[nyc_paramedic]

Why not just use aplay with your various hires files to help sort this out?

e.g.

aplay -v -D plughw:0,0  xxxx.wav or  aplay -v -D hw:0,0  xxxx.wav

that at least should clarify what voaygae MDP/ALSA can handle with your DAC.

He'll have to use plughw. Also, only .wav files need apply for this particular problem.  And, turn down the volume reaaaal low before using this command.

[nyc_paramedic]

Hello all,

I just wanted to thank Nick for starting a great and informative thread, and everyone else for contributing.  Only a few weeks ago I discovered VortexBox,...

Welcome. I hear really nice things about Vortexbox. It now comes with Alsa 1.0.24, so UAC2 devices work out of the box. Personally, I prefer to rip my CD's manually, especially my classical music.

I'm cheating a little, which may in the end prove detrimental to the sound quality, but last night, instead of an Alix board, I bought an HP thin client.  The board is cluttered with an integrated video chipset, and it has a ridiculous number of usb ports, but the Atom cpu should make it versatile, and it has a fair amount of memory.  I'll load it with voyage and use it as an mpd server first.  I'm also curious about the cics memory player approach, so I'll likely purpose the new box to that too, after I get a feel for how the Voyage/mpd configuration sounds.  I may opt to get the Alix then, so I can do side-by-side comparisons.  I may pick up a few more thin clients, too, depending on how this one fares in the shoot out.  They go fairly cheap at auction, and I have friends and family who would love to have one.

Just be aware that Voyage MPD on HP Thin Client might not sound like Voyage MPD on Alix. Anyway, give it a shot.

[ted_b]

He'll have to use plughw. Also, only .wav files need apply for this particular problem.  And, turn down the volume reaaaal low before using this command.

Ok, so a couple questions:
1)  what does this command do
2)  do i need to specify the device when typing -D even if I have only one device?
3)  why plughw and not hw?

[ted_b]

root@voyage:~# aplay -v -D plughw:0,0  wav/Jimmy Cobb Quartet/Jazz in the Key of Blue (24-192)/06 - Jimmy Cobb Quartet - Jazz in the Key of Blue (24-192) - I Had the Craziest Dream.wav
-bash: syntax error near unexpected token `('
root@voyage:~# aplay -v -D plughw:0,0  wav/Norah Jones/Feels Like Home/01 - Norah Jones - Feels Like Home - Sunrise.wav
wav/Norah: No such file or directory


Both songs (first one is 24/192, second is redbook) play fine via Mpod, and play in their native sample rates, with DAC set to UH1.  Both syntaxes were taken directly from the mpd.log entries but I guess aplay needs more info cuz it doesn't like them.

[nyc_paramedic]

root@voyage:~# aplay -v -D plughw:0,0  wav/Jimmy Cobb Quartet/Jazz in the Key of Blue (24-192)/06 - Jimmy Cobb Quartet - Jazz in the Key of Blue (24-192) - I Had the Craziest Dream.wav
-bash: syntax error near unexpected token `('
root@voyage:~# aplay -v -D plughw:0,0  wav/Norah Jones/Feels Like Home/01 - Norah Jones - Feels Like Home - Sunrise.wav
wav/Norah: No such file or directory


Both songs (first one is 24/192, second is redbook) play fine via Mpod, and play in their native sample rates, with DAC set to UH1.  Both syntaxes were taken directly from the mpd.log entries but I guess aplay needs more info cuz it doesn't like them.

You have to specify an absolute path, i.e., on my box, /mnt/nfs/mpd_music/... .wav

[nyc_paramedic]

Ok, so a couple questions:
1)  what does this command do
2)  do i need to specify the device when typing -D even if I have only one device?
3)  why plughw and not hw?

1. It's a simple command player for Alsa. http://en.wikipedia.org/wiki/Aplay If you can play an audio file with aplay and not with mpd, then maybe it's an mpd issue.

2. yes.

