[Nova40]

Yup... it's a bummer.  So should I start looking for a new DAC, or can I achieve 192/24bit with SPDIF or something?

 

[Anda]

I you like the sound you can get a Gustard U12. Are you located in the UK (based on your Cambridge and B&W gear)?
 
It's available here:
http://www.ebay.co.uk/itm/GUSTARD-U12-Native-32bit-384kHz-XMOS-DAC-0-1ppm-U10-upgrade-Black-Silver-/261596415436

 

[Nova40]

Actually I'm in the United States, but that's a good price and looks pretty nice.  Can this be used at 384/32bit under linux without a driver?  Also, I'm considering going for a balanced dac design, but I don't think it'll fit my current budget... 

 

[Anda]

Ok, I think you can get it slightly cheaper from Asia. Honestly don't know about 32/384. My spdif input is limited to 24/96, so I am not able to test it properly.

 

[Nova40]

Yeah, my goal is just to be able to play 24/192 because I have quite a number of flac files that are 24/192.  But so far I've heard that you need a driver to do that.  How certain are you that the XMOS chip in the DacMagic Plus can play 24/192 under linux?

 

[undersys]

  Yeah, my goal is just to be able to play 24/192 because I have quite a number of flac files that are 24/192.  But so far I've heard that you need a driver to do that.  How certain are you that the XMOS chip in the DacMagic Plus can play 24/192 under linux?

 
Like I said a few a posts back you can use the Linux in kernel USB2 support for that resolution.
You do not need a "driver" for USB2 audio compliant devices.
My Audio-GD just works at 24/192+ resolutions.

 

[Anda]

Just to confirm it, this is the output when playing a 24/192 flac file on my U10:

Spoiler: Warning: Spoiler!

$ cat /proc/asound/card1/stream0 
XMOS XMOS USB Audio 2.0 at usb-0000:00:1d.7-2, high speed : USB Audio
 
Playback:
  Status: Running
    Interface = 1
    Altset = 1
    Packet Size = 247
    Momentary freq = 192000 Hz (0x18.0000)
    Feedback Format = 16.16
  Interface 1
    Altset 1
    Format: S32_LE
    Channels: 2
    Endpoint: 1 OUT (ASYNC)
    Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000
    Data packet interval: 125 us
 
Capture:
  Status: Stop
  Interface 2
    Altset 1
    Format: S32_LE
    Channels: 2
    Endpoint: 2 IN (ASYNC)
    Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000
    Data packet interval: 125 us

 

[Nova40]

What do I have to change to get kernel USB2 support working?

 

[undersys]

  What do I have to change to get kernel USB2 support working?

 
Kernel option for USB audio is "CONFIG_SND_USB_AUDIO" (USB1 and 2)
I assume your kernel already has USB2 support.
I would also assume that "CONFIG_SND_USB_AUDIO" is already part of your kernel. 
Since your getting audio playback via USB 1.0.
 
 
It seems the DacMagic is operating in USB 1.0 audio mode, hence the limit on playback.
See this page for details on the limits :
http://www.thewelltemperedcomputer.com/KB/USB.html
 
 
Maybe a newer kernel is required to support the DacMagic 100 ? 
Have you asked Cambridge ?

 

[watchpocket]

So I have four different mpd clients: Cantata, GMPC, Glurp and mpc. 
 
All of them except Cantata will play .dsf (DSD) files.  I've looked through the settings of each of these clients to see how they might differ from one another, and I cannot figure out what might be different about Cantata such that it refuses to play .dsf and/or .dff files. 
 
My DAC is the Geek Out 1000.
 
I've played .dsf files successfully in any one of the three clients that can do that, then closed that client, unplugged and re-plugged the Geek Out (makiing sure it was selected in sound-settings), killed mpd and re-started it, and then opened Cantata to see if it would play .dsf.  I've changed the order in which I made these steps several times.  I'm using ALSA, Pulseaudio is turned off. ReplayGain is turned off.
 
All I get with Cantata are sample rates of either 44100 or 48000.
 
My OS is Ubuntu-MATE 14.04 LTS on a System76 Wild Dog desktop unit.
 
Anyone have a clue why all the other clients work fine but Cantata?  Do I need a special plugin to play .dsf / .dff files in Cantata?

 

[julian67]

So I have four different mpd clients: Cantata, GMPC, Glurp and mpc. 

All of them [COLOR=0000FF]except[/COLOR] Cantata will play .dsf (DSD) files.  I've looked through the settings of each of these clients to see how they might differ from one another, and I cannot figure out what might be different about Cantata such that it refuses to play .dsf and/or .dff files. 

My DAC is the Geek Out 1000.

I've played .dsf files successfully in any one of the three clients that can do that, then closed that client, unplugged and re-plugged the Geek Out (makiing sure it was selected in sound-settings), killed mpd and re-started it, and then opened Cantata to see if it would play .dsf.  I've changed the order in which I made these steps several times.  I'm using ALSA, Pulseaudio is turned off. ReplayGain is turned off.

All I get with Cantata are sample rates of either 44100 or 48000.

