
Just another reason NOT to upgrade to Win11 then...
Except this update breaks things in Win10 too...
Well.... Linux it is then...
The list of known issues in the Windows January 14 update continues to grow with USB audio device users the latest to be hit. 2025 is not going well for Microsoft and its regular Windows update. Shortly after the release of the security update, KB5050009, the company admitted to issues with Citrix Session Recording software. …
I agree. I have changed one of my Win10 machines that is not capable of upgrading to Win11 (Lynnfield/Nehalem-gen CPU) to Ubuntu 2404 and I use an external DAC. Even installed the optional Ubuntu Audio Studio package.
Not able to figure out whether it is outputting 16bit, 24bit, or 32 bit and 44.1, 48, 88.2 or 96kHz. Installed all the Pipewire stuff, then Pavucontrol etc. Figured out the frequency output but not the bit rate. Searching the web is a mess of instructions for ALSA, PulseAudio and PireWire. Whereas in Windows this is dead easy as it offers the different bit depths and sample rates in a drop down list.
Sure I probably need to go through some conf file but audio is at least easier in Windows with a GUI.
simpler solution :)
install macrium reflect, my goto for backup software
attach an external or internal drive
quick restore to previous state after ms shenanigans, surprisingly quick and easy procedure
easy to create a macrium iso usb in case of major microsoft shenanigans :)
ps the fully free version can still be downloaded legally and allows the iso creation too
Well.... Linux it is then...
No scheisse, Sherlock. So, looks like this weekend, I:
* update to Linux Mint 22.1
* update to Wine 10
* download Arturia's Software Center, which in turn can download V Collection 10, and see if any of the soft-synths work.
Yeah, I've threatened to do this before, but round-tuit, and so on. However, the fact that Micros~1 has figured out how to fuck-up external sound cards adds a new sense of urgency for an organized but rapid escape!
Another reason not to use USB 1.0?
Hopefully, in this day and age, USB sound devices use USB 2.0? (While many devices still use PCIe v1 or v2, when newer specs exist: "if it ain't broke, don't fix it" ? - My answer: USB 1.0 was born "broke".)
Of course, MS might be spinning things: if the issue actually does affect USB 2.0 devices, and someone decided to "shorten" (for "clarity's sake", yeah, that's the ticket) the phrase "USB 1.0 and USB 2.0" to just "USB 1.0" (PR 101: minimize impact), then, there's more to this story than meets the eye.
The USB 1.0 audio spec is well defined, manages up to 24-but depth and 96KHz sampling, and important in Win 10, did not require an audio driver, so was plug and play. It is also the spec that works out of the box with consoles (Nintendo Switch etc).
For some reason, Win 10 did or used to require manufacturer-specific audio drivers for the USB 2.0 audio, so was at the mercy of what was supplied. There is a generic Microsoft supplied USB 2.0 driver that does higher (24-bit, 192 KHz) but works with Win11 upwards (from memory, I used an Astell & Kern DAC with another Win ARM machine that could not use an X64 driver from the A&K website).
I am perfectly happy with 24/96 audio and the DAC is set to this, hence my media server is based in Win 10 but the machine will not boot as of this weekend.
So hasta la Vista to MS and time to switch a bit earlier to Linux than I had planned.
Hopefully, in this day and age, USB sound devices use USB 2.0? (While many devices still use PCIe v1 or v2, when newer specs exist: "if it ain't broke, don't fix it" ? - My answer: USB 1.0 was born "broke".)
I've got a very long blog post(/rant) on this topic that I'm still trying to reduce to something actually readable by the sort of half-tech-literate audience that might benefit from it. Some of the takeaways:
1. Many USB audio devices are UAC class 1, which dates to roughly the original USB (or USB 1.1, I forget right now) spec.
2. USB 1,2,3... are not speeds. USB LS, FS, HS, SS (low, full, high, super) are. USB number is not actually that useful in knowing the capabilities, leading to:
3. If a device is USB-C it is by definition USB 2 or higher, USB 2 was revised to include the USB-C connector. A USB-C device can never be USB 1 even if...
4. ... 2+3= a USB-C device may still only be capable of USB FS communication. UAC-1 uses FS.
5. USB FS is plenty of bandwidth for reasonable audio, particularly uni-directional, as in the comment above.
6. But USB FS communications when on a controller managing high speed devices result in scheduling that drops the available bandwidth for isochronous streams quite a bit below even the FS bandwidth.
7. Pretty much every USB-C headphone or headphone adapter is: a. USB-UAC1, and b. Pushing ridiculous sample rates and depths for playback, I've met some that are also trying to do 24bit 48kHz stereo input on a lapel mic. Put this all together and, if it doesn't fall over by itself, it quickly does if it has to share with, for example, a webcam or a GSM modem even when the nominal available bandwidth at even USB HS should easily be enough.
The model in USB-1 era for this kind of thing was multiple ports with their own controllers if you wanted to run the high bandwidth profiles in UAC1, but since USB2 the tendency is lots of ports on a single controller. At the same time cute little dongles for USB-C now support the kind of profiles only pro hardware used to, plug it into a phone and there's no way to choose (in linux you can if you're willing to dig about in pipewire configs, but audio menus wont let you, maybe windows can do similar). USB3 and newer USB2 devices at least have xHCI controllers, which can schedule a bit more robustly, worst of all worlds is if you happen to have USB 2 era EHCI but without a dedicated OHCI/UCHI controller for the USB-1 mode.
Okay, you can now see why that blog post is unmanageable.
USB has always been marketed as "simple" for the user. Having grown up with computers before USB was invented (or even the "PC"), I've always known USB was not simple "under the bonnet" and has many traps for the unwary[*]. You just revealed to me that's it's even more of a minefield then a ever dreamt! Thanks for that :-)
* I've seen users with cascaded USB hubs trying to "get stuff working" and having varying degrees of success in the early days, when 2 USB ports was all you got on the PC. Best one was probably the person with a keyboard with a USB port on it for a mouse (actually an internal hub, of course) and had a 4 port hub plugged into that and an external USB HDD on the end and wondered why it was so slow :-) (Well, it was the days of 1.0/1.1 and fecking slow for an external HDD anyway!)
According to Microsoft, "You are more likely to experience this issue if you are using a USB 1.0 audio driver based DAC in your audio setup."
Notice how Microsoft doesn't take ownership of the problem they caused. This is all "YOUR" problem because "YOU" are using particular driver.
It would have been so easy to write something like "Systems using a USB 1.0 audio based DAC for their audio setup...." No, instead Microsoft decided to personalize the problem and directly call out the individual users as being the problem. "YOU" decided to run this USB DAC. "YOU" are using a 1.0 audio driver. This is all in "YOUR" audio setup.
Really speaks to their mindset about how they hold their users in such high regard.
> Don't use an audio DAC in the connection process.
Unclear here. ALL "audio" beyond BIOS boops goes through some form of DAC. On $399 PCs the DAC is a sliver on the system chip; on a $3,999 audiophile workstation it may be a hand-honed boutique chip on an add-on (or it may be a $2 chip in a $999 case).
Is the problem USB? (Shocked!) Mobo bus connect? PCI? Bluetooth?