3. It has to do with S32_LE and conversion from S24_LE in this case, I think.

[ted_b]

So, Nick and I have been IM'ing today and I have a call into Antelope.  As stated earlier, my redbook-to-24/192 (see below) songs play just fine in UH1 and don't play at all in UH2.  Here is successful UH1 Aplay results with a 24/192 file.

root@voyage:/mnt/tunes/wav# aplay -v -D hw:0,0 equinox04.wav
Playing WAVE 'equinox04.wav' : Signed 24 bit Little Endian in 3bytes, Rate 19200 0 Hz, Stereo
Hardware PCM card 0 'Zodiac Gold' device 0 subdevice 0
Its setup is:
stream : PLAYBACK
access : RW_INTERLEAVED
format : S24_3LE
subformat : STD
channels : 2
rate : 192000
exact rate : 192000 (192000/1)
msbits : 24
buffer_size : 96000
period_size : 24000
period_time : 125000
tstamp_mode : NONE
period_step : 1
avail_min : 24000
period_event : 0
start_threshold : 96000
stop_threshold : 96000
silence_threshold: 0
silence_size : 0
boundary : 1572864000
appl_ptr : 0
hw_ptr : 0

[Eudyptes]

root@voyage:~# aplay -v -D plughw:0,0  wav/Jimmy Cobb Quartet/Jazz in the Key of Blue (24-192)/06 - Jimmy Cobb Quartet - Jazz in the Key of Blue (24-192) - I Had the Craziest Dream.wav
-bash: syntax error near unexpected token `('
root@voyage:~# aplay -v -D plughw:0,0  wav/Norah Jones/Feels Like Home/01 - Norah Jones - Feels Like Home - Sunrise.wav
wav/Norah: No such file or directory

That's just a basic syntax error on the CLI, use double quotes around your file name when it has spaces etc in the file name, e.g.

aplay -v -D plughw:0,0  wav/Jimmy Cobb Quartet/Jazz in the Key of Blue (24-192)/06 - Jimmy Cobb Quartet - Jazz in the Key of Blue (24-192) - I Had the Craziest Dream.wav
bash: syntax error near unexpected token `('

aplay -v -D plughw:0,0  "wav/Jimmy Cobb Quartet/Jazz in the Key of Blue (24-192)/06 - Jimmy Cobb Quartet - Jazz in the Key of Blue (24-192) - I Had the Craziest Dream.wav"
wav/Jimmy Cobb Quartet/Jazz in the Key of Blue (24-192)/06 - Jimmy Cobb Quartet - Jazz in the Key of Blue (24-192) - I Had the Craziest Dream.wav: No such file or directory

The first fails becasue of syntax error, second one fails as I don't have your file, otherwise it should be OK if you have the file path name correct.

Yes, I forgot there's no volume control with aplay, but your example seems to show that the Antelope DAC handles 24bit/192KHz fine, but what happended with higher res files, what was the  aplay ouptut?

During playback over a USB connection this command can give some additional info:

cat /proc/asound/card0/stream0

Here's an example from audio playback on my cheap USB DAC which is card1 on my system (your DAC is card0).

cat /proc/asound/card1/stream0
Burr-Brown from TI               USB Audio CODEC  at usb-0000:00:02.0-1, full s : USB Audio

Playback:
  Status: Running
    Interface = 1
    Altset = 1
    URBs = 3 [ 8 8 8 ]
    Packet Size = 192
    Momentary freq = 44100 Hz (0x2c.199a)
  Interface 1
    Altset 1
    Format: S16_LE
    Channels: 2
    Endpoint: 2 OUT (ADAPTIVE)
    Rates: 32000, 44100, 48000
  Interface 1
    Altset 2
    Format: S16_LE
    Channels: 1
    Endpoint: 2 OUT (ADAPTIVE)
    Rates: 32000, 44100, 48000
  Interface 1
    Altset 3
    Format: S8
    Channels: 2
    Endpoint: 2 OUT (ADAPTIVE)
    Rates: 32000, 44100, 48000
  Interface 1
    Altset 4
    Format: S8
    Channels: 1
    Endpoint: 2 OUT (ADAPTIVE)
    Rates: 32000, 44100, 48000
  Interface 1
    Altset 5
    Format: U8
    Channels: 2
    Endpoint: 2 OUT (ADAPTIVE)
    Rates: 32000, 44100, 48000
  Interface 1
    Altset 6
    Format: U8
    Channels: 1
    Endpoint: 2 OUT (ADAPTIVE)
    Rates: 32000, 44100, 48000


PS From the Voyage MPD web page http://linux.voyage.hk/voyage-mpd

Voyage MPD is based on Debian 6.0 "Squeeze" and Voyage Linux 0.7. It contains the following features:

    * MPD 0.16
    * latest ALSA driver that supports USB Audio Class 2 (allowing 24bit and up to 192Khz sample rates)
    * 2.6.33 real-time kernel
    * lirc and lirc driver for remote control
    * WebGUI modified from Meshlium Manager System


