|
|
Post by NigelB on Jan 6, 2024 21:48:10 GMT
Yes, and since it is the very beginning of the physical signal chain it seem to have a really big impact if the down stream signal chain is already improved, i found. That is at least my way of trying to explain how this £35 wifi extender could have a larger impact than my two £300 ethernet cables had, at least individually. (I was very surprised by those cables as well and found them worth the money...) Now when that first cable is fed with a much cleaner source (wifi extender->LAN Silencer) that cable is showing its shielding properties by being able to maintain that lower noise signal into the switch, to give the switch less to work with. That is at least my amateur explanation of this...? I should say that i don´t have the best switch of course... But the improvement was so big that i would be very surprised if this was not noticed also with an expensive switch in place. But I could be wrong... Apologies, but what does this signal chain now look like (from the WiFi Extender onwards)?
Could you possibly summarise it along these lines?
- Make/model WiFi Extender > 3m Cat6 cable > make/model switch > 1m Cat 5 cable > make/model streamer (for illustrative purposes only!)
Many thanks
|
|
|
|
Post by MartinT on Jan 6, 2024 21:48:36 GMT
Why EtherRegen? I know you have one but other makes and models are available, including award-winning ones... Because I only recommend what I know and respect. The EtherREGEN is the best switch I know, designed by John Swenson from the ground up and it works very well. |
|
|
|
Post by NigelB on Jan 6, 2024 21:53:21 GMT
Why EtherRegen? I know you have one but other makes and models are available, including award-winning ones... Because I only recommend what I know and respect. The EtherREGEN is the best switch I know, designed by John Swenson from the ground up and it works very well. That's cool, I do the same. It would be good to know which other audio-optimised switches you have heard it up against.
|
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 6, 2024 22:00:06 GMT
Yes, and since it is the very beginning of the physical signal chain it seem to have a really big impact if the down stream signal chain is already improved, i found. That is at least my way of trying to explain how this £35 wifi extender could have a larger impact than my two £300 ethernet cables had, at least individually. (I was very surprised by those cables as well and found them worth the money...) Now when that first cable is fed with a much cleaner source (wifi extender->LAN Silencer) that cable is showing its shielding properties by being able to maintain that lower noise signal into the switch, to give the switch less to work with. That is at least my amateur explanation of this...? I should say that i don´t have the best switch of course... But the improvement was so big that i would be very surprised if this was not noticed also with an expensive switch in place. But I could be wrong... Apologies, but what does this signal chain now look like (from the WiFi Extender onwards)?
Could you possibly summarise it along these lines?
- Make/model WiFi Extender > 3m Cat6 cable > make/model switch > 1m Cat 5 cable > make/model streamer (for illustrative purposes only!)
Many thanks
Sure, I have most of it in my signature. The first ethernet cable is 5m (expensive...but i got it used) and the second one is 0,75m. (I think they are both CAT8, in CAT terms) I hope that is all and clear enough? Otherwise let me know! |
|
|
|
Post by NigelB on Jan 6, 2024 22:18:16 GMT
Apologies, but what does this signal chain now look like (from the WiFi Extender onwards)?
Could you possibly summarise it along these lines?
- Make/model WiFi Extender > 3m Cat6 cable > make/model switch > 1m Cat 5 cable > make/model streamer (for illustrative purposes only!)
Many thanks
Sure, I have most of it in my signature. The first ethernet cable is 5m (expensive...but i got it used) and the second one is 0,75m. (I think they are both CAT8, in CAT terms) I hope that is all and clear enough? Otherwise let me know! Is the TP-Link the extender then? I know, I know, I should scroll back and check...
If so, is this correct?
TP Link TL-WR902AC v4 WiFi Extender->IFI LAN iSilencer->5m Audioquest Vodka->Netgear gs105e(IFI Ipower X)->0.75m Wireworld Starlight 8->Yamaha WXC-50
If so, the proxomity of your sweitch to your streamer is spot on, but have you tried an unshielded cable vs the Wireworld? The trouble with a shielded cable is that IF (and most manufacturers don't say) the shield is grounded at both ends then this shield can carry noise and override the galvanic isolation which is a good part of the benefit of even a generic switch. The only branded cable I know which is explicit on this is the Melco C100 which is grounded at only one end and this end is installed at the switch. Might be worth you trying a 0.75-1m U/UTP cable for pocket money to see what this does to the sound?
|
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 6, 2024 22:20:40 GMT
Yes, it is.
|
|
|
|
Post by MartinT on Jan 6, 2024 22:26:36 GMT
It would be good to know which other audio-optimised switches you have heard it up against. I said "ground up", not audio optimised. I don't know any other ground-up designed switches for audio, which also feature a moat and allow for an external clock. |
|
|
|
Post by NigelB on Jan 6, 2024 22:31:04 GMT
It would be good to know which other audio-optimised switches you have heard it up against. I said "ground up", not audio optimised. I don't know any other ground-up designed switches for audio, which also feature a moat and allow for an external clock. I'm not convinced ground up is inherently a good thing (it isn't in automobile design, for example; muhc better to take something proven and improve what matters). But perhaps I would say that.
