[16:41] <JensStutte[m]> Hi! I have a question regarding Behringer USB interfaces: I have a UMC202HD and a UMC404HD. Both are working fine alone, but if I try to use them together bridged, one of the two will have an enormous feedback on all its input channels, making them unusable for recording. Only the one chosen as Master in studio controls will work normally. Instead a bridged setup between one of the bBehringers and a Creative E-MU 0202 works just
[16:41] <JensStutte[m]> fine. Poking a bit around I found, that for both Behringer interfaces QUIRK_FLAG_GENERIC_IMPLICIT_FB is set, which seems to have to do with feedback (just looking at the names). Did any of you experience the same and/or have a workaround? Thank you so much, Jens
[16:41] <JensStutte[m]> s/bBehringers/Behringers/, s/QUIRK_FLAG_GENERIC_IMPLICIT_FB/QUIRK\_FLAG\_GENERIC\_IMPLICIT\_FB/
[16:53] <Eickmeyer[m]> Hi Jens Stutte ! I have the exact same interfaces, and have never had an issue. If you're experiencing feedback, make sure you have everything patched correctly, i.e. in Carla's patchbay or in another patchbay.
[16:55] <JensStutte[m]> Thank you! At least that sounds as if there is hope... I usually setup things in studio control and then use Ardour for recording without any additional setup. Will take a look at Carla. 
[16:55] <Eickmeyer[m]> Carla should be able to run alongside Ardour just a visual patchbay or jack plugin host.
[16:56] <Eickmeyer[m]> It's not a DAW like Ardour.
[17:28]  * JensStutte[m] uploaded an image: (59KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.org/WdzffhkNrWzqEVSsnjXRKcwa/grafik.png >
[17:29] <JensStutte[m]> Hmm, to me this happens already with Carla. If I patch an input from the UMC404 to the output, it is highly distorted and overdriven.
[17:33] <OvenWerks> JensStutte[m]: this may be a zitta-ajbridge bug
[17:33] <OvenWerks> make sure you have the latest version or you can manually use alsa_in and Alsa_out
[17:34] <OvenWerks> I don't know if our backports PPA has a newer version of zitta-ajbridge or not, but check that first.
[17:35] <OvenWerks> some times zitta-ajbridge fails to detect the audio device properly, this may be the problem but I think it has been fixed already
[17:44] <JensStutte[m]> My version is 0.8.4-1, that seems to be the latest?
[17:52] <JensStutte[m]> Which kernel version are you on? I see that [the wuirk flags for Behringer](https://github.com/torvalds/linux/blame/81ff855485a366a391dc3aed3942715e676ed132/sound/usb/quirks.c#L2072) have been added ca. 8-9 months ago only. I am on 5.19.0-1018-lowlatency here.
[17:52] <JensStutte[m]> s/wuirk/quirk/
[18:05] <JensStutte[m]> Ah, reading that bug they say Behringer changed something, so older devices did not need that quirk: "I can confirm that the older model UMC404HD with bcdDevice= 1.12 do work perfectly as a Class Compliant device.  It seems Behringer has changed something." Which bdcDevice version do you see with `lsusb -v` ? Mine is 1.35 (new) on the UMC404 but 1.12 (old) on the UMC202 - that seems as if it could result in conflicting settings if
[18:05] <JensStutte[m]> they share the same driver (I have no idea how this all works, so do not bother with me if I say something stupid)?
[18:08] <JensStutte[m]> * Ah, [reading that, * that bug](https://bugzilla.kernel.org/show_bug.cgi?id=215934) they
[18:08] -ubottu:#ubuntustudio- bugzilla.kernel.org bug 215934 in Drivers "Behringer UMC 404 HD : clock source 41 is not valid and audio stuttering (confirmed for bcdDevice = 1.35)" [Blocking, Resolved: Code_fix]
[18:29] <Eickmeyer[m]> Jens Stutte: What kernel are you running? (uname -r in a terminal)
[18:29] <JensStutte[m]> I am on 5.19.0-1018-lowlatency
[18:30] <Eickmeyer[m]> Meaning it should be fixed in your kernel. If you tweaked some settings internally on yours, it may have messed things up.
[18:30] <JensStutte[m]> No, it is a plain studio installation I made a week ago, no tweaks
[18:31] <Eickmeyer[m]> (the status on that bug is RESOLVED CODE_FIX)
[18:32] <JensStutte[m]> Yeah, I am just wondering if the combination of one old and one newer device is somehow broken. Could you check for your bcdDevice numbers?
[18:32] <Eickmeyer[m]> I don't currently have both devices connected, and don't have time to do that today. Sorry.
[18:33] <JensStutte[m]> Sure, no problem. I think I have enough of that for today, anyways. Thanks for the fast response!
[18:33] <Eickmeyer[m]> zita-ajbridge handles the connections, and should automatically differentiate though.
[18:34] <Eickmeyer[m]> OvenWerks: Do you know anything about this?
[18:34] <JensStutte[m]> Is there a way to see what it sets somewhere?
[18:34] <Eickmeyer[m]> (just pinged the person that might know)
[18:36] <Eickmeyer[m]> Hmmm... no newer version of zita-ajbridge
[18:38] <Eickmeyer[m]> And it's the latest.
[20:56] <JensStutte[m]> OK, just to add another data point: If I select no USB device as master, I can see zita-a2j -j USB2,0,0-in -d hw:U192k,0,0 -r 48000 -p 1024 -n 2 -c 100 -I 0 processes for both devices, while if I select one of the Behringers to be the master, there is no such process for it. Those that use zita-a2j will distort, so if I do not select any of them as master, all 6 input channels are distorted.
[21:06] <JensStutte[m]> And if I use the Creative E-MU 0202 as master and one of the Behringers as additional device, it will still distort. Only if it is a master, it will work for me. So it seems zita-a2j is somehow the culprit here? Is there anything else I can debug?
[21:06] <JensStutte[m]> * still distort, too. Only
[23:59] <OvenWerks> JensStutte[m]: have you tried not to add a device with studio-controls and to instead from a terminal use alsa_in (with a similar command line but no -r 4800)?