Possible remedy to instability with USB interfaces - a dedicated USB card

USB headsets to digital audio workstation software...
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Possible remedy to instability with USB interfaces - a dedicated USB card

Postby kc2rgw » Mon Jun 27, 2022 6:58 pm

It's pretty early, just a week or so in, so anecdotal rather than definitive...

I've always had odd issues with USB interfaces, a Focusrite 8i6 and now my Motu M2 where when you start up with Voicemeeter and DAW or when you stop and start various SDR software that fairly often you get grungy jittery audio where the interface bitrate or sampling rate hasn't negotiated properly.

I was wondering if all the junk I have on the USB busses on my PC, multiple radios, SDR receivers, mouse and keyboard and on and on were fighting for bandwidth and possibly disrupting things.

I put this generic card from Amazon in, a PCIe USB 3.0 card to try and run exclusively for my audio interface.

https://www.amazon.com/dp/B09KLVRBQ2

It was plug and play with Win 10 Pro.

So far, again, a bit anecdotal with all the variables with a DAW and Voicemeeter and multiple SDR platforms, but it's all been rock stable for me since doing this.

I've been running 48kHz and 256 buffers to give room for stability. I'll have to try pushing the buffers smaller now and see if it holds together or not.
User avatar
w-u-2-o
Posts: 5526
Joined: Fri Mar 10, 2017 1:47 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby w-u-2-o » Mon Jun 27, 2022 7:02 pm

Interesting!
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby Tony EI7BMB » Mon Jun 27, 2022 9:19 pm

I think I want one :D Edit: ordered from Amazon, will update with any findings.
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby kc2rgw » Tue Jun 28, 2022 11:28 am

The card works in linux as well just plug and play with Ubuntu 22.04.
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby rbduck » Tue Jun 28, 2022 4:46 pm

Are you using one of the USB 3.0 ports or one of the USB-C ports. I have tried several interfaces. On none of them can I get below 70ms on the RingBuffer for stable receive. To transmit clear audio I have to put the RingBuffer in auto mode. If I transmit in any mode with faster timing everyone says I sound like I have RF in the shack. Of course I know I'm sure I don't. I have I have tried all the methods on this forum to try and obtain better results. This includes the different firmware offerings. I never thought about trying a dedicated USB card. Or, I may have reached the limits of inherent stability of the computer.
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby kc2rgw » Wed Jun 29, 2022 12:46 am

Just using one of the type A ports as the interface doesn’t really support or need USB 3.0.

It probably won’t get you below 70ms. I have no need for that with my operating. Considering the transition of TX and RX getting that low is quite a task.

I run the auto mode for the ring buffers, far less trouble.

What this is doing for me is not having to bring up Voicemeeter and test the audio, then bring up Reaper and test the audio, then bring up Thetis or EESDR or fldigi and retest the audio.

Right now it all just starts and works consistently. Of course that’s just my system and there are a lot of variables.
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby Tony EI7BMB » Thu Jun 30, 2022 3:49 pm

Just installed mine using USB C to USB C cable from MOTU M4 to new PC card and working fine so far. I am running 64 buffer size and manual buffer latencies of 10ms.
User avatar
rbduck
Posts: 327
Joined: Tue Dec 03, 2019 1:49 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby rbduck » Thu Jun 30, 2022 4:47 pm

Tony EI7BMB wrote:Just installed mine using USB C to USB C cable from MOTU M4 to new PC card and working fine so far. I am running 64 buffer size and manual buffer latencies of 10ms.


Tony,
I know you will update us farther down the line in time. Just for a comparison were you able to run the 64 buffer size and 10ms latency before the new card? Thanks
73
Ruben
NB4R
Apache-Labs Anan 7000DLE MKII Black -- Thetis 2.10.3.6 dev_2 -- Windows 11
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby Tony EI7BMB » Thu Jun 30, 2022 5:01 pm

I was able to do those settings beforehand Ruben,with the occasional run away maybe every few days
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby kc2rgw » Sat Jul 02, 2022 1:00 pm

The results so far are that things are definitely more stable than they were. I've only had one glitch on startup and it was on Windows with Thetis.

I've not tried pushing the buffers smaller yet, trying to observe with the known configuration first. Still testing at 48kHz and 256 buffers on the Motu M2.

I did re-test linHPSDR and the audio buffering issues are still there with the clicking/snapping noise, so no joy there.

It's been rock solid on linux with my Sun SDR2 DX.as well.
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby kc2rgw » Sat Jul 02, 2022 2:42 pm

So this morning on Windows I set up for 48kHz with 64 buffer. It glitched a few times with getting Reaper to play nice.

Then Thetis did its thing where I had to toggle the sampling rates around a few times, toggle VAC on and off, before it would "sync" and play ball.

It's solid as I'm running it, but it's that pain with the startup and getting everything to "agree" to get moving.
K4IBC
Posts: 64
Joined: Sun Apr 09, 2017 7:31 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby K4IBC » Sat Jul 02, 2022 3:10 pm

I am wondering if you have tried different USB cables? Even though the Motu has a USB C port, it is a USB 2.0 protocol.

On some of the Audio forums I saw a recommendation to use a Samsung USB C to A cable. I don't know if maybe a high quality cable from like USBGear or Showmecables might help.

Also clicking and popping with the Motu has been associated to drivers and firmware. Upgrade them to latest versions. I know drivers don't apply to Linux but firmware might help.
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby Tony EI7BMB » Mon Jul 04, 2022 9:35 pm

Very stable here. I've not had to reset since i installed the card.
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Re: Possible remedy to instability with USB interfaces - a dedicated USB card

Postby kc2rgw » Tue Jul 05, 2022 11:48 am

K4IBC wrote:I am wondering if you have tried different USB cables? Even though the Motu has a USB C port, it is a USB 2.0 protocol.

On some of the Audio forums I saw a recommendation to use a Samsung USB C to A cable. I don't know if maybe a high quality cable from like USBGear or Showmecables might help.

Also clicking and popping with the Motu has been associated to drivers and firmware. Upgrade them to latest versions. I know drivers don't apply to Linux but firmware might help.


Yeah I've tried a few cables and yes the drivers and firmware are up to date.

In linux it works fine with Reaper, Jack, fldigi, EESDR3, Audacity etc the snaps and pops are specific to linHPSDR unfortunately.

The update is that it has added some stability but it hasn't been a complete magic bullet. At 48kHz and 256 in Windows it's solid, when I pushed to smaller buffer sizes it still has issues on initial start up of Thetis to "settle down" and negotiate the bit rate. Once toggled a bunch, when it clears up, it keeps going but the small buffers still introduce some instability. I don't really notice much of a practical difference in the smaller buffer sizes but it was just an experiment.

Return to “Digital ("Virtual") Audio”