Audiolinux Server configurations, Software, Hardware, and Listening Impressions

[bodiebill]

@hifi25nl

hi Piero, my audio PC has a 10GB Intel X520-D1 fiber network card. This card appears to work well with some Linux distro's, but not with all. Before I buy a license: will it work with Audiolinux?

[bodiebill]

57 minutes ago, renan said:

Well thank for your comment.

 

Finnally with usb3 key all interaction for installation are quicker. And know only 6 minutes to boot in RAM root. Not yet very low but much better than 20 minutes.

Thanks

 

Is that for the LXQt version?

Here, with a MX-ES 3.0 usb stick, AL Mini loading into RAM takes 2'30" after pushing the start button.

[bodiebill]

I guess the speed of the drive is the bottleneck, so the main thing here.

[bodiebill]

Using a similar setup here:

Windows control PC => AL server with HQPlayer embedded => AL endpoint with NAA

with the difference that I do upsample (to DSD) and therefore I need a powerful server. I am actually upgrading now to take advantage of HQP's cpu-demanding EC modulators.

 

2 hours ago, cat6man said:

In my system, the difference in playing back WAV versus FLAC is significant.

I have noticed the same. If the explanation is less processing, and therefore less pollution, on the server (which seems a likely explanation), then that means that the server plays an important role for the SQ. In which case I find it interesting that you are using the stock SMPS for it.

 

Quote

My choices seem to be either a NUC with AudioLinux running HQPlayer-embedded or a NUC running Jussi’s HQPlayer image directly. I’m planning on trying both and comparing them. Q: has anyone compared the AudioLinux/HQP combo versus running the NUC off Jussi’s image?

In the process of trying both myself. I have not noticed big differences in SQ. However, I may need AL to optimize some things if I go the EC route. Also, it is convenient to monitor temperatures and such.

 

Quote

2. Which version of AudioLinux? I had assumed that I wanted to run headless but I see people moving back and forth between lqxt and headless.

I am using the command-line version of AL (which is called headless, I don't know why as I also run the LXQt headless, with remote access to the GUI). It means less processes, so less pollution in theory. The AL menus are all conveniently accessible. I have a spare stick with the LXQt version for occasional gparted use.

 

Quote

4. H/W configuration, part 2: Any pros/cons for USB stick versus m.2 optane to use for loading ramroot? From what I read, I concluded that m.2 optane and no internal HDD/SSD is quieter and sufficient to run HQPlayer-embedded.

I am using a new 16GB optane m.2 stick in my endpoint. It loads AL into RAM in just over a minute. Initial impression (first half hour) is a bit cold/harsh as others have noted, but that seems to improve considerably after some hours of music. I was not expecting that as I play from RAM, but we know so little...

[bodiebill]

On 10/22/2019 at 8:49 PM, cat6man said:

time for a report after heading down this rabbit hole with you all.

...

Jussi's hqplayerd image on usb stick > old baseline system

hqplayer embedded under AL on usb stick > Jussi's hqplayerd image on usb stick

hqplayer embedded under AL in ramroot >> hqplayer embedded under AL on usb stick

hqplayer embedded under AL in ramroot w/ Apacer 4G ram >> hqplayer embedded under AL in ramroot

...

is this really all about power?

 

Interesting! Are you using ECC or normal RAM?

I once compared the two and could not hear the difference. But I never tried Apacer.

 

My recent experience with ramroot is similar: a real improvement, and surprisingly enough also on the AL HQPe server (not only on the NAA endpoint).

 

Another thing I tried recently and suprised me: SBooster BOTW ECO MkII as well as cheap Chinese Studer900 power supplies sound considerably better than HDPlex 100W for the endpoint. (I am using HDPlex 200W for the server as this requires more power.)

[bodiebill]

11 hours ago, Boomboy said:

@hifi25nl is a 32gb USB stick still ok for use on version 2.0 ? 

 

@hifi25nl And is a 16GB USB stick too small for version 2.0 and/or 4.0?

[bodiebill]

I just installed Audiolinux 2.0. At bootup I see:

   SHA256 validated
   Failed to acquire RNG protocol: Not Found

Not sure whether this is a problem.

[bodiebill]

13 minutes ago, hifi25nl said:

No, this is a systemd... excess of information, see this bug report:

https://bugs.archlinux.org/index.php?do=details&action=details.addvote&task_id=63698

 

"it looks like both messages will be suppressed in future versions of systemd"

 

Thanks Piero.

