Volumio 3 RC1 - Open Beta Testing

I had problems with Qobuz (not using volumio) when playing from more than one device with my single user account.

Hi Guys, I have reported a V3RC1 UI problem with Qobuz which is annoying here : Qobuz on Volumio - #9 by fabriceo
Please take a look, this doesn’t seem to be extraordinary bug and all virtuoso users who have paid for Qobuz are impacted. Many thanks for supporting us!
fabriceo

Hi

Latest version 3.163 seems not as good as previous.
Music (webradio or spotify) would stop after sometimes. System irresponsive to do anything, will take time to come back online.
Log http://logs.volumio.org/volumio/aT2fyL9.html

Hardware
Raspberry Pi 3
DAC Allo Piano 2.1 + Kali reclocker
Wifi dongle TP-Link … realtek (don’t remember the model)
mSata SSD from Suptronics not recognized.

Hope the logs helps you

Seems that there is little attention to it, but it relates to the same issue I have raised on 24/10 in this thread.
Afterwards, submitting a few logs (w/o particularly raising the issue on the forum) and attempting to push an individual message to support on the forum I decided to examine it (I have 4 active Volumio devices and 3 DACs to test around) and as a Premium user I felt this is a nasty issue of losing the anticipated quality of the sound…

Issue:
a) irrespective of DSD settings (direct or DoP) and resampling (off or up to 768kHz target), USB output is always PCM limited to max 384kHz
b) this is observed on 3.* versions, when logged in as Superstar
c) no problem at the same setup used but logged out / operating as Basic user
d) no issues with 2.9* versions at Superstar or Basic

Hypothesis:
Based on my tests, there is a guess that the issue might be related to the implementation of the multi-room sync playback in Volumio 3.
For multi-room sync ALSA is deployed via conversion (creating loopback devices for PCM streams rather than HW) which cannot be disabled. And this might limit Premium users to lower quality than Basic users.

See tests & logs below to illustrate my idea.


Here are 2 test logs to illustrate it with the following setup
Setup: Raspberry Pi 4 8Gb + Topping D90SE DAC (ESS9038PRO) on USB (max native DSD512, PCM768). Only one device is active on the network (to avoid confusion with multi-room).
Fresh boot, all plugins disabled, output → USB, DSD direct, no mixer, no resampling. Live log enabled, then all setup reconfirmed and played test sequence.
10 test files in a sequence (PCM 44.1->88.2->176.4->352.8->44.1 MQA, then DSD64->128>256->512, them PCM 768)
First part of log - Superstar user / then logged out (Basic) for the second.

1) Volumio 3.159 test log:
http://logs.volumio.org/volumiobuster/650lpx5.html

Results confirm b+c issue (see above):
Anything = OK (i.e. DAC received on USB bitstream that matches the file correctly) for the Basic user (when not logged in).
For Superstar:
= OK: for PCM 44.1-352.8
= TO CONSIDER: 44.1 MQA played back at 44.1 and MQA was not available for the unfold (Volumio 2.917 did the same unfolding → 352.8)
= FAILED: DSD output always PCM (352.8k for DSD64, 284k for DSD128-256-512 and PCM768)
Note: when upsampling is enabled later, it always maxes out at 384kHz (irrespective of 768kHz target rate enabled)


2) Volumio 2.917 test log:
http://logs.volumio.org/volumio/AK775hy.html

Results confirm d (above)
= ALL TESTS OK (including MQA unfolded correctly for both Premium and logged-off/Basic users)


Few extra tests

The same results are achieved using other DAC - Topping E50 (ESS9068, USB max native DSD512, PCM768). Did not attach logs for these to keep it shorter.
Test of another Pi device (Pi 400 4Gb) - same results on both DACs.
Test of x86 setup - same results again…


When tested MoOde 7.60 (5.4.7 kernel based) on the same Pi 4 + Topping D90SE and E50 setup - discovered that:

  • when multi-room is enabled, loopback device is created and the setup works at PCM up to 384kHz as a limit (and MoOde states it in in-app help)
  • when multi-room is disabled, none of these limitations occur

CONCLUSIONS

  1. the synced multi-room implementation might be the issue
  2. as multi-room is available for Premium users only, they unwillingly fall victims to this implementation:
  • Basic users do not get Manifest UI (which I really love) and multi-room, but they get audiophile sound :slight_smile:
  • Premium users get UI & multi-room but due to the latter, they do not get the option to enjoy the best sound
    This is a shame from the perspective of Premium user’s perspective - likely a discerning and audiophile crowd…
    3) a quick fix for this trade-off between quality vs features is an option to switch off multi-room is recommended to be made available for Premium users
    As a Premium user, I would prefer Manifest UI + good sound over less frequently used synced multi-room. If I enable it, I would accept lower quality knowingly.
  1. In a long-term, obviously a proper fix would be appreciated anyway…

I hope this helps the developers to do the fantastic job they do for us - Volumio is great and I hope it will become even better. :v:

2 Likes

Volumio 3 now supports multi-channel output via HDMI. Works fine on my Mini PC with Intel UHD 600 graphics . Is it possible to display not only the file format that is being played, but also the number of channels?

Manifest is really awesome!
I would really appreciate it if multiroom can be switched off in the settings.
Many do not use it and do not need the option permanently available in the UI.

Option, would be great but is not urgent:
It is also not always necessary that I can switch from every player to everyone that is in operation in the network. I would appreciate it if this can also be switched off in the settings.

volumio 3.163 : you can disable enable and set latency in sources settings :wink:

1 Like

Interesting observation, but I guess it doesn’t apply to my case as when I was logged in it was as a basic user, since I really only use the MPD part of the Volumio and control it via ncmpcpp or android app.

Nevertheless, i tried logging out completely, and the problem remained. The stream is always converted to PCM, but what is different to what your observation is that in my case the DSD gets upsampled as high as 1536000

so for
DSD512 and DSD1024 the rate is 1536000
DSD256 the rate is 1411200
DSD64 the rate is 352800

the current stream0

#/proc/asound/card5#cat stream0
DENAFRIPS USB HiRes Audio at usb-0000:01:00.0-1.2, high speed : USB Audio                                                                                                                                                                                                                                      
                                                                                                                                                                                                                                                                                                               
Playback:                                                                                                                                                                                                                                                                                                      
  Status: Stop                                                                                                                                                                                                                                                                                                 
  Interface 1
    Altset 1
    Format: S32_LE
    Channels: 2
    Endpoint: 1 OUT (ASYNC)
    Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000, 705600, 768000, 1411200, 1536000
    Data packet interval: 125 us
    Bits: 32
  Interface 1
    Altset 2
    Format: SPECIAL DSD_U32_BE
    Channels: 2
    Endpoint: 1 OUT (ASYNC)
    Rates: 44100, 48000, 88200, 96000, 176400, 192000, 352800, 384000, 705600, 768000, 1411200, 1536000
    Data packet interval: 125 us
    Bits: 32

setting dop to yes or no doesnt make any difference.

