Jump to content
IGNORED

HQPlayer's Network Audio Adapter


Recommended Posts

Yeah I don't know, I wanted to use your optimized NAA without needing to install HQP OS or minimal Ubuntu for the NAA. 

 

I added the default route on the NAA and even disabled the 2nd network adapter on the bios 

 

The thing is there is connectivity I can ping from one to another, do you think that something very specific on the networkaudiod process might create the conflict? I am going to try another image just to see 

 

 

Link to comment
27 minutes ago, luisma said:

And is it normal HQPe remains showing the previous allo Bridge NAA even though is disconnected? 

 

It shows what ever is "Current" because what is being used at the moment cannot be detected. So the currently selected device is just shown from the configuration file. What ever else than "Current" is actually detected on the network. That is lovely side effect of stateless web interfaces.

 

59 minutes ago, luisma said:

Yeah I don't know, I wanted to use your optimized NAA without needing to install HQP OS or minimal Ubuntu for the NAA. 

 

I added the default route on the NAA and even disabled the 2nd network adapter on the bios 

 

The thing is there is connectivity I can ping from one to another, do you think that something very specific on the networkaudiod process might create the conflict? I am going to try another image just to see

 

For me things work without issue when everything is connected to my regular home network.

 

ping doesn't really tell much because it works on top of basic peer specific IP addresses. You'd need to test multicast between devices to figure out.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

Thanks Jussi, on the HQPE running Ubuntu I see with tcpdump coming from the NAA

 

13:50:00.541597 IP6 fe80::72e9:4cff:fe68:3676 > ff02::16: HBH ICMP6, multicast listener report v2, 2 group record(s), len

 

That specific address iff02::16 is the Multicast Listener Discover (MLD) used by routers to discover multicast listeners.

 

I think what's happening is that since there is no  router in my chain (for isolation purposes) my HQPE Ubuntu just discards the packet and doesn't respond with "I'm a listener" 

 

I can try to do the workaround and I am actually looking for it to see if I could add the MLD within Ubuntu, would it be too much to ask for your image to include ff02::1 to address all nodes in the local segment? of course if this doesn't create too much overhead for you, I can add a fake router as well but since you build your own image I was wondering if in the IPv6 stack during kernel compiling or on your NAA app you could add that option without taking the chance of breaking anything else.

 

I know you don't want to get too much specific with your images. If that's the case then no worries I will find a workaround or I can move in the AL direction (which I have been trying to avoid as I trust you capable of building a good minimal image for your NAA software)

 

I am assuming things here that may not be accurate, apologies in advance if these are inaccurate.

Thanks

 

 

 

Link to comment

It could be also IPv4 side lacking configuration. When IPv6 is enabled, both HQPlayer and NAA operate in dual-stack mode supporting both IPv4 and IPv6 (since IPv6 is backwards compatible with IPv4). If you disable IPv6, then only traditional IPv4 is used. There is no option at the moment to operate in IPv6-only.

 

Both my HQPlayer OS image and NAA image use the IPv4 + IPv6 dual stack mode by default.

 

IPv6 multicast ICMP's are sent by kernel to manage multicast subscriptions across smart switches etc.

 

P.S. Point-to-point wired Ethernet reduces isolation, having a switch there increases the isolation.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
37 minutes ago, Miska said:

There is no option at the moment to operate in IPv6-only.

Then that's the issue, there is no ip4 dhcp server in this config

 

39 minutes ago, Miska said:

 

P.S. Point-to-point wired Ethernet reduces isolation, having a switch there increases the isolation

I have powered both computers from isolated LPS, adding a switch "non LPS ready" thought it will add ripple and other EMI to the path 

 

Thanks for clarifying 

Link to comment
4 hours ago, luisma said:

I have powered both computers from isolated LPS, adding a switch "non LPS ready" thought it will add ripple and other EMI to the path 

 

As long as you don't use shielded cables, ethernet is isolated at both ends using transformers. Since the signaling is differential, any common mode noise is canceled in the transformers. If you have a switch in the middle, you double the number of isolation barriers between HQPlayer computer and NAA.

 

NAA is designed for normal home network infra and works well in such environment. I'd say 90% of the problems are from cases where someone is trying to use point-to-point connections - which is not a network anymore.

