I had some quiet time with my system this weekend, so I thought I would try reinstalling my SOtM modified Ethernet switch. (With Uptone LPS-1 power) I got it set up and running last night, and this morning tried some direct comparisons with the Arcam again.
Long term members of this forum with very good memories might recall that I have a bit of history with Ethernet switches. A while ago I bought an AQVOX audiophile switch, and after much A/B testing versus a standard switch, I concluded that it did not offer any improvement and I sent it back for refund. My SOtM switch is not the all conquering sNH-10G, but a far more modest D-Link switch, modified by SOtM to accept a reference clock signal. Modest maybe, but there was a time when the ultra tinkerers on Computer Audiophile were raving about this switch. For me, I had removed the SOtM switch from my system after changing my rack layout, I needed the extra ports offered by the 8 port TP-Link switch (the SOtM switch is 6 port), and this also saved a bit of hassle that I had with the layout of the kit by losing the extra clock cable. Following on from my AQVOX experience, I was not that bothered if it was installed or not.
However, now I can direct A/B switch to the Arcam as a reference, I though it was worth giving another try. So to listening. This was actually very interesting, whereas before I could switch between the Arcam and the SOtM kit with very little difference but maybe a slight nod to the Arcam, with the SOtM modified switch in place it seamed to be more track dependant. With the SOtM kit there was now a far better ambience to the sound, more coherent sound staging, this was a positive. But on the flip side, the SOtM kit seamed just a tiny fraction harsher in the highs, not by much, but there was something niggling me. No longer was there preference for the Arcam over the SOtM kit, now it was track dependant. I then had an idea. With the 6 port SOtM switch in the system, I needed to run an Ethernet cable to the 8 port TP-Link in order to have enough outlets for the AV and other hifi kit. The signal path was router to SOtM switch, to sMS-200Ultra, with an ethernet cable to the from the SOtM switch to the TP-Link switch for the other kit. As an experiment, I changed this to router to TP-Link switch, to SOtM switch, to sMS200Ultra, basically running two ethernet switches in series. I just did this on a hunch, but I am glad that I did, no more was the preference between the Arcam and the SOtM kit track dependant, but the whole sound was now far nicer and more coherent, with the niggling harshness gone, and this for just adding a cheap and cheerful TP-Link back in the chain. Not the result I was expecting, quite baffling in fact. I am now wondering if the TP-Link switch just happens to be good for noise isolation, hence explaining why I could not find much difference between it and the AQVOX switch. Then add the SOtM switch in the chain, I get the goodness that the CA tweakers were picking up in the past. Just a theory, from a technical perspective none of this stuff makes an awful lot of sense! Although surprising, this is an extremely pleasing result, I shall be leaving the system like this for a while, and see how I find things over time.
I think I have gone from being a little sceptical about the benefit of switches to being rather more intrigued about the possibilities. I wonder what that EtherRegen might bring to the party?
Long term members of this forum with very good memories might recall that I have a bit of history with Ethernet switches. A while ago I bought an AQVOX audiophile switch, and after much A/B testing versus a standard switch, I concluded that it did not offer any improvement and I sent it back for refund. My SOtM switch is not the all conquering sNH-10G, but a far more modest D-Link switch, modified by SOtM to accept a reference clock signal. Modest maybe, but there was a time when the ultra tinkerers on Computer Audiophile were raving about this switch. For me, I had removed the SOtM switch from my system after changing my rack layout, I needed the extra ports offered by the 8 port TP-Link switch (the SOtM switch is 6 port), and this also saved a bit of hassle that I had with the layout of the kit by losing the extra clock cable. Following on from my AQVOX experience, I was not that bothered if it was installed or not.
However, now I can direct A/B switch to the Arcam as a reference, I though it was worth giving another try. So to listening. This was actually very interesting, whereas before I could switch between the Arcam and the SOtM kit with very little difference but maybe a slight nod to the Arcam, with the SOtM modified switch in place it seamed to be more track dependant. With the SOtM kit there was now a far better ambience to the sound, more coherent sound staging, this was a positive. But on the flip side, the SOtM kit seamed just a tiny fraction harsher in the highs, not by much, but there was something niggling me. No longer was there preference for the Arcam over the SOtM kit, now it was track dependant. I then had an idea. With the 6 port SOtM switch in the system, I needed to run an Ethernet cable to the 8 port TP-Link in order to have enough outlets for the AV and other hifi kit. The signal path was router to SOtM switch, to sMS-200Ultra, with an ethernet cable to the from the SOtM switch to the TP-Link switch for the other kit. As an experiment, I changed this to router to TP-Link switch, to SOtM switch, to sMS200Ultra, basically running two ethernet switches in series. I just did this on a hunch, but I am glad that I did, no more was the preference between the Arcam and the SOtM kit track dependant, but the whole sound was now far nicer and more coherent, with the niggling harshness gone, and this for just adding a cheap and cheerful TP-Link back in the chain. Not the result I was expecting, quite baffling in fact. I am now wondering if the TP-Link switch just happens to be good for noise isolation, hence explaining why I could not find much difference between it and the AQVOX switch. Then add the SOtM switch in the chain, I get the goodness that the CA tweakers were picking up in the past. Just a theory, from a technical perspective none of this stuff makes an awful lot of sense! Although surprising, this is an extremely pleasing result, I shall be leaving the system like this for a while, and see how I find things over time.
I think I have gone from being a little sceptical about the benefit of switches to being rather more intrigued about the possibilities. I wonder what that EtherRegen might bring to the party?
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
