Thanks, that makes sense now — “Roon-only testing” means HQPlayer NAA should be disabled to avoid any possible exclusive-access conflict with Roon Bridge.
I tested again with HQPlayer NAA disabled in the Services tab, so the endpoint was running with Roon Bridge only, and the behavior is still the same.
After the DA2/C2700 is turned off and then back on, playback fails in the same way.
I also submitted this result as feedback: 3ae6a3df75a69ddc.
Anyways, I’ve found an issue…. your DAC keeps the USB connection ‘alive’ when going into standby. So when I put my DAC in standby, it disconnects from the USB. So RoPieee knows when a DAC is disconnected. Which means I also can detect when a device is powered back on (and should receive a reset).
This cycle is absent with your device, and that’s the reason why it isn’t working….
I need to think about this, because for now I don’t see a solution.
Thanks for confirming. That matches what I observed too.
In my testing, the connection still appears alive from both sides, but when actual playback / USB audio communication is attempted, the failure shows up.
The Korean McIntosh distributor told me they have already escalated the question to the US side, but they have not received a reply yet. They also confirmed that the latest DA2 firmware version is V5.12. My DA2 is up-to-date.
After work, I’m going to capture as much low-level logging as I can and investigate this further.
Thanks again for taking the time to investigate this so thoroughly.
I’ve now forwarded the report and logs again to the McIntosh distributor and will wait to hear back from them.
I understand that this appears to be on the McIntosh side, but if you ever find a practical host-side workaround for this kind of standby/resume behavior, it would be very helpful in real-world use.
Also, I sent a small token of appreciation earlier. Thank you again for your time and help.
Well, that is/was my intent. What I build now for you to test is rather generic. I even prepared the UI elements to make it configurable.
Problem is that is relies on detecting that the device disconnects as well. And that’s not the case, so I also technically don’t see a solution for this issue (expect scanning for the specific issue in the logs and resetting based on that, but that is something I’m not willing to do).
Thanks again for taking the time to investigate this so thoroughly.
I’ve now forwarded the report and logs to the McIntosh distributor and will wait to hear back from them.
I understand your point, and I really appreciate the time you spent looking into it.
If anything new comes back from the McIntosh side, I’ll share it.
Thanks for the suggestion.
I’m a bit cautious about hard power-cutting the C2700 directly rather than using standby first, even if it likely has protection built in.
Also, I haven’t actually tested yet whether a full power-off (rather than standby) would avoid the issue.
Hi @DDPS ,
May I ask which DA2 firmware version you are using?
Also, when you tested RoPieee with the DA2 connected, did you happen to try a standby → power on → playback again cycle?
I’m seeing a reproducible issue there on my side and was curious whether you saw the same.
@Chris.Ha one more thing…. we can test if the mechanism works by just pulling the USB cable.
So do the same as before, except you, after bringing your C2700 back online, also pull the USB cable and put it back again. That should trigger the reset mechanism.
Thank you for continuing to look into this. I also wanted to mention that I still haven’t received any response from McIntosh yet.
I tested the USB pull/replug case on the 3372 test image with the following results:
RoPieee ON, playback OK
Feedback: 2f49edfb8353f55e
C2700 standby / ON, playback fails
Feedback: d3d024c493b23029
I then physically unplugged the USB cable, waited about 30 seconds, plugged it back in, waited about 1 minute, and tried playback again
Result: playback still fails
Feedback: 5f12b07730af8680
So on my unit, a physical USB disconnect/reconnect after the standby-resume failure did not recover playback.
For power, I’m using an LHY Audio RPI Pro powered directly from a Shunyata Research DENALI 6000/S v2 via a Jorma Design Trinity Plus power cable. For reference only, that same outlet and power cable have worked fine with other equipment in my setup.
Could you please check whether the reset-on-connect path actually triggered in the last feedback log?
I’m using a 1.5m USB 2.0 Type A-to-B cable from 10Hz Audio, a Korean company. The manufacturer provides impedance and insertion-loss data and specifies 26AWG conductors with dual shielding.
Also, this is a newly purchased USB A-to-B cable. Playback is stable before the standby/on event, and manual USB reset has recovered playback before, so from my side I haven’t seen signs of a persistent cable issue so far.