Official moOde 8 support thread

[Tim Curtis]

Hello everybody,
since the functionality of the "loopback" option was released I have noticed some anomalies. After many tests, with various releases, software components, etc, I can now be sure that there is a problem related to enabling this option.
The last test I did today with a scratch configuration (8.0.2).
Configuration:
Raspberry pi: 4 - 4GB,
DAC hat: I tried with some cards,
no additional software components.
Symptom:
significant increase in CPU load and slight increase in memory occupation.
Repeatability:
with the "loopback" option set to "ON", starting to listen to a radio, the system freezes after some hours.
During this period of time, keeping under control with "TOP" you notice that the "MPD" process grows (in CPU and Memory occupation).
With the same configuration and the same radio station (Jazz24), but with the "loopback" in "OFF" the CPU and Memory load (even after 3 hours) remain stable at the same value.
It is essential that there are no pauses during playback: if you "pause" or "stop" or change the source, the problem clears up..
I tried to search if there was already a "thread" on this problem, but I did not find anything: please excuse me if I am wrong.
Please let me know if it is appropriate to open a different thread.
Thank you all for your attention and greetings,
Francesco

No issues on my end after 4 1/2 hours. 

   

Code:

pi@sig:~ $ moodeutl -m
^CU: 1.2 GHz | LOAD: 2% | TEMP: 46°C | RAM_USED: 20% | DISK_USED: 38% | DISK_FREE: 4.3G | FPM_POOL: 12 workers  

[fdealexa]

Hello everybody,
since the functionality of the "loopback" option was released I have noticed some anomalies. After many tests, with various releases, software components, etc, I can now be sure that there is a problem related to enabling this option.
The last test I did today with a scratch configuration (8.0.2).
Configuration:
Raspberry pi: 4 - 4GB,
DAC hat: I tried with some cards,
no additional software components.
Symptom:
significant increase in CPU load and slight increase in memory occupation.
Repeatability:
with the "loopback" option set to "ON", starting to listen to a radio, the system freezes after some hours.
During this period of time, keeping under control with "TOP" you notice that the "MPD" process grows (in CPU and Memory occupation).
With the same configuration and the same radio station (Jazz24), but with the "loopback" in "OFF" the CPU and Memory load (even after 3 hours) remain stable at the same value.
It is essential that there are no pauses during playback: if you "pause" or "stop" or change the source, the problem clears up..
I tried to search if there was already a "thread" on this problem, but I did not find anything: please excuse me if I am wrong.
Please let me know if it is appropriate to open a different thread.
Thank you all for your attention and greetings,
Francesco

No issues on my end after 4 1/2 hours. 

Code:

pi@sig:~ $ moodeutl -m
^CU: 1.2 GHz | LOAD: 2% | TEMP: 46°C | RAM_USED: 20% | DISK_USED: 38% | DISK_FREE: 4.3G | FPM_POOL: 12 workers  

Thank you very much, Tim, for your answer.
I have to investigate better to create something that could be replicated... any way, my situation is as follow:
Best regards and thanks,
Francesco


  Prove mpd01.zip (Size: 1.57 MB / Downloads: 4)

  Prove mpd02.zip (Size: 1.07 MB / Downloads: 2)

[Tim Curtis]

Here is with Jazz24 and Loopback on. No issues after almost 5 hours. It's fresh r802 on a Pi-3B+ playing though an old IQaudiIO Pi DAC+.

   

The system in my earlier post was an r801 -> r802 updated system running on an Allo USBridge SIG (Pi-CM3+) playing through an Allo Revolution USB DAC. 

Another system I have that uses Loopback is a Pi-4 1.1 2GB It functions as a dedicated Multiroom Sender and plays all day long.

[Pbnmjk]

Hello, how can I avoid the upsample to work in the original bitrate only?

I have my music in 24 bit, but it's upsampled to 32 to return to 24 in the DAC, obviously it's unnecessary.

Thank you in advance.

[Tim Curtis]

You DAC's USB interface might only accept 32-bit.

[Pbnmjk]

You DAC's USB interface might only accept 32-bit.

Ok, I understand... thank you for your time.

[Tim Curtis]

I should add that the the 24 -> 32 bit conversion is just simple zero-padding thats done by ALSA and does not affect the bit-perfectness of the audio.

[Pbnmjk]

I should add that the the 24 -> 32 bit conversion is just simple zero-padding thats done by ALSA and does not affect the bit-perfectness of the audio.

My DAC is a Bryston BDA-1 and it accept 24 bit only.

[Tim Curtis]

It would be odd if it's USB receiver only accepted 24 bit word length.

What is the output from the command below? It prints the formats that Linux/ALSA has detected for the device.

Code:

moodeutl -f

[Pbnmjk]

It would be odd if it's USB receiver only accepted 24 bit word length.

What is the output from the command below? It prints the formats that Linux/ALSA has detected for the device.

Code:

moodeutl -f

S32_LE