Best Ethernet Cards for Streaming

[jabbr]

5 hours ago, TomJ said:

At the end of the day it is important what noise will be carried through USB to the DAC. Is there any information if clearfog has put effort here?

 

That would be a fruitful area to measure. You could look at different endpoints. You could look at different network inputs e.g. copper vs fiber and see the effect on the noise at the USB output.

[jabbr]

5 hours ago, Jud said:

In that light, @TomJ isn’t wrong to be interested in the noise a fiber to copper converter inside a switch may be putting out. It wouldn’t be a bad thing to do some measuring, and also not a bad thing to put out information about various best practices for keeping electrical noise generated by network components from affecting the eventual output.

As I said just above ... the common mode noise at the USB output is what I'd be interested in.

[jabbr]

22 hours ago, TomJ said:

At the end of the day it is important what noise will be carried through USB to the DAC. Is there any information if clearfog has put effort here?

 

You know, if you can do your common mode noise measurements on the USB output signal, this measurement could be a reference for a whole host of factors, including endpoint hardware, power supplies, network hardware etc etc etc -- it would be really great, and at this point we could discuss the intricacies of common mode noise testing. You could even measure the USB CMN at the port vs at the end of various USB cables.

 

People could also repeat it. I do strongly suspect that "SQ" would correlate.

[Superdad]

1 hour ago, jabbr said:

it would be really great, and at this point we could discuss the intricacies of common mode noise testing.

 

@JohnSwensonwill be including a bunch of information on this very subject as part of something we are preparing to help guide @TomJ's investigations.  

The place the CM noise matters most is at the differential between the DAC clock's input pin and the ground plane. And the frequencies examined are important also (so far Tom/Eric has been looking much too broadband; 1MHz and far below--especially way down at AC line frequency and multiples--matter more).

That's all I'll comment on at the moment. Will leave the rest to JS.

[jabbr]

1 minute ago, Superdad said:

 

@JohnSwensonwill be including a bunch of information on this very subject as part of something we are preparing to help guide @TomJ's investigations.  

The place the CM noise matters most is at the differential between the DAC clock's input pin and the ground plane. And the frequencies examined are important also (so far Tom/Eric has been looking much too broadband; 1MHz and far below--especially way down at AC line frequency and multiples--matter more).

That's all I'll comment on at the moment. Will leave the rest to JS.

 

No doubt. Some actual information is better than none!

 

Ultimately the inputs th the DAC, no just the clock, matter. There is perhaps too much obsession with the clock itself. In any case that measurement is a hella lot more difficult to make, than simply the USB outputs and we have been waiting for measurements for X years now...

[Markus8]

On 8/8/2022 at 9:09 PM, Superdad said:

 

@JohnSwensonwill be including a bunch of information on this very subject as part of something we are preparing to help guide @TomJ's investigations.  

The place the CM noise matters most is at the differential between the DAC clock's input pin and the ground plane. And the frequencies examined are important also (so far Tom/Eric has been looking much too broadband; 1MHz and far below--especially way down at AC line frequency and multiples--matter more).

That's all I'll comment on at the moment. Will leave the rest to JS.

Many are eagerly awaiting any findings, insights that could be helpful to the overall HiFi community. Could you give as a short update if @TomJhas received/is working with you on that?

[Superdad]

On 9/16/2022 at 2:25 AM, Markus8 said:

Many are eagerly awaiting any findings, insights that could be helpful to the overall HiFi community. Could you give as a short update if @TomJhas received/is working with you on that?

I know.  I’m traveling on a 3-week vacation this month and @JohnSwenson is still hard at work on the 3 test boards for EtherREGEN Gen2–and I am loath to distract him from that. Still, we have several times discussed providing at least some of the promised input soon. We really appreciate everyone’s patience in this. 

[Markus8]

On 9/17/2022 at 11:34 AM, Superdad said:

I know.  I’m traveling on a 3-week vacation this month and @JohnSwenson is still hard at work on the 3 test boards for EtherREGEN Gen2–and I am loath to distract him from that. Still, we have several times discussed providing at least some of the promised input soon. We really appreciate everyone’s patience in this. 

Will the Gen.2 apart from functional upgrades like 1GB on B port also provide audible improvements?

[Superdad]

10 hours ago, Markus8 said:

Will the Gen.2 apart from functional upgrades like 1GB on B port also provide audible improvements?