Moat doesn't appear to be a technical term in circuit design; I think John means galvanic isolation.
An external clock capability would be a bonus if it was a quiet clock but only if using an external clock powered down the internal clock.
|
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 6, 2024 22:31:55 GMT
Sure, I have most of it in my signature. The first ethernet cable is 5m (expensive...but i got it used) and the second one is 0,75m. (I think they are both CAT8, in CAT terms) I hope that is all and clear enough? Otherwise let me know! Is the TP-Link the extender then? I know, I know, I should scroll back and check...
If so, is this correct?
TP Link TL-WR902AC v4 WiFi Extender->IFI LAN iSilencer->5m Audioquest Vodka->Netgear gs105e(IFI Ipower X)->0.75m Wireworld Starlight 8->Yamaha WXC-50
If so, the proxomity of your sweitch to your streamer is spot on, but have you tried an unshielded cable vs the Wireworld? The trouble with a shielded cable is that IF (and most manufacturers don't say) the shield is grounded at both ends then this shield can carry noise and override the galvanic isolation which is a good part of the benefit of even a generic switch. The only branded cable I know which is explicit on this is the Melco C100 which is grounded at only one end and this end is installed at the switch. Might be worth you trying a 0.75-1m U/UTP cable for pocket money to see what this does to the sound?
No, i have not tried that or heard this before. Hmm, thanks for sharing! I could test it i guess. |
|
|
|
Post by NigelB on Jan 6, 2024 22:35:39 GMT
Is the TP-Link the extender then? I know, I know, I should scroll back and check...
If so, is this correct?
TP Link TL-WR902AC v4 WiFi Extender->IFI LAN iSilencer->5m Audioquest Vodka->Netgear gs105e(IFI Ipower X)->0.75m Wireworld Starlight 8->Yamaha WXC-50
If so, the proxomity of your sweitch to your streamer is spot on, but have you tried an unshielded cable vs the Wireworld? The trouble with a shielded cable is that IF (and most manufacturers don't say) the shield is grounded at both ends then this shield can carry noise and override the galvanic isolation which is a good part of the benefit of even a generic switch. The only branded cable I know which is explicit on this is the Melco C100 which is grounded at only one end and this end is installed at the switch. Might be worth you trying a 0.75-1m U/UTP cable for pocket money to see what this does to the sound?
No, i have not tried that or heard this before. Hmm, thanks for sharing! I could test it i guess. There was a thread last year on Audiohpile Sound (I think) where someone asked about this and the consensus was that unshielded is the safesat option unless you know your shielded cable is grounded only at one end. I've never tried it myself yet, it's on the to-do list, but have stuck with unshielded Cat6 (U/UTP) by default at this final step; I have shielded Cat 6 (F/UTP) on the 10m run from router to switch. |
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 6, 2024 22:41:32 GMT
But unshielded goes right against all common sense for me now, when I have concluded how important shielding is to preserve a low noise floor... :-).
I hear you, and sort of understand, but it feels a bit weird. But I have been surprised before, on the other hand.
It feels like something Wireworld and Audioquest should have thought about, in my mind, since my cables are not entry level. The whole point with these ethernet cables is to sit between a switch and a streamer i am thinking. Just feels very unlikely that they would miss that aspect, in my mind.
Edit:
To me, it feels like they where discussing non "audiophile grade" ethernet cables, now when i think about it? I am all for tweaking and getting away with cheaper gear, but I just learnt the importance of shielding when ethernet cables are fed with a clean signal (at least i am pretty sure about that logic).
|
|
|
|
Post by NigelB on Jan 6, 2024 23:14:42 GMT
But unshielded goes right against all common sense for me know, when I have concluded how important shielding is to preserve a low noise floor... :-). I hear you, and sort of understand, but it feels a bit weird. But I have been surprised before, on the other hand. It feels like something Wireworld and Audioquest should have thought about, in my mind, since my cables are not entry level. The whole point with these ethernet cables is to sit between a switch and a streamer i am thinking. Just feels very unlikely that they would miss that aspect, in my mind. Edit: To me, it feels like they were discussing non "audiophile grade" ethernet cables, now when i think about it? I am all for tweaking and getting away with cheaper gear, but I just learnt the importance of shielding when ethernet cables are fed with a clean signal (at least i am pretty sure about that logic).
