11-Aug-2019, 12:51
@Confused ,
Answer 1: Don't ask me, I know nothing about networking. I often joke that a lot of things in audio work because of voodoo and things like how many chickens you sacrifice and whether you do it at the dark of the moon all play a part. There's no way I can give you a technical explanation, much less one that's likely to be convincing.
Answer 2: You say that you would have thought that a router or switch would help to insulate the Devialet from the noise of the NUC. What kind of noise are we talking about? My understanding is that there's no noise added to the actual digital signal but that electrical noise gets carried by the connection and can be transmitted to the circuits of the Devialet. Routers/switches are supposedly noisy, that's one of the reasons that people buy things like audiophile grade ethernet switches and use ethernet to optical and optical to ethernet connections to provide a level of isolation that a router or switch doesn't provide. I've got no idea whether or not those claims are correct but it seems there are a lot of people who accept that they are correct. The switch I've been using isn't audiophile grade and I've also been using the normal switch mode power supply with it, not an LPS.
Basically, a slightly lower noise floor is the only thing I can think of which could make a difference. The improvement I hear is an increase in clarity and definition, the sort of thing you notice when the noise floor drops for some reason and that also inclines me to the lower noise floor explanation. If I am right and I do hear a difference, the explanation for the difference could be something else.
As I said, I think it's a small difference in this case but I think there is a difference and it is an improvement. Before I tried the method in Danny Dulai's instructions I tried a couple of different settings for the manual IP addresses, hit and miss attempts at getting a direct connection to work because I couldn't find any clear instructions about how to do it. The first such attempt worked with some improvement but also some aspects of the sound were slightly worse, the second method I tried was better without the problems of the first. Danny's settings were the third lot of settings I tried and they worked better for me than either of my 2 attempts so I now recommend his approach. I'll add that he said in the thread I provided a link for that he isn't certain that there will be a difference and that he doesn't recommend the approach so it's definitely one of those things that people need to try for themselves and see what they think.
In the end it all comes down to the fact that we can make a direct connection work. We really can't eliminate a router, after all the NUC/Nucleus needs a network connection so it's going to connect back to a router somewhere but we can avoid having a router or switch between the NUC/Nucleus and the Devialet and depending on how your network is set up that may remove one network device which is a potential source of noise from your system. In my case, it lets me remove the switch that both my Nucleus+ and Devialet connected to. It also results in the NUC/Nucleus to Devialet connection only carrying traffic related to the actual music stream since all traffic between the NUC/Nucleus related to the internet connection that Roon uses goes via the server's network connection and isn't carried by the connection carrying the music stream. Whether or not the direct connection delivers an improvement in sound quality and how much of an improvement it makes is up to the user to decide, along with whether they think whatever improvement they get justifies putting up with the process required to get it working and the fact that they lose a wired network connection to the Devialet.
Answer 1: Don't ask me, I know nothing about networking. I often joke that a lot of things in audio work because of voodoo and things like how many chickens you sacrifice and whether you do it at the dark of the moon all play a part. There's no way I can give you a technical explanation, much less one that's likely to be convincing.
Answer 2: You say that you would have thought that a router or switch would help to insulate the Devialet from the noise of the NUC. What kind of noise are we talking about? My understanding is that there's no noise added to the actual digital signal but that electrical noise gets carried by the connection and can be transmitted to the circuits of the Devialet. Routers/switches are supposedly noisy, that's one of the reasons that people buy things like audiophile grade ethernet switches and use ethernet to optical and optical to ethernet connections to provide a level of isolation that a router or switch doesn't provide. I've got no idea whether or not those claims are correct but it seems there are a lot of people who accept that they are correct. The switch I've been using isn't audiophile grade and I've also been using the normal switch mode power supply with it, not an LPS.
Basically, a slightly lower noise floor is the only thing I can think of which could make a difference. The improvement I hear is an increase in clarity and definition, the sort of thing you notice when the noise floor drops for some reason and that also inclines me to the lower noise floor explanation. If I am right and I do hear a difference, the explanation for the difference could be something else.
As I said, I think it's a small difference in this case but I think there is a difference and it is an improvement. Before I tried the method in Danny Dulai's instructions I tried a couple of different settings for the manual IP addresses, hit and miss attempts at getting a direct connection to work because I couldn't find any clear instructions about how to do it. The first such attempt worked with some improvement but also some aspects of the sound were slightly worse, the second method I tried was better without the problems of the first. Danny's settings were the third lot of settings I tried and they worked better for me than either of my 2 attempts so I now recommend his approach. I'll add that he said in the thread I provided a link for that he isn't certain that there will be a difference and that he doesn't recommend the approach so it's definitely one of those things that people need to try for themselves and see what they think.
In the end it all comes down to the fact that we can make a direct connection work. We really can't eliminate a router, after all the NUC/Nucleus needs a network connection so it's going to connect back to a router somewhere but we can avoid having a router or switch between the NUC/Nucleus and the Devialet and depending on how your network is set up that may remove one network device which is a potential source of noise from your system. In my case, it lets me remove the switch that both my Nucleus+ and Devialet connected to. It also results in the NUC/Nucleus to Devialet connection only carrying traffic related to the actual music stream since all traffic between the NUC/Nucleus related to the internet connection that Roon uses goes via the server's network connection and isn't carried by the connection carrying the music stream. Whether or not the direct connection delivers an improvement in sound quality and how much of an improvement it makes is up to the user to decide, along with whether they think whatever improvement they get justifies putting up with the process required to get it working and the fact that they lose a wired network connection to the Devialet.
Roon Nucleus+, Devilalet Expert 140 Pro CI, Focal Sopra 2, PS Audio P12, Keces P8 LPS, Uptone Audio EtherREGEN with optical fibre link to my router, Shunyata Alpha NR and Sigma NR power cables, Shunyata Sigma ethernet cables, Shunyata Alpha V2 speaker cables, Grand Prix Audio Monaco rack, RealTRAPS acoustic treatment.
Brisbane, Qld, Australia
