The EtherREGEN thread for various network, cable, power experiences and experiments

[kennyb123]

3 minutes ago, matthias said:

 

OK, this is his comment;

 

"I discarded it as a contender until a few days later while cleaning up my desk stumbled upon these OEM SFP module samples a supplier send me to try, these are 1310nm 20km range modules designed for usage in sensitive equipment, they draw lower then usual power and are designed for reduced EMI."

 

Matt

Yup and I should have typed 1310 instead of 1350, as 1350 doesn’t exist.  I was just browsing 1310nm SFPs on Amazon.  I’m tempted to purchase a couple to hear if they sound any better.

 

 

[matthias]

1 minute ago, kennyb123 said:

Yup and I should have typed 1310 instead of 1350, as 1350 doesn’t exist.  I was just browsing 1310nm SFPs on Amazon.  I’m tempted to purchase a couple to hear if they sound any better.

 

All clear now but please be aware that Emile did not cross the ER as nearly all folks here do. 

He stayed with his connections on the A-side only as this was superior sounding with the Extreme.

 

Matt

 

[ray-dude]

8 minutes ago, kennyb123 said:

Yup and I should have typed 1310 instead of 1350, as 1350 doesn’t exist.  I was just browsing 1310nm SFPs on Amazon.  I’m tempted to purchase a couple to hear if they sound any better.

 

 

 

https://planetechusa.com/product/mgb-tlx-mini-gbic-lx-module-20km/

 

Later Emile posted that he thought this might be the same OEM module he has (unconfirmed).  Interestingly, a wide temperature part.

 

I'm not (yet) setup for fiber ethernet, but looking forward to joining the party soon...

 

[PYP]

24 minutes ago, ray-dude said:

I'm not (yet) setup for fiber ethernet, but looking forward to joining the party soon...

 

Likewise.  Should receive a Sonore opticalModule later this month.  The oM's optical out will feed the eR's optical A side. My setup is simple so nothing else connected to the eR.  Currently using the eR with copper ethernet A > B with excellent results.

 

Just received an SFP and 1 meter fiber cable (via Amazon).  I've seen fiber before, but looking at it now vs. all the huge and inflexible cables in my audio system, it just seems so cool.  Like the first time you try keyless entry to your car and wonder why your front door, computer login and everything else doesn't work that way.  

 

Just received a JS-2 and letting it settle in with the eR.  The second rail of the JS-2 will feed the oM.  Down the rabbit hole!  😎

[kennyb123]

35 minutes ago, ray-dude said:

 

https://planetechusa.com/product/mgb-tlx-mini-gbic-lx-module-20km/

 

Later Emile posted that he thought this might be the same OEM module he has (unconfirmed).  Interestingly, a wide temperature part.

 

 

Taking a shot in the dark, but my curiosity got the best of me.  I ordered a pair of the following SFPs and will have them tomorrow.  I picked a model from Amazon vs the above as return will be easier and free.  I will try these without attenuators initially and compare to the Startech with attenuators.  
 

https://www.amazon.com/dp/B0713ZTQB8/ref=cm_sw_r_cp_tai_ahhqEbE7SR679

[kennyb123]

6 hours ago, Jud said:

By the way, I wanted to mention: I think the sound with the inexpensive Monoprice Cat 8 cable I am using is great, but I want to see if the Ghent will improve my impression of the sound at all. (I have Monoprice Cat 8 with shields tied between cable modem and router, and between ER and microRendu. Between router and ER I have Monoprice Slimrun Cat 6a with plastic connectors.)

How does the Monoprice CAT8 compare to the Cable Matters CAT8 in terms of flexibility?  I moved my CM to use between my wall jack and the upstream FMC.  It would be helpful to have a more flexible cable there.

[lmitche]

34 minutes ago, kennyb123 said:

How does the Monoprice CAT8 compare to the Cable Matters CAT8 in terms of flexibility?  I moved my CM to use between my wall jack and the upstream FMC.  It would be helpful to have a more flexible cable there.

The Monoprice Cat 8 is equally or even less flexible than the CM Cat 8.

