On 3/11/25 3:59 PM, Alexandre Ratchov wrote:
On Tue, Mar 11, 2025 at 02:41:32PM -0400, Geoff Steckel wrote:
I have 2 USB DACs. One uses a Burr-Brown chip and the other
one is a Modi 3+ (custom circuit) from Schiit Audio.
Both work under Linux.
The Schiit seems to accept playback data but no analog output appears.
Both are claimed by uaudio(4).
Note the lack of a outputs.dac= line for the unhappy unit.
Well, there are many devices with no controls at all, so the lack of
output.dac is not wrong. But a mute control with no level control
would be surprising design choice.
I'd be very surprised since the manufacturer claims truly wonderful
audio and many other virtues. It's supposed to be plug-and-play
compatible with any other DAC.
Linux shows a volume control. That could be implemented anywhere
in its 1,001 layers between the user program and hardware
so that's not definitive.
I downloaded the audio v.2 usb spec. usbdevs may have printed something
useful.
There are quite a few differences in the USB interface information
especially V1 vs V2 format.
Is this problem discussed in an online thread somewhere or would
tech@ be interested? I'm not familiar with the details of USB audio (yet).
Would defining UAUDIO_DEBUG help?
yes, I'd be intested in seeing the dmesg output when the device is
plugged.
uaudio0 at uhub3 port 3 configuration 1 interface 1
"Schiit Audio Schiit Modi 3+" rev 2.00/1.00 addr 5
uaudio0: class v2, high-speed, async, channels: 2 play, 0 rec, 1 ctls
^^ ^^^^^
audio1 at uaudio0
compared to
uaudio1 at uhub3 port 4 configuration 1 interface 1
"Burr-Brown from TI USB Audio CODEC" rev 1.10/1.00 addr 6
uaudio1: class v1, full-speed, sync, channels: 2 play, 2 rec, 3 ctl
^^ ^^^^
audio2 at uaudio1
usbdevs output:
D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs=
vs |||||||
D: Ver= 2.00 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs=
I didn't see any obvious code in uaudio.c that checked that class #.
I found a place where some device capabilities are decoded but
didn't see where a volume/mixer control would show up.
The spec should give the exact bit indicating its presence.
thanks
Geoff Steckel