My OS is Ubuntu-MATE 14.04 LTS on a System76 Wild Dog desktop unit.

Anyone have a clue why all the other clients work fine but Cantata?  Do I need a special plugin to play .dsf / .dff files in Cantata?

Maybe you should attach your ~/.asoundrc and your working mpdconf and any cantata config files so people have a chance to help.

In any case gmpc is the only GUI MPD client which supports all of MPD's features so I'd suggest that anyone needing anything more than basic features and simple metadata support should just stick with gmpc (or a console client such as ncmpc/ncmpcpp/mpc). I've been using MPD for quite a few years and I think I must have tried every GUI client at least once. Sooner or later, even with more ambitious or extended clients like Cantata, you find that a feature is missing and only gmpc can do what you want. And on a GTK desktop like Mate gmpc surely makes more sense than a QT client?

 

[yay101]

Basically it comes down to how mpd works, Cantata probably hasn't been updated to allow adding dsd to playlists, realistically one of the very few things mpd actually listens to the frontend for.

 

[kazcou]

For those who want directly use there SACDISO with mpd, there is a branch (http://git.musicpd.org/cgit/manisiutkin/mpd.git)
who support it and also DVDAISO. But you have to compile it.
 
 

 

[watchpocket]

Maybe you should attach your ~/.asoundrc and your working mpdconf and any cantata config files so people have a chance to help.

 
Why certainly. I didn't attach them originally since the other clients work fine, didn't think the problem would be with the configs but you never know, the problem can always be with the configs and any perverse portmanteau-like constructions they may contain.
 
> And on a GTK desktop like Mate gmpc surely makes more sense than a QT client?
 
I do heart and I do use gmpc but what problems is using a QT client actually expected to pose on Ubuntu?  Cantata is a stellar program but for this one issue and I don't know for sure yet, but I'll bet the Cantata developers simply haven't made accommodations for delivering or facilitating DSD.  If not, I'll be fiing a bug report with those lo-rez luddites.
 
asoundrc:
 

 [size=8px]pcm.!default { type plug slave.pcm "hw:1,0"; # slave.pcm "plughw:G1V5,0; }[/size]

 
.mpd/mpd.conf:
 

 [size=8px]### Config file for mpd (Music Player Daemon) v. 0.19.9 ### I'm using mpd with GMPC (Gnome Music Player Client) v. 11.8.16 on Ubuntu-MATE 14.04 LTS ### Last update: Friday  2015-03-27 10:35:53 PM  EDT ### This is version 0.3 of this file as I have more to learn about mpd ### This setup is for using mpd as a "user service," not as a "system service." ### See <https://help.ubuntu.com/community/MPD> ### See <https://bbs.archlinux.org/viewtopic.php?pid=1017873#p1017873> (post number 5) ### See <http://lacocina.nl/detect-alsa-output-capabilities> ### See mpd example config file: /etc/mpd.conf music_directory "/home/rj/Music" # Don't use ~/ playlist_directory "/home/rj/.mpd/playlists" # db_file "/home/rj/.mpd/mpd.db" database { plugin "simple" path "/home/rj/.mpd/mpd.db" } log_file "/home/rj/.mpd/mpd.log" pid_file "/home/rj/.mpd/mpd.pid" state_file "/home/rj/.mpd/mpdstate" # Where settings are maintained across re-boots & crashes. sticker_file "/home/rj/.mpd/mpd.sticker.db" # A database that manages dynamic information attached to songs. auto_update "yes" # Enables automatic update of MPD's database when files in music_directory are changed. zeroconf_enabled "no" # Don't publish 'service information.' max_playlist_length "90000" max_output_buffer_size "16384" bind_to_address "127.0.0.1" # Sets the address for the daemon to listen on. Careful attention should be paid if this is assigned # to anything other than the default. This setting can deny access to control of the daemon. port "6600" log_level "verbose" # For now. # Available settings are "default", "secure" or "verbose". The "verbose" setting is recommended # for troubleshooting, though can quickly stretch available resources on limited hardware storage. save_absolute_paths_in_playlists "yes" # Enables MPD to create playlists in a format usable by other music players. metadata_to_use "artist,album,title,track,name,genre,date,composer,performer,disc,comment" # Defines a list of tag types that will be extracted during the audio # file discovery process. # Input: input { plugin "curl" } # Decoder plugins: decoder { plugin "dsf" # Decodes DSF files containing DSDIFF data (e.g. SACD rips). [/size] [size=8px] enabled "yes" } decoder { plugin "dsdiff" # Decodes DFF files containing DSDIFF data (e.g. SACD rips). enabled "yes" lsbitfirst "no" # Decode the least significant bit first. Default is 'no'. } decoder { plugin "wildmidi" config_file "/etc/timidity/freepats.cfg" } # Audio ouput block for ALSA output: audio_output { type "alsa" name "Geek Out HD Audio 1V5, USB Audio" device "plughw:G1V5,0" format "*:*:*" dsd_usb "yes" # ESSENTIAL. Without this setting, I get -NO- DSD playback on the Geek Out. # <http://comments.gmane.org/gmane.comp.audio.musicpd.devel/2248> mixer_type "none" auto_resample "no" auto_channels "no" auto_format "no" replay_gain_handler "none" } # Audio output block for http: # audio_output { # type "httpd" # (Do this part later.) # } # Character Encoding: filesystem_charset "UTF-8" # If file or directory names don't display correctly for your locale, you may need to modify this setting. id3v1_encoding "UTF-8" # This setting controls the encoding that ID3v1 tags should be converted from. ### end of file ### [/size]