 

Although this is not the appropriate thread to discuss EtherREGEN Gen2, I can confirm that our new model will incorporate a number of technical improvements. These include:

--Much lower jitter clock synthesizers (moving away from the SilLabs/Skyworks parts);

--The world's best sine>square wave converter chip (benefiting those using an external reference clock; the first EtherREGEN did not have such a chip at all);

--All different Ethernet switch chip and PHY transceiver set

--Different pre-regulators for all the LT3042/3045 LDOs

--Simplified clock buffer architecture.

 

Development is proceeding apace.  Yet it will be a little while before the whole design comes together onto one board (with isolators and moat) as we have learned from the past that it is more efficient to first create separate test boards for each new section--and to first get the wiring and programming right for each.  So right now we have separate clock synth test board, 'A' side switch chip and PHY test board, and the beginnings of the 'B' side test board.

 

It is impossible to answer the question of "audible improvements" just yet. We will know for sure once the design is unified onto one board--and that will be quite close to actual production; then all you users will get to decide for yourselves.  

It is important to understand that UpTone products are developed with careful technical analysis (and John's 3.5+ decades of engineering expertise) and are not just a tossing together of off-the-shelf chips and designs sprinkled with fairy dust and wishful thinking. And while we try to be a little cost conscious, there are an awful lot of expensive parts of the board--it really adds up.  

So yes, since the internal measurements (ground-plane-noise, jitter, signal integrity, isolation, etc.) will most all be improved, we expect this to translate to better SQ.  We will see hear. That's all we can do. 

[Markus8]

Thank you very much @Superdadfor your insights and the interesting approach showcasing your philosophy and commitment. 
 

I am sure many that look forward to it will be excited and very much appreciate all the hard work you / the team put into it to make this a great successor of the highly renowned gen. 1 edition. 

[tgb]

Hi TomJ,

I just see your research works yesteday : great !

This is true research we would like to see more often : we don't understand something, we look for explanations thru experiments. We end up to conclusions (expected / unexpected is not the issue there), and then we keep going looking for info / explanations & so on.
Congrats !
I'll follow up your work on the German site.

 

I want to recall here 2 things :
- a "famous" forum with "science" in the title is making a great damage to "science" by keeping saying weird stuff like "all DAC sounds the same", any electronic process in the digital chain that runs above the audio spectrum (the tiny 0Hz - 20kHz) has "no" effect on SQ ; all their statements based upon old-fashioned measurements (relevant on analogue audio signal, but irrelevant on digital signal).
- TomJ, you did a great job of investigation (basic & "common" research work), you spend time & money & use your brain (a bit too I guess 😊 ), although you take time to present nicely your investigations, you have to face posts that are pure bashing by people that don't have any idea how a basic "smps noise" looks like. Nice era...  Though era for enthousiast hobbyists, "easy era" for bashers. 
I know that "topic" since I've been "banned" recently from the 2 major frenchies audio forums for talking about "things we don't have to talk about" ("bits are bits", we don't have to say "it seems bits are bits is not so clear", since we have SQ improvement thru tweaks on the digital chain on : clocking & power supply (and PDN at PCB level by extend) (PDN = Power Delivery Network)

 

I made a lot of tweaks on my digital setup, leading to night/day SQ improvements.
I don't use NIC inside a computer, I prefered another way : a do-everything laptop (unmodified) followed by tweaked interfaces to "clean" the signal before reaching the end-point (active speakers with AES input, Dynaudio Core59, highly recommended, amazing gears).

 

I confess I use my measurements devices (scope Siglent SDS-2504X+, nanoVNA etc) only on the PDN issues at PCB level.
But by lack of time (or lazy) I did not do measurements on Ethernet interfaces (I use combination of tweaked FMCs & switches), nor measured the AES signal coming out of my heavy tweaked USB-to-spdif interface (Singxer SU-1) => I'll do that ! 
We'll see if my scope finds something 😀  For sure I'll have a big headache if, like you, the scope says that my tweaks higher the distortion on the signal path ‼ Although I'm more than happy with a night-day SQ improvement ! 😄
Rgds

[TomJ]

Hi RGDS,

 

thanks for compliments!

 

Regarding ASR, I also think that this is not really science. It is only the application of known measurement methods, but not research. I'm an analytical guy myself, but I can't just ignore the sound differences in Ethernet and also in DACs because I can't explain them. Then you have to start researching - that's my understanding of science.