I'm sure you've seen this document:

http://www.antelopeaudio.com/SUPPORT/Manuals/Linux%20Ubuntu%20OS%20&%20Zodiac.pdf

Which contains this for part of the suggested MPD conf

audio_output {
type
"alsa"
name
"Zodiac Gold"
device
"plughw:1,0"
auto_resample "no"
user_mmap
"yes"
period_time 0
buffer_size 0
period_size 0
mixer_device "default"
mixer_control "PCM"
mixer_index "0"}

But the plughw:1,0 is only there as in Zodiac's example they are using the DAC with a m/board that has inbult sound (see page 3), hence the DAC is device number is 1 and not  0 as in your case. But why plughw as opposed to hw? This might suggest that at certain bit/frequency audiofile formats ALSA needs to do some kind of format conversion for the DAC to play the file.   Hence I previously suggested experimenting with the "aplay" command.

It also says on page 3:

NOTE: If your version of the Linux kernel does not support USB Audio
Class 2, then you must install the latest Linux kernel (from version >=
2.6.35 onwards) to be able to use USB Audio Class 2.
The benefit of using Audio Class 2 is that it enables you to playback audio
through the Zodiac, up to a sample rate of 384kHz.
(See Zodiac Manual for further information on USB Audio Class)

I don't know if this is correct,  nor if the Voyage MPD kernel 2.6.33 was been compiled with a configuration that includes support USB Audio Class 2. I think it must, as how else have you managed to play 24bit/192KHz audio on the Alix?

On other point of interest in this document is on page 8. Did you notice in the screenshot of the DeadBeeF player that there is a 24bit/384KHz listed?  Do this mean they played this hirez file on this particular version of Linux over the Zodiac Gold DAC, or not?

[Eudyptes]

I'm cheating a little, which may in the end prove detrimental to the sound quality, but last night, instead of an Alix board, I bought an HP thin client.  The board is cluttered with an integrated video chipset, and it has a ridiculous number of usb ports, but the Atom cpu should make it versatile, and it has a fair amount of memory.  I'll load it with voyage and use it as an mpd server first.  I'm also curious about the cics memory player approach, so I'll likely purpose the new box to that too, after I get a feel for how the Voyage/mpd configuration sounds.  I may opt to get the Alix then, so I can do side-by-side comparisons.  I may pick up a few more thin clients, too, depending on how this one fares in the shoot out.  They go fairly cheap at auction, and I have friends and family who would love to have one.

If you progress to an Alix board, feedback on the comparison with your thin client would be useful.  For those on a budget  a cheap thin client might be workable alternative assuming it has around the same RAM & flash as the Alix.

It's interesting to note that the Bryston BDP-1 digital player uses a mini-itx Alix board and the Auraliti PK100 uses an Intel D945GSEJT  min-itx. So they both have unnecessary and possibly electrically noisy VGA chips.  While the output of both use coax digital via a modified ESI Juli@ s/card and not USB as per the Alix,  the designers must have considered the presence of additional electronics not to be deterimental to SQ as the s/card is said to provide some "isolation" from the m/broads electrical noise and both these m/boards can be powered from an internal PSU of the makers own design.

I have only seen internal photos of the  Bryston BDP-1 digital player which has a purpose built PSU for clean 12v, 5v (and 3,3V ?) to power the Alix mini-itx board.  Whether a low noise PSU is really crucial, or cost effective, for something like the ALix 2D I don't know.  I guess it depends on the rest of your gear and if you're trying to squeeze the last ounce of performance out of your system.

Makers of quality audio equipment always put effort into power supplies whether it's analogue or digital, but for computer based devices there are as yet no ready made audio grade type PSUs other than using a decent linear PSU for m/boards that only require around 12v/5A. But as the m/boards electronics will split these voltages down for other components and maybe introduce more noise again, is this a waste of time?

[ted_b]

here is what happens when I try to play a 2L recording, originally DSD dff file, converted to 24/384 (it plays on MAC server beautifully in UH2 as 24/384k)

First via UH1:

root@voyage:/mnt/tunes/wav# aplay -v -D hw:0,0 2L384.wav
Playing WAVE '2L384.wav' : Signed 24 bit Little Endian in 3bytes, Rate 384000 Hz, Stereo
Warning: rate is not accurate (requested = 384000Hz, got = 192000Hz)
         please, try the plug plugin
Hardware PCM card 0 'Zodiac Gold' device 0 subdevice 0
Its setup is:
  stream       : PLAYBACK
  access       : RW_INTERLEAVED
  format       : S24_3LE
  subformat    : STD
  channels     : 2
  rate         : 192000
  exact rate   : 192000 (192000/1)
  msbits       : 24
  buffer_size  : 96000
  period_size  : 24000
  period_time  : 125000
  tstamp_mode  : NONE
  period_step  : 1
  avail_min    : 24000
  period_event : 0
  start_threshold  : 96000
  stop_threshold   : 96000
  silence_threshold: 0
  silence_size : 0
  boundary     : 1572864000
  appl_ptr     : 0
  hw_ptr       : 0

