Behringer X32 - glitch problems

classic Classic list List threaded Threaded
16 messages Options
Reply | Threaded
Open this post in threaded view
|

Behringer X32 - glitch problems

Niklas Reppel
Hi,

I've been playing around with an X32 recently, and it starts glitching really fast.

Like, if i start JACK on my internal sound card (2ch) and SuperCollider with 8 to 24 channels, routing all channels to the 2 outputs, everything's fine.

But if I use JACK on the X32 with 8 channels and SuperCollider with 8 channels, it starts glitching as soon as the load increases just a little,
using the same programs in the background (ScSynth, SBCL (Common Lisp) using Incudine for OSC scheduling).

I've found some settings that work ok-ish, but I still have to be careful not to play too many things at the same time. Using the internal sound card,
there's much more leeway.

I'm working on Arch Linux, using the latest RT kernel with all exploit-fixing options disabled (https://make-linux-fast-again.com/). No, I'm not using these
options for my day-to-day work :D .

Has anybody encountered this before ? And maybe knows some way to mitigate this ?

Best,
Niklas
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
Hmm I didn't see the display, and just had my laptop connected via USB, so
I only noticed audio dropouts (they don't correlate to xruns, btw).

Could firmware be an issue here ?


Mac wrote on 09.10.2019 20:58 (GMT +02:00):

>
>
> I experienced something similar last week. The display was flashing
> periodically.
>
>
> I rebooted (the X32) and haven't seen it happen again... 🤔
>
>
> Just a note: not the latest firmware, so not sure. And, it had 2 S16's on
> it.
>
>
> On Wed, Oct 9, 2019 at 2:49 PM <[hidden email]
> <mailto:[hidden email]> > wrote:
>
>> Hi,
>>
>> I've been playing around with an X32 recently, and it starts glitching
>> really fast.
>>
>> Like, if i start JACK on my internal sound card (2ch) and SuperCollider
>> with 8 to 24 channels, routing all channels to the 2 outputs,
>> everything's fine.
>>
>> But if I use JACK on the X32 with 8 channels and SuperCollider with 8
>> channels, it starts glitching as soon as the load increases just a
>> little,
>> using the same programs in the background (ScSynth, SBCL (Common Lisp)
>> using Incudine for OSC scheduling).
>>
>> I've found some settings that work ok-ish, but I still have to be careful
>> not to play too many things at the same time. Using the internal sound
>> card,
>> there's much more leeway.
>>
>> I'm working on Arch Linux, using the latest RT kernel with all
>> exploit-fixing options disabled
>> (https://make-linux-fast-again.com/). No, I'm not using
>> these
>> options for my day-to-day work :D .
>>
>> Has anybody encountered this before ? And maybe knows some way to
>> mitigate this ?
>>
>> Best,
>> Niklas
>> _______________________________________________
>> Linux-audio-user mailing list
>> [hidden email]
>> <mailto:[hidden email]>
>> https://lists.linuxaudio.org/listinfo/linux-audio-user
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Paul Davis
In reply to this post by Niklas Reppel


On Wed, Oct 9, 2019 at 12:49 PM <[hidden email]> wrote:
Hi,

I've been playing around with an X32 recently, and it starts glitching really fast.

Like, if i start JACK on my internal sound card (2ch) and SuperCollider with 8 to 24 channels, routing all channels to the 2 outputs, everything's fine.

But if I use JACK on the X32 with 8 channels and SuperCollider with 8 channels, it starts glitching as soon as the load increases just a little,
using the same programs in the background (ScSynth, SBCL (Common Lisp) using Incudine for OSC scheduling).

What period size and sample rate?
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user

_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
The samplerate was 44100, period size 1024. Going up to 2048 made it
worse.