 

Another question: when I start/enable NAA (networkaudiod) on AL 2.0, it is shown as running, however it is not found by HQPlayer Desktop or HQPlayer embedded. Any clue what could be wrong?

[bodiebill]

5 minutes ago, lmitche said:

Piero's hard work is evident throughout the product. I am grateful to Piero for this effort.  Well done!

 

Agreed! Many thanks, Piero!

[bodiebill]

deleted

[bodiebill]

Thanks Piero, starting ramroot has become even faster 🙂

 

Just one thing: when cloning to another drive the cloning script is changing the UUID of the UEFI partition, but the ramroot configuration file still has the previous one. So I had to fix this by following your advice to edit the ramroot.conf file so that it has

   mounts_zram=''

(empty instead of wrong UUID).

Maybe you can modify the script in the next menu?

[bodiebill]

19 hours ago, hifi25nl said:

@bodibill

I discovered that problem some days ago, so I added a line to fix this to clone script in previous menu version. Re-checking the line I see I have forgotten something...  just uploaded version 207 with the fix!

 

 

Thanks Piero. What is the easiest way to upgrade to version 207? Can it be done from the current 206 menu? When I try updating the Audiolinux menu, it stays at 206. Or do I need to download the latest 2.1.0?

[bodiebill]

After '6 Update Audiolinux menu' the result is still 'Audiolinux menu is now version 206'.

[bodiebill]

Thanks, I tried that and it still says 'Audiolinux main menu - version 206' on top.

Should I download the latest image? That takes about two hours, hence my question.

[bodiebill]

Yes, strange. To be exact:

4 INSTALL/UPDATE menu

6 Update audiolinux menu

enter password

Result: 'Audiolinux menu is now version 206'

The strange thing is that this happens whatever password I try (audiolinux, audiolinux0 or the password given to me by Piero in an email).

Maybe Piero can help out?

 

[bodiebill]

it worked now, I used the wrong password... thanks for your patience 🙂

[bodiebill]

21 hours ago, lmitche said:

After writing this I realized that my configuration was less than optimal. After fixing that oversight, the AL2.0 real-time, non-BFQ config on both the Roonserver and endpoint running Roonbridge beats all other options. It is the most natural and coherent of the options tested here today.

 

Thanks, wanted to try this configuration, however run into a problem:

I am running ramroot mode. When rebooting and trying to say no to copying the system to ram, I can press 'n' as much as I like, but nothing happens, so it goes ahead to copy to ram again. I tried several usb ports and several keyboards. @hifi25nl Any idea?

[bodiebill]

I have a two PC setup with AL + HQPe on a server and AL + NAA on an endpoint. Whatever the order in which I power up the PC's, I always have to stop all audio services on the server and start HQPe before my control point recognizes HQPe. Is there a way to make them handshake automatically?

[bodiebill]

Seems to work, thanks!

[bodiebill]

In my case hqplayerd runs under AL headless (GUI-less) version. Not sure how I should make the service dependent on network interfaces being up.

 

(So far Piero's 'hack' works fine.)

[bodiebill]

I have setup AL with ramroot. However, it often fails to load into RAM. It then says:

RAMROOT: Active: no - Configured: yes

Often I have to reboot a second time to get: 

RAMROOT: Active: no - Configured: yes

 

Any idea?

[bodiebill]

On 4/20/2020 at 7:16 PM, lmitche said:

When I had this problem here is what Piero recommended:

 

Please reboot answering N at boot prompt and as root type:

mkinitcpio -p linux-rt
or
mkinitcpio -p linux-rt-bfq

depending on which kernel you use

then reboot once more allowing ram boot.

 

Thanks Imitche. Tried that a few days ago and occasionaly it still booted 'normally' (not from RAM). Not a big deal as a reboot corrects that, but maybe I will try what Piero recommended to me many months ago:

 

from command line:

sudo ramroot -D
sudo ramsave [when in RAM mode]
reboot
 

and then enable ramroot from scratch.

[bodiebill]

9 hours ago, hifi25nl said:

For your reference full readme on ramroot is here:

https://github.com/arcmags/ramroot/blob/master/README.rst

 

If you can is better to boot without ramroot answering N to boot prompt. 

 

The latter is not possible, possibly because the current HW setup (mouse/keyboard) is different from when it was setup.

[bodiebill]

30 minutes ago, cat6man said:

fyi..........new 2.4.1 version available for download.

 

everyone must be at home downloading it as the download is going pretty slow here, est. 2-4 hours

 

Thanks for the heads up. Here in The Netherlands it downloaded in less than a minute.