Audio clipped on start of playback

[sparkcml]

Using DisplayPort output on a Dell Precision 3260 running CPU i5-12500 - Windows 10 - connected to an DisplayPort screen input with speakers.  I've updated the video drivers to the latest available, and even tried the Intel Generic Video Drivers.  I've updated the BIOS to the August 2024 release from Dell with the Intel fixes.

The problem is - when I play a very short WAV file, the first 200ms (guessing) of the audio is missing, and if the WAV file is only 250ms long, then nothing is heard.  WAV file is attached.

I don't think this is a new problem.  I think there's a few posts on the topic, and I've tried a few fixes, but nothing appears to resolve the issue.

Can you have the DEV team play the WAV file using Windows Media Player on a computer with an DisplayPort attached monitor that has speakers?

If you get sound, I would be surprised.  If you don't get sound, then play the file in a loop by pressing the Media Player loop button.  Then you'll hear the sound.

What I suspect (me guessing again) is there's some sort of power-save processing going on, on the audio-output used by DisplayPort, and it takes a few milliseconds to wake up.  Once it's working, it continues to work for perhaps a second, then needs startup time again to play audio.

I've duplicated this issue on a few different computers, so it's not the hardware.  I do R&D for our company, and wouldn't ask for help without doing the obvious - new cables, 2nd computer, different monitor...etc.

I appreciate any direction you can provide, and confirmation that this problem is reproduced in the Intel Labs.

(Edit: - I tested DisplayPort to an HDMI monitor - and >> some audio << is heard, but not all.  Looping plays all audio.  I direct-connected the audio-out jack on the PC to the audio-in jack on the monitor, and set RealTek as the audio-out on the PC, and all the audio is heard, so it's an Intel audio-out over displayport issue.)

(2nd Edit: - I tested DisplayPort to an HDMI monitor on an old Dell Optiplex 7050, Windows 10, Intel Core i7-7700, running the Microsoft audio driver 10.0.17763.1 and there is NO audio lost.  It plays fine.)

[sparkcml]

hello?

[SageOne]

Its a digital rights management issue. In July microsoft 's updates tripped a known bug in intel and realtek and Dolby drivers. They all trying to do the same thing at the same time and failing. System restart works for the 1st service but as soon as the 2nd starts they crash. solution is to reinstall all original oem drivers one by one while fiddling with the HECI, HDCP, and Dolby settings till you find a combination that works. Its happened twice this year and once in 2023.

Remember windowed apps start and stop these services when you go in and out of full-screen. This in turn crashes those services also.

[sparkcml]

Not out of the question.  It is possible DRM is interfering with the startup of audio.  But, as stated, this issue doesn't happen with RealTek audio-out.  At the moment, we have to use the PC 3.5mm Audio-out, and link it to the DisplayPort Monitor 3.5mm audio-in, bypassing the DisplayPort audio path.  In some cases, DisplayPort-out to HDMI-in on the monitor has better results (but not every case).

Wish I could find a generic Microsoft audio-out driver that would work.  The lack of Intel response is an indicator that they're not seeing the post, or they've seen the post, and have no opinion on this issue because they know it's their bug.

A simple "We see your post.  We're investigating" or "We see your post.  We don't have a problem"...anything is better than nothing.

[sparkcml]

More details:

removing Intel drivers (defaulting to Microsoft generic drivers) - audio works

Testing process with driver versions attached in the rtf document

requested SSU outputs during failed audio attached.

[VARADHARAJAN]

@sparkcml 

I checked the SSU report, and I found Realtek's dedicated audio chip embedded in MB, But you trying to get audio from monitor's built-in speaker right? 

[sparkcml · Accepted solution]

Correct.  We're using the DisplayPort-out audio path.  The Realtek audio-out onboard has no issue.  It will play the clip start-to-finish.  It's only when trying to use built-in Monitor speakers via HDMI or DisplayPort audio path that we see the issue.  Using 3.5mm PC-Out to Monitor 3.5mm input (bypassing HDMI or DisplayPort audio) - all audio is heard.

Thank you for looking.

[VARADHARAJAN]

I am using an external speaker from Realtek audio chip,   Will you try to  install  previous sound driver which built in graphics driver, 

If you have a previous graphics driver, extract using winrar app, go to  graphics folder,   install  this file MSHdaDac.inf using  device manager

[sparkcml]

Please read the steps followed in the "Audio Missing Steps Tried.rtf" document.  Audio works with Microsoft Generic Drivers.  Audio also works when using RealTek audio output.  Audio fails using Intel Audio Drivers (auto downloaded), and fail when using Intel Generic Audio drivers.  It also fails using Dell's installed Intel drivers

This is not a new issue.  It's also seen here: 

https://community.intel.com/t5/Graphics/UHD-730-HDMI-audio-cutting-out-issue/m-p/1342396

There are more examples.  Most people are using a work-around.  Not all people can use a work-around.

If there's a silent audio stream being sent, mixed audio will be heard from the start.  If the audio path is quite for over a second or two, the startup audio is clipped for about 200ms before audio is heard.  The community suspects there's a wake up time for the audio path (perhaps DRM interference, or power-save interference).  Intel will need to fix the drivers being released.

I will wait for an official verification of the bug in their R&D Labs, and their official driver fix.

Thank you

[VARADHARAJAN]

@sparkcml 

You can prevent the auto-downloading of  all drivers  from Windows update by adding registry key

[sparkcml]

Yes - this is another workaround.  This also prevents bug-fixed, and security-fixed drivers from being automatically installed.

I can't condone reducing security or reliability.

The correct "solution" is for Intel to duplicate the issue, and for their programmers to resolve the issue, and provide the general software update thru normal channels.

 I'm sure they will resolve the issue.  It's in their process queue to review.

[JedG_Intel]

Hi sparkcml,

Thank you for posting at Intel Community Forums.

I sent an email so I can provide you with further assistance. Please check your inbox if you have the time so we can determine the best course of action.

Best regards,

Jed G.

Intel Customer Support Technician

[sparkcml]

Official Intel Response received:

~~~~~~~~~~~~

Hi Steve,

Thank you for your ongoing patience as I work to resolve your case number. I apologize for the delay in our response in the previous days due to high volume of cases our services are currently dealing.

After investigation, I would like to inform you that this is a well known issue. Our team cannot promise a fix anytime soon but if there is one, I will note this in the community thread and in future driver release notes. If you don't have any other inquiries regarding to your system, let us know if we can close the case.

Should you have any further questions, you may reply to this email. I will be waiting for your response. In case I haven't heard from you, I will be sending you a follow-up message on October 29, 2024.

Thank you for choosing Intel.

Best regards,

Hannah

Intel Customer Support Technician

~~~~~~~~~~~

Yes - this is a known issue.  No, this is not an acceptable response.  No - it's not likely I'll ever see a true fix.

Life goes on....40+ years of IT, and I saw this response weeks ago.  Intel either has interest in fixing their bug, or they don't.

I expect more.  Oh well.  If you found this post informative, Thank our R&D department (me)...lol.  Have a wonderful day!