The Sablon is a wonderful ethernet cable. I have it throughout my system.....from the router to a GigaFoil, then to an Innuos PhoenixNet switch/reclocker, then out to my Lumin X1 and Roon NUC. Can't recommend them highly enough.
Funny you should describe the system the way you did - I am currently testing the GigaFOIL before the switch by my streamer and after the switch by my streamer to see which sound I prefer. Did you try the same? I notice yours is before your switch.
I have indeed. I've actually tried it many different ways: A Gigafoil before the switch, two Gigafoils before the switch, one after the switch, one before AND after the switch and then into the X1 or the NUC. There was no doubt I noticed blacker backgrounds with one or two before the switch, and decided for complexity's sake - as I'm powering them with linear power supplies - that just a single one before the switch was good enough. I still have my extra Gigafoil, so I might try the second one again one of these days.
I find the experimenting the most fun part of this hobby.
I can't tell you the number of ignoramuses I deal with on our videos who have never actually listened with their own ears, but will argue needlessly a point they theoretically think to be true as that's what they read somewhere.
Bravo for experimenting!
What the Gigafoil is doing is removing high-source leakage impedance current, which causes threshold jitter. This is because the leakage current doesn't pass on an optical cable (photons don't carry current). What a run of optical will not do is reduce or remove phase noise. My experience is you want the switch with the best clock (e.g. a SOtM with the sCLK-EX clock board upgrade or the EtherREGEN with the Crystek CCHD-575 clock) closest to the "endpoint" (network bridge, streamer, or in this case, Bob's X1) so that it can reduce any phase noise induced by the "cheaper" clocks from the lower-spec switches upstream.
Also, Bob, as you have an X-1, you don't need a Gigafoil or a series of switches in between your server and the X1. You can simply use a run of LC/LC fiber from an OpticalModule Deluxe or a SOtM sNH-10 switch's optical transceiver SFP cage right into the back of your X1. I've personally found that less devices in the "path" sounds better, and is more reliable. Just some suggestions for your consideration. Cheers.
Makes sense. My thinking was that I didn't like having a high end switch, then a "good" cable to the GigaFOIL with my "best" cable to the Rose. I *think* logically I'd want the best cable going from the switch to the streamer uninterrupted, meaning put the GF right before it.
I don't know - will be trying it out later today to see which I like better.
The reason the better switches matter is because they have better clocks
Michael, you've read John's white paper. You know from reading that that John really knows WTF he's doing with respect to all this.
I was impressed not only with what he covered, but that he did it in a way that even a non-science guy like myself could easily understand. I'm glad you sent me that link to read.
Yep, that sounds like it'd be the best path:
Upstream devices-Gigafoil->Better Ethernet cable (e.g your Soblon)->Best switch->Best Ethernet cable->Rose.
The reason the better switches matter is because they have better clocks in them and better isolation transformers in the RJ45 jacks. Good switches like the EtherREGEN's RJ45 use 12 isolation transformers per RJ45 jack and they are designed so there is no leakage current that can pass between each RJ45 port. The ER also used an "isolation moat" design to prevent leakage current from A side getting through to B side. And, they benefit from good power supplies.
Michael, you've read John's white paper. You know from reading that that John really knows WTF he's doing with respect to all this.
The other noise factor no one is discussing here is common-mode noise. Michael, you'll find out all about the impact this has when you compare the Sigma Ethernet cable I sent you with it's C-Mode filter compared to any other Ethernet cables.
Cheers, buddy.
I really like the Sigma cable but to be perfectly honest I think in my system it’s interchangeable between the Shunyata Sigma and my AQ Diamonds. I have two sigmas and all the rest are the AQ Diamonds.
I do think there are some very subtle differences that in my system I’m likely the only one that would pick up on them , both are truly excellent cables.
I think the Shunyata Omega might be the answer from where I am at but I just can’t reconcile the cost. It’s just one of those items I can’t seem to bring myself to purchase at the price.
The guys in this thread are talking about Fibre Media converters . I went that route with linear power supplies on both converters … all in. It was a dramatic improvement over the long run of standard cat 6 I had been using….
But
Subsequent to that I found Ediscreation . I purchased one of his switches and two of his Fibre Box II optical isolation devices
There is a lot going on in my listening space so I now run a long run of WireWorld starlight to the first Fibre Box then from there into the Ediscreation switch…. From the switch it goes to my Innuos server and from there through another fibre box II …. Then off to the Auralic Aries G2.1
Anyway, long story longer. If you want and excellent way to handle optical isolation I cannot recommend these Ediscreation Fibre Box II’s enough. A one box solution with fibre converters, high quality clock and an onboard linear power supply.
These have been crucial to the upgraded ethernet streaming in my house. More so than the bazillions spent on ethernet cables. It’s a high performance and super easy/elegant way to isolate components from upstream noise.
All, just my two cents
What the Gigafoil is doing is removing high-source leakage impedance current, which causes threshold jitter. This is because the leakage current doesn't pass on an optical cable (photons don't carry current). What a run of optical will not do is reduce or remove phase noise. My experience is you want the switch with the best clock (e.g. a SOtM with the sCLK-EX clock board upgrade or the EtherREGEN with the Crystek CCHD-575 clock) closest to the "endpoint" (network bridge, streamer, or in this case, Bob's X1) so that it can reduce any phase noise induced by the "cheaper" clocks from the lower-spec switches upstream.
Also, Bob, as you have an X-1, you don't need a Gigafoil or a series of switches in between your server and the X1. You can simply use a run of LC/LC fiber from an OpticalModule Deluxe or a SOtM sNH-10 switch's optical transceiver SFP cage right into the back of your X1. I've personally found that less devices in the "path" sounds better, and is more reliable. Just some suggestions for your consideration. Cheers.
Woot ! Woot !
Just received some Shunyata power and ethernet cables to try from a friend!
Will report how they compare to the SOtM and Sablon Ethernet cables.
View attachment 31983
View attachment 31984
I know your advice is for Bob, but it applies to me as well. I'm reading about what works for you and others and I'm eagerly awaiting Michael's results. I find all of it very enlightening.
Beginning to think that instead of investing in switches, isolators, cables...I might be better served by saving my pennies for a LUMIN X1 or P1 so I can run fiber right into the back of the unit with fewer steps in the path - only one ethernet to optical conversion. Finally addressing my room would be a cost-effective way to improve the sound as I save up.
(Hmmm, how many pennies would it take to get to $$$$$$?)
Beginning to think that instead of investing in switches, isolators, cables...I might be better served by saving my pennies for a LUMIN X1 or P1 so I can run fiber right into the back of the unit with fewer steps in the path - only one ethernet to optical conversion. Finally addressing my room would be a cost-effective way to improve the sound as I save up.
Wow Michael, that PS-10 is on my short list for the office system!