Ultralite AVB

Talk about your MIDI interfaces, microphones, keyboards...

Moderators: MattKingUSA, khz

Musinux
Established Member
Posts: 59
Joined: Sun Jan 14, 2018 8:38 pm
Has thanked: 3 times
Been thanked: 3 times

Re: Ultralite AVB

Post by Musinux »

alcomatt wrote:
. I am quite busy right now so it might take me a while.
I came across that patch as well.

Since I'm too lazy to build a patched module, I tried adding

Code: Select all

options snd_usb_audio index=3 quirk_alias=07fd0005:168600dd
and

Code: Select all

options snd_usb_audio index=3 quirk_alias=07fd0005:154e1003
to /etc/modprobe.d/sound.conf (not both at the same time of course),
which should work the same way as loading the patched module would.

The first one works just like unquirked.
With the second, it was nearly impossible to switch samplerates in jack and,
after playing around with it too much, the Ultralite locked up (in a crazy new way).

If I'd build a patched module, I'd play around with the delay amount (as suggested in your URL as well).

Edit: In the case somebody wants to try other usb ids with the modprobe method, you might have to remove or edit the "index=3" part.
grep https://github.com/torvalds/linux/blob/ ... b/quirks.c for USB_ID.
User avatar
Be.
Established Member
Posts: 55
Joined: Mon Mar 16, 2015 4:51 am
Been thanked: 6 times

Re: Ultralite AVB

Post by Be. »

alcomatt wrote:Has any of you manged to get a decent performance with jack2? I keep on having xruns all over the place, regardless what I do. With Jack1 however, the performance is solid and I can 0 xruns with buffer sizes down to 64. I would have moved back to jack1 but sadly Bitwig only supports jack2 and now I am in a right kerfuffle... this catch-22 situation drives me mad
Interesting. I have not tried Jack1.
alcomatt wrote:I may have a solution to this but not 100% sure yet. I have been checking my kernel logs and I kept on seeing the below error reported from the device:

Code: Select all

usb 3-4: clock source 1 is not valid, cannot use
This lead me to http://guillaumeplayground.net/teac-ud- ... mment-8186 and it looks like it is worth a shot. I will try writing a quirks options for MOTU hardware IDs and see if it makes a difference. I am quite busy right now so it might take me a while. If any of you are experienced in C then you're more than welcome to try
Cool, I would be happy to test patches. Maybe if we're lucky that will magically fix the crackling too?
Ulrich
Established Member
Posts: 13
Joined: Thu Jan 25, 2018 4:21 pm

Re: Ultralite AVB

Post by Ulrich »

Anyone tried the patch ? Does it fix the issue when plugging the card ?
dsreyes1014
Established Member
Posts: 166
Joined: Mon Dec 24, 2012 1:42 pm
Has thanked: 1 time

Re: Ultralite AVB

Post by dsreyes1014 »

Has anyone tried this patch? I'm curious to see if this solves an issue for me with the TouchConsole beta firmware where it doesn't playback correctly.
Musinux
Established Member
Posts: 59
Joined: Sun Jan 14, 2018 8:38 pm
Has thanked: 3 times
Been thanked: 3 times

Re: Ultralite AVB

Post by Musinux »

dsreyes1014 wrote:Has anyone tried this patch?
There is no patch as of yet.

See my other post, if you want to try, you can use the quirks option.

Either way (compiling a patched kernel or using quirks), I think just changing the VID:PID won't be enough.
You'll have to change the delay parameter to what the ultralite avb needs.

Just be sure to backup all your settings, because the ultralite avb can crash while testing this...
Musinux
Established Member
Posts: 59
Joined: Sun Jan 14, 2018 8:38 pm
Has thanked: 3 times
Been thanked: 3 times

Re: Ultralite AVB

Post by Musinux »

There's a new firmware update for the Ultralite AVB.

https://cdn-data.motu.com/downloads/aud ... B16.update

Changelog:

v1.3.4+603 (Release Date 2018-06-20)

Introduces new Touch Console™
Fixes an intermittent issue that could cause audio glitch on USB input
Improves audio startup timing

Besides the ui changes, the "bit-crusher" effect is different now. it still happens without xruns, but it's softer and it fades into complete silence, returning to normal for some time.
After some testing, I can say it's worse for me (happens more often and longer than before).

