Is there a limit to the length of ethernet? Thinking of temporarily bridging past the ethernet over power but it would need 30m+ of cable.
30 m should not be a problem. I think absolute limit is 100 m.
Correct, by passing the power over Ethernet has to be worth trying, when my network was playing up due to a 'noisy' device the ID41 was always the device to suffer most and drop off the network, MS600's in contrast were rock solid.
I concur with this. The only problem I have encountered is with Develo ethernet over mains to MS200 as a zone. However disappearance (cause unknown) has only happened twice in approximately 12 months and a hard reboot of MS200 has fixed the issue.
Hector
AV: Thorens TD160B, SME, ShureV15MK2, 500, 598, 861v4, DSP7200.1, DSP3300HC Samsung UE55, SkyHD. DINING: MS600, AC200, DSP5200SEUG. STUDY 504, Roon Nucleus + RevB, 808v6, 5200SE, Mac & PC (Touch Screen), Prime+PS Grado GS1000i HORIZONTAL: M80, MS200, Red Rose Spirit + Sonus Faber Toys, iPad Mini A capella to Zydeco
Doesn't this admirable and systematic testing rather confirm that the problem is the BBC end of things? When I have had problems (on iPlayer TV) they have often been brief and reconnection has occurred without my intervention. The fact that the non-BBC stream continued when R4 didn't seems conclusive.
I suspect (but have no data or knowledge to support this) that there is some kind of dynamic authentication going on somewhere between leaving the BBC and arriving (or not) at my ID41. And that somehow this authentication does not survive very small network disconnects caused (possibly) by my ethernet over power plugs. That hypothesis fits the facts but I have no way of confirming it empirically.
For example, I just loaded 10 CDs into Sooloos without problem while getting interrupts on R4 - all devices connected to the same dumb switch after the power plugs except the MD600.
In the spirit of Spinaltap, I would like to confirm that I am not involved in the corrupt theft of a global sport.
I get a few ID41 dropouts - maybe one a month or one every two months. Enough to be a mild irritation but no worse. Ethernet over power. Solved by restarting the ID41 by holding down the repeat button on the 808 when in standby.
I have to say that I find ethernet over power plugs not reliable (any make) and I no longer use them because modern beam shaping wifi is better, even in my old house with plenty of solid walls. If you have any dimmers in your home then in my experience things are even worse (even if the dimmers are set on 100%), not to mention mains transients causing things to briefly topple, or imperfectly screwed together mains cabling with the possibility of adding highly unpredictable spikes or momentary gaps. I have recently implemented a new wifi router, a TP Link VR900 v2 which has been reviewed as being especially good on 2.4GHz with standard wireless N adapters and I can only confirm the reviews: my Sooloos system has been completely reliable since the TP Link replaced my (Sagemcom) BT/Plusnet VDSL router.
Inspired by a comment from Ratbert (thanks Russ) about noisy devices, I switched off and unplugged the two EoP devices and switched off the 818v2/ID41. I then plugged in the EoP devies into the "other" plug outlet of the twinned power points. Then switched on distant (into the power) EoP first, then close EoP (out of power), then finally the 818v2/ID41.
I have since run Sooloos non-stop for 48 hours including streaming Radio 4 non-stop for 23 hours and palying many streamed music files. I put the 818v2/ID41 into standby for a few hours and started streaming R4 again - which has now been running for ~15 hours.
The trick (compared with just rebooting everything) appears to be the unplugging and replugging of the EoP devices (possibly the other gang is electrically cleaner and less noisy?) and then the sequence of power activation.
In terms of the observed behaviour, I am led to understand that BBC R4 uses "multicasting" to reduce load on its source servers. I think that means that it provides single feeds to major network nodes and domestic streaming is served from those nodes, including re-synching when there are interrupts. That process seems to fail when the local EoP connection is interrupted. Streaming from foreign radio stations or from web sites does not use multicasting so does not suffer the same breaks I was getting from BBC R4.
Ultimately the EoP connection had become so unstable/noisy that everything was breaking down.
Fingers crossed for the moment.
In the spirit of Spinaltap, I would like to confirm that I am not involved in the corrupt theft of a global sport.