[lmitche]

On 2/8/2020 at 3:14 PM, sahmen said:

 

Lol : You're certainly not alone, if that is any consolation... My needs are many, but now, I am limiting them to one, for the meantime : a simple 1 meter ethernet cable to connect the B-side of my ER to my Metrum Acoustics Ambre streamer. Reading this forum and others, I have learnt about the relative benefits of the following options, but there is sadly no consensus, and I get the feeling there isn't going to be one, any time soon :

 

1. Audioquest vodka (and maybe diamond?)

2. Wireworld Starlight 8

3. Supra Cat8

4. Shunyata Sigma (or  Alpha)

5. Nordost Hemdall (or Blue Heaven)

6. SOtM dCBL Cat7 (and Cat6)...etc etc

 

I am sure there are others that I am leaving out, but the list at least provides some indication as what I am faced with, at the moment.  In an ideal world, i'd simply get all of them and try them one by one, in order to ferret out and keep the one which sounds best with the ER and my other gear, and return all the rest, but it doesn't look like there is some way to try out all these expensive cables without breaking the bank (Will the cable company have all of them for rent, for example?).

 

On the other hand, I have already replaced the generic cheapo RJ45 cable inside my Metrum Ambre Roon bridge with a 20cm length SOtM dCBL Cat6 cable, and that replacement made a very good, and exciting impact on the sound, so something tells me that I should simply get a 1m version of the same cable to connect the ambre with the B-side of the ER and be done with all my present torment of second guessing cables.  Except that this last thought itself leads to another set of questions : What if one of the other cables on the list would sound better? Is it ever a good idea to mix brands of cables on the B-side of the ER (i.e. whatever cable I choose is going to have to interact directly with the SOtM dCBL Cat6 inside the Ambre, so if mixing cables on the b-side is.not a good idea, then the best approach would either to use another dCBL Cat6 cable to complement the one already inside the Ambre, or to use another brand of cable for the two purposes : connecting the Ambre with the ER, and replacing the RJ45 cable inside the Ambre)...

 

So far, I am inclined to settle for the 1m SOtM dCBL Cat6 cable not necessarily because I know it is the best option, but it seems to be the easiest, cheapest, and most pragmatic choice, however is pragmatism a good reason to choose a cable, given how expensive this one is going to be (i.e. $350)?... As you can see, I have learnt how to torture myself with such endless questions...

 

I thought having an audio enthusiast hobby was supposed to be fun all the way....😀

 

By the way speaking of pragmatism, I am already using a Cable Matters Cat8 cable between the ER and the Ambre, and although it sounds fine, and leaves nothing obvious to complain about, I have got all the forces of audio nervosa and confirmation bias in my body protesting against that choice, and trying to convince me that the Cable Matters Cat8 simply can't be the best I can do under the present circumstances... And what can a poor guy like me do to quiet down such mighty forces?...😘

 

Seriously though, I still want to make one choice among the above and be done with this phase of the quest.

Hi Sahmen,

 

Thanks for your response. There are people here that won't cross the line into audiophile Ethernet cables, or any cables, priced in the hundreds or thousands of dollars.

 

I am one of them. The Cable Matters and MonoPrice cables are good enough for me. My Lush 1 cable is the only exception here.

[Jud]

52 minutes ago, lmitche said:

The Cable Matters and MonoPrice cables are good enough for me.

 

Yes, I'm very happy with what I'm hearing. I have this nagging curiosity about whether a well shielded cable without the shield tied to ground at the connectors will do a good job, particularly upstream of the ER, so I'm trying the Ghent. But that's as expensive as I go.

[lmitche]

28 minutes ago, Jud said:

 

Yes, I'm very happy with what I'm hearing. I have this nagging curiosity about whether a well shielded cable without the shield tied to ground at the connectors will do a good job, particularly upstream of the ER, so I'm trying the Ghent. But that's as expensive as I go.

So the grounded shields on the Cat 8  cables made me ask which device had a path to ground in various parts of the chain. It turns out on the server side I had two power supplies with ground shunts, so I removed one and SQ improved, not hugely so, but a nice jump.

 

Tomorrow I'll remove or swap the shunt on the other supply to determine if no shunt or if the placement of the shunt is best in one spot or the other.

 

Eventually I'll jssg360 the cat 8 as well.

 

It never stops.

[Jud]

4 hours ago, lmitche said:

Eventually I'll jssg360 the cat 8 as well.

 

Yes, with cables as inexpensive as that, if you can roll your own it's a nice experiment to do.

[soares]

22 hours ago, soares said:

Interesting conversation gentlemen. I use my Pink Faun lan isolator immediately after my router and before the Zen. It makes a difference. To be honest I never tried it after the eR and before the rendu. But if it would affect negatively SQ one might have a case against using cables with isolation after eR. No sure what is the type of isolation used in Sigmas, but several people reported to increase SQ. So I am looking forward to hearing your views on the products your expecting Jud! And yes it’s a tragedy, so many lives...

