Occasional Audio Glitches as USB Interface

Would you mind to test if you can route an instrument via the QC into a DAW and send it back via the Neural DSP ASIO driver at e.g. 64 samples buffer size without problems ?

I’m asking because on a optimized Windows 10 computer the Neural DSP ASIO driver is not reliable today. The test if your computer is working fine is ASIO4ALL - if there is no problem with ASIO4ALL the problem is not the computer configuration - if this is not correct I would like to know. But because it is working for you I’m asking about e.g. your routing.

Thank you all for the thread here, I’m glad I found it. I also just reported the Bitcrusher audio glitching to support today. I thought I was going crazy a bit until I read through all this.

I get about 2-4 minutes on average until the audio glitches, and then it stays glitchy until I toggle some audio settings in my DAW, which forces it to reset. Then it’s clean again for a few minutes until it glitches out.

It took me a while to realize where the issue was because I had been monitoring directly from the QC instead of the DAW, which always sounded fine. So I wouldn’t hear the glitches until after I had recorded a clip, which made me think the problem was the DAW.

My setup is less common though, I’m using Ubuntu Linux 20.04 as my OS, and Bitwig Studio 4.0.1 as my DAW. I kinda don’t expect to get a ton of support given my setup, but it’s at least another data point showing that this is a real issue across multiple environments.

I’m using Windows 10 and Xubuntu Linux 20.04 with kxstudio, too.
Because the QC should work as a class compliant USB 2.0 audio device
it should work with Linux out of the box, if the driver would work.
I’ve mentioned this a few times in support communication but so far no reaction.

At least in this linked article they are saying it’s class compliant:

I had the problems on my Ubuntu 20.04 install, which I use for work. Today I tried again and the issues are gone. I guess some update in the Ubuntu packages fixed the issue for me. Nothing else changed. I am happy now, but unfortunately can not give any tips as to what exactly was the culprit.

Would you mind to write about e.g.: how long tested, software used, audio routings (multichannel), audio settings?

1 Like

I have the same problem on my macbook and my laptop. I thought the ASIO4All driver would fix the problem on the PC, but unfortunately after a while the USB connections to and from the Quad Cortex are cut off. I spent hours / days trying to find a solution as I really need these USB connections to access GigPerformer. I finally modified all my QC patches and I use an external audio interface (an EVO4 in my case) and I connected the I / O to send and return the QC.
I am looking forward to a patch for this version 1.1 of CorOS…Please go ahead!

If you’ve managed the replicate the issue on two separate devices with the same QC, I think you should contact their support and tell them, if you haven’t already.

They got back to me about my case, describing how they are trying to find specific cases they can better troubleshoot. They mentioned specifically trying to find customers who were able to replicate the issue on two separate devices using the same QC, so your case might be particularly helpful if they can get more details.

You’re right. I just sent them a detailed message and I’m waiting for there reply.

For what it’s worth to anyone following along, I also managed to reproduce the issue with the same QC in two separate environments last night, so sent all the details to Neural.

First environment: Ubuntu Linux 20.04, using Bitwig Studio 4.0.1. ALSA driver included in the kernel.

Second environment: Windows 10, latest updates, using a fresh install of Reaper 6.34, and fresh install of the Neural ASIO driver.

In the Win10 environment I tried a variation of sample sizes, with Safe Mode on and off - but in all cases the audio would eventually get corrupted. I think that’s about as thorough as I can be to rule out an environmental issue, so hoping that might help narrow things down.

Nice. I couldn’t really reproduce it other than run more CPU-heavy plugins which triggers it. I think the fact that it resets after using the safe mode on/off should give a clue though as to where the problem lies.

On another note for those that have suggested other solutions:

  1. ASIOFORALL is a generic sub-standard windows driver. While it may be useable this shouldn’t be a substitute for the main driver you’re supposed to be using.
  2. Amending to a higher buffer size - Raising the buffer size to suit the interface is like acknowledging the interface can’t handle lower buffer sizes with lower latency. It should be able to do 128 just fine and even lower.
  3. USB - The QC has many functions but I think it’s USB and recording functionality is a key reason as to why many people buy it. It’s kind of like buying it and you can’t load up presets or something similar. It should be usable as a standard interface and should be top notch.

Im using my other interface now with the QC going into it. Works fine but not exactly prize 1. Hopefully this gets fixed soon.

I’m having a hard time with the same problem. All I could do was shoot USB statistics and send it to the support team when there was a problem. I hope this bug is resolved as soon as possible. I really hope.

I had the same issue. I was messing around over the weekend, and flipped the MIDI over USB button off and back on in settings. It seems to be working without glitching.

2 Likes

Thank you :slight_smile: i solved the issue in your way

Sorry for the late feedback. I experimented a lot with jack, and different polling settings, and master/slave settings (most of which I have no clue of, just tried out some stuff I found on google referring to similar issues with USB interfaces). In the end I uninstalled Jack again. The glitches were gone without further changes after regular os-updates.

I am using standard ubuntu 20.04 with pulseaudio. The QC is recognized as “Analog Output - Quad Cortex” with no special config what so ever. On the Input-side its “Multichannel Input”, which I am not using yet. I will send an update if I find any issues with that.

I’ve a quite similar setup and I’ve no problems with a ZOOM UAC2 USB2 audio interface thus I assume the QC should work here without uninstalling jack.

Are you routing audio from the QC to e.g. a DAW for processing and back from the DAW to the QC for processing and output with a sample buffer size of 64 at 48KHz, or what’s your application?

I have been trying to hook up a usb ‘in’ from my MacBook pro - which is running fishman tripleplay synth pads… it worked for a while, then it started doing the bit crusher routine - a digital nasty glitchy distorted noise. Does it almost continually and I cant fix it. Anyone got the answer?

my conclusion after having read this thread and after heavy testing

  • switch on and of the ASIO Save Mode (driver control panel)
  • switch on and off the USB MIDI functionality (QC)
  • if you want to have a reliable connection from the start, performance ready - use analog audio
  • report your issue and wait for an update

Not that I think this is cool. If someone can provide a real solution today I’m all ears.

Got an email from the support:

"Hello everyone,

Gonzalo here from Neural DSP.

We’re working as hard as we can to address the USB audio glitches that a very small percentage of users are experiencing. Please stay tuned for updates and of course we’ll let you know as soon as we have news. We’ve found a couple of issues that could be related to the USB glitches, which can hopefully help us to resolve these problems, but it is still too early to confirm anything. We will let you know how this evolves.

If you have any questions or additional information, please don’t hesitate to contact us.

Thank you for your support and for bearing with us."

Hey. Has anyone noticed any changes to this issue since the release of Coros 1.2?

Bumping this up. Has anyone tested recording / using the QC as an interface since the release of 1.2? Just wondering without changing my whole setup whether any changes have been made to the bitcrushing issue?