[USB-Audio - RODECaster Pro, playback] Sound is distorted (20.04)

[Oliver Tacke]

I got a Rodecaster Pro USB device that has been causing trouble before and already lead to quirk handling in the kernel for the sample rate, see https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/sound/usb/format.c?h=v5.8#n412

The Rodecaster is running fine on my laptop, but on my desktop, only audio input is working. The audio output is distorted. It's played kind of very slowly. Interestingly, if the source is a video file, the video plays slower, e.g. YouTube in a browser. I have never seen anything like that before.

I have tried to figure out what the cause might be and I think I have reached the end of where I can get to on my own. I have documented most things at https://ubuntuforums.org/showthread.php?t=2461568

Putting my last step it in a nutshell: I have booted both my computers with a plain Ubuntu 21.04 from a pen drive (it uses kernel version 5.11.0-16-generic) to make sure it's not some different configuration that I have set up over the years on my systems (both Ubuntu 20.04 with kernel 5.8.18). I attached the Rodecaster. It runs fine on my laptop, but not on my desktop (used the output test from).

I then detected the card number with

aplay -l

I killed pulseaudio with

pulseaudio -k

I then played the same sample (encoded in 48000 Hz in PCM signed 32 bit little endian format) using

aplay -f S32_LE -c 2 -r 48000 -D hw:x ~/sample.wav

where x is the appropriate card number and the rest of the arguments meet the Rodecaster's narrow specification. It's playing fine on the laptop, but distorted on the desktop. As far as I understand using aplay this way without a plugin should send the file directly to the device, so I assume there's something on the way from aplay to the Rodecaster that's interfering (kernel or even some piece of hardware) - and that definitely is where my expertise ends (rather a little before :-)).

I'll gladly contribute any information that could be useful and I will willingly help to debug this problem if it may be related to the kernel. If there's anything I can do, please let me know.

Cheers,
Oliver

ProblemType: Bug
DistroRelease: Ubuntu 20.04
Package: alsa-base 1.0.25+dfsg-0ubuntu5
Uname: Linux 5.8.18-050818-generic x86_64
NonfreeKernelModules: blackmagic_io nvidia_modeset nvidia
ApportVersion: 2.20.11-0ubuntu27.16
Architecture: amd64
CasperMD5CheckResult: skip
CurrentDesktop: ubuntu:GNOME
CurrentDmesg: Error: command ['dmesg'] failed with exit code 1: dmesg: read kernel buffer failed: Operation not permitted
Date: Wed May 5 18:28:46 2021
InstallationDate: Installed on 2020-05-15 (355 days ago)
InstallationMedia: Ubuntu 20.04 LTS "Focal Fossa" - Release amd64 (20200423)
PackageArchitecture: all
SourcePackage: alsa-driver
Symptom: audio
Symptom_AlsaPlaybackTest: ALSA playback test through plughw:Pro failed
Symptom_Card: RODECaster Pro - RODECaster Pro
Symptom_Type: Digital clip or distortion, or "overdriven" sound
Title: [USB-Audio - RODECaster Pro, playback] Sound is distorted
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 10/30/2020
dmi.bios.release: 5.17
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: A.40
dmi.board.asset.tag: To be filled by O.E.M.
dmi.board.name: B550-A PRO (MS-7C56)
dmi.board.vendor: Micro-Star International Co., Ltd.
dmi.board.version: 1.0
dmi.chassis.asset.tag: To be filled by O.E.M.
dmi.chassis.type: 3
dmi.chassis.vendor: Micro-Star International Co., Ltd.
dmi.chassis.version: 1.0
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrA.40:bd10/30/2020:br5.17:svnMicro-StarInternationalCo.,Ltd.:pnMS-7C56:pvr1.0:rvnMicro-StarInternationalCo.,Ltd.:rnB550-APRO(MS-7C56):rvr1.0:cvnMicro-StarInternationalCo.,Ltd.:ct3:cvr1.0:
dmi.product.family: To be filled by O.E.M.
dmi.product.name: MS-7C56
dmi.product.sku: To be filled by O.E.M.
dmi.product.version: 1.0
dmi.sys.vendor: Micro-Star International Co., Ltd.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Alexander List (alexlist)]

I can confirm this problem on Ubuntu 21.04 (hirsute) running on a Dell XPS 13 7390 2-in-1

[Brett Callaughan (bcallaughan)]

I alos have the same issue, I have found that input via the Roadcaster Pro is fine, so I use the systemboard sound card with headphones for the output.

[Oliver Tacke (otacke)]

@Brett Not an option for my use case, unfortunately, but good to know that more and more report the issue here.

