread priority (1-7) and setting it to 1, rather than the default 7.  This is the opposite of what I've read on various forums.

I also had to bump up the "Hardware buffer in MS - event mode" to 75.  I tried up to 100, but then foobar would have a melt down and do this weird, glitchy, fast forward playback.

Other settings that I'm using are:
Using MMCSS set to Pro Audio
Buffer length set to 350ms (minimum determined by the ctrl-alt-delete method)(http://www.head-fi.org/t/539638/how-to-set-perfect-buffer-lengths)

Everything else is default and no active DSP's.

I still get the occasional glitch, so there is obviously a DPC latency issue on the computer, which I can't fix due to the PC being locked down.

WASAPI output component version 3.0 final

Reply #102
You say you have tried just about anything but have you tried keeping the output buffer longer and playback priority high? Put the playback thread priority back to 7 and pump the output buffer to something like 5000 ms. Low output buffer offers no benefits in music playback. And unless you have a very special output device with broken drivers there is absolutely no difference in the playback between event and push WASAPI outputs.

WASAPI output component version 3.0 final

Reply #103
I've found Event Mode drastically clearer and more open sounding compared to Push Mode, so I've tried just about everything possible to get it working glitch free.
Don't waste your time on hunting figments of your imagination. If "push" mode or DirectSound don't have any glitches, use those, preferably the latter.

Buffer length set to 350ms (minimum determined by the ctrl-alt-delete method)(http://www.head-fi.org/t/539638/how-to-set-perfect-buffer-lengths)
There is zero benefit of having a short buffer for music playback other than that changing DSPs on-the-fly will have the changes kick in earlier. If you are experiencing glitches, try to increase buffer size to something reasonable like a couple of seconds to avoid those. The guide on head-fi is plain wrong:
Quote
Having the lowest latency possible is important for your audio.
This is a lie, at least with respect to audio playback.

I still get the occasional glitch, so there is obviously a DPC latency issue on the computer, which I can't fix due to the PC being locked down.
Too bad, this might also be important to solve.
It's only audiophile if it's inconvenient.

WASAPI output component version 3.0 final

Reply #104
If I set a long buffer, the pauses increase and take longer for playback to recover.  But yeah, I agree that low latency shouldn't be that important for playback unless you're simultaneously recording or using DSP's.  I'll try some further tests.

I probably should do an ABX test or something, but at least on my work machine there is an easily discernible difference in sound quality between modes.  Maybe an RMAA test would prove the point, however it's not the main purpose of my post, so please ignore my previous comment on sound quality (I should have known better).  On my home PC there is no significant differences between the two modes.  More than likely my work PC's sound or usb drivers are broken, but there's nothing I can do.

WASAPI output component version 3.0 final

Reply #105
I did a "reset page" for the advanced preferences.  This seemed to help quite a bit, so maybe my settings were corrupted or something.  At 350 ms output buffer, the stutters were still there, but only after a few songs played, not straight away.  I put the buffer up to 3000 ms and everything seems pretty good now.

WASAPI output component version 3.0 final

Reply #106
didnt update this component for a long time. After i did it i found some problem: when next track has different format (96khz/24bit instead of previously played 44.1/16) playback stops with error message: "Unrecoverable playback error: Device invalidated" or "wrong arguments (or somehow like that)", but if wait 2-3 seconds and manually begin playback of new file, it plays fine. I didnt noticed this problem before. Using USB sound card E-MU 0204, Win8.1 x64.

p.s. maybe somehow add pause between different source tracks?

WASAPI output component version 3.0 final

Reply #107
I have no problems switching from 16/44.1 to 24/96 using WASAPI.

Are there any newer drivers for your E-MU 0204?
Who are you and how did you get in here ?
I'm a locksmith, I'm a locksmith.

WASAPI output component version 3.0 final

Reply #108
nope, they are old as this soundcard. And no any hope for update. WASAPI ver. 2.1 make it same, looks like i get it after win8.1 update. Seems like it really need some delay between tracks like this.

WASAPI output component version 3.0 final

Reply #109
Does the 0204 support ASIO output? Maybe that's the way to go?
Who are you and how did you get in here ?
I'm a locksmith, I'm a locksmith.

WASAPI output component version 3.0 final

Reply #110
only ASIO4ALL, but it plays fine only 16/44.1. In higher resolution it begins play cracks, noise and more over the music.
Best of all 0204 handles by DS output, but it begins to glitch when i work in visual studio, so i need WASAPI.

also, is it possible to get sources of this foo_out_wasapi? Intersting in some moments


WASAPI output component version 3.0 final

Reply #112
LatencyMon says my pc isnt good for realtime audio. But playing music - isnt real-time task. Dont want install beta drivers - files from it smaller than the original driver, so i dont believe them. Also, problem isnt in playback, it between different tracks. Maybe, device configured for new track after buffer is filled with new pcm data (async, or maybe something like that). Or hardware problem - device really need some time to switch something

WASAPI output component version 3.0 final

Reply #113
I would be looking in to the reason why LatencyMon is reporting your PC is struggling with realtime audio. That would explain why your getting glitches when your using ASIO.

First thing i would try is the beta driver, if it doesn't fix it then just roll back to the latest stable version. Other than that i don't think your going to be about to the the source code for foo_out_WASAPI since it is maintained by the creator of Foobar2000.
Who are you and how did you get in here ?
I'm a locksmith, I'm a locksmith.

