Thanks for confirming the firmware, and for sticking with this. I want to step back, because your testing has actually cleared a lot of ground, it’s not Tidal or your internet (the track was always fully buffered server-side), it’s not the Arcam (zero dropouts across every event), it’s not DNS, and the Wi-Fi test cleared the Silent Angel switch. Solo playback has never dropped either.
What’s left is very specific: it only happens when both Bluesound units are in the group and the stream is high-resolution, and they drop together. That’s a grouped-sync-under-load pattern, so I’d like to test the two variables we haven’t cleanly isolated yet. In order:
Disable Sample Rate Conversion on every zone in the group, including the Arcam. A grouped zone plays the highest common stream, so with the Arcam still upsampling to 24/192, your network is likely still carrying 24/192 to the Bluesound units even though you disabled it on them. Turn off all upsampling/DSP everywhere, group them, and play. If this holds, we've found both the cause and a workaround.
Remove the 100ms zone grouping delay you set on the two Bluesound devices (set it back to 0), then test the group again. It directly affects grouped playback timing, and we've never tested with it reverted.
If it still drops, group only the two Bluesound units together (no Arcam), DSP off, that tells us whether the two BluOS devices staying in sync with each other is the trigger.
Since this began right after a firmware update on an otherwise unchanged system, I'm also flagging it internally to look at what changed around that time. Take these one at a time whenever you're home, no rush.
I am back from vacation and I have been doing some testing, and its a bit of a mistery. I have been getting many dropouts even with all changes, but I have done something different now, I have switched of the streaming to the Bluesound edge, I am using it now as pure amp (connected with a RCA cable from the preout ARCAM to the back of the edge- so physical old school connection, not internet and no steaming). This seems to be working much better, almost no dropouts, but do get skipping of track. I did get the message “tidal is responding slowly” when that happens. Could it be the the issue with the core? can you check the error on Wed 12th Aug, morning around 10-11 CET and afternoon around 14:50 CET?
also for the record I have disconnected the silent Angel rounter and connected a cisco router, just in case.
Thanks for letting us know, we were able to locate the latest log set. The pattern is clear: in the grouped zone, the Pulse Flex 2i is experiencing dropouts. This may be related to the extra strain of grouping it with the Arcam.
Does the issue still happen when the Arcam and Flex 2i are not grouped? Also, how is the Flex 2i connected in your network setup? Please let us know what you see.
hi @noris I feel we are going around in circles… I have been in touch with @benjamin and we went through a thorough analaysis. the flex is wired. and to add to my latest message, I have been listening for weeks now where the bluesound edge is disconnected, and had 0 dropouts, so this cannot be the Flex2i. @benjamin can you pls advise how to proceed?
We went through the full set (July 26 – August 18). Two things stand out, and one of them changes what we’ve been chasing.
The upsampling is still active on the FLEX 2i. Across the whole capture, nearly every track sent to that zone is being converted to 24/192, a 16/44 TIDAL FLAC is going out at roughly 9 Mbit/s, right at the speaker’s ceiling. Garden is doing the same. I know you removed this back in June, so something has either reverted it or is overriding it, likely a DSP preset or a Max Sample Rate setting in Device Setup.
The dropouts happen with a full buffer. At the moment each one occurs, Roon Server shows the audio already decoded and buffered at 100%. That means the music had arrived, your connection, TIDAL, and DNS weren’t the problem at those moments. The “Track Stopped Due to Slow Media” message you’ve been seeing is logged after the stream stops, so it describes the outcome rather than the cause.
So the SRC test we discussed in June hasn’t actually run yet under the conditions we intended. Could you do the following?
Open Settings → Audio → FLEX 2i- Rooms → DSP and send me a screenshot. Also check Device Setup → Max Sample Rate and clear it if it’s set to 192 kHz. Same for Garden.
Start a track and screenshot the signal path, it should read 16/44 straight through, with no conversion.
With that confirmed, group Arcam + FLEX 2i only (leave Garden out) and play TIDAL FLAC for an hour or so.
Separately, when you get a chance, please put POWERNODE EDGE back on Ethernet, directly to the router rather than through the Silent Angel. Its timing numbers over Wi-Fi are by far the worst in your system, and it’s making any three-zone test hard to read.
Thanks for the screenshots, those are exactly what we needed, and the news is good: Sample Rate Conversion is off. Nice work! That was the change that mattered, and it’s now confirmed in place rather than just assumed.
So the test we’ve been building toward is ready to run now. Please leave your settings exactly as they are and do this:
Group only the Arcam AVR11 + FLEX 2i-Rooms (leave the Garden / Powernode Edge out).
Play TIDAL for about an hour.
If it drops out, note the time and let us know.
That’s the whole ask. If it plays clean for an hour, we’ve isolated this to stream bandwidth under grouped sync, and we know what to do next. If it still drops with SRC off, then bandwidth was never the cause and we can stop chasing it, which is just as useful a result.
Two smaller things, whenever it’s convenient, neither one should hold up the test above:
Powernode Edge back on Ethernet, direct to your router. Its Wi-Fi timing numbers are the worst in your system by a wide margin, and we’d rather not have that variable in play once the Edge rejoins the group.
Tone controls flat on the Bluesound units. Your Flex 2i is applying Treble +5.0 dB and Bass −1.5 dB on the device itself, outside of Roon. That won’t cause dropouts, but +5 dB of treble is enough to push the amp into clipping on loud material, just something to keep in mind.
One last thing worth checking: the DSP panel you sent is the grouped zone view, and it lists Crossfeed, Procedural EQ, and Headphone EQ. But your signal path also shows a 5-band Parametric EQ, which isn’t in that list, meaning it’s almost certainly set on the FLEX 2i-Rooms zone individually. That’s fine and you don’t need to change it, but it’s the same kind of hidden per-zone setting that caused the June Sample Rate Conversion test to look done when it wasn’t. Worth knowing it’s there.
We know this has been a long road. Your testing has genuinely eliminated a lot: TIDAL, your internet connection, DNS, the Arcam, and the Silent Angel switch are all cleared. This next hour of listening narrows it further either way.
I have followed your instructions and played some music, it was ok for a whole but then on saturday 5 sept at 11:42 CET, I had the first stop, a bit later I had the scond one (not sure when as I was out side, but it was not later than 12:20 CET), could youu check the reason for it. at 14:21, I expreienced the same drop 3 sec after the start of the song and 5 min later the same, this is the problem, not anything else.
and multiple dropouts on sunday 6 sep between 9:00 and 10:00 cet.
one more thing to mention, I was playing music to all 3 devices but the Edge was on mute, not sure if its just a volume change or it doesnt recieve the stream, the dops started when I unmute the Edge. Not sure its related, and its been pluged in wired already a while back.
I have been in touch with 
