YES kernel #1 works on Audiolab CDQ8200 in 16/44 but the CDQ8200 must be in USB "Combability Mode" not in 24/96...
Kernel #2 did not work and I don't have a USB Hub. And I need the (24/96) for my HD tracks.
The sound from kernel #1 is sadly not near a TT 3.0 with a good spdif cable... but i'm loving that it is working :-) Good work!
Results 21 to 30 of 311
-
2012-02-05, 02:52 #21Junior Member
- Join Date
- Jul 2010
- Posts
- 19
-
2012-02-05, 04:34 #22Senior Member
- Join Date
- Apr 2005
- Posts
- 6,932
Last edited by Triode; 2012-02-05 at 04:37.
-
2012-02-05, 04:46 #23Senior Member
- Join Date
- Apr 2005
- Posts
- 6,932
Yes for hot plugging to work the jive application would need to support restarting with a new output interface. As it is at boot time it claims the output device and so we need to have the usb dac available at that time.
Hence I think anything which doesn't mess with the jive app needs to assume the dac is there when you start up touch. [so I've ruled out hotpluging as something to target for the moment]
The playback solution uses the updated jive_alsa backend which supports sending data direct to the dac, bypassing the plug layer.
I supsect this is more of a problem with the device name and bit depth - try editting /etc/squeezeplay/userpath/settings/SqueezeboxFab4.lua direct to change the sample size and output device name then reboot? The USB Audio Applet is responsble for setting this - if you can find the right things to parse out of /proc/asound/ to get the settings then I will uptdate.I tried kernel #2 out with an old adaptive (2706 based) DAC and it did not work at all, the Touch just continually rebooted (whether plugged into the hub or direct). I haven't had time to do any debugging on this. I haven't had time to try this DAC out with kernel #1.
-
2012-02-05, 07:20 #24Junior Member
- Join Date
- Feb 2012
- Posts
- 9
Kernal #2 on Music Streamer II working without a hub
It looks like the title is not part of the quoted text so you could not see the dac is a Music Streamer II when you wrote this note. The dac is async and only has a usb input. The firmware is 1.7.
Perhaps the Squeezebox Touch firmware is important. I got on the nightly channel for the server software to try to solve a scanning problem so I have the latest beta server software. It updated the Touch firmware to 7.7.2-r9597.
Here is what USB Audio Output says:
Playback
Status: Running
Interface: 1
Allset = 1
URBs = 2(84)
Packet Size = 600
Momentary freq = 44100 (0x2c.199a)
Interface 1
Allset 1
Format: 0x20
Channels: 2
Endpoint: 1 OUT (ASYNC)
Rates: 96000,88200,48000,44100,32000
Interface 1
Allset: 2
Format: 0x2
Channels: 2
Endprint: 1 OUT (ASYNC)
Rates: 96000,88200,48000,44100,32000
The above is for MOG, CD's ripped to WAV files and stored MP3 files.
The Momentary freq changes to 96000 Hz (0x60.0000) for the 96k file that comes with the Music Streamer II.
I still do not have any pops or clicks with any file format I have tried.
Fred
-
2012-02-05, 07:24 #25Senior Member
- Join Date
- Apr 2005
- Posts
- 6,932
So I suspect you are lucky there then - with the momentary frequency set at the sample frequency and not changing it looks like it is working without receiving any feedback messages. However this is probably only ok as your dac's clock is close to the same frequency as the internal clock SBT is using to calculate how much data to send. I suspect you do get clicks/pops but they are so far apart you don't notice... If you had a hub in the way then I would expect the momentary frequency to update.
-
2012-02-05, 09:58 #26Junior Member
- Join Date
- Feb 2012
- Posts
- 9
Kernal #2 on Music Streamer II working without a hub
I installed a powered USB hub. The DAC is not recognized unless the hub is powered on. If the hub is powered on when the Touch is rebooted there is no need to worry about turn on sequences. Everything can be hooked up and there does not seem to be any problem.
On the file formats that gave a Momentary Freq of 44000 they now will switch to 44500. They stay at 44000 maybe 90 percent of the time.
Results are similar for the 96k file except it cycles from 96000 to 96250.
Clicks and pops seem to be a thing of the past.
Fred
-
2012-02-05, 17:24 #27Senior Member
- Join Date
- Jun 2006
- Posts
- 1,460
It works with the samplebit set to 16. This can be found in the stream0 (which I presume you are already looking at) in the format:
0x2 = S16_LE
0x6 = S24_LE (format used by Touch internal drivers, 4 bytes with 24 bits in the bottom 3 bytes)
0x20 = S24_3LE (most 24 bit USB DACs, 3 bytes)
BTW the streamer II works fine with the normal kernel, just install the USB Audio applet.
John S.
-
2012-02-05, 22:27 #28Senior Member
- Join Date
- Jun 2006
- Posts
- 1,460
I tried this with TT3.0, it took some work to get the two to work together. The best bet seems to be to install TT3.0 first. Turn off the display blank so you can see what you are doing with Triodes changes. Set the buffer to 5000 (does not work for me with much lower). Turn off the priority changes, something in there seriously messes up the Streamer II. I did this by editing the file:
/etc/init.d/rcS.local
Comment out the line that sets the priorities.
Then install Triodes applets. Unplug the power, plug in the hub and DAC, turn power back on, use USB Audio to select the DAC, it will reboot and should work.
The results are quite a bit better than without TT3.0, it was a pretty shocking difference to me. Interestingly with the internal DAC the daemon killer made a big difference, but with the Streamer II on the hub it hardly made any difference at all.
John S.
-
2012-02-06, 14:15 #29Senior Member
- Join Date
- Apr 2005
- Posts
- 6,932
That's interesting - do you have a usb isolator to hand to test if that makes any difference? Trying to guess what the interaction is with the dac if the clock really is entirely local to the dac...
I will add something to the USB output applet for 16 bit.
Also kernel #2 only contains a couple of changes over the default kernel. Its likely with dacs that don't also use HID that the default kernel will work in place of kernel #2, they will get a divide by 0 backtrace in dmesg but I suspect they will work. Other dacs need one of the other changes I've make for kernel #2Last edited by Triode; 2012-02-06 at 14:25.
-
2012-02-08, 14:40 #30Senior Member
- Join Date
- Apr 2005
- Posts
- 6,932

Reply With Quote

