Article: This Just In: Microsoft Launches Native Class 2 USB Audio Support. Wait, What?

[Phil C]

Thanks for the answer Chris.  

 

BTW, Microsoft is now pushing the Creators Update patch. One of my PCs downloaded it as part of the May monthly update and is awaiting restart to complete installation.

 

Microsoft is now very aggressive with the monthly updates.  For Win 10 at least, there is no longer a menu of patches from which to pick and choose.  All patches are now bunched into a monthly update, so we have a Hobson's choice to install the Creators Update patch as part of the monthly update or to not do the monthly update at all (rendering the computer more vulnerable to viruses and other nasties).    

[csx31]

Hi, Chris,

 

I have enjoyed reading your reviews of the Berkeley Alpha DAC and Alpha USB, and have read that you presently use their Reference DAC.  Berkeley provides a custom USB driver for use with a Windows PC.  

 

In your opinion, do you believe the new Windows 10 USB audio class 2 driver would offer an audio performance upgrade over the Berkeley-provided driver?  Your thoughts on this are appreciated.

 

Possibly others on this forum use Berkeley DAC's and Alpha USB converters, have used the new Windows driver, and can comment.

 

 

[The Computer Audiophile]

27 minutes ago, csx31 said:

Hi, Chris,

 

I have enjoyed reading your reviews of the Berkeley Alpha DAC and Alpha USB, and have read that you presently use their Reference DAC.  Berkeley provides a custom USB driver for use with a Windows PC.  

 

In your opinion, do you believe the new Windows 10 USB audio class 2 driver would offer an audio performance upgrade over the Berkeley-provided driver?  Your thoughts on this are appreciated.

 

Possibly others on this forum use Berkeley DAC's and Alpha USB converters, have used the new Windows driver, and can comment.

 

 

 

I don't believe there should be an audio performance upgrade, but I'm guessing it may work better. The manufacturer supplied driver i really old and has to be massaged just to get it installed on Windows 10.

[JackoUK]

"There is no such thing as a class driver specification."

Sorry mansr but you are manifestly wrong on this crucial point: please see ...

http://www.usb.org/developers/docs/devclass_docs/

Note just one title from this International Standards body:

"APPROVED CLASS SPECIFICATION DOCUMENTS"

[JackoUK]

For those who are concerned about the impact of upgrading to a new version of Windows 10 I refer back to my post with the Powershell command:

Export-WindowsDriver -Online -Destination D:\DriverBackup

Good practice (or should I say 'sound' practice, get it?) might be to run this command from a working Windows 10 installation before upgrading ... and if the upgrade disturbs working drivers, then re-install from the old configuration in the backup folder.

[JackoUK]

Although one might want to delay upgrading to the Windows 10 Creators Update ...

... once you've got there, Microsoft have extended the ability to control further changes:

http://www.intowindows.com/how-to-delay-or-defer-updates-up-to-365-days-in-windows-10/

Don't forget too that you can rollback to the previous installation if you don't like the upgrade result.

 

My view is that one should be looking to upgrade ... but at a convenient time (now better supported) ... since MSFT are improving Windows 10 free of charge. Consider for example that flac support came with the first version of W10, and now USB class 2 audio with the Creator's Update. If the audio industry decides to add functionality in web browsers then it is a pound to a penny that MSFT will deliver this via Edge and other newbies via Groove.

 

It's hypocritical to lambast MSFT for not supporting flac or USB audio class 2 ... if you then say you aren't prepared to upgrade for free!

 

That doesn't mean you should jump off a cliff ... but recent changes to updating make the upgrade process more convenient and manageable. Since I enjoy a lot of computing I treat maintenance of my PC's as regularly as the maintenance of my house, car, ...

[mansr]

1 hour ago, JackoUK said:

"There is no such thing as a class driver specification."

Sorry mansr but you are manifestly wrong on this crucial point: please see ...

http://www.usb.org/developers/docs/devclass_docs/

Note just one title from this International Standards body:

"APPROVED CLASS SPECIFICATION DOCUMENTS"

That's all about hardware, the thing you connect to a USB port. Drivers are software.

[JackoUK]

"In your opinion, do you believe the new Windows 10 USB audio class 2 driver would offer an audio performance upgrade over the Berkeley-provided driver?"

 

As I have been trying to persuade another contributor the IT theory is that class drivers are interoperable: one can switch them without any effect whatsoever.

 

However we are talking about 2 companies developing a piece of software (the driver) ... and every piece of software contains errors. Most likely Berkeley subcontracted the development of the driver to a software house. So the question is "Did the subcontracted software house do a better job than Microsoft?"

Answer: <your opinion here>.... or would it be better to try a test or two?

 

My guess is:

- your CD laser has read the next PCM 16-bit digital sample as 0000000000101011 (that should have been decimal 42 but it's actually 43 because you have dust on the CD)

- the error correcting process in your CD software has corrected this to 0000000000101010

- your music player software has called WASAPI or ASIO code and asked it to transfer the value 0000000000101010  without alteration to the USB audio driver

- the USB audio driver has sent 0000000000101010 to your DAC

- your DAC plays 0000000000101010 at the right time because it is pretty good

Some errors occur on the way now and then ... but you never hear them ... in the same way that when you attend a live concert the performers are so good that even when they make a mistake you at not knowledgeable or attentive enough to notice them ... or you do on rare occasions and ... so what?

 

0000000000101011

0000000000101010

0000000000101010

0000000000101010

0000000000101010

 

What performance upgrade can you do to 0000000000101010 ?

 

[OK, so I have over-egged the process.]