WASAPI output component version 3.0 final

Reply #114
i have installed beta drivers and my problem has been solved after few reboots, so stay on wasapi renderer

asking for sources not to fix them, now digging in microsoft`s example wasapi out for win8, and this one are cracking too much on start or end playback

WASAPI output component version 3.0 final

Reply #115
Is there any reason why foobar2000 WASAPI plugin supports only exclusive mode? I really, really dislike this "exclusive" part, I use WASAPI because I have sound issues when using DirectSound (sound crackling and skipping). Other programs like AIMP3 support WASAPI Shared mode (http://i.imgur.com/qqpdpyG.png), which is perfect for me, but I don't want to change my audio player that I use since 2004. Is there any way I can use foobar2000 with WASAPI non-exclusive mode?

WASAPI output component version 3.0 final

Reply #116
DirectSound uses WASAPI shared mode behind the scenes. There's no need for foobar2000 to reimplement it. Don't use too small buffers or touch priority settings and there won't be glitching.

WASAPI output component version 3.0 final

Reply #117
What does this error signify:  Unrecoverable playback error: Invalid argument

I have foobar into my AVR via SPDIF.  I'm using v 3.2.3 of the component and the v 1.3.4 of Foobar.  Playback is set to WASAPI (event):  SPDIF OUT (2 creative SB X-Fi).


EDIT:  If I add in SOX and reample to 48Hz the error goes away on the problem files.  But, playback is glitchy (even with a huge buffer set) with pops/clicks on seeking and start/end. 




As a second, essentially unrealated question, is it possible to have the foo-out WASAPI not let the driver know it's out of data to play i.e continue to send silence?  If not, would it be possible to introduce such an option?  Reason being that I (and I have seen other people experiencing this via google searches) have been experiencing pops/clicks/glitches at the start and end of songs.  It might be a resampling issue but it might also be an issue with the PCM stream stopping.  Have tested on my X-Fi soundcard, integrated HD audio on mobo and also on the HMDI audio chip of my graphics card.

WASAPI output component version 3.0 final

Reply #118
WASAPI Output Component version 3.2.3, Foobar2000 version 1.3.4
Windows 8 x64, Realtek High Definition Audio on board,S/PDIF optical to external DAC
I have encountered the following error: 'Unrecoverable playback error: Unsupported stream format: 88200 Hz / 24-bit / 2 channels'
Obviously, this is due to the fact that my embedded card’s driver (Realtek, maybe it’s actually a generic Microsoft’s driver) provides for only a limited number of choices for sampling frequencies supported by a receiver (external DAC): 44.1; 48; 96; 192. See Figure 1 below.

Fig 1
This issue has been already raised with regards to WASAPI Output Component version 3.2.1. (See http://www.hydrogenaud.io/forums/index.php...6805&st=50)
The proposed solution was to
Quote
Just go in foobar2000 option then in Advanced/Playback/WASAPI check "Attempt to play unsupported stream formats”

However, in the current version of the component there is no longer such an option. (See Figure 2.)

Fig 2
I understand that a possible solution is to resample, e.g. with the Sox. However, I prefer to stream unchanged at a sampling rate actually supported by my DAC. 88200 Hz / 24-bit / 2 channels.
Please, advice what happened to the "Attempt to play unsupported stream formats” option of the component. How to stream unaltered  88200 Hz?

WASAPI output component version 3.0 final

Reply #119
Pavlo, OK, you DAC supports 88200. But does your realtek supports 88200? I guess no.

WASAPI output component version 3.0 final

Reply #120
Pavlo, OK, you DAC supports 88200. But does your realtek supports 88200? I guess no.


Realtek does support it. See http://www.realtek.com.tw/products/product...&ProdID=284

I have tried both most up-to-date drivers for audio on the motherboard vendor site and on the Realtek site. But of no avail. The Windows dialog does not list 88200 as a choice.

WASAPI output component version 3.0 final

Reply #121
Pavlo did you manage to fix this in the end? I have the same problem.

WASAPI output component version 3.0 final

Reply #122
http://www.head-fi.org/t/550087/realtek-al...15#post_7455952

Quote
IOW, the generic Windows driver recognizes the hardware capability, but Realtek's own driver intentionally disables one of its output modes.

The Realtek driver includes all the EQ and sound effects software, but I'm not sure I need that, so I uninstalled all audio drivers and let Windows install its own driver a 2nd time, and now I have 88.2kHz playback capability.

...but, slightly later:

Quote
I did try it to uninstalling the realtek drivers and indeed the default Microsoft driver gives an 88.2 kHz option. However when I use foobar2000 with the wasapi driver to playback a 88.2 kHz file (which it was first refusing todo) my receiver is reporting 96 kHz so it seems like the Microsoft driver is doing a sample rate conversion.


At least you can try MS drivers.

WASAPI output component version 3.0 final

Reply #123
Hi Sir,

I have installed WASAPI component, there are 2 options in Output:
1. WASAPI (event)
2. WASAPI (push)

Which plays better sound?
What is the

WASAPI output component version 3.0 final

Reply #124
There're sudden pause sometimes with WASAPI event mode after upgrade my computer to windows 10, I don't know why.

Foobar2000 behaves quite well on windows 8.1. Is there anything that we should do to avoid such issues, or it's just the driver not mature for windows 10?

I have a schiit Bofrost USB Audio DAC, with a asynchronous usb connection to the computer.

Edit: Manually installed the driver for windows 8.1, now it works perfect. It seems the 'windows 10' specified new driver has serious bugs.