USB 2.0 vs 3.0: Noticeable Sound Quality Difference

[omega66]

Hello everyone,

I’m facing a strange issue with my audio setup and I’d appreciate your insights.

My system consists of:

• Amplifier: Yamaha A-S501
• Speakers: Mission MX-3
• DAC: Topping E30 II Lite (separately powered)
• Source: PC running Windows 11, connected to the DAC via USB

Until recently, I was using a regular PC and the sound was excellent.

I switched to a fanless industrial HTPC (no-name, wanted to get rid of noisy fans), but to my surprise the sound quality dropped noticeably. Music now feels compressed and lacking dynamics.

To troubleshoot, I tested with a laptop and the sound was again very good. The only difference I noticed:

• On the PC/laptop I was using a USB 3.0 port
• On the HTPC I initially used a USB 2.0 port

When I tried the HTPC’s USB 3.0 port, the sound improved significantly, close to the original PC quality.

I always thought it was completely irrelevant which device sends the signal via USB to the DAC, since USB is digital and the DAC is the one converting the digital signal into analog audio. So I’m a bit confused.

Has anyone else experienced such differences in audio quality depending on USB port version? Could this be related to bandwidth, power delivery, or something else in the USB implementation?

Thanks in advance for any advice or explanations!

 

[nerdstrike]

You can stream 4k content over properly implemented USB2 with no limit on the data. You've got bandwidth up the wazoo for merely streaming music. I would assume the external DAC was installed as a new device when you plugged into another port, and whatever happened there resulted in a change in volume mixing or something of that sort.

It's not immediately obvious but (on Windows at least) if you connect a device to each USB port in turn, it will be registered as a new device each time. Device manager hides them away, but they're there in the registry, the same hardware four times or whatever. That triggers new driver setup and would likely give you a clean slate of settings.

We might suppose that there's a difference in the Topping E30 as to how it represents itself through its USB controllers, since I assume your USB 3 is via type-C cable and not the extra-tall USB-B connector... There are a lot of layers to what was once termed "plug and pray".

 

[StormBreaker]

Whoa, USB 3.0 actually boosting audio feels like unlocking hidden FPS—never thought port choice could affect the sound grind!

 

[obsessive9119]

It has been a long time since I used windows as a source, but have you made sure you’re getting bit perfect audio out of Windows? It’s possible (tho unlikely I’d think) that Windows is downsampling the audio going via one usb port for some reason.

 

[Vincent Kars]

USB audio is done using UAC2 (USB Audio Cass 2).
It set the bus to isochronous mode and uses High Speed (480 Mb/s).
This applies to both USB2 and USB3 as both support High Speed.
Switching between a USB2 or 3 port won't make a difference for the data received.
Can't explain the difference.

Can you tell them reliable apart in a unsighted test?

 

[omega66]

I haven’t set up bit‑perfect audio on any of my devices - the PC, the HTPC, or the laptop - and yet it still sounds different.

 

[kazuviking]

never thought port choice could affect the sound grind!

Spoiler it DOESNT.
My hunch is windows applying sound processing on the usb2.0 port while not on the 3.0.

 

[omega66]

USB audio is done using UAC2 (USB Audio Cass 2).
It set the bus to isochronous mode and uses High Speed (480 Mb/s).
This applies to both USB2 and USB3 as both support High Speed.
Switching between a USB2 or 3 port won't make a difference for the data received.
Can't explain the difference.

Can you tell them reliable apart in a unsighted test?

Between the USB 2.0 and USB 3.0 ports on my HTPC the difference is very easy to notice.

 

[omega66]

Does the way a USB port is implemented matter? Could one port be ‘cleaner’ than another? Is that even a thing when we’re talking about a digital interface like USB?

 

[respice finem]

I assume the 3.0 is simply louder. Maybe 2.0 is "native" (MB) and 3.0 via third party chip?
It shouldn't alter the volume, since it's digital signal and not analog, but who knows...
Also possible IMHO: As it's an industrial PC, it might have non-standard options, such as per-port volume settings.
If all else fails, simply use 3.0, problem solved.

