[LAU] Motu M64 Madi interface (not) class compliant?

[Anders Hellquist]

The problems we had with glitching audio and possibly also channel hopping
seems to affect Apple and windows users too to a degree according to
thrrads at motunation.



That means there are likely bugs in the firmware that are causing
this issue and the Linux problem might not even be a Linux issue.

Since there is now a better way to debug this using the newly discovered
root access, there might be workarounds or other nice things in the pipe.

/Anders

On Mon, Apr 19, 2021, 20:00 Fernando Lopez-Lezcano <nando at ccrma.stanford.edu>
wrote:

> On 4/18/21 12:09 AM, Peter P. wrote:
> > * Paul Davis <paul at linuxaudiosystems.com> [2021-04-18 03:43]:
> >> On Sat, Apr 17, 2021 at 4:22 PM Peter P. <peterparker at fastmail.com>
> wrote:
> >>
> >>> * Peter P. <peterparker at fastmail.com> [2021-03-12 18:43]:
> >>>> Hi list,
> >>>>
> >>>> can report that the Motu M64 Madi interface does work with a limited
> >>>> channel count of 24 for input and another 24 for output (instead of 64
> >>>> each). So if you are coming across a venue or employer that already
> has
> >>>> one, you are good to go. Sample rate needs to be set on the device
> >>>> before starting jack. Of course, buying the card for oneself for so
> much
> >>>> money getting only one third of the channels seems useless linux users
> >>>> IMHO.
> >>>
> >>> I can add now that the card does not work reliably in class compliant
> >>> mode with external sync clocks. It will randomly switch back and forth
> >>> output channels in banks of eight, cause clicks and distortion and
> >>> aliasing artifacts.
> >>
> >> The bank switching is a known problem with other MOTU USB devices, and
> has
> >> been "solved" by using specific firmware releases (e.g. for the MOTU
> >> Ultralite AVB) which may not be available anymore or for your specific
> >> device.
>  >
> > Thanks Paul, that is good to know! Does anyone have more info on this
> > back switching and associated firmware updates?
>
> (the old firmware versions are very old - they used to allow up to 64
> channels in class compliant mode - and do not run in the newest hardware
> that you can buy today)
>
> There is a VERY long thread about this here:
>
>
> https://linuxmusicians.com/viewtopic.php?f=6&t=18046&sid=3586a531a296e4134cd4c7418f429395
>
> A lot of information - I was just looking at new pages at the end since
> last time I looked, and there is a way to get a telnet root login into
> the boxes, ha ha - running linux internally of course (I expect this
> will be gone in the next firmware release :-)
>
> Anyway - not really a resolution to the issue in the thread, but lots of
> ideas to try.
>
> I am running a couple of motu interfaces using a patched kernel that
> enables the use of the alternate usb (proprietary) endpoints and that
> works mostly fine (64 channels over usb with current firmware, 16A & 8M
> tested with this). But the patch does not patch over current kernels and
> a LOT has changed in the audio usb driver, I still have to figure out
> where to do things now. Sigh.
>
> -- Fernando
>
>
> >> I do not believe that the aliasing is caused by bank switching, but by
> >> (possibly) an ALSA driver bug. It generally goes away within a few
> minutes
> >> or on device restart.
> > But wouldn't that driver bug affect all class compliant interfaces then?
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user at lists.linuxaudio.org
> https://lists.linuxaudio.org/listinfo/linux-audio-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.linuxaudio.org/archives/linux-audio-user/attachments/20210420/f78b8b19/attachment.html>