[Peabody (peabodygerald)]

Same problem on Ubuntu 21

[Oliver Tacke (otacke)]

@Peabody Yes, I have ran into the same problems on 21.04 and reported that separately at https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/1927337

[Severi Härkönen (sharkodev)]

Same prob on Ubuntu 20.04.3

[Oliver Tacke (otacke)]

@all who commented: I am not sure if there's some threshold that needs to be met before this issue gets more attention by someone who could look into it, but I assume it helps if you confirm that this bug affects you at the top of this page, so the "hotness" score rises.

[Oliver Tacke (otacke)]

Rumor has it that the my USB ports powered by an AMD chipset might be the culprit. I have no clue whatsoever how this could be fixed in software, but I am now waiting for a PCIe USB card with a different chipset to check if there's a (preliminary) fix in hardware at least.

[Oliver Tacke (otacke)]

I have now built in a separate PCIe board with USB ports that do not rely on AMD chipsets. The RODECaster Pro now works perfectly. It' sad that I lost one port, but at least there's a new trace to where the cause might be. Hopefully, this can be fixed in software, too.

[Heidi Harding (katzekat)]

Similar to #4 I am using the Rodecaster pro for the XLR mic input as this works fine. I connected a 3.5 mm TRS male to male cable from my Dell Latitude's output (which outputs audio just fine) and connected it to the mobile phone input on the rodecaster pro. That way I can still use the headphone output from the rodecaster. It works pretty well and it means I don't have to keep swapping headphones around just for the one machine that has this issue.

[Cijay (cijay9)]

Not sure about this rumor about AMD chipset being the culprit.
Two of my computers (running Ubuntu 20.04 with the 5.11.0-40-generic kernel) have the problem of distorted sound when I use the Rodecaster Pro as audio output.
The first is a desktop using an AMD Rysen 9 and the other is a Dell XPS laptop using a 10ᵉ generation Intel Core i7.

@otacke, when you use a regular USB port of your motherboard or when you use a USB port of the PCI card, have you noticed any difference in the Pulseaudio or Alsa configurations?

[Oliver Tacke (otacke)]

@cijay The AMD chipset as the culprit is the best hypothesis so far. I have not checked for differences in the configuration, but only been enjoying the Rødecaster on a non-AMD chipset port.

[Oliver Tacke (otacke)]

@cijay Amendmend, since you're mentioning your computer's CPUs - that's not what I was referring to when I mentioned an AMD chipset, but the USB controller chipset that's also made by AMD.

[Cijay (cijay9)]

My bad, I thought that the brand of the main USB controller was directly correlated to the motherboard chipset. (In my case, the Intel PC includes an Intel USB controller and the AMD an AMD USB controller).

In any case, if you have the possibility to post a log of your different configurations, it might help us (dmesg, pacmd list-sinks output), thanks!

[Oliver Tacke (otacke)]

@Cijay

Sure, if you're interested in anything particular, just let me know. There's full bug report output attached to the original bug report ticket for my old setup.

Hope those two help:

dmesg | grep -e usb3 -e usb\ 3-1 > dmesg.txt
pacmd list-sinks output > pacmd.txt

[Jacob Pfeifer (jacobpfeifer)]

Hey guys, I was the one who started the amd chipset rumor. We've been talking about the issue over at https://forum.level1techs.com/t/rodecaster-pro-bad-audio-out-solved-ish/170601.

At first I thought it was an amd vs intel issue just due to the hardware I had on hand that worked/didn't work. After someone popped in mentioning their intel device having the issue and a pcie card they bought also having the issue it seems like the issue is with all usb3.2 gen2x2 host controllers, or at least that's my best guess. I picked up a different pcie usb card than the person had trouble with and that was 3.2 gen2x2 capable. I can reproduce the issue on both it, and my motherboard usb ports. However, using a usb3.0 pcie card the issue goes away.

One strange thing to me about all of this is that the rodecaster pro is negotiating usb2 480mbps connections, so i'm not sure why a later usb spec would mess with that. It's also worth noting that the issue seems to be related to the usb host controller itself as using ports capable of different speeds/going through usb hubs or directly connecting doesn't change the behavior. I also don't have a usb 3.1 gen2 host controller to test to see if the issue is truly only with usb 3.2 gen2x2 controllers.

Debugging any further than that is proving difficult though. I did a diff on both my lsusb and pw-dump output and the only changes in the two were paths/ids which are expected to change.

[Oliver Tacke (otacke)]

@Jacob Awesome research! Thanks!

[Alve (alve-larsson)]

Can confirm this bug is happening on, Manjaro, Debian(bullseye), Fedora
But works just fine on Windows using the exact same hardware setup(And ports)

On a AMD 5950x, MSI MAG X570S TOMAHAWK
Got a sound sample recorded the output directly on the road caster:

Hope it helps!