 

[omega66]

I don’t think it’s necessarily louder - on the contrary, the impression of being louder comes from 2.0 because the vocals feel somewhat “crammed” together with the instruments. In the end I’ll stick with the 3.0 port, but now I’m already wondering if there’s a difference between the 3.0 port on the HTPC and the one on the laptop, even though in a blind test I couldn’t tell them apart. Either way, it remains a mystery to me - I would have never imagined such a scenario.

 

[boswell]

windows can do audio lots of ways and it is hard to find which it is using, try using EAPO (and PEACE) it can be configured to go to a specific dac via usb (also has eq/dsp) see below for some background and summary at end of original post is ok

Ending the Windows Audio Quality Debate

A debate has been long raging in the audio community that Windows isn't fit for use in a high resolution audio system. Sceptics counterargued that bits are bits and Windows audio degradations are a figment of an over-active audiophile imagination. I decided to investigate this as I kept...

www.audiosciencereview.com

 

[respice finem]

What might/should help - using ASIO if the driver can do it.

USB in mainboards is sometimes strange - I had a board which "saw" a DAC only on one port (of eight IIRC).

 

[omega66]

windows can do audio lots of ways and it is hard to find which it is using, try using EAPO (and PEACE) it can be configured to go to a specific dac via usb (also has eq/dsp) see below for some background and summary at end of original post is ok

Ending the Windows Audio Quality Debate

A debate has been long raging in the audio community that Windows isn't fit for use in a high resolution audio system. Sceptics counterargued that bits are bits and Windows audio degradations are a figment of an over-active audiophile imagination. I decided to investigate this as I kept...

www.audiosciencereview.com

I’ll read the material, thank you.

 

[omega66]

What might/should help - using ASIO if the driver can do it.

USB in mainboards is sometimes strange - I had a board which "saw" a DAC only on one port (of eight IIRC).

I’ve installed the ASIO driver in the meantime, but I only use it in foobar. In this particular case with ASIO, I assume there shouldn’t be any difference anymore? Should I risk reconnecting it to USB 2.0 and see if in foobar + ASIO it still sounds bad? Or maybe I should just leave it as it is, since I already got it working (someone mentioned earlier that when you change the USB port it basically sees it as a new device)? I really wouldn’t want to reinstall the drivers again...

 

[Mike_Trueman]

Did you check the windows bitrate and sample rate on the device connected?

 

[omega66]

Unfortunately, I didn’t check on the 2.0 port. On the 3.0 port I’m connected to now, the default was 24‑bit and 48 kHz. I switched it to 16‑bit and 44.1 kHz, but I didn’t notice any difference here.

 

[Koeitje]

Whoa, USB 3.0 actually boosting audio feels like unlocking hidden FPS—never thought port choice could affect the sound grind!

I just realised that higher sampling rates are like high fps audio.

Just without the advantage that a high fps has on a screen

 

[omega66]

windows can do audio lots of ways and it is hard to find which it is using, try using EAPO (and PEACE) it can be configured to go to a specific dac via usb (also has eq/dsp) see below for some background and summary at end of original post is ok

Ending the Windows Audio Quality Debate

A debate has been long raging in the audio community that Windows isn't fit for use in a high resolution audio system. Sceptics counterargued that bits are bits and Windows audio degradations are a figment of an over-active audiophile imagination. I decided to investigate this as I kept...

www.audiosciencereview.com

I read the article from the link along with the opinions of the other users who replied to me, and I understand that the difference could only come from the software side (Windows) rather than the hardware (USB 2.0 vs 3.0 port). So is it correct to assume that in ASIO mode there should be no difference at all between the 2.0 and 3.0 ports? At first, I believed the HTPC had a hardware issue.

 

[pablolie]

USB audio is done using UAC2 (USB Audio Cass 2).
It set the bus to isochronous mode and uses High Speed (480 Mb/s).
This applies to both USB2 and USB3 as both support High Speed.
Switching between a USB2 or 3 port won't make a difference for the data received.
Can't explain the difference.

Can you tell them reliable apart in a unsighted test?

indeed, the bits are identical. As you said, could be a difference between UAC1 and UAC2 (timing?)... but I'd be very surprised.

The most likely cause this would be some analog noise issue. And it's USB3.0 that could be noisier (high speed devices sharing?)... but that should be eminently measurable with the right gear, which I'd urge the OP to do. Audacity with a test track and something like a MOTU M2 are likely allies here. I'd be curious about the results.