Cheers Jorge

Just to confirm that the Pink Faun after the eR has a SQ negative impact. So back to place it was before.

[kennyb123]

On 2/9/2020 at 1:02 PM, kennyb123 said:

Taking a shot in the dark, but my curiosity got the best of me.  I ordered a pair of the following SFPs and will have them tomorrow.  I picked a model from Amazon vs the above as return will be easier and free.  I will try these without attenuators initially and compare to the Startech with attenuators.  
 

https://www.amazon.com/dp/B0713ZTQB8/ref=cm_sw_r_cp_tai_ahhqEbE7SR679

 

These 1310nm SFPs arrived today.  I swapped them in for the Startech 1550nm SFPs but left the attenuators.  Unlike with the 1550nm SFPs, these didn't run me out of the room with a too-hot treble.  I thought the treble overall sounded a bit drier though.  I will let them settle in a bit though before doing any serious listening.

 

It still amazes me that there are some who prefer the Startech 1550nm SFPs without attenuators.  I'm not sure if any of those cases they were using an ER.    I could be wrong but I don't believe Emile used attenuators when he tried this Startech SPF in the ER and found it to "perform very poorly".  Could it be that the ER is more sensitive to the ill effects of over saturation than his typical FMCs?  

[ray-dude]

2 minutes ago, kennyb123 said:

It still amazes me that there are some who prefer the Startech 1550nm SFPs without attenuators.  I'm not sure if any of those cases they were using an ER.    I could be wrong but I don't believe Emile used attenuators when he tried this Startech SPF in the ER and found it to "perform very poorly".  Could it be that the ER is more sensitive to the ill effects of over saturation than his typical FMCs?  

 

IIRC, Emile was looking for a lower power SFP for the ER, which is why he tried the OEM 1310nm/20km units that he had on hand (no attenuators for either SFP, IIRC).  Alas, difficult to find power consumption specs for these SFPs (unless someone has access to OEM spec sheets?)

 

If Emile's hypothesis about power consumption bears out, that would imply more sensitivity related to the power network in the ER than sensitivity to over saturation.  Alas, difficult to confirm without have a bunch of SFPs that draw different amounts of power (and decoupling wavelength, laser power, etc)

 

 

 

 

[kennyb123]

6 minutes ago, ray-dude said:

 

If Emile's hypothesis about power consumption bears out, that would imply more sensitivity related to the power network in the ER than sensitivity to over saturation.  Alas, difficult to confirm without have a bunch of SFPs that draw different amounts of power (and decoupling wavelength, laser power, etc)

 

 

Great observation, thanks.  And yeah, really difficult to find power consumption specs for SFPs.

[ray-dude]

32 minutes ago, kennyb123 said:

 

Great observation, thanks.  And yeah, really difficult to find power consumption specs for SFPs.

 

It could all be related: overstaturation on receiver draws max power, different wavelength SFPs having different power load, etc.  Emile's hint is the only hypothesis I've seen for the root cause of the differences people are hearing with SFPs, but it was guess work.

 

That being said, even with plain old cat 8, my ER scaled very nicely from the stock SMPS to LPS 1.2 to PHL SR4.  That seems to hint that more/better power available, the happier the ER.  A SFP that is drawing a significant amount of power from the board (1W seems to be the spec limit) on top of the background power consumption could be tipping the SQ.

 

For the solder inspired, the most definitive test might be to hack a SFP to externally power it (relieving the power load on the main ER board).  Looking at the pinouts, it looks like host sensing is on the ground lines, not the power lines, so we might be lucky on the power handshake.

 

https://en.wikipedia.org/wiki/Small_form-factor_pluggable_transceiver#Signals

 

 

[kennyb123]

18 minutes ago, ray-dude said:

 

It could all be related: overstaturation on receiver draws max power, different wavelength SFPs having different power load, etc.  Emile's hint is the only hypothesis I've seen for the root cause of the differences people are hearing with SFPs, but it was guess work.

 

 

I just ran across the page that follows when reading up on the above.  First time I've seen attenuator values specified - interesting that the recommendation aligns with what my ears told me.

 

1000BASE-ZX is the standard that applies to the Startech 1550nm SFPs.  The new SFPs that I just received are 1000BASE-LX.  