Started with:

jackd -P98 -dalsa -dhw:2 -p1024 -n8 -r44100
scsynth -u 57110 -z 256 -i 0 -o 8 -m 262144



Paul Davis wrote on 09.10.2019 22:38 (GMT +02:00):

>
>
> On Wed, Oct 9, 2019 at 12:49 PM <[hidden email]
> <mailto:[hidden email]> > wrote:
>
>> Hi,
>>
>> I've been playing around with an X32 recently, and it starts glitching
>> really fast.
>>
>> Like, if i start JACK on my internal sound card (2ch) and SuperCollider
>> with 8 to 24 channels, routing all channels to the 2 outputs,
>> everything's fine.
>>
>> But if I use JACK on the X32 with 8 channels and SuperCollider with 8
>> channels, it starts glitching as soon as the load increases just a
>> little,
>> using the same programs in the background (ScSynth, SBCL (Common Lisp)
>> using Incudine for OSC scheduling).
>
>
> What period size and sample rate?
>
>
>> [hidden email]
>> <mailto:[hidden email]>
>> https://lists.linuxaudio.org/listinfo/linux-audio-user
>
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Paul Davis


On Wed, Oct 9, 2019 at 3:21 PM <[hidden email]> wrote:
The samplerate was 44100, period size 1024. Going up to 2048 made it
worse.

Started with:

jackd -P98 -dalsa -dhw:2 -p1024 -n8 -r44100
scsynth -u 57110 -z 256 -i 0 -o 8 -m 262144

try running at 48kHz with 3 periods and see if that helps.



_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Thomas Ebeling


On 10/9/19 11:24 PM, Paul Davis wrote:


On Wed, Oct 9, 2019 at 3:21 PM <[hidden email]> wrote:
The samplerate was 44100, period size 1024. Going up to 2048 made it
worse.

Started with:

jackd -P98 -dalsa -dhw:2 -p1024 -n8 -r44100
scsynth -u 57110 -z 256 -i 0 -o 8 -m 262144

try running at 48kHz with 3 periods and see if that helps.
My X32 Rack runs flawlessly on arch using 48kHz, 1024 and 2 periods. Though I had to sort out USB issues first by running lsusb -t and find the least crowded bus.

Cheers

Bollie


_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
Thanks for the hints,
I'll try some different samplerate settings the next time I have access to the device.



Thomas Ebeling wrote on 09.10.2019 23:53 (GMT +02:00):

>
>
> On 10/9/19 11:24 PM, Paul Davis wrote:
>
>>
>> On Wed, Oct 9, 2019 at 3:21 PM <[hidden email]
>> <mailto:[hidden email]> > wrote:
>>
>>> The samplerate was 44100, period size 1024. Going up to 2048 made it
>>> worse.
>>>
>>> Started with:
>>>
>>> jackd -P98 -dalsa -dhw:2 -p1024 -n8 -r44100
>>> scsynth -u 57110 -z 256 -i 0 -o 8 -m 262144
>>
>>
>> try running at 48kHz with 3 periods and see if that helps.
>>
>>
> My X32 Rack runs flawlessly on arch using 48kHz, 1024 and 2 periods.
> Though I had to sort out USB issues first by running lsusb -t and find the
> least crowded bus.
>
> Cheers
>
> Bollie
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
In reply to this post by Niklas Reppel
Hmm I have to check, I think i deactivated audit as well.

I haven't done any real profiling, but to be honest I don't think there's
a huge difference, no. From a purely intuitive assessment, it takes a little
more load to cause xruns.

Best,
Niklas


Ralf Mardorf wrote on 10.10.2019 00:02 (GMT +02:00):

> > https://make-linux-fast-again.com/
>
> Hi,
>
> for testing purpose you might want to disable audit by boot parameter
> 'audit=off', too. If you should use the kernel config of the AUR tarball
> ( https://aur.archlinux.org/packages/linux-rt/ ) , audit is
> enabled.
>
> https://lists.archlinux.org/pipermail/arch-general/2018-September/045580.html
> http://lists.jackaudio.org/pipermail/jack-devel-jackaudio.org/2019-July/002037.html
>
> Btw. does it make a noticeable difference for rt audio performance on
> your machine, if you disable those mitigations? Did you compare the
> performance with and without mitigations?
>
> Regards,
> Ralf
>
>
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Paul Davis
The other thing to check, as usual with USB devices, is ... every single USB port. Sometimes a device will work very badly on one port and completely fine on another. This is true even on Apple machines. Amazing, but true.