I'm still not using quirks or a patched kernel.
User avatar
Be.
Established Member
Posts: 55
Joined: Mon Mar 16, 2015 4:51 am
Been thanked: 6 times

Re: Ultralite AVB

Post by Be. »

Well, that's anticlimactic. :/

I've gotten much better speakers since before (Genelec 8030As) and just tried the Ultralite AVB for the first time with them. I can clearly hear now that the RME Babyface Pro sounds better, so I'll likely sell my Ultralite AVB. Too bad I already broke the screen carrying it in a backpack so I'll have to reduce the price a lot...

EDIT: Sweetwater still gave me a refund minus a $15 restocking fee! Woo! If you're in the US, I highly recommend buying your music gear from Sweetwater.
Last edited by Be. on Sat Nov 24, 2018 6:56 am, edited 2 times in total.
Musinux
Established Member
Posts: 59
Joined: Sun Jan 14, 2018 8:38 pm
Has thanked: 3 times
Been thanked: 3 times

Re: Ultralite AVB

Post by Musinux »

Be. wrote:Well, that's anticlimactic. :/

I've gotten much better speakers since before (Genelec 8030As) and just tried the Ultralite AVB for the first time with them. I can clearly hear now that the RME Babyface Pro sounds better, so I'll likely sell my Ultralite AVB. Too bad I already broke the screen carrying it in a backpack so I'll have to reduce the price a lot...
Just to raise the value of our Interface a little (before I sell mine too in the long run :D ):
After the update it doesn't crack as often when starting or stopping something!


OT:
Have you compared the Babyface Pro with the non-Pro Babyface by accident? I think my non-Pro Babyface is a little little worse sounding than the Ultralite with my Adam A7Xs...
The Fireface UCX would be the RME-Interface of my choice, mainly because of the stand-alone/midi remote-control feature.
But it's expensive and sadly some of those features seem pretty non-working if you google for it.

OT2:
Since I couldn't overread your signature regarding Mixxx, thank you for this great piece of software! Although I don't really use it much myself, a friend of mine loves Mixxx and shows/tells everybody how great it is.
User avatar
Be.
Established Member
Posts: 55
Joined: Mon Mar 16, 2015 4:51 am
Been thanked: 6 times

Re: Ultralite AVB

Post by Be. »

Musinux wrote:After the update it doesn't crack as often when starting or stopping something!
That's nice for home/studio use, but any chance of crackling makes the device unsuitable for performance.
Musinux wrote:Have you compared the Babyface Pro with the non-Pro Babyface by accident? I think my non-Pro Babyface is a little little worse sounding than the Ultralite with my Adam A7Xs...
No, I have the Babyface Pro, not the original Babyface.

I didn't think this would be as annoying as it is, but after setting it up for performance a few times, the requirement to plug the Ultralite AVB into its own power supply is quite annoying. It's more annoying that the cord for the AC adapter it is shipped with is not very long. This makes me appreciate the Babyface Pro being USB bus powered. I was hoping the MOTU 624 would be able to be bus powered since it uses USB 3 which can supply more current, but when I asked MOTU support about it they said it still requires its own power supply. :/
Musinux wrote:Since I couldn't overread your signature regarding Mixxx, thank you for this great piece of software! Although I don't really use it much myself, a friend of mine loves Mixxx and shows/tells everybody how great it is.
Thanks :)
vth
Established Member
Posts: 17
Joined: Wed Jul 18, 2018 6:21 am

Re: Ultralite AVB

Post by vth »

Has anyone experimented with downgrading their AVB firmware to 1.2.8+378 (September 2016 release)?

