Improve Raspberry 4 streamer

[somebodyelse]

At the PC I can adjust the USB to transmit up to 384khz. Is this somehow also possible/needed with the Raspi? I am using Picoreplayer and Raspi 3B V1.2.

Are you aiming to play everything at native sample rate, up to the maximum the DAC supports, or to convert everything to the highest sample rate? The former is easy - go to the 'Squeezelite Settings' tab and make sure the 'Output setting' is using one starting 'hw:' as suggested when you click 'more>' next to it. You can click on one in the list rather than typing it in. The latter is a bit more involved - see the 'Upsampling setting' entry in the squeezelite settings. You'll want to refer to the squeezelite man page resampling section, and possibly to the sox documentation, to understand the options available.

 

[cbxbiker61]

The Pi4 has a couple of significant advantages over the Pi3b+. Non-shared USB and a better wifi chip. The Pi3b+ had advantages over it's predecessors. A better wifi chip.

I have pretty much all of the Pi2-Pi4 variants. I usually try to recommission my older boards to do whatever tasks they are well suited to. I recommissioned a 3b+ to act as a music streamer with a hat and I ended up with minor audio glitches here and there. After swapping out the 3b+ to a 4 the audio glitches were gone. I don't know exactly why it glitched with the 3b+, it really doesn't matter much why. The conclusion was simply to use the Pi4.

 

[Gekel]

Is there an audible difference using USB or SPDIF (with Digi2 Hat) to go into an external DAC (SU-9)?
At the PC I can adjust the USB to transmit up to 384khz. Is this somehow also possible/needed with the Raspi? I am using Picoreplayer and Raspi 3B V1.2.

Yes there is. Someone showed the effect in a YT video and it was audible, even in a blind test. If you push raw audio data through different encodin/decoding mechanisms there is a good chance there is a difference in the quality.

here we go:

- it's SPDIF vs USB. Ok, not pi vs. pi, but you can hear a difference which wouldn't exist if both player would use the same cable type (USB) and data encoding (DSD or PCM).

 

[chefffe]

Hearing a difference must not be .... its good against bad. What would be a clear result is audible better bass or better highs.
The Web is full of statements that the Pi3 is better USB audio performer than the Pi4.

 

[abdo123]

Yes there is. Someone showed the effect in a YT video and it was audible, even in a blind test. If you push raw audio data through different encodin/decoding mechanisms there is a good chance there is a difference in the quality.

here we go:

- it's SPDIF vs USB. Ok, not pi vs. pi, but you can hear a difference which wouldn't exist if both player would use the same cable type (USB) and data encoding (DSD or PCM).

i don't really hear a difference, and even if there were a difference we wouldn't be able to tell over 192kbps mp3 quality (Youtube).

unless it's tiny differences in volume, which is probably the case.

 

[officialsm]

Yes there is. Someone showed the effect in a YT video and it was audible, even in a blind test. If you push raw audio data through different encodin/decoding mechanisms there is a good chance there is a difference in the quality.

here we go:

- it's SPDIF vs USB. Ok, not pi vs. pi, but you can hear a difference which wouldn't exist if both player would use the same cable type (USB) and data encoding (DSD or PCM).

Not clear to me if with Rasp 4 will be the same. Spdif hat will be ever better versus usb or not?

 

[Foxenfurter]

I don't know about Moode but I know with LMS & Picoreplayer that when a new player is setup the volume is not at 100% and this obviously has an audible impact.

I have tried a load of USB tweaks over the years including product from ifi for separating signal from power and intona for galvanically isolating the signal. This was before ASR as no-one was actually measuring these things so I got sucked in by all the subjective nonsense.

I could not hear any difference when blind testing, and only measured a difference using the intona on an old ESI USB audio interface which improved noise by 10db, but none of the replay mechanisms showed a difference. I have sold all these devices and have a tidier setup as a result.

 

[Gekel]

i don't really hear a difference, and even if there were a difference we wouldn't be able to tell over 192kbps mp3 quality (Youtube).

unless it's tiny differences in volume, which is probably the case.

I didn't hear it when listening to my speakers here at the computer, but there was a notable difference when using my headphones (DT770 pro), but now as I watch the same video again, I can't reproduce this outcome.