On Thu, Oct 10, 2019 at 1:21 AM <[hidden email]> wrote:
Hmm I have to check, I think i deactivated audit as well.

I haven't done any real profiling, but to be honest I don't think there's
a huge difference, no. From a purely intuitive assessment, it takes a little
more load to cause xruns.

Best,
Niklas


Ralf Mardorf wrote on 10.10.2019 00:02 (GMT +02:00):
> > https://make-linux-fast-again.com/
>
> Hi,
>
> for testing purpose you might want to disable audit by boot parameter
> 'audit=off', too. If you should use the kernel config of the AUR tarball
> ( https://aur.archlinux.org/packages/linux-rt/ ) , audit is
> enabled.
>
> https://lists.archlinux.org/pipermail/arch-general/2018-September/045580.html
> http://lists.jackaudio.org/pipermail/jack-devel-jackaudio.org/2019-July/002037.html
>
> Btw. does it make a noticeable difference for rt audio performance on
> your machine, if you disable those mitigations? Did you compare the
> performance with and without mitigations?
>
> Regards,
> Ralf
>
>
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user

_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
That's interesting, I never noticed that.

Thomas noted earlier that it's a good idea to look for the least busiest bus,
I wonder if that's related? On the other hand, I usually don't have any other
hardware connected except the interface ...

Best,
n

Paul Davis wrote on 10.10.2019 18:01 (GMT +02:00):

>
>
> The other thing to check, as usual with USB devices, is ... every single
> USB port. Sometimes a device will work very badly on one port and
> completely fine on another. This is true even on Apple machines. Amazing,
> but true.
>
>
> On Thu, Oct 10, 2019 at 1:21 AM <[hidden email]
> <mailto:[hidden email]> > wrote:
>
>> Hmm I have to check, I think i deactivated audit as well.
>>
>> I haven't done any real profiling, but to be honest I don't think there's
>> a huge difference, no. From a purely intuitive assessment, it takes a
>> little
>> more load to cause xruns.
>>
>> Best,
>> Niklas
>>
>>
>> Ralf Mardorf wrote on 10.10.2019 00:02 (GMT +02:00):
>> > > https://make-linux-fast-again.com/
>> >
>> > Hi,
>> >
>> > for testing purpose you might want to disable audit by boot parameter
>> > 'audit=off', too. If you should use the kernel config of the AUR
>> tarball
>> > ( https://aur.archlinux.org/packages/linux-rt/ ) , audit
>> is
>> > enabled.
>> >
>> >
>> https://lists.archlinux.org/pipermail/arch-general/2018-September/045580.html
>> >
>> http://lists.jackaudio.org/pipermail/jack-devel-jackaudio.org/2019-July/002037.html
>> >
>> > Btw. does it make a noticeable difference for rt audio performance on
>> > your machine, if you disable those mitigations? Did you compare the
>> > performance with and without mitigations?
>> >
>> > Regards,
>> > Ralf
>> >
>> >
>> >
>> _______________________________________________
>> Linux-audio-user mailing list
>> [hidden email]
>> <mailto:[hidden email]>
>> https://lists.linuxaudio.org/listinfo/linux-audio-user
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Max
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Max
On 11.10.19 10:13, [hidden email] wrote:
> Thomas noted earlier that it's a good idea to look for the least busiest bus,

Yes, that's the same advice basically.

> I wonder if that's related? On the other hand, I usually don't have any other
> hardware connected except the interface ...

You may think that, but in fact the trackpad, keyboard, fingerprint
reader etc. might internally be USB devices which are on the same bus.
You may see this with the lsusb command.

I had a look on the board schematic of my laptop and noticed that USB
actually has completely unused buses. In the hope the docking station
would expose those I got one only to be disappointed that the docking
station was just acting as another hub to one of the already accessible
buses :/ With some hardware hacking it might be possible to access the
other buses, I opened the laptop to have a look but in the end wasn't
adventurous enough to actually try it :)

m.