 

https://www.cisco.com/c/en/us/products/collateral/interfaces-modules/gigabit-ethernet-gbic-sfp-modules/datasheet-c78-366584.html

[kennyb123]

I do hear a difference between the TRENDnet 1000BASE-LX SFP (no attenuators) and the Startech 1000BASE-ZX SFP (with a 10 dB attenuator on each receiver). It’s not a difference that leaves one thinking one performs poorly relative to the other though.  Most apparent to me was a difference in the upper frequencies.  Highs are a bit more prominent with the TRENDnet.  This brings out a bit more detail.  I’m tempted to say that it slightly shifted the balance on guitar strings and voices where there was a bit less midrange presence with the TRENDnet.  These are very subtle differences and I haven’t done enough listening yet to be able to say which is more correct.

 

The attenuators tended to reveal a bit more to me after swapping one in for another that had been in place for a few days at least.  While I have had the Startech SFPs for a while, my system got quite a better over the weekend after shifting a few things around.  I’ll have to spend more time with each of these to be able to reach any firm conclusions.  I guess I’m actually somewhat relieved that the differences were pretty subtle.  

[austinpop]

4 minutes ago, kennyb123 said:

I do hear a difference between the TRENDnet 1000BASE-LX SFP (no attenuators) and the Startech 1000BASE-ZX SFP (with a 10 dB attenuator on each receiver). It’s not a difference that leaves one thinking one performs poorly relative to the other though.  Most apparent to me was a difference in the upper frequencies.  Highs are a bit more prominent with the TRENDnet.  This brings out a bit more detail.  I’m tempted to say that it slightly shifted the balance on guitar strings and voices where there was a bit less midrange presence with the TRENDnet.  These are very subtle differences and I haven’t done enough listening yet to be able to say which is more correct.

 

The attenuators tended to reveal a bit more to me after swapping one in for another that had been in place for a few days at least.  While I have had the Startech SFPs for a while, my system got quite a better over the weekend after shifting a few things around.  i’ll have to spend more time with each of these to be able to reach any conclusions.  I guess I’m actually somewhat relieved that’s the differences were pretty subtle.  

Thanks for your experiments! Extremely valuable data.

[HeeBroG]

11 hours ago, ray-dude said:

That being said, even with plain old cat 8, my ER scaled very nicely from the stock SMPS to LPS 1.2 to PHL SR4.  That seems to hint that more/better power available, the happier the ER.  A SFP that is drawing a significant amount of power from the board (1W seems to be the spec limit) on top of the background power consumption could be tipping the SQ.

 

 

 

Hi Ray,

 

Is that improvement using ER with streaming/NAS or playing "local" files?

 

Did you find ER better with PHL SR4 than LPS 1.2?

Do you think it's due to higher current capacity of SR4?

 

G

[ray-dude]

4 hours ago, HeeBroG said:

 

Hi Ray,

 

Is that improvement using ER with streaming/NAS or playing "local" files?

 

Did you find ER better with PHL SR4 than LPS 1.2?

Do you think it's due to higher current capacity of SR4?

 

G

Hi Geoff

 

My power/cable tests with ER where with Tidal and content from my Mac mini file server.  I have an end point NUC on the B side, and a Roon server NUC on the A side.  All my ER tests were with content going over the moat. 
 

I did find the ER improved as I went from SMPS to LPS1.2 to PHL SR4.  Very easy to hear in my system.   I have found that the SR4 has bigger impact on my tX-ultra SE, so that is where I usually run it. 

 

I have a pretty limited stable of power supplies to test, so I can’t be definitive on what factor may be driving things.  I’m looking forward to Rajiv’s findings once he finishes his eval and reports his findings. 

[Nenon]

My experience on cables / power supplies.

I have a first batch etherREGEN, which has been running non-stop ever since I received it. 

 

The power supply definitely makes a difference. LPS-1.2 is an improvement over the stock SMPS here. Charging the LPS-1.2 with another good LPS (JS-2 for example) is even better. I am using a different LPS, which I find to be even better in my system. The point is - the ehterREGEN scales with better power supplies, and that is easy to hear. 

