Very newbie question: how do I tell if minimstreamer is runnng?

[nickc]

(23-12-2022 17:45)simoncn Wrote:  When transcoding other stream types, wav24 transcoding does provide benefits compared with wav transcoding. I listen to BBC radio AAC 320k streams and using aac:wav24 on these streams creates true 24-bit samples where the low-order bits contain audio information. This is because aac:wav24 transcoding uses floating-point calculations with more than 24 bits of precision. In contrast, aac:wav transcoding only produces 16-bit samples.

This caught my eye, especially now the BBC streams have changed, and I've not got my head fully round bits (pun intended!). Sourcing a BBC R3 stream that still works, VLC shows the codec is ADTS, the bitrate is consistently above 320kb/s, the sample rate is 48000 Hz and there are 32 bits per sample. This is without any transcoding by MinimStreamer. So would there still be a gain in transcoding to wav?

[simoncn]

Thanks for asking about this. Are you using VLC as a UPnP renderer in combination with MinimServer/MinimStreamer or are you using VLC to consume the BBC streams directly? Also, what device are you using to run VLC? This information will enable me to provide a more helpful answer to your question.

[nickc]

Sorry for the delay in replying - only just getting back up to speed after three days without power following storm Ciaràn!

I'm running VLC on a Linux desktop (Manjaro Gnome) and entering the stream URL directly into VLC. I did this in order to discover the exact specifications of the stream. I'm assuming that the stream has identical properties when it's being accessed by MinimStreamer running on a RaspberryPI, and hence my question; but perhaps my assumption is wrong and it's a dynamic relationship between a stream source server and a local server accessing it, so the specs of the steam received locally could vary? As you can probably tell, I'm at the limits of my understanding of these things!

[simoncn]

Thanks for this clarification and sorry to hear about your outage caused by storm Ciaràn.

When you enter the stream URL directly into VLC, VLC is streaming AAC data from the internet, decoding this AAC data into LPCM and playing the LPCM to a DAC in the local device. In this "all in one" case, VLC can perform decoding computations with 32-bit precision and this is why you are seeing VLC displaying 32 bits per sample.

When you are using MinimStreamer to stream AAC data from the internet, transcode AAC to WAV and send the transcoded WAV data to a separate player device (renderer), the situation is different. If you are transcoding AAC to 16-bit WAV, the LPCM samples sent to the renderer will have 16-bit precision. When transcoding AAC to 24-bit WAV, the LPCM samples will have 24-bit precision and when transcoding to 32-bit WAV (if your renderer can handle this), the LPCM samples will have 32-bit precision. In this context, more precision is better.

From the above, it would seem that 32-bit local playback by VLC is the best option. However, this doesn't take into account the benefit you get from doing the streaming/decoding on a different device from the playing/rendering device. In my experience, this provides significant audible benefit because the electrical noise generated by streaming/decoding doesn't interfere (or interferes much less) with the playing to DAC part of the process.

Some would disagree about whether this separation makes any difference and I don't want to start a debate about this here. The relevant point for this discussion is that transcoding to WAV by MinimStreamer (if performed) should lose as little information as possible from the AAC data in the original stream.

[nickc]

Thanks for this very helpful explanation.

If I've understood you correctly then, the quality of the DAC chip in the computer where I'm using VLC is an entirely separate matter from the capability of the software rendering VLC delivers to the DAC? (And thus the discussions that go on re the quality of DACs are focused on different issues?)

It's not practical for me to use my desktop running VLC as the primary source feeding our two main sound systems in the house, and I tend to agree with you about the benefit of streaming and rendering on different devices. It was reading your comment about that elsewhere in the forum that helped me decide to install a second RaspberryPi to do the rendering (using Volumio), and that has definitely proved a good move. For me, living in France, it's had the added benefit of being able to run a VPN on the original (streaming/decoding) Pi.

From your explanation I'm still not clear though what happens to those BBC streams on my system (MinimServer/Streamer on RPi > Volumio on on RPi > Arcam SR-250) when I don't ask MS to do any transcoding. The reason I tried the stream in VLC was to get the hard info on the stream that so far I can't find a way of doing from Volumio or the Arcam.

Obviously I need to experiment and see if I can hear a difference if I add decoding to MS. (I have to wait a suitable opportunity as I'm not the only listener in the house.) My ears are not as good as they were once!

[simoncn]

(06-11-2023 20:10)nickc Wrote:  Thanks for this very helpful explanation.