 

HQPlayer OS image still usually works in such environments too, because it bridges multiple ethernet adapters together. Essentially creating a switch inside the HQPE computer and thus joining NAA to rest of the network. But this is not recommended configuration as it puts unnecessary extra strain on the HQPE computer. It is better to use a good switch instead since switches are specialized hardware and software just for that single purpose and generally do better job at it.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

Hi @Miska
Using NAA (3.5.4-38) with DietPi (6.19.7) on ALLO USBridge (Sparky Kernel version 3.10.38) to feed my DAC. 
I've recently discovered something weird when playing (PCM / DSD whatever the resolution):
dmesg output (about every second): delay: estimated 0, actual X 


DietPi ssh terminal: 

> dmsg -wH 

DAC ON

[Feb. 4 16:03] usb 4-1.1: new high-speed USB device number 7 using aotg_hcd 
[+0,000025] aotg_hcd aotg_hcd.1: device addr: 0x00000082 
[+0.080071] aotg_hcd aotg_hcd.1: <DISABLE EP> ep0 index 0 from ep[in] 
[+0.019897] aotg_hcd aotg_hcd.1: device addr: 0x00000082
[+0.000719] usb 4-1.1: New USB device found, idVendor = 16d0, idProduct = 071a 
[+0,000013] usb 4-1.1: New USB device strings: Mfr = 1, Product = 2, SerialNumber = 3 
[+0,000008] usb 4-1.1: Product: Combo384 Amanero 
[+0,000006] usb 4-1.1: Manufacturer: Amanero Technologies 
[+0,000007] usb 4-1.1: SerialNumber: 413-001 
PLAY

[Feb. 4 16:04] iso_packets: 0, bInterval: 1, urb_interval: 1, reg_con: 0x4 
[+0.000160] iso_packets: 0, bInterval: 6, urb_interval: 32, reg_con: 0x4
[+0,198317] delay: estimated 0, actual 177 
[+0,999997] delay: estimated 0, actual 177
[+1.000004] delay: estimated 0, actual 177 
[+0,999999] delay: estimated 0, actual 177
[+1.000126] delay: estimated 0, actual 177 
[+0,999998] delay: estimated 0, actual 177 
[+1.000002] delay: estimated 0, actual 177 
[+0,999999] delay: estimated 0, actual 177 
[+1.000124] delay: estimated 0, actual 177

.... and so on until PAUSE

 

Sound is perfect, no issue !

No such messages when playing from RoonBridge .

I've looked at https://www.raspberrypi.org/forums/viewtopic.php?t=15204#p407215 seems same messages output using MPD ?

Any idea why these messages ?
 

I've cross posted to ALLO support but don't know yet where networkaudiod logs are located. 

Thank you ...

ROON > HQPlayer > USBridge > T+A DAC8 DSD > NAD M22 (Ncore Hypex) > Harbeth SLH5+
Setup details

Link to comment

Using my Ubuntu NAA with my Ubuntu HQPe
Funny that the HQPe log gets refreshed when the web interface gets refreshed not when the networkaudiod NAA boots but working good. This is the HQPE log.

======================================================================

  2019/02/04 19:57:16 asoundlib version: 1.1.3
  2019/02/04 19:57:16 Network Audio IPv6 support enabled
  2019/02/04 19:57:16 Discovery from ::
& 2019/02/04 19:57:17 Discovered network audio: name='naa01' version='Signalyst Network Audio Daemon 3.5.5'  @[fe80::72e9:4cff:fe68:3676%eno1]:43210
& 2019/02/04 19:57:17 Discovered network audio: name='naa01' version='Signalyst Network Audio Daemon 3.5.5'  @[fe80::72e9:4cff:fe68:3676%eno1]:43210
  2019/02/04 19:57:18  Network endpoint: Combo384 Amanero: USB Audio (hw:CARD=Amanero,DEV=0)
  2019/02/04 19:57:18  Network endpoint: HDA Intel HDMI: HDMI 4 (hw:CARD=HDMI,DEV=10)
  2019/02/04 19:57:18  Network endpoint: HDA Intel HDMI: HDMI 0 (hw:CARD=HDMI,DEV=3)
  2019/02/04 19:57:18  Network endpoint: HDA Intel HDMI: HDMI 1 (hw:CARD=HDMI,DEV=7)
  2019/02/04 19:57:18  Network endpoint: HDA Intel HDMI: HDMI 2 (hw:CARD=HDMI,DEV=8)
  2019/02/04 19:57:18  Network endpoint: HDA Intel HDMI: HDMI 3 (hw:CARD=HDMI,DEV=9)
  2019/02/04 19:57:18  Network endpoint: HDA Intel PCH: ALC662 rev3 Analog (hw:CARD=PCH,DEV=0)
  2019/02/04 19:57:18  Network endpoint: HDA Intel PCH: ALC662 rev3 Digital (hw:CARD=PCH,DEV=1)
- 2019/02/04 19:57:18 Network engine stopping...

