@
mdconnelly , @
audio_engr
How long will it take? How long is a piece of string? Sorry but I couldn't resist that old one.
Devialet have Matthieu Pernot doing 2 appearances at Oxford Audio Consultants and Hidden Systems in the UK on the 17th and 18th of June, a bit under 2 weeks away. He has to be expecting questions on this so hopefully he'll have some answers. Even more hopefully he'll want to say "It's fixed" which would mean we'll see a fix before then. Well, actually I'll be he'd love to say "It's fixed" but he may not and we may still be waiting and he'll just be fielding the questions with nothing to offer.
At least we know that if we haven't seen something from Devialet before then, someone is going to ask him the question on the 17th and we will hear back from someone who was present just what his answer is. The 17th is 3 days short of 4 months since the release of the Roon Ready update and they really should have made some sort of response to owners by now.
A couple of weeks back I posted a method of making a direct ethernet connection from a Roon Nucleus to a Devialet with no intervening switch between Nucleus and Devialet. This is the "experiments" @
audio_engr mentioned and it's been very helpful having two of us trialling things as I've worked at it. I started trying to come up with a way of doing this in the hope of getting better sound quality, not because of the problem we're discussing here because I wasn't noticing the problem we're discussing here. I do think it delivers improved sound quality.
I now have found a better way of handling the manual IP addresses required to make that work, discovered on one of the Roon forums and provided by Danny Dulai of Roon. It also works for Intel NUC's running ROCK, in fact it was posted in a thread relating to Intel NUCs which is why I initially missed it. I can't say it totally solves the problem and the best results I've got with it involve using a USB2 to ethernet adapter for the connection to the Devialet but my understanding is that a USB adapter limits the connection to 10/100 ethernet. I occasionally notice what may be a small "micro dropout", almost like a very brief drop in volume and last night, streaming something from Tidal, I started getting a "file is loading slowly from Tidal" error message and stuttering sound. That error message said that the issue could be a Tidal or network issue so it may well have just been that rather than the issue that concerns us here. Danny's connection method can be found in this thread if you scroll down to the post from Danny which is highlighted in yellow:
https://community.roonlabs.com/t/rock-du...d/26222/11
Unfortunately this method isn't going to work for everyone because not everyone uses a NUC or Roon Nucleus and for those who do it requires your Devialet to be connected directly, ie not over your network, to the NUC/Nucleus and you still need your normal network connection for the NUC/Nucleus. That arrangement may not be physically possible for everyone using a NUC or Nucleus and, as I said, I think I'm noticing very brief and minor issues anyway, even using a 10/100 USB2 to ethernet adapter for the connection. On the plus side this method isolates the music signal traffic between NUC/Nucleus and the Devialet completely from traffic on your network and delivers improved sound quality to my ears. It's not perfect but it works very well apart from that occasional "micro dropout" I mentioned and as I said, it works best when you avoid a gigabit connection between the NUC/Nucleus and Devialet so it's basically just another option for some of those dropping their connection speed to 10/100 in order to get around the problem.
This method is available to anyone who wants to try it but at present I would recommend using a USB2 to ethernet adapter rather than a USB3 adapter. As I said, I think it still doesn't offer a perfect solution but I'd say it's a 98 or 99% perfect solution with a 10/100 ethernet connection between NUC/Nucleus and Devialet and a bit less good than that in my system with a USB3 to ethernet adapter. I can't guarantee that it will be equally problem free for anyone else in relation to the problem we're concerned about here. I do think there has to be something about different network paths and things like switches which relates to why some people get the problem and some people don't. I definitely think it's a Devialet problem but I do think there's something different in different people's networks which is determining whether or not they get the problem and how badly it affects them if they get it.