If I've understood you correctly then, the quality of the DAC chip in the computer where I'm using VLC is an entirely separate matter from the capability of the software rendering VLC delivers to the DAC? (And thus the discussions that go on re the quality of DACs are focused on different issues?)

Yes, these are different. In practice, there is very little variation in the output sent to the DAC by different software renderers. There will be much more variation in the electrical noise generated by this rendering and between the quality of the digital/analogue conversion performed by different DACs.

Quote:It's not practical for me to use my desktop running VLC as the primary source feeding our two main sound systems in the house, and I tend to agree with you about the benefit of streaming and rendering on different devices. It was reading your comment about that elsewhere in the forum that helped me decide to install a second RaspberryPi to do the rendering (using Volumio), and that has definitely proved a good move. For me, living in France, it's had the added benefit of being able to run a VPN on the original (streaming/decoding) Pi.

From your explanation I'm still not clear though what happens to those BBC streams on my system (MinimServer/Streamer on RPi > Volumio on on RPi > Arcam SR-250) when I don't ask MS to do any transcoding. The reason I tried the stream in VLC was to get the hard info on the stream that so far I can't find a way of doing from Volumio or the Arcam.

As I understand it, your setup is:

internet stream -> RPi1(MinimServer) -> RPi2(Volumio) -> Arcam SR-250

If this is correct, how are the RPi2/Volumio and Arcam SR-250 connected?

If MinimServer on RPi1 is not doing transcoding, the stream is going through the following processing stages:

internet stream: HLS -> RPi1 decodes HLS to AAC -> RPi2 converts AAC to LPCM -> Arcam converts LPCM to analogue audio

If MinimServer on RPi1 is doing transcoding to WAV, the stream is going through the following processing stages:

internet stream: HLS -> RPi1 decodes HLS to AAC, then transcodes AAC to LPCM -> RPi2 passes on LPCM -> Arcam converts LPCM to analogue audio

Quote:Obviously I need to experiment and see if I can hear a difference if I add decoding to MS. (I have to wait a suitable opportunity as I'm not the only listener in the house.) My ears are not as good as they were once!

It is worth trying this experiment. I will be interested to hear what you find.

[nickc]

Quote:As I understand it, your setup is:

internet stream -> RPi1(MinimServer) -> RPi2(Volumio) -> Arcam SR-250

If this is correct, how are the RPi2/Volumio and Arcam SR-250 connected?

Yes, that's correct, and RPi2/Volumio connects to the Arcam by HDMI

Quote:It is worth trying this experiment. I will be interested to hear what you find.

Thanks for your further explanation of the processes at work. I'll post back when I've carried out the experiment, though as I implied earlier, the fact that I may not hear a difference doesn't necessarily mean there isn't one! I'll have to try again next time I have a visitor known to have sharp ears.

[simoncn]

The other factor that could come into play when you do the comparison is the "weakest link" effect. The audio signal passes through a number of stages on the way through your setup:

1) RPi1 receives HLS from your router and decodes it to AAC
2) RPi1 sends AAC over your network to RPi2
3) RPi2 converts AAC to LPCM
4) RPi2 transmits LPCM over HDMI to Arcam
5) Arcam converts LPCM to analogue audio
6) The analogue audio is amplified and played though your speakers

One of these stages is the weakest link in terms of sound quality and slight improvements in other stages might not make a noticeable difference to what you hear.

[nickc]

(07-11-2023 13:24)simoncn Wrote:  The other factor that could come into play when you do the comparison is the "weakest link" effect. The audio signal passes through a number of stages on the way through your setup:

1) RPi1 receives HLS from your router and decodes it to AAC
2) RPi1 sends AAC over your network to RPi2
3) RPi2 converts AAC to LPCM
4) RPi2 transmits LPCM over HDMI to Arcam
5) Arcam converts LPCM to analogue audio
6) The analogue audio is amplified and played though your speakers

One of these stages is the weakest link in terms of sound quality and slight improvements in other stages might not make a noticeable difference to what you hear.

Yes indeed! And in my setup there's an added complication in that the Arcam SR-250, as well as sending the analogue signal to the speakers in the living room/projector room, also sends it via RCA connection (with a very long cable) to an old Arcam A90 in "zone 2", which is another living room, and that Arcam outputs to speakers in that room as well as sending the signal to a Teac amplifier which feeds the speakers in the kitchen! All three sets of speakers and the amplifiers are of good quality, best I've been able to afford at the time, but quite honestly, given this Heath Robinson arrangement, I'm continually surprised at how consistently good the music sounds, and of course it's perfectly in sync as one moves between rooms.