 

DC cables also make a difference here. Not as much as power supplies, but they do. I've tried a bunch of different wires / cables. The one that sounds best in my system is this one: 

 

It's actually quite shocking how much of a difference power supplies and DC cables make on network equipment. I made two identical cables - one with JSSG360 shielding and one without. And I did A/B comparison with them connected to my cable modem. I could clearly hear the differences when streaming Qobuz (no difference for local files). Let me repeat that - my cable modem! Two identical cables with different shielding! You don't have to believe me on that one. I would not believe you if you told me that. But for some reason it is easy to hear in my system. 

 

I have over 20 different transceivers, including the Startech and PlanetTech ones Emile recommended. Most of them are left overs I did not buy. But I haven't had the chance to compare any of them yet. And I have a JCAT Signature LAN cable coming my way to test. 

 

 

[lmitche]

15 minutes ago, Nenon said:

My experience on cables / power supplies.

I have a first batch etherREGEN, which has been running non-stop ever since I received it. 

 

The power supply definitely makes a difference. LPS-1.2 is an improvement over the stock SMPS here. Charging the LPS-1.2 with another good LPS (JS-2 for example) is even better. I am using a different LPS, which I find to be even better in my system. The point is - the ehterREGEN scales with better power supplies, and that is easy to hear. 

 

DC cables also make a difference here. Not as much as power supplies, but they do. I've tried a bunch of different wires / cables. The one that sounds best in my system is this one: 

 

It's actually quite shocking how much of a difference power supplies and DC cables make on network equipment. I made two identical cables - one with JSSG360 shielding and one without. And I did A/B comparison with them connected to my cable modem. I could clearly hear the differences when streaming Qobuz (no difference for local files). Let me repeat that - my cable modem! Two identical cables with different shielding! You don't have to believe me on that one. I would not believe you if you told me that. But for some reason it is easy to hear in my system. 

 

I have over 20 different transceivers, including the Startech and PlanetTech ones Emile recommended. Most of them are left overs I did not buy. But I haven't had the chance to compare any of them yet. And I have a JCAT Signature LAN cable coming my way to test. 

 

 

Great to read that you share the same crazy SQ observations from JSSG360 shielding on DC cable. It is indeed hard to believe.

 

My Planetech SFPs arrive on Friday. I am very curious to hear the SQ impact, if any.

 

 

[austinpop]

6 hours ago, Nenon said:

It's actually quite shocking how much of a difference power supplies and DC cables make on network equipment.

 

5 hours ago, lmitche said:

Great to read that you share the same crazy SQ observations from JSSG360 shielding on DC cable. It is indeed hard to believe.

 

Yup, right there with you guys!

[lpost]

On 1/27/2020 at 6:12 PM, GryphonGuy said:

 

Hi lpost,

 

Yes I did try all copper Ethernet and quite a few different combinations of same brand before and after the ER and different brands before and after the ER. Cables in Cat 6A, Cat7 and Cat8 were used and they varied with the amount of shielding on the cables from none (UTP) to unshielded cable but foil shielded pairs (U/FTP) and fully shielded (S/FTP). Any form of shielding was better than unshielded before the ER. The most pleasing copper connection was Cable Matters Cat8 before the ER and Tera Grand Cat7 after the ER BUT the OM5 fibre cable's transparency (with the 1310nm transceivers) and airiness trounced the copper connection in my system. Of course YMMV. 

 

Unfortunately I do not have an OM4 fibre optic cable long enough to try it to see if it is something special about the OM5 cable but the reason the OM5 cable was introduced, according to sales literature, was to get greater distance using 850nm transceivers.

 

It was a surprise to me too that the wavelength of the transceivers mattered especially since the cable is supposedly optimised for the 850nm wavelength signals. It was not that subtle that the 1310nm transceivers opened the air around the instruments in the mix. The transceivers are from the Fiberstore custom made for my ubiquity networking gear.

 

Regards

GG

 

I've got a long-since retired HP 10Ge dual SFP+ nic (found it in a cabinet in our storage room at the office) that I first need to be sure Archlinux has drivers for...if it does, I'll try some of the 1310nm and perhaps 1550nm transceivers and patch.  It appears that OM3-OM5 support 850-1300nm.  I may also try 100Mb SFP 1310nm as I've got fiber to my server thus the ER is B-A and fed at 100Mb.  Lots of combos to try, awesome that fs.com has transceivers for so cheap.

 

I'm thinking these low power, only 2KM, 100Mb could be the ticket.  Shouldn't need attenuators but who knows how many ma they draw, could be less efficient that newer but could be less as it operates so 'slowly'.

 

https://www.fs.com/products/23583.html