(it plays choppy and half speed)

So I tried the plughw as per the instructions above:

root@voyage:/mnt/tunes/wav# aplay -v -D plughw:0,0  2L384.wav
Playing WAVE '2L384.wav' : Signed 24 bit Little Endian in 3bytes, Rate 384000 Hz, Stereo
aplay: set_params:1116: Unable to install hw params:
ACCESS:  RW_INTERLEAVED
FORMAT:  S24_3LE
SUBFORMAT:  STD
SAMPLE_BITS: 24
FRAME_BITS: 48
CHANNELS: 2
RATE: NONE
PERIOD_TIME: 125000
PERIOD_SIZE: 48000
PERIOD_BYTES: 288000
PERIODS: 4
BUFFER_TIME: 500000
BUFFER_SIZE: 192000
BUFFER_BYTES: 1152000
TICK_TIME: 0

( No music)

When going to UH2 (the mode that actually plays this file via MAC/Pure Music into this Antelope Gold DAC)

root@voyage:/mnt/tunes/wav# aplay -v -D hw:0,0 2L384.wav
Playing WAVE '2L384.wav' : Signed 24 bit Little Endian in 3bytes, Rate 384000 Hz, Stereo
aplay: set_params:1116: Unable to install hw params:
ACCESS:  RW_INTERLEAVED
FORMAT:  S24_3LE
SUBFORMAT:  STD
SAMPLE_BITS: 24
FRAME_BITS: 48
CHANNELS: 2
RATE: 384000
PERIOD_TIME: (113776 113777)
PERIOD_SIZE: 43690
PERIOD_BYTES: 262140
PERIODS: (4 5)
BUFFER_TIME: (455109 455110)
BUFFER_SIZE: 174762
BUFFER_BYTES: 1048572
TICK_TIME: 0

(no music.  This UH2 mode has played nothing to date via Linux.)

Summary:
1)  I am happy with 24/192 playback on my Alix (especially with Hynes power supply) and however can't seem to get anything higher.
2)  I have a call into Antelope US tech support to ask why the call UH1 "USB Class 1" when it allows 480Mbits and 24/192k playback, two major tenants of USB Class 2 Audio.  Conversely, they call UH2 "USB Class 2" yet it is incompatible with Voyage MPD to-date.

[Eudyptes]

What looks to be hapening with a 24/384KHz file is that first ALSA detects a rate greater than the max 192KHz, but does not attempt a sample rate conversion (down sample) because you addressed the DAC as a "hw" device, and it suggests using a plug plugin which basically means addressing the DAC as a "plughw" device. The audio result is as you describe.

But when you use "plughw" in the aplay command, ALSA appears unable to deal with the 384KHz rate, hence no audio ouput.

The one thing you did not show is what happnes in UH2 mode when you use "plughw" in the aplay command. I assume the result is the same as using "plughw" in the aplay command in UH2 mode.

[ted_b]

The one thing you did not show is what happnes in UH2 mode when you use "plughw" in the aplay command. I assume the result is the same as using "plughw" in the aplay command in UH2 mode.

I assume you meant it would be the same as plughw in UH1 mode, but regardless, your wish is my command. 

root@voyage:/mnt/tunes/wav# aplay -v -D plughw:0,0  2L384.wav
Playing WAVE '2L384.wav' : Signed 24 bit Little Endian in 3bytes, Rate 384000 Hz, Stereo
aplay: set_params:1116: Unable to install hw params:
ACCESS:  RW_INTERLEAVED
FORMAT:  S24_3LE
SUBFORMAT:  STD
SAMPLE_BITS: 24
FRAME_BITS: 48
CHANNELS: 2
RATE: 384000
PERIOD_TIME: (113776 113777)
PERIOD_SIZE: 43690
PERIOD_BYTES: 262140
PERIODS: (4 5)
BUFFER_TIME: (455109 455110)
BUFFER_SIZE: 174762
BUFFER_BYTES: 1048572
TICK_TIME: 0