So maybe I got caught by the placebo effect last time I tried it. I don't have any objective ways to measure it and I don't have the time to build such a device. So take my last comment with a grain of salt.

 

[Gekel]

And just to add another comment: even if there is a difference between the spdif/usb, it is so small that in no way I would spend any additional money to build "the better one" of the two. The sound quality is the same.

 

[officialsm]

Do you think i can imporve Raspberry perofrmance adding a reclocker hat like Ian Canada's like this:
https://www.audiophonics.fr/en/dac-...d-reclocker-low-noise-low-jitter-p-15350.html

And on top adding an hat to output with usb/spfid/coaxial to dac bypassing Rasp usb.
Do you think could works?

 

[somebodyelse]

Do you think i can imporve Raspberry perofrmance adding a reclocker hat like Ian Canada's like this:
https://www.audiophonics.fr/en/dac-...d-reclocker-low-noise-low-jitter-p-15350.html

And on top adding an hat to output with usb/spfid/coaxial to dac bypassing Rasp usb.
Do you think could works?

I don't think it's intended to be used that way, and it wouldn't make any difference to a well implemented DAC via those interfaces anyway. What exactly do you think is wrong with the Pi's USB output?

 

[officialsm]

I don't think it's intended to be used that way, and it wouldn't make any difference to a well implemented DAC via those interfaces anyway. What exactly do you think is wrong with the Pi's USB output?

something says the way i told is best to have a less shout output

 

[somebodyelse]

Well implemented DACs are built to work well with the normal range of noise from USB. Archimago published Raoul Trifan's measurements from the different USB ports on several computers, showing quite how much variation there is even between different ports on the same computer. The DAC measurements on this site use an ordinary USB port on the PC running the software, not some special low noise port. These measurements show how the same DAC measures when fed from a Pi, an Odroid SBC, and a Microsoft Surface tablet - the differences are similar to those you would get between runs on the same equipment.

There are some DACs like the Modi 2 that are sensitive, but this is obvious from the measurements. The solution is to get a better designed DAC.

The other thing that's often incorrectly blamed on 'noisy USB' is the modern presentation of the 'ground loop.' The stray ground current used to come mostly from mains frequency sources, presenting as hum or buzz. Now the stray ground current can be modulated by computer activity (mouse movement, gpu activity etc.) so it sounds different, but the fix is the same, and not specific to the Pi or USB.

 

[officialsm]

Reading links you have posted seems to be better to underclock Raspberry cpu to have less output noise? Correct?

 

[somebodyelse]

Reading links you have posted seems to be better to underclock Raspberry cpu to have less output noise? Correct?

If you're referring to the 'CRAAP' settings, the sarcasm in that section may be lost in translation. Underclocking will reduce power consumption so the Pi will run cooler and your power bill will be slightly lower, but the audio performance won't change any more than it did with the different computers.

 

[Kagemusha]

Yes there is. Someone showed the effect in a YT video and it was audible, even in a blind test. If you push raw audio data through different encodin/decoding mechanisms there is a good chance there is a difference in the quality.

here we go:

- it's SPDIF vs USB. Ok, not pi vs. pi, but you can hear a difference which wouldn't exist if both player would use the same cable type (USB) and data encoding (DSD or PCM).

He is also the guy recommending to use the audioques dragonfly sayng improves the sound a lot, and we know now that DAC measures badly.

 

[officialsm]

Reading on web someone says that using i2s output from Rasp to dac (if it will accept it), instead.of usb, will improve a lot sound quality. To do this you will need a board on top. Have you experience or tested this?

 

[somebodyelse]

Reading on web someone says that using i2s output from Rasp to dac (if it will accept it), instead.of usb, will improve a lot sound quality. To do this you will need a board on top. Have you experience or tested this?

I assume you're talking about using i2s to an external DAC, rather than using a hat DAC.
https://www.audiosciencereview.com/...face-better-for-dacs-than-s-pdif-or-usb.7105/
Short answer - it won't improve measured performance. It's also easy to mess up the implementation, so could easily end up measuring worse than USB.

 

[officialsm]

I bought a linear power supply with dc output to power my Rasp. Could you suggest a cable to better feed it?
To use or not ferrite magnets?

 

[Gekel]

You can put rings around it, dance around it, do whatever you want. It won't change the quality of the sound output of the pi. All you do is to buy electronic jewelry for your amp.