And this is the output on the NAA, when it starts and gets discovered.
root@naa01:~# networkaudiod
[networkaudiod] (1687): networkaudiod Copyright (C) 2011-2018 Jussi Laako / Signalyst. All rights reserved.
[networkaudiod] (1687): asoundlib version: 1.1.3
[networkaudiod] (1687): discovery from [fe80::96c6:91ff:fe1b:25f%enp2s0]:33413
[networkaudiod] (1687): discovery from [fe80::96c6:91ff:fe1b:25f%enp2s0]:33413
[networkaudiod] (1687): connection from [fe80::96c6:91ff:fe1b:25f%enp2s0]:34642
[networkaudiod] (1687): Found ALSA device: hw:CARD=HDMI,DEV=3 - HDA Intel HDMI: HDMI 0
[networkaudiod] (1687): Found ALSA device: hw:CARD=HDMI,DEV=7 - HDA Intel HDMI: HDMI 1
[networkaudiod] (1687): Found ALSA device: hw:CARD=HDMI,DEV=8 - HDA Intel HDMI: HDMI 2
[networkaudiod] (1687): Found ALSA device: hw:CARD=HDMI,DEV=9 - HDA Intel HDMI: HDMI 3
[networkaudiod] (1687): Found ALSA device: hw:CARD=HDMI,DEV=10 - HDA Intel HDMI: HDMI 4
[networkaudiod] (1687): Found ALSA device: hw:CARD=PCH,DEV=0 - HDA Intel PCH: ALC662 rev3 Analog
[networkaudiod] (1687): Found ALSA device: hw:CARD=PCH,DEV=1 - HDA Intel PCH: ALC662 rev3 Digital
[networkaudiod] (1687): Found ALSA device: hw:CARD=Amanero,DEV=0 - Combo384 Amanero: USB Audio
[networkaudiod] (1687): begin disconnection
[networkaudiod] (1687): ALSA backend uninitialized
[networkaudiod] (1687): disconnected [fe80::96c6:91ff:fe1b:25f%enp2s0]:34642

======================================================================

 

All good so far.

 

Now same hardware booting from your latest image from USB it is a mystery to me why it doesn't work, doing this again.

Oh WAIT ... now it does work on this hardware? what?

 

======================================================================

Log of HQPE
  2019/02/04 20:11:41 asoundlib version: 1.1.3
  2019/02/04 20:11:41 Network Audio IPv6 support enabled
  2019/02/04 20:11:41 Discovery from ::
& 2019/02/04 20:11:42 Discovered network audio: name='naa' version='Signalyst Network Audio Daemon 3.5.5'  @[fe80::72e9:4cff:fe68:3676%eno1]:43210
& 2019/02/04 20:11:42 Discovered network audio: name='naa' version='Signalyst Network Audio Daemon 3.5.5'  @[fe80::72e9:4cff:fe68:3676%eno1]:43210
  2019/02/04 20:11:43  Network endpoint: Combo384 Amanero: USB Audio (hw:CARD=Amanero,DEV=0)
  2019/02/04 20:11:43  Network endpoint: HDA Intel PCH: ALC662 rev3 Analog (hw:CARD=PCH,DEV=0)
  2019/02/04 20:11:43  Network endpoint: HDA Intel PCH: ALC662 rev3 Digital (hw:CARD=PCH,DEV=1)
- 2019/02/04 20:11:43 Network engine stopping...
  2019/02/04 20:12:07 Web configuration written
% 2019/02/04 20:12:09 Reloading configuration...
- 2019/02/04 20:12:09 Control ended from [::ffff:127.0.0.1]:37734
- 2019/02/04 20:12:09 Network engine stopping...