UH2 even hates Linux redbook:

root@voyage:/mnt/tunes/wav# aplay -v -D plughw:0,0  ike01.wav
Playing WAVE 'ike01.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
aplay: set_params:1116: Unable to install hw params:
ACCESS:  RW_INTERLEAVED
FORMAT:  S16_LE
SUBFORMAT:  STD
SAMPLE_BITS: 16
FRAME_BITS: 32
CHANNELS: 2
RATE: 44100
PERIOD_TIME: (125011 125012)
PERIOD_SIZE: 5513
PERIOD_BYTES: 22052
PERIODS: (3 4)
BUFFER_TIME: 500000
BUFFER_SIZE: 22050
BUFFER_BYTES: 88200
TICK_TIME: 0

[Eudyptes]

Ouch! UH2 with redbook fails?  I wonder what the Zodiac is doing in this mode.

[ted_b]

Ouch! UH2 with redbook fails?  I wonder what the Zodiac is doing in this mode.

Yes, as I stated when this began, UH2 is no go for Alix in any sample rate.  No biggie really, in that I was never really expecting greater than 24/192 on Alix, and that UH1 works flawlessly from redbook to 24/192 and every sample rate in between.....but the fact that the MAc easily handles UH2 and is also driverless (i.e no installed drivers, just shows up) is a head scratcher.

[jrebman]

Hi Ted,

Just because there isn't a separate downloaded usb driver doesn't necessarily mean that the driver wasn't updated as part of an OS upgrade at sometime in the past.  After all this, I'm really guessing this is what has happened.



-- Jim

[Eudyptes]

....but the fact that the MAc easily handles UH2 and is also driverless (i.e no installed drivers, just shows up) is a head scratcher.

I guess we'd have to say this is one case where OSX has the lead over Linux at the moment.  Voayage MPD 0.7 is based on the lastest version of ALSA 1.0.24 and if you look at the changes to USB audio between v1.0.23 and v.1.0.24  as listed here:

http://www.alsa-project.org/main/index.php/Detailed_changes_v1.0.23_v1.0.24

a quick page search on UAC2 ( USB audio class 2) results in lot of hits.  It's clear Linux USB audio is still developing in this, and other areas.  With more hirez USB consumer grade solutions becoming available Linux will have to catch up sometime.

[ted_b]

Antelope help desk just responded that they have a firmware update almost ready that will allow 24/384 (UH2) capability.  Stay tuned. 

[Eudyptes]

Antelope help desk just responded that they have a firmware update almost ready that will allow 24/384 (UH2) capability.  Stay tuned. 

That's intriguing, as so far what I've read about Linux, ALSA and UAC2,  is that will do up to 24bit/192Khz and not beyond.   

[ted_b]

Here is the give and take on my helpdesk dialog with antelope re: USB Class 1 and 2:

Regarding your questions about USB modes and Classes:
USB mode,classes and versions are 3 totally different things.

UH1 mode is Called USB Class 1 because its compliant with USB Audio Class1 (not to be confused with USB VERSION 1.0)
UH2 mode is Called USB Class 2 because its compliant with USB Audio Class2 (not to be confused with USB VERSION 2.0)

so in general terms USB Class 1 IS NOT USB Version 1.0 and/or USB Class 2 IS NOT USB Version 2.0

Thank you
Best regards!
1 Days Ago

tedmbrady -- Yesterday, 1:30 PM
Thanks. I fully understand the differences between USB Class 1 and USB Version 1.0. But isn't it true that USB Class 1 support is only up to 24/96 and 12Mbits? UH1 goes significantly higher.

5 Hours Ago 
techsup (Antelope) -- Today, 8:07 AM
Dear Ted,
again USB Class is totally different from USB "Version".So in USB 1.0 you get the 24/96 12Mbits/Sec. But USB Audio Class 1 can be used either in USB 1.0 or USB 2.0.[/b]

I'm not sure I agree with their definitions; maybe it's me.

[Eudyptes]

Antelope make the correct distinction between "USB audio class" and the more general USB (host to device) spec as in USB 1, USB 2 and now USB 3. 

They are also correct to say a "USB audio class 1" device could be used over USB 1 or USB 2, but I would have expected that to max out at 24/96 but as you say it goes higher, which seems odd.

And as you posted from their own docs/manual this is what they write about UH1 and UH2:



Surely, someone somewhere at Antelope has got their wires crossed.

Antelope's email description of UH1 seems to correspond to UF1 in their manual, which leaves me wondering about the rest.

I agree with you, it's a bit of a muddle.