It took me a while to get my head around this but once you understand how a network switch does its audio magic (even one that wasn't designed for audio at all will give some benefit), it makes sense. Worth reading around and of course experimenting.
|
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 6, 2024 23:24:44 GMT
I might read it but i have read "everything" already it feels like, from all angles. I feel very comfortable with my logic right now :-) and that wireworld made a big difference, when i bought it.
If that last cable was a problem then I would not have noticed such a big improvement when implementing a better WiFi extender before it in the chain. That just doesn´t make any sense to me.
|
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 7, 2024 0:13:17 GMT
I read it now, since i am curious after all...
I can´t say that there was any information there that made me want to try it. It was to me conflicting logic and also mixed conclusions from different people. (and a language and don´t appreciate)
|
|
|
|
Post by orange55 on Jan 7, 2024 8:42:54 GMT
It would be good to know which other audio-optimised switches you have heard it up against. I said "ground up", not audio optimised. I don't know any other ground-up designed switches for audio, which also feature a moat and allow for an external clock. The SOtM switch meets that criteria. |
|
Tobias
Rank: Quartet
Posts: 328
Member is Online
|
Post by Tobias on Jan 7, 2024 9:20:11 GMT
I think this "Non-shielded last ethernet cable" logic is easily explained, now after thinking about it.
"...but have you tried an unshielded cable vs the Wireworld? The trouble with a shielded cable is that IF (and most manufacturers don't say) the shield is grounded at both ends then this shield can carry noise and override the galvanic isolation which is a good part of the benefit of even a generic switch. The only branded cable I know which is explicit on this is the Melco C100 which is grounded at only one end and this end is installed at the switch. Might be worth you trying a 0.75-1m U/UTP cable for pocket money to see what this does to the sound?
This can only happen if you have too much noise in the source end of that last cable (the switch), and the cable is bad for its purpose. Otherwise, if there is a very low level of noise in the switch, there is "no" noise to get transferred from that switch into the streamer on that shielding layer.
When you already have a low noise floor in the switch then you must protect it, with good shielding, into the streamer.
Those guys who claim this, about a not shielded last cable, doesn´t yet understand the importance of having a low enough noise floor already in the switch, i would say.
|
|
|
|
Post by MartinT on Jan 7, 2024 10:29:08 GMT
The SOtM switch meets that criteria. Thanks for letting me know. I have never come across one. |
|
|
|
Post by orange55 on Jan 7, 2024 11:13:13 GMT
|
|
|
|
Post by MartinT on Jan 7, 2024 11:39:15 GMT
Thanks. In view of the current performance I already achieve with the EtherREGEN, and trusting John Swenson's designs and circuit description, I shall be ordering the EtherREGEN Gen 2 when it's released this year.
|
|
|
|
Post by NigelB on Jan 7, 2024 13:29:27 GMT
I think this "Non-shielded last ethernet cable" logic is easily explained, now after thinking about it. "...but have you tried an unshielded cable vs the Wireworld? The trouble with a shielded cable is that IF (and most manufacturers don't say) the shield is grounded at both ends then this shield can carry noise and override the galvanic isolation which is a good part of the benefit of even a generic switch. The only branded cable I know which is explicit on this is the Melco C100 which is grounded at only one end and this end is installed at the switch. Might be worth you trying a 0.75-1m U/UTP cable for pocket money to see what this does to the sound? This can only happen if you have too much noise in the source end of that last cable (the switch), and the cable is bad for its purpose. Otherwise, if there is a very low level of noise in the switch, there is "no" noise to get transferred from that switch into the streamer on that shielding layer.When you already have a low noise floor in the switch then you must protect it, with good shielding, into the streamer. Those guys who claim this, about a not shielded last cable, doesn´t yet understand the importance of having a low enough noise floor already in the switch, i would say. This is incorrect. The role of a switch is to break the noise chain by forwarding only the digital part of whatever inputs (data plus noise) it receives. You write as if it does this regardless of the output cable connected to it. What others are saying, and I am inclined to agree with, is that the (any) switch performs its role because it acts as a more or less effective (according to design) "moat" separating the input signal from the output signal. The input signal is accompanied by noise which enters the chassis alongside it;
a shielded cable which is grounded at both ends is like a bridge over the moat, allowing the noise to cross to the streamer in a way an unshielded (or shielded but grounded at only one end) cable does not.
I suspect that if you took your Wireworld apart, you might indeed find that the shield is grounded only at one end.
|
|