 
.config/cantata/cantata.conf:
 

 [size=8px][General] storeCoversInMpdDir=true storeLyricsInMpdDir=true currentConnection=- stopOnExit=true version=1.5.2 showMenubar=true showFullScreen=false mainWindowSize=@Size(1513 1001) showPlaylist=true playStream=false splitterState=@ByteArray(\0\0\0\xff\0\0\0\0\0\0\0\x2\0\0\x1|\0\0\x2:\x1\0\0\0\x1\x1\0\0\0\x1) page=ContextPage contextZoom=-1 contextSplitterState=@ByteArray(\0\0\0\xff\0\0\0\0\0\0\0\x3\0\0\x2\x6\0\0\x1\xe8\0\0\x1\xf7\x1\0\0\0\x6\x1\0\0\0\x1) contextAutoScroll=false contextTrackView=2 hiddenStreamCategories=@Invalid() searchCategory=artist showTimeRemaining=false hiddenOnlineProviders=@Invalid() forceSingleClick=true startHidden=false lyricProviders=lyrics.wikia.com, lyricstime.com, lyricsreg.com, lyricsmania.com, metrolyrics.com, azlyrics.com, songlyrics.com, elyrics.net, lyricsdownload.com, lyrics.com, lyrics$ inhibitSuspend=true storeBackdropsInMpdDir=true libraryArtistImage=true libraryView=icontop libraryYear=true albumsView=icontop albumSort=artist-year folderView=detailedtree playlistsView=detailedtree playListsStartClosed=true streamsView=detailedtree onlineView=detailedtree groupSingle=true useComposer=false filteredOnly=false showDeleteAction=true devicesView=detailedtree searchView=table playQueueView=grouped playQueueAutoExpand=true playQueueStartClosed=false playQueueScroll=true playQueueBackground=1 playQueueBackgroundOpacity=69 playQueueBackgroundBlur=0 playQueueBackgroundFile= playQueueConfirmClear=true playQueueSearch=true infoTooltips=true touchFriendly=true showStopButton=true showCoverWidget=true showRatingWidget=true useSystemTray=true showPopups=true minimiseOnClose=true startupState=show fetchCovers=true lang= hiddenPages=@Invalid() sidebar=17 monoSidebarIcons=false splitterAutoHide=false httpInterface=lo cdAuto=true cddbHost=freedb.freedb.org cddbPort=8880 paranoiaFull=true paranoiaNeverSkip=true useCddb=true wikipediaIntroOnly=true contextDarkBackground=true contextAlwaysCollapsed=false contextBackdrop=1 contextBackdropOpacity=30 contextBackdropBlur=0 contextBackdropFile= contextSwitchTime=1000 wikipediaLangs=en:en contextSlimPage=artist playQueueHeaderState="@ByteArray(P,\x95\xf\x1\0\0\0\f\0\0\0N\0\0\0\0\0\0\x1\x83\0\0\x1\xbd\0\0\x2\x42\0\0\0l\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\f\0\0\0\0\0\0\0\$ searchHeaderState="@ByteArray(P,\x95\xf\x1\0\0\0\t\0\0\0\x1d\0\0\0\x1b\0\0\0#\0\0\0\0\0\0\0\a\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\t\0\0\0\0\0\0\0\x1\0\0\0\x2\0\0\0\x3\0\0\0\x4\0$ mainWindowCollapsedSize=@Size(611 86) [Connection--] host=/home/rj/.local/share/cantata/mpd/socket port=6600 dir=~/Music passwd= coverName= streamUrl= [PreferencesDialog] size=@Size(891 1051) [PlayQueuePage] searchActive=true [DevicesPage] searchActive=false [DynamicPage] searchActive=false [PlaylistsPage] searchActive=false [FolderPage] searchActive=false [AlbumsPage] searchActive=false [LibraryPage] searchActive=false [Jamendo] format=ogg [PodcastSettingsDialog] size=@Size(550 160) [StreamProviderListDialog] size=@Size(466 876) [Magnatune] membership=streaming download=mp3 username= password= [Scrobbling] enabled=false loveEnabled=true[/size]

 

[watchpocket]

So it turns out that Cantata has a "basic" configuration mode where you can't set the host and port.  To set those, you need to got to Edit -> Preferences -> Collection, and create a new Collection with your preferred host and port settings. (Otherwise it's set to a local path socket.)  Once that's done, Cantata will allow mpd to play .dsf and .dff files at their own native, higher sample rates without re-sampling them down to 44100 or 48000.