  2019/02/04 20:12:09 libDSP version 20.6.0
  2019/02/04 20:12:09 HQPlayer Engine version 4.10.0
  2019/02/04 20:12:09 Number of processor cores: 2
  2019/02/04 20:12:09 DSP thread pools disabled
  2019/02/04 20:12:09 Pipelined DSP disabled
  2019/02/04 20:12:09 Audio engine: network
  2019/02/04 20:12:09 Network Audio IPv6 support enabled
  2019/02/04 20:12:09 Discovery from ::
& 2019/02/04 20:12:10 Discovered network audio: name='naa' version='Signalyst Network Audio Daemon 3.5.5'  @[fe80::72e9:4cff:fe68:3676%eno1]:43210
& 2019/02/04 20:12:10 Discovered network audio: name='naa' version='Signalyst Network Audio Daemon 3.5.5'  @[fe80::72e9:4cff:fe68:3676%eno1]:43210
  2019/02/04 20:12:11  Network endpoint: Combo384 Amanero: USB Audio (hw:CARD=Amanero,DEV=0)
  2019/02/04 20:12:11  Network endpoint: HDA Intel PCH: ALC662 rev3 Analog (hw:CARD=PCH,DEV=0)
  2019/02/04 20:12:11  Network endpoint: HDA Intel PCH: ALC662 rev3 Digital (hw:CARD=PCH,DEV=1)
  2019/02/04 20:12:11 Discovered 1 Network Audio Adapters
  2019/02/04 20:12:11 Input set channels: 2 (2)
+ 2019/02/04 20:12:11 Connect to [fe80::72e9:4cff:fe68:3676%eno1]:43210
  2019/02/04 20:12:11 Network format: 32000/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 44100/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 48000/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 88200/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 96000/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 176400/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 192000/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 352800/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 384000/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 705600/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 768000/32/2 [pcm] 
  2019/02/04 20:12:11 Network format: 2822400/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 3072000/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 5644800/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 6144000/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 11289600/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 12288000/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 22579200/1/2 [dsd] 
  2019/02/04 20:12:11 Network format: 24576000/1/2 [dsd] 
  2019/02/04 20:12:11 Output set channels: 2 (2)
  2019/02/04 20:12:11 Hardware volume control disabled
  2019/02/04 20:12:11 Volume max: 0
  2019/02/04 20:12:11 Volume min: -60
  2019/02/04 20:12:11 Volume scaler: 0.75
  2019/02/04 20:12:11 Set volume: -30
  2019/02/04 20:12:11 Automatic rate family enabled
  2019/02/04 20:12:11 DirectSDM disabled
  2019/02/04 20:12:11 SDM conversion parameters (5, 0)
  2019/02/04 20:12:11 DSD 6 dB gain enabled
  2019/02/04 20:12:11 Initial parallel threads: 4
  2019/02/04 20:12:11 Number of logical cores: 4
& 2019/02/04 20:12:11 Set transport (240): 
  2019/02/04 20:12:11 Set filter: 9
  2019/02/04 20:12:11 Set dither: 5
  2019/02/04 20:12:11 Set oversampling: 9
  2019/02/04 20:12:11 Set modulator: 6
  2019/02/04 20:12:11 Set filter: 6 / 9
  2019/02/04 20:12:11 Set dither: 5
  2019/02/04 20:12:11 Set oversampling: 7 / 9
  2019/02/04 20:12:11 Set modulator: 6
  2019/02/04 20:12:11 Audio engine is normal
  2019/02/04 20:12:11 IntegratorM: IIR
  2019/02/04 20:12:11 Audio engine SDM mode enabled
  2019/02/04 20:12:11 Requested output rate: 12288000
  2019/02/04 20:12:11 Input set sampling rate: 12288000 (12288000)
  2019/02/04 20:12:11 Automatic output rate switching enabled
& 2019/02/04 20:12:11 Convolution disabled
  2019/02/04 20:12:11 Control active volume range: -60 - 0 dB
  2019/02/04 20:12:11 Enable UPnP Renderer functionality for interface: auto
  2019/02/04 20:12:11 Adding interface 'eno1' for UPnP Renderer
  2019/02/04 20:12:11 Adding interface 'wlp2s0' for UPnP Renderer
% 2019/02/04 20:12:11 ...configuration reloaded
  2019/02/04 20:12:17 asoundlib version: 1.1.3
  2019/02/04 20:12:17 Network Audio IPv6 support enabled
  2019/02/04 20:12:17 Discovery from ::
- 2019/02/04 20:12:19 Network engine stopping...

