I am getting glitches when playing over I2s to a Pontus 15th over I2s with a Rivo+. This does not happen over USB.
The glitch does not happen in every track but quite often. It sounds like something is played back at very high speed. The glitch happens only once in a track and usually within the first 20 seconds of playback. I am using a 50cm HDMI cable that is rated for 8K playback.
I am using the Denafrips pin out on the Rivo+ and mode 6 from the Denafrips manual as instructed. Can you help me troubleshoot this?
Hello,
Does the issue occur with specific playback sources such as TIDAL Connect, or does it happen across all sources?
In the meantime, please try disabling the Multiroom feature from the Sources menu and test the playback again.
Additionally, it would be very helpful if you could send us a log right after the issue occurs. This will allow us to investigate further.
Finally, please follow up this conversation via the direct email support channel : support at volumio dot org
Thank you!
Best
Davide
VOLUMIO TECH SUPPORT
Done.
I found it also happens over coaxial. Not just I2S, also when playing the Volumio app, Tidal or web radio.
Hi, I have the exact same issue. Additionally DSD playback also does not work over I2S, there are just sound fragments. Over USB connection DSD playback is working, however USB has other issues like in case I used another input on Pontus before, then switched back to USB, playback seems to be ongoing, but there is no sound. After repeated restarts of Rivo+ , the sound gets restored. Any hints?
In Roon this little whoop at the start of a track when switching formatsā¦can be remedied by adding some resync delay in the device settings. 2000ms works in my setup
I have the old Rivo , which I like, I had a DDC in line before heading to my ā¦via I2s ā¦Cyan2 , every couple of days Iād boot up , see web radio āplayingā on the app but no music ā¦Iād have to reboot everything waiting patiently, sometimes it would work sometimes not . Not sure if it was related directly to your issue ā¦
ā¦Recently I took the DDC out going back to AES directly to the DAC ā¦which is a marked improvement over USB ā¦so far so good
Hello, I am using the Rivo+ with an Ares 15th connected by I2S.
On the Rivo+ side I selected āDenafripsā as the HDMI I2S mode.
On the Ares 15th I am using the #6 setting for the I2S pinout configuration. The reason that I am doing this is that I saw on the internet that someone suggested to use these settings for the Pontus 15th.
Can anyone confirm if I am doing the correct thing? Or let me know the best setting to use.
On another point, if I try skipping to the next song in shuffle mode, the new track sometimes pauses and I get a red pop-up msg ā¦ if I hit the play button the new track plays and the msg box disappears ā¦ has anyone else had this happen?
This hasnāt been addressed in the latest Firmware update dated april 24 or it wasnāt solved. Still glitching.
Hello,
weāre investigating on the root cause of Denafrips issues, please be patient.
Weāll keep you informed
@GrumpyOne please try the #5 setting on the DAC pinout configuration
Thanks
Davide
VOLUMIO TECH SUPPORT
@Marc050 did you manage to solve the issue in the end? Iām having the same problem with I2S to a Pontus 15th
I really hope thereās a way to fix it, because other than this everything sounds amazing!
I am also waiting for this to be resolved. Everything else is working.
No solution yet, I havenāt tested for a while but the firmware updates donāt mention it. So have to use USB.
Thatās a real shame. Thanks a lot for keeping me updated!
This is how it works for me:
- Denafrips setting #6 (setting #6 is PS Audio compatible)
- Rivo+ HDMI PS Audio mode
PCM - OK
DSD - OK
Channel polarity - OK
Internet radio - OK
Music library - OK
Soundcloud - OK
Continuing testingā¦
So maybe I should use PS audio mode instead of denafrips. Those glitches are likely time domain corrections since it sounds like super fast playback. I am using Denafrips mode on the Rivo+ and mode six on the Pontus 15th. Maybe I should be using 8 instead for Denafrips in the Rivo+.
Denafrips setting #6, Rivo PS Audio
I have been researching this problem with Gemini. Finally I noticed the Glitches only happened with a bit depth change, not with a bit rate change. As long as the bit depth stays the same between tracks, there is no problem. This is wat Gemini concluded:
Blockquote Technical Analysis: Bit-Depth Switching Glitches (Rivo & Denafrips Pontus 15th)
Observation: The āchipmunkingā glitch (audible buffer acceleration) occurs specifically when switching between 16-bit and 24-bit tracks (or vice versa) during playback via Tidal Connect. It does not occur when only the sample rate changes while the bit depth remains constant. The glitch typically manifests 20ā30 seconds into the new track.
The Technical Issue: Word Length Alignment This indicates that the issue is not related to clock jitter or speed, but rather to data architecture (word length alignment) during the digital handshake between the Rivo and the DAC.
- Frame Packaging: In both I2S and Coaxial protocols, audio data is usually transmitted within 32-bit āpacketsā or subframes. The actual audio data (16 or 24 bits) must be precisely aligned (padded) within these frames.
- FPGA Reconfiguration: When the Rivo switches from 16-bit to 24-bit, the Pontusās FPGA must instantaneously reconfigure its buffer to interpret the bits at the correct position.
- The 30-Second Drift: Because the glitch occurs significantly after the track starts, it suggests that upon a bit-depth change, the DAC initially āmisinterpretsā the bit alignment. This causes a minute, cumulative timing error (drift) in the FIFO buffer. After roughly 20ā30 seconds, the buffer reaches a critical threshold (āoverflowsā or loses lock), forcing the FPGA to rapidly process the accumulated data to resynchronize, resulting in the audible āchipmunkā effect.
Root Cause: Software/Driver Implementation This behavior is characteristic of a software or driver-level bug within the ALSA driver (Volumioās audio engine) or the Tidal Connect implementation (shairport/tidal-connect daemon).
- The Rivo may not be sending a ācleanā re-initialization command to the hardware output during bit-depth transitions.
- Alternatively, the Tidal Connect implementation bypasses Volumioās internal resampling/processing engine, preventing the user from forcing a āFixed 24-bitā output to stabilize the handshake.
Conclusion: This appears to be a logical synchronization conflict between the streamerās output driver and the DACās FPGA buffer management during bit-depth transitions.
It also happens with volumio app, not just in Tidal connect, unless you set a fixed bit depth with the resampling option, but that isnāt applied to Tidal Connect. So a work around is set the bit depth fixed to 24 bit (you can fix the sample rate too but it should not matter for this issue) and use the Volumio App instead of the Tidal app.
Hopefully Volumio 4 will bring an end to this problem.
1 Like
again, do not take for granted whatever AI claims.
the frequency/bitdepth switching must be handled at DAC level to avoid glitches during the transition. For example, I spent a considerable amount of time on this subject, when developing the Volumio Preciso, at FW level, to make sure the format switching was handled properly, with no glitches.
On Linux systems, the audio USB driver is a standard and unified one (UAC2), the very same for all the USB DACs.
There is no such a thing as āre-initialization commandsā that must be sent to the DAC.
Some specific DACs require dedicated quirks, but itās the responsibility of the DAC manufacturer to make sure those quirks are included in the mainline Linux kernel, because they know the HW they produce and they know what kind of quirks are required.
Conclusion: my suggestion is to contact the customer support of your DAC manufacturer, and ask if there is a FW update or a specific quirk to be applied to solve the problem you reported.
If no firmware updates will be provided and no USB quirks will be included by the DAC manufacturer, ending up on the mainline Linux kernel, itās very unlikely that the problem will be solved with Volumio4, being the problem at DAC level.
This is not an USB issue. It only happens with sp/dif I2S