 

OTOH ... maybe you know something about the subcontractor software development team used by Berkeley or the team MSFT used for UAC2?

 

 

 

[rickca]

1 hour ago, JackoUK said:

My view is that one should be looking to upgrade

The fix released on 5/9 that made HQPlayer crash was a quality (security) update, not a feature update like Creators Update.  Quality updates can only be deferred for 30 days.  Feature updates can be deferred up to 365 days.

 

A lot of the new features in Windows 10 are incomplete and buggy.  Microsoft is rolling out a lot of infrastructure that they haven't even exploited yet.  The recently announced fall Creators Update gives you a good idea of where they are trying to go.  

 

I agree with you for general computing requirements, but not so much on a PC primarily used for audio.

[The Computer Audiophile]

3 hours ago, JackoUK said:

"In your opinion, do you believe the new Windows 10 USB audio class 2 driver would offer an audio performance upgrade over the Berkeley-provided driver?"

 

As I have been trying to persuade another contributor the IT theory is that class drivers are interoperable: one can switch them without any effect whatsoever.

 

However we are talking about 2 companies developing a piece of software (the driver) ... and every piece of software contains errors. Most likely Berkeley subcontracted the development of the driver to a software house. So the question is "Did the subcontracted software house do a better job than Microsoft?"

Answer: <your opinion here>.... or would it be better to try a test or two?

 

My guess is:

- your CD laser has read the next PCM 16-bit digital sample as 0000000000101011 (that should have been decimal 42 but it's actually 43 because you have dust on the CD)

- the error correcting process in your CD software has corrected this to 0000000000101010

- your music player software has called WASAPI or ASIO code and asked it to transfer the value 0000000000101010  without alteration to the USB audio driver

- the USB audio driver has sent 0000000000101010 to your DAC

- your DAC plays 0000000000101010 at the right time because it is pretty good

Some errors occur on the way now and then ... but you never hear them ... in the same way that when you attend a live concert the performers are so good that even when they make a mistake you at not knowledgeable or attentive enough to notice them ... or you do on rare occasions and ... so what?

 

0000000000101011

0000000000101010

0000000000101010

0000000000101010

0000000000101010

 

What performance upgrade can you do to 0000000000101010 ?

 

[OK, so I have over-egged the process.]

 

OTOH ... maybe you know something about the subcontractor software development team used by Berkeley or the team MSFT used for UAC2?

 

 

 

 

The company is called Thesycon and it has done drivers for many companies. 

 

I'm not sure what you are trying to say with your example.

[scan80269]

This Microsoft UAC2 driver is not as "universal" as I was led to believe.  So far, two of my DACs don't work with this driver:

 

* Schiit Yggdrasil - I have a post about this

 

* Anedio D2 - No yellow bang in Device Manager so driver loaded OK, but both Foobar2000 and HQPlayer choked up trying to play any track in WASAPI mode.

 

For both these DACs, the manufacturer supplied Windows driver represents the only working driver option.

 

[mansr]

5 hours ago, scan80269 said:

This Microsoft UAC2 driver is not as "universal" as I was led to believe.  So far, two of my DACs don't work with this driver:

 

* Schiit Yggdrasil - I have a post about this

 

* Anedio D2 - No yellow bang in Device Manager so driver loaded OK, but both Foobar2000 and HQPlayer choked up trying to play any track in WASAPI mode.

 

For both these DACs, the manufacturer supplied Windows driver represents the only working driver option.

Add Steinberg UR242 to that list. The device is displayed in the control panel and applications, but it doesn't work at all.

[Dr Tone]

Working fine so far with the Ayre Codex.  PCM and DSD DOP.

[KDinsmore]

Worked fine for my Wyred 4 Sound Dac2v2SE. However I started getting clicking sounds from my ripped SACD's? I've read about that before but never experienced it. Switched back to ASIO and it plays fine again. Anyone have an answer?

[mansr]

1 hour ago, KDinsmore said:

Worked fine for my Wyred 4 Sound Dac2v2SE. However I started getting clicking sounds from my ripped SACD's? I've read about that before but never experienced it. Switched back to ASIO and it plays fine again. Anyone have an answer?

What format are you playing your SACD rips in?

[The Computer Audiophile]

Also, what app are you using?

[KDinsmore]

Hi Chris,

JRiver latest version.

 

Keith

[KDinsmore]

Using ISO and dsf

[balasiv]

Hello everyone,

I'm a Program Manager on the Windows Audio team and wanted to respond.

Yes - the USB Audio 2 class driver was released in the Creators Update. We have added more features and fixed bugs in the Fall Creators update which should address some devices not working with the class driver. We're striving to make the driver as useful as possible - if you encounter something that doesn't work, please report it via the Feedback hub as described in this article [link]. Windows Insider builds may have additional fixes - I encourage everyone to test out Windows Insider builds and let us know if something isn't working (so we can fix it before the public release).

 

The OP is correct that the class driver does not support ASIO. Recently added features (like MATCH_FORMAT) make it possible for an app to output high fidelity audio (no sample rate conversions) without resorting to exclusive mode (which I assume is what "WASAPI playback" alludes to).

 

In the event of a third party driver being present, it will be preferred over the inbox class driver. Follow these steps to switch over to the Microsoft class driver (can be reverted at any time).

 

I hope this is helpful to understand native USB Audio 2 support in Windows.

 

~Bala Sivakumar

[The Computer Audiophile]

Hi Bala - Thanks so much for posting here at CA. We value open communication between our community and developers / manufacturers.

 

Thanks for the links as well. Good stuff.