I recently bought a used UltraLite AVB despite having read this thread, with the intention of reselling it if I can't make it work for me. My unit came with the latest 1.3.4+603 firmware already installed. Initially on my setup the bitcrusher effect reliably kicked in after around 13 minutes of playback and it didn't seem to go away until I killed and restarted my audio server (Jack and PA behaved the same in this regard). Inspired by this thread: http://forum.cakewalk.com/tm.aspx?m=3682836&p=2 I downgraded my firmware to the suggested version and I haven't been able to reproduce the crackling issue since.

Granted, I haven't had the time to test it very thoroughly, but in addition to simple playback tests I've done some stress testing with plugin-induced xruns and nothing strange seems to happen. Before downgrading, simply playing from Spotify through PulseAudio was enough to eventually initiate the crackling for me.
Musinux
Established Member
Posts: 59
Joined: Sun Jan 14, 2018 8:38 pm
Has thanked: 3 times
Been thanked: 3 times

Re: Ultralite AVB

Post by Musinux »

vth wrote:Has anyone experimented with downgrading their AVB firmware to 1.2.8+378 (September 2016 release)?
Yup, I tried at least following firmware versions :
1.2.5+308
1.2.8+378
1.2.9+479
1.3.2+520
1.3.4+558
1.3.4+603
1.3.4+577
1.3.4+603

For Me, the time when the bitcrusher effect kicks in varies too much to correlate it to different firmware versions.

I'm using 1.2.9+479 now, but I gave up testing it with what I really want (Digital Room Correction 24/7).

Sometimes it took like 3 hours+ to "bitcrush", if you tried it much longer than that, I'd have a go and try my luck again.
vth
Established Member
Posts: 17
Joined: Wed Jul 18, 2018 6:21 am

Re: Ultralite AVB

Post by vth »

Musinux wrote:
vth wrote:Has anyone experimented with downgrading their AVB firmware to 1.2.8+378 (September 2016 release)?
Sometimes it took like 3 hours+ to "bitcrush", if you tried it much longer than that, I'd have a go and try my luck again.
I have used the interface without issues for the last three days or so, with sessions easily spanning over three hours, though obviously that's not continuous playing or recording time. The first night after downgrading, I left my current Bitwig project playing in a loop for 8+ hours and while everything was normal in the morning, I've really no way of knowing if it did glitch out sometime during the night.
vth
Established Member
Posts: 17
Joined: Wed Jul 18, 2018 6:21 am

Re: Ultralite AVB

Post by vth »

I also just happened to come across this interesting ALSA patch notification from two days ago: https://www.spinics.net/lists/alsa-devel/msg80348.html
If the audio device is externally clocked and set to a rate that does
not match the external clock, the clock will never be valid and we cannot
set the rate successfully. To fix this, allow a rate change even if
the clock is initially invalid, and validate again after the rate is
changed.

This fixes problems with MOTU UltraLite AVB hardware over USB.
vth
Established Member
Posts: 17
Joined: Wed Jul 18, 2018 6:21 am

Re: Ultralite AVB

Post by vth »

Unfortunately, I can now confirm that while downgrading the firmware to 1.2.8 did solve the problem I was initially having with the current firmware revision, it didn't fix all glitch issues for me. While working on a project today, recording and playback worked normally for a good while, but eventually all input channels got crushed. Playback continued to work even after that.
dsreyes1014
Established Member
Posts: 166
Joined: Mon Dec 24, 2012 1:42 pm
Has thanked: 1 time

Re: Ultralite AVB

Post by dsreyes1014 »

Unfortunately for me the Ultralite even on my Mac would get the occasional xrun on playback with moderate buffer sizes as compared to a newly bought MADIface pro. I wanted to really like the Ultralite but I’ve lost data during recording sessions a couple of times and support seems to never respond when I reach out concerning firmware/driver issues. MADIface on both Mac and Linux work so reliably I’d never look back. It’s just my preference but reliability has become top priority for me. I’d highly recommend the MADIface/Babyface Pro even on Linux for those that can afford it as an alternative. It’s really solid.
Post Reply