> Paul Davis wrote on 10.10.2019 18:01 (GMT +02:00):
>>
>>
>> The other thing to check, as usual with USB devices, is ... every single
>> USB port. Sometimes a device will work very badly on one port and
>> completely fine on another. This is true even on Apple machines. Amazing,
>> but true.
>>
>>
>> On Thu, Oct 10, 2019 at 1:21 AM <[hidden email]
>> <mailto:[hidden email]> > wrote:
>>
>>> Hmm I have to check, I think i deactivated audit as well.
>>>
>>> I haven't done any real profiling, but to be honest I don't think there's
>>> a huge difference, no. From a purely intuitive assessment, it takes a
>>> little
>>> more load to cause xruns.
>>>
>>> Best,
>>> Niklas
>>>
>>>
>>> Ralf Mardorf wrote on 10.10.2019 00:02 (GMT +02:00):
>>>>> https://make-linux-fast-again.com/
>>>>
>>>> Hi,
>>>>
>>>> for testing purpose you might want to disable audit by boot parameter
>>>> 'audit=off', too. If you should use the kernel config of the AUR
>>> tarball
>>>> ( https://aur.archlinux.org/packages/linux-rt/ ) , audit
>>> is
>>>> enabled.
>>>>
>>>>
>>> https://lists.archlinux.org/pipermail/arch-general/2018-September/045580.html
>>>>
>>> http://lists.jackaudio.org/pipermail/jack-devel-jackaudio.org/2019-July/002037.html
>>>>
>>>> Btw. does it make a noticeable difference for rt audio performance on
>>>> your machine, if you disable those mitigations? Did you compare the
>>>> performance with and without mitigations?
>>>>
>>>> Regards,
>>>> Ralf
>>>>
>>>>
>>>>
>>> _______________________________________________
>>> Linux-audio-user mailing list
>>> [hidden email]
>>> <mailto:[hidden email]>
>>> https://lists.linuxaudio.org/listinfo/linux-audio-user
>>
> _______________________________________________
> Linux-audio-user mailing list
> [hidden email]
> https://lists.linuxaudio.org/listinfo/linux-audio-user
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
In reply to this post by Thomas Ebeling
Hmm, this is interesting, i just checked the USB ports in my Thinkpad T440s with lsusb -t.

Turns out, it has 3 buses, one USB3.0 (i suppose) and two 2.0 buses.

There's nothing on the 3.0 bus, but if i connect my interface, it still won't connect
to that bus because it's a 2.0 interface (testing with my UMC1820 at home, but I assume the X32 will be the same).

Now, one hub seems to be there for the (rather useless) smart card reader exclusively, while
ALL external (physical) ports are wired to the one bus that also has the memory card reader and other things
attached.

That means, not matter where i connect the interface, there's no way to connect to the least busy bus.

I wonder if there's a way to expose the unused buses ? Like, disconnect the smart card reader and attach
an USB hub? Or, if i connect an external USB 3.0 hub, will it wire external devices to the 3.0 internal bus?

Anybody got experience here ?

Best,
Niklas

Thomas Ebeling wrote on 09.10.2019 23:53 (GMT +02:00):

>
>
> On 10/9/19 11:24 PM, Paul Davis wrote:
>
>>
>> On Wed, Oct 9, 2019 at 3:21 PM <[hidden email]
>> <mailto:[hidden email]> > wrote:
>>
>>> The samplerate was 44100, period size 1024. Going up to 2048 made it
>>> worse.
>>>
>>> Started with:
>>>
>>> jackd -P98 -dalsa -dhw:2 -p1024 -n8 -r44100
>>> scsynth -u 57110 -z 256 -i 0 -o 8 -m 262144
>>
>>
>> try running at 48kHz with 3 periods and see if that helps.
>>
>>
> My X32 Rack runs flawlessly on arch using 48kHz, 1024 and 2 periods.
> Though I had to sort out USB issues first by running lsusb -t and find the
> least crowded bus.
>
> Cheers
>
> Bollie
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Niklas Reppel
In reply to this post by Niklas Reppel
Well, the problem seems to be that there's no dedicated physical USB 3.0 port on my machine, the physical
port act as the one or the other depending on what you connect.

