I am currently tuning my Piano 2.1 DAC and combined it with the Kali again. I really hate the variable delay the buffer causes, making it impossible to watch movies with it, but the improvement in sound is undeniable. I just switch the NAD amp between HDMI (DAC on TV) and the Piano 2.1 on the amp by pressing a button for now.
So yesterday I actually started reading the Piano 2.1 DAC documentation and learned about an option called dual-mono (it was set to this by default, but directly switched to 2.0 when I started using it). I connected the RCA cables to both outer connectors (as described in the documentation) and switched to dual-mono. I do not know exactly why, but the sound is MUCH better while it was already very good before (combined with the Kali). If you have this DAC you should really try this setting out.
I am not sure what happened. If it is just louder (does not appear to be) or the dynamics improved, but the difference seems clear to me. Still figuring this out. I really thought it could not get better before with just the 2.0 > and I still have to do the DRC.
So yesterday I actually started reading the Piano 2.1 DAC documentation and learned about an option called dual-mono (it was set to this by default, but directly switched to 2.0 when I started using it). I connected the RCA cables to both outer connectors (as described in the documentation) and switched to dual-mono. I do not know exactly why, but the sound is MUCH better while it was already very good before (combined with the Kali). If you have this DAC you should really try this setting out.
I am not sure what happened. If it is just louder (does not appear to be) or the dynamics improved, but the difference seems clear to me. Still figuring this out. I really thought it could not get better before with just the 2.0 > and I still have to do the DRC.
I agree, and when using the Moode player the sound improves further if 'glb-mclk' and 384kHz upsampling are enabled.
Edit: The above assumes that Kali is being used.
Edit: The above assumes that Kali is being used.
Last edited:
I already use glb-mclk. I do not use Moode player, but my own Raspbian Stretch installation with Mpd, Kodi (HDMI output, not DAC), Brutefir and Mosquitto. I overclocked my rpi2, but will soon get a rpi3. My Mpd is using SoX set to high quality. What alsa output format is your mpd using (mpd.conf)? Something like: "384000:32:2"? I wonder if my rpi2 can keep up with that while doing the resampling and DRC as well 😛. But yeah, more samples per second for the DAC to work with.
Update: I tried setting it to 32bit/384k, but I do not get any sound 🙁
Last edited:
32bit/192k works, the Kali turns on the 192k LED. The Kali has no LED for 384k it seems, not sure if that matters. Only the 48k family (of which 384k is a multiple), the lock and power LEDs are on. When I select 384k the buffer full LED sometimes flashes as well. Any idea why I do not get any sound when selecting 384k? Do I have to modify / patch the kernel?
Also. How do I know for sure that upsampling sound from for example a 44.1k source to 384k will get better results from the DAC? Is the DAC not already doing something like this? Any tests showing this? Not sure if my ears are good enough to hear the difference and do not have a analyzer to measure it myself 😛
Also. How do I know for sure that upsampling sound from for example a 44.1k source to 384k will get better results from the DAC? Is the DAC not already doing something like this? Any tests showing this? Not sure if my ears are good enough to hear the difference and do not have a analyzer to measure it myself 😛
Last edited:
I'm currently using SoX resampling 384000:24:2.
At 384kHz the internal filter of the DAC is bypassed, I believe. Whether the difference can be heard is of course a subjective matter.
I don't know why yours won't play 384 - your posts suggest that you know more about such things than I do!
Edit: MPD settings:
M P D S E T T I N G S
Version = 0.20.10
Volume control = hardware
ALSA device = hw:0
SoX resampling = 384000:24:2
SoX quality = very high
SoX multithreading = on
Audio buffer (kb) = 4096
Buffer before play = 10%
Output buffer size (kb) = 81920
Volume normalization = no
DSD over PCM (DoP) = no
Replay gain = off
Last edited:
Thank you. It does not work with these settings as well. The PCM seems to correctly switch when playing audio (at the hw level only?):
Code:
access: RW_INTERLEAVED
format: S32_LE
subformat: STD
channels: 2
rate: 384000 (384000/1)
period_size: 16384
buffer_size: 65536Somewhere in the line it is not getting accepted because it plays but I hear no sound.
I did some research and found that someone tried to get a patch accepted to add 384k support to ALSA / Linux:
[alsa-devel] [PATCH] ALSA: Add rate defines for 352k8 and 384k
But apparently one of the core ALSA maintainers wants that person to rewrite / cleanup unrelated code as well before accepting the patches.
I will try to apply those patches and compile my own kernel and libasound if needed using the RPi toolchain.
Last edited:
- Status
- Not open for further replies.