Issues with Windows sharing audio - massively distorted sound

[concealed identity]

Hi all, hoping people more knowledgeable about computers and audio can maybe help me solve this issue - I've looked for answers to this on reddit and elsewhere but haven't been able to find anything.

I'm using the Novation Ultranova as my audio interface in Windows 10 64-bit.

Everything seems to be working fine when I am in my DAW or regularly using Windows, as long as in my DAW I have the option "Release Audio Driver in Background" selected - meaning when the DAW is not in focus, it will not play audio.

Ideally, I'd like to have this option unchecked, so I can play from the DAW while in other programs, and vice versa.

The issue is that when this is unchecked, audio from other Windows programs (including Windows system sounds) are EXTREMELY distorted and garbled. This also happens if Windows plays a sound when it pops up a window while I'm working in my DAW.

I assumed this was an issue of bitrates not lining up, but everything seems to be set to the same as far as I can tell. Here is how things are configured:

- If I access the Novation USB Audio Control Panel from within my DAW, the sample rate/format is at 44100/16-bit (if I access it from Windows, it does not list or give me the option to change these).

- My project in the DAW is also set to 44100/16, with "Release Audio Driver in Background" unselected (so that it continues to play when the DAW is not in focus).

- If I access my sound settings and device properties in Windows, I have "Allow applications to take exclusive control of this device" unselected, and "Default format" for when running in shared mode set to 2-channel, 16-bit, 44,100hz.

- On my Ultranova itself, in the Global settings, I have the Default Rate set to 44.1KHz.

So as far as I can tell, everything is lined up, but any sound not coming from my DAW is unbearably loud and garbled. I've tried this with two different DAWs (latest versions of Cubase 12 and Studio One 6).

I'm on Windows 10 64-bit, and I've updated both Windows and my audio driver to latest versions.

I don't have another computer or audio interface to cross reference this with, but do you have any idea what I could be missing here? Could it simply just be a fundamental issue with the Novation drivers? I'm struggling to come up with anything else.

Thanks for reading all of that - I'd really appreciate any help!

[BertKoor]

I'd blame the Ultranova trying to be a jack-of-all-trades, and thus is not perfect in any of these trades. Programming a proper audio driver is a costly business, not all companies (read: very few) are good at that.

You probably can use the built-in audio chip of your computer with less of these issues. Route its (headphone?) outputs to the analog inputs of your Ultranova.

Another thing worth experimenting with is try to use a different driver model in your DAW. Some let you have the choice: WDM, WASAPI, ASIO... Experiment with that, if you can.

[lfm]

On my windows 10 laptop if I rightclick on speaker symbol on task bar there is a menu.
- Open volume mixer, one option says
- there you can set level of various sources

I have a vague memory I chose some option in Sound setting to get this alternative if not there for you from start.

[Pictus]

You have Ultranova and onboard sound, try disabling one of them.
Test both options.

BTW, some good tweaks for lower DPC latency and better USB/LAN.
They will not fix your problem.
viewtopic.php?p=8667750#p8667750

[concealed identity]

You have Ultranova and onboard sound, try disabling one of them.
Test both options.

Thanks, but this unfortunately does nothing - the issue happens because both my DAW and Windows/other apps are using Ultranova, and for some reason can't share it properly.

[concealed identity]

On my windows 10 laptop if I rightclick on speaker symbol on task bar there is a menu.
- Open volume mixer, one option says
- there you can set level of various sources

I have a vague memory I chose some option in Sound setting to get this alternative if not there for you from start.

Thank you, but I'm not sure I understand what exactly you're suggesting here. Do you mean I should just mute all the other sound sources every time?

[concealed identity]

I'd blame the Ultranova trying to be a jack-of-all-trades, and thus is not perfect in any of these trades. Programming a proper audio driver is a costly business, not all companies (read: very few) are good at that.

You probably can use the built-in audio chip of your computer with less of these issues. Route its (headphone?) outputs to the analog inputs of your Ultranova.

Another thing worth experimenting with is try to use a different driver model in your DAW. Some let you have the choice: WDM, WASAPI, ASIO... Experiment with that, if you can.

It very well could be the driver, but the fact that I can't find evidence of anybody else having this problem is giving me the slim hope that this is user error. Unfortunately ASIO is my only option.

[Goony]

Not many ASIO drivers allow multi app usage its one of the limitations.

[T-CM11]

Not many ASIO drivers allow multi app usage its one of the limitations.

The OP isn't trying to use ASIO in multiple applications.

[BertKoor]

No, but while ASIO uses it, it's exclusive. No other app can access it (eg through WDM)

I see no other option for the OP than to invest in a decent audio interface with more mature drivers.

[T-CM11]

No, but while ASIO uses it, it's exclusive. No other app can access it (eg through WDM)

I see no other option for the OP than to invest in a decent audio interface with more mature drivers.

Or use integrated sound (if available) for non-ASIO purposes.

[BertKoor]

That's what I adviced already: plug the analog output of built-in sound of the PC into analog inputs of the keyboard/interface and use that for monitoring. Only requires a stereo 3.5mm -> RCA cable plus a pair of RCA -> 1/4" TS adapters.

[T-CM11]

That's what I adviced already: plug the analog output of built-in sound of the PC into analog inputs of the keyboard/interface and use that for monitoring. Only requires a stereo 3.5mm -> RCA cable plus a pair of RCA -> 1/4" TS adapters.

That's an option, maybe*. For my second (DAW/General usage/Games) PC, I have a Soundblaster Z and a Steinberg UR22C. They both go to different inputs in my active speakers (AudioEngine A5+).
The OP hasn't mentioned the "output situation" though (speakers/amp/headphones/mixer?)

*I don't know if the external input on the Ultra goes through the synth section or not. Sometimes you need to open the amp (e.g. with an envelope) to hear sound.

[lfm]

On my windows 10 laptop if I rightclick on speaker symbol on task bar there is a menu.
- Open volume mixer, one option says
- there you can set level of various sources

I have a vague memory I chose some option in Sound setting to get this alternative if not there for you from start.

Thank you, but I'm not sure I understand what exactly you're suggesting here. Do you mean I should just mute all the other sound sources every time?

No, just set level so sum of them all does not create overs.
- what goes via Sound in Windows, browser usually do that
- what daw produce

If both set to peak at -6dB, you will reach 0dB and no overs.

I get that all the time
- I listen to a pod locally on computer
- goes with browser to a site that play audio by default, some YT channel or whatever

This raises total level by 6 dB or so.

So I turned nominal level down.

I had this for daw usage too, that I had two outputs
- one for phones and one for speakers
- but made mistake in RME totalmix to also route them there in soundcard to same output
Result was overs by many dB

You often run daw to close of 0 dB peaks out, so not much is needed to tilt over to more than that, if having more sources.

So using hardware faders to turn that down a bit, -6dB or so, so final output with another source that also do close to -6dB or even 0dB you can keep total so not distorted.

[concealed identity]

Thanks for the advice, everybody!