All physical ports have the SuperSpeed symbol, but none of them automatically connects to the
(internal) USB 3.0 root hub if I connect a USB 2.0 device, so I assume there's some logic that
detects whether the connected device is 2.0 or 3.0 and connects it accordingly. Like, when I connect
a USB 3.0 storage device, lsusb shows it as connected to the 3.0 bus.

So I my guess would be, if I use a dedicated (external) USB 3.0 hub, it would be detected as a 3.0 device
and routed to the (internal) 3.0 bus, and thus the devices connected to the external hub would be as well.
But, of course, that's just speculation. I might just get a 3.0 hub and try.


Ralf Mardorf wrote on 13.10.2019 18:37 (GMT +02:00):

> On Sun, 13 Oct 2019 18:19:15 +0200 (CEST), [hidden email] wrote:
> >Or, if i connect an external USB 3.0 hub, will it wire external
> >devices to the 3.0 internal bus?
>
> USB 3 is backward compatible. Occasionally I connect an USB 2 Scarlett
> 18i20 2nd gen to a GA-B85M-D3H mobo's USB 3 slot of an Arch Linux
> machine. This works without issues and allows lower latency, than is
> possible with a PCIe slot assembled RME HDSPe AIO on the same machine.
> Even if the USB 3 port shouldn't be the best for an audio device on
> your machine, it at least should sort of work, perhaps causing xruns.
> That it doesn't work at all, might be a pointer to an issue of your
> computer.
>
>
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

>64 Channels @ 96k ?

Manuel Haible
 
Hi,
 
 what is currently the way to go for rock solid
> 64 output channels @ 96k + some input channels (24 maybe)
?
 
Multiple RME HDSPe MADI cards? MADIfx?

+ Alsa + Jack ?

Would external clocking even be possible somehow?


 
 
 

 
 
 
 
 
 

_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

Re: Behringer X32 - glitch problems

Len Ovens
In reply to this post by Niklas Reppel
On Sun, 13 Oct 2019, [hidden email] wrote:

> Hmm, this is interesting, i just checked the USB ports in my Thinkpad
> T440s with lsusb -t.

> That means, not matter where i connect the interface, there's no way to
> connect to the least busy bus.

Which driver is servicing them? if it an xhci driver... turning that off,
either in bios (what I did) of via software may let the ohci driver work
and split them up a bit. (I at least got two doing this) This basically
makes your machine USB2 and not USB3. However, it may allow you to
separate at least some of you USB parts. On this machine with xhci running
every single plug ended up on the same USB bridge (all 10 of them!) but
turning xhci off in bios split the plugs into two bridges (front and back
in my case) and allowed me to give my USB audio a higher priorty while
leaving my mouse and printer at lower priority. (the mouse has since been
replaced with a ps/2 mouse) This may help:
https://www.systutorials.com/241533/how-to-force-a-usb-3-0-port-to-work-in-usb-2-0-mode-in-linux/


--
Len Ovens
www.ovenwerks.net
_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user
Reply | Threaded
Open this post in threaded view
|

more then 64 channels at 96k?

Manuel Haible
In reply to this post by Manuel Haible
Hello,

I am wondering what are the possibilities of HQ audio with a large channel-count in Linux?
 
 What is the way to go for more than 64 output channels @ 96k
plus some input channels?

Several RME HDSPe PCIexpress driven together with Alsa and Jack?
One RME MadiFX card?
 
Any experience here?


Is it possible to run different channels with a different sample-rate?

Would external clocking be possible, too?


On a laptop there is no way to archive this? For example with a Sonnet Thunderbolt to external PCI adaptor? Thunderbold is not supported by Linux, is it?

There is no solution with the standart ethernet-port at this high sample rate?
Motu AVB is limited to 48k as class compilant USB?
 
 
I'd be happy about any answer *

Bests, Manuel

_______________________________________________
Linux-audio-user mailing list
[hidden email]
https://lists.linuxaudio.org/listinfo/linux-audio-user