here is the mpd.log releveant part

Dec 05 16:37 : playlist: play 0:"NAS/PVE1/DSD/Vexel do minante - DSD Piano instrumental music - 20210-2019 (Classical, Audiophile, test)/01-Waver-DSD1024.dff"                                                                                                                                                 
Dec 05 16:37 : client: [1] command returned 0                                                                                                                                                                                                                                                                  
Dec 05 16:37 : decoder_thread: probing plugin dsdiff                                                                                                                                                                                                                                                           
Dec 05 16:37 : playlist: queue song 1:"NAS/PVE1/DSD/Vexel do minante - DSD Piano instrumental music - 20210-2019 (Classical, Audiophile, test)/02-Babel-DSD512.dff"                                                                                                                                            
Dec 05 16:37 : decoder: audio_format=6144000:dsd:2, seekable=true                                                                                                                                                                                                                                              
Dec 05 16:37 : alsa_output: opened hw:5,0 type=HW                                                                                                                                                                                                                                                              
Dec 05 16:37 : client: [1] process command "idle"                                                                                                                                                                                                                                                              
Dec 05 16:37 : client: [1] command returned 1                                                                                                                                                                                                                                                                  
Dec 05 16:37 : alsa_output: buffer: size=384..131072 time=250..85334                                                                                                                                                                                                                                           
Dec 05 16:37 : alsa_output: period: size=192..65536 time=125..42667
Dec 05 16:37 : alsa_output: default period_time = buffer_time/4 = 85333/4 = 21333
Dec 05 16:37 : client: [1] process command "status"
Dec 05 16:37 : alsa_output: format=S32_LE (Signed 32 bit Little Endian)
Dec 05 16:37 : alsa_output: buffer_size=131072 period_size=32768
Dec 05 16:37 : output: opened "Denafrips Ares II" (alsa) audio_format=1536000:32:2
Dec 05 16:37 : soxr: soxr engine 'double-precision'
Dec 05 16:37 : soxr: samplerate conversion ratio to 0.25
Dec 05 16:37 : output: converting in=6144000:dsd:2 -> f=6144000:16:2 -> out=1536000:32:2
Dec 05 16:37 : exception: Failed to open mixer for 'Denafrips Ares II': no such mixer control: PCM```

Hi
most likely the problems with DSD files and limited sampling rate is due to the “type plug” used in the alsa device in file /etc/asound.conf

# There is always a plug before the hardware to be safe
pcm.volumioOutput {
    type plug
    slave.pcm "volumioHw"
}

because “plug” cannot passthrough the native dsd and has limitations regarding formats or bitrate.

2 Likes

In addition to post #622 Nov 20, 11:23 PM I update to the latest version but it still no t working.
The DLNA servers are not discovered (just few time random) the cusious thing that if I set some favorite from dlna server (in the rare times when the servers was visible) they are still working if the server is not visible. So, the connection from the dlna to volumio is worknig but the discovery function not. If the dlna discovery will be no working in the new version 3 should be at least a way to add manually a dlna server root address or IP in some way?

Perfect and thank you very much!
Indeed, great switches!

Anyone provide some feedback on why I receive no metadata on the local device, RC1 3.163? 3.152 was the last build that was working for me. Remotely via HTTP works fine, just no metadata on the local device’s display.

I have re-flashed, checked DNS, and FW w/ out success. Running RC1 3.163, RPI 4B, IOT subnet, and no FW ACLs. Also, a yearly SS subscriber.

Log: http://logs.volumio.org/volumio/jFiHZVq.html

Thanks

1 Like

Hi,

Device: raspberry4B , HifiBerry DAC+, WaveShare 5inch HDMI LCD
Volumio: 3.1.63, Thu 02 Dec 20210 06:37:00 PM CET
Plugins: PeppyMeter for Volumio 0.2.1, TouchDisplay 1.3.0, Spotify 2.0.3

on some webradio stations i get an Error :

image

This happens on 128kbit sources, but not for all. Other resolutions seem to work .

logfile: http://logs.volumio.org/volumio/q17Y3Hf.html

Can anybody tell me, what’s going wrong?

Thx, Carsten

Hi,
1.spotify is a premium feature? I need premium account in order to use spotify plugin?
2. If I subscribe for only one month then subscribe again after 3 months, my saved configuration, will be lost, web radio… Etc?
Thanks.

1-Spotify is a free feature.
2-no idea…

ohh, is today the day (7th Dec) Volumio 3 is released?
Maybe time difference but not up yet, unless there is a download link or will it appear as an update in the Google app store?
Should be good, thanks to the devs and testers!

1 Like

In China it is, in Europe not yet.

Here the countdown clock is running: :wink:

I could upgrade my V2 already yesterday to V3.165!
Is this another beta or the final release?

Are you sure it wasn’t 3.163 (the latest beta)?

2 Likes