======================================================================

 

 

Cool time to do some Ubuntu vs Miska's image comparison

 

 

Link to comment

Hard to evaluate with wife asking for options for 14th Feb dinner, trying to listen without hurting her feelings and answering pre canned phrases

Anyways sounds kind of the same with the image and Ubuntu, in Genesis Abacab bass and drums are more upfront and the instrument separation is better with Miska's image, in Direstraits Private Investigations the guitar chords sound more crisp with my Ubuntu, further testing is required

 

Link to comment

Switching to the other hardware, roon controls are passed to HQPE and it seemed to be working but no audio at all, rediscovered the NAA same thing. 

Turned off the DAC and turned back on, NAA re initialized the DAC and now I have audio again

 

This little fanless Celeron I wanted to try with all the BIOS customizations and stripped of everything sounds very good

Thank Jussi

 

 

Link to comment
18 hours ago, Miska said:

 

No idea why you get those messages. Are you using same output format (rate) with Roon Bridge?

 

3.5.4 is a bit old version, I would also recommend to update to latest 3.5.5 build.

 

Yes @Miska 
Same rate with NAA and Roon Bridge (tested with PCM 44.1 and DSD 11.28)

 

- NAA: each seconds 

delay: estimated 0, actual xxx

- Roon Bridge: No messages

I will update to 3.5.5 and let your know ...

ROON > HQPlayer > USBridge > T+A DAC8 DSD > NAD M22 (Ncore Hypex) > Harbeth SLH5+
Setup details

Link to comment

 

For quite a while I have been using a SOtM SMS-200 as an NAA with Roon and HQPlayer into an Audiolab MDAC+ and have been able to upsample to DSD 128 or output DSD 64, 128 and 256 files using DirectSDM.

 

I'm not sure what has change but I recently notice everything was being output as PCM even DSD files when Direct SDM is selected. I am unable to select SDM output from the drop menu in the main GUI window. 

 

Any help in determining what has caused this is greatly appreciated.

Link to comment
1 hour ago, dm68 said:

For quite a while I have been using a SOtM SMS-200 as an NAA with Roon and HQPlayer into an Audiolab MDAC+ and have been able to upsample to DSD 128 or output DSD 64, 128 and 256 files using DirectSDM.

 

I'm not sure what has change but I recently notice everything was being output as PCM even DSD files when Direct SDM is selected. I am unable to select SDM output from the drop menu in the main GUI window. 

 

Any help in determining what has caused this is greatly appreciated.

 

I see two possibilities. Either kernel update to sMS-200 has removed native DSD support for your DAC, or you need to have "DoP" selected as "SDM Pack" method.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
On 2/5/2019 at 1:30 AM, Miska said:

 

No idea why you get those messages. Are you using same output format (rate) with Roon Bridge?

 

3.5.4 is a bit old version, I would also recommend to update to latest 3.5.5 build.

 

Hi @Miska, just updated NAA from 3.5.4 to 3.5.5 (USBridge / DietPi 6.21).

Still strange dmesg like delay: estimated  X, actual Y messages playing to NAA at any rate and no such messages with RoonBridge.

Anything else to investigate ?

> root@USBridge:~# dmesg -wH

[févr. 7 17:21] delay: estimated 1411, actual 0
[  +1,436056] delay: estimated 0, actual 706
[  +0,999999] delay: estimated 0, actual 706
[  +0,999996] delay: estimated 0, actual 706
[  +1,000126] delay: estimated 0, actual 706
[  +1,999999] delay: estimated 0, actual 706
[  +1,000124] delay: estimated 0, actual 706
[  +1,000003] delay: estimated 0, actual 706
[  +0,999997] delay: estimated 0, actual 706
[  +1,000000] delay: estimated 0, actual 706
[  +3,119467] delay: estimated 1411, actual 0
[  +0,878039] delay: estimated 0, actual 706
[  +3,000125] delay: estimated 0, actual 706
[  +0,999997] delay: estimated 0, actual 706
[  +1,000000] delay: estimated 0, actual 706

and so on ...

 

ROON > HQPlayer > USBridge > T+A DAC8 DSD > NAD M22 (Ncore Hypex) > Harbeth SLH5+
Setup details

Link to comment
3 hours ago, volpone said:

 

Hi @Miska, just updated NAA from 3.5.4 to 3.5.5 (USBridge / DietPi 6.21).

Still strange dmesg like delay: estimated  X, actual Y messages playing to NAA at any rate and no such messages with RoonBridge.

Anything else to investigate ?

 

 

I don't know, I have not seen such on my devices yet...

 

Something from the kernel, so probably something hardware related.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...