[Edit: The problem went away when I rolled back to an older realtime kernel, but I'll leave this here for reference]
I have a fresh installation of Ubuntu 17.10 with almost nothing extra installed yet besides Bitwig Studio, Cadence and a few other things from the KX Studio repositories.
But I also tried the bug after a fresh install with nothing else than Bitwig Studio installed, it's the same.
Not sure if this information can be added afterwards: but I installed the 3.19.8 low latency kernel, and the problem disappeared. On 17.1 I can't use the current low latency kernel because the system becomes very unresponsive to the point of being hard to even log in. So I'm not sure if this is just a matter of difference between the low latency and regular kernels.
I have a TC Electronics Impact Twin Firewire audio interface.
It is connected to a Iocrest SY-PEX30016 PCI-E card, which has a Texas Instruments XIO2213BZAY chipset.
The ALSA snd-dice drivers work, but with glitches.
There is a periodic, glitch in the sound that sounds like a part of the stream is missing, or something. It is steady in timing, nothing seems to have an effect on it.
Recording seems to work. All outputs and inputs, also the ADAT outputs seem to work, altough I could not test if the ADAT outputs feature the glitch too.
Seemingly randomly there might be a short maybe around 0.3 second total break in the sound.
I'm not getting xruns at all.
I tried also Ubuntu Studio live USB, it has the exact same glitch. I also tried a Native Instruments Audio2DJ USB interface, with that there is no problem with the sound on my computer.
I tried also with Carla and ZynAddSubFx, and also with the live USB of Linux Mint (current version), and the same glitch is present.
I tried also another Firewire controller card with a XIO2200A Texas Instruments chip, it features the same glitch, but running a different speed.
I had a KX Studio 14.x installation a few years ago, where the interface was working without a hitch. But then I was running with FFADO.
So to me it seems that the kernel integration of snd-dice has some problem with timing with a TC Electronics interface?
I could not get FFADO to run with Jack now so I can't say if the FFADO implementation would work.
[SOLVED] Periodic glitch with snd-dice ALSA drivers
Moderators: MattKingUSA, khz
- Capoeira
- Established Member
- Posts: 1321
- Joined: Tue May 12, 2009 1:01 pm
- Location: Brazil
- Has thanked: 3 times
- Been thanked: 2 times
Re: [SOLVED] Periodic glitch with snd-dice ALSA drivers
Hi, are you still using this interface? snd-dice never worked for my Konnekt 24d (those glitches). Didn't care too much since Fffado works; but now I am trying out Pipewire and it doesn't support Ffado