 

I'm heavily involved in my cable measurements at the moment. Differences can be seen nicely here. Also in the GND of the streamer, where the noise takes its further run in the audio chain.

 

I will compare a JCAT Ethernet card with my Intel i350, which I also presented here at the beginning of my research journey, and publish it.

 

I am curious what you will find with your research. 

 

Eric

 

[di-fi]

On 10/1/2022 at 9:20 AM, tgb said:

I made a lot of tweaks on my digital setup, leading to night/day SQ improvements.
I don't use NIC inside a computer, I prefered another way

Hi @tgb, maybe you could update your AudioSystem info? I like night/day SQ improvements!

 

On 10/1/2022 at 9:20 AM, tgb said:

I know that "topic" since I've been "banned" recently from the 2 major frenchies audio forums for talking about "things we don't have to talk about"

If Tom did talk about things we do not have to talk about then maybe you can also do here? Why would you even try on a forum like ASR or what other forum you cryptically refer to? Maybe AS is a better place?

 

Just to be clear, I encourage you to share your experiences here. 

Thanks!

[tgb]

On 7/30/2022 at 5:22 PM, TomJ said:

https://www-open--end--music-com.translate.goog/forum/privatforen/thomas-michael-rudolph-tmr/651284-messungen-von-ethernet-infrastruktur-nur-lesen/page2?_x_tr_sl=de&_x_tr_tl=en&_x_tr_hl=de&_x_tr_pto=wapp#post658415

 

Hope this works now!

 

hi TomJ

Thanks for the link to your work.

 

I have a (stupid?) question : why did you perform jitter measurement (eye-diagram) using a home-made "music" sample ?

Why didn't you simply use a music track (you know as "good recording", high DR for instance) ?

 

My point is that by using your sample, the eye-diagram is the picture of "steady-state".

Steady-state because you sample is "simple" / very short in time, and replicated n times.

But using a "music" sample, you would have a fully random sample (thru measurements), closer to real streaming, because that way the UUT woudl be in constant transient-state (which is the state of its real life :-) )

What's your feedback about that clever (or stupid) idea ?  :-)

 

[TomJ]

Hi TGB,

 

I am not sure if we are talking about the same thing.
The jitter measurements are made by me on Ethernet 100base tx. It is completely irrelevant what data is sent, since this does not change anything. The eye pattern is not a representation of audio signals.

 

Eric

[itomonje]

On 10/9/2021 at 1:56 PM, TomJ said:

Hi all,

 

Since I am looking for the cause of the different sound of LAN cables and switches, I had bought on Ebay Tektronik TDS784A 1GHz oscilloscope.

To get a good 100Base TX reference for my cable measurements I wanted to check which network card produces the best Ethernet signal and also sounds best.

 

NIC in my Test

 

• Intel X540 T2 10GbE Dual Server NIC
  • The question was if a 10GbE card produces a better 100Base TX
• Supermicro AOC-STG-i2T 10GbE Dual Server NIC
  • The question was if a 10GbE card produces a better 100Base TX and how does the Supermicro competes with the Intel
• Intel PRO/1000 PT Dual Gigabit Server NIC with 82571EB chipset
• Intel i350 T2 Dueal Gigabit  Server NIC
  • Question: How does the i350 card compete with the older 82571EB card?
• Intel i210 Desktop 1 Port Nic
  • very small card that doesn't have much power conditioning and isn't that complex
  • Question: means less complexity a better sound?
• TP-LINK TG-3468 1 Port Nic
  • This card was already praised in other threads here
  • very small card that doesn't have much power conditioning and isn't that complex
  • Question: means less complexity a better sound?

 

Test Setup

• Host system for the NICs
  • Gigabyte Z97N-WIFI motherboard with WLAN deactivated, PCIe slot and HDD raid system
  • Windows home Server 2011, which is actually a Windows Server 2008
• Cable
  • 1 meter of a shielded Gore ePTFE CAT cable, which is my current reference and with best measurements so far and best signal.
  • Shielding only connected to the NIC
• End device and measuring point
  • A discarded Asus WIFI router on which the measurement was made at the RJ45 connector's release points - in front of the router's transformer. So no Magjack.
  • The router produces a very distinctive signal, so that one can quickly recognize in which auto MDI-X mode the router and the card are.

• Measuring instruments

  • Oscilloscope: An old Tektronix TDS784A 1Ghz

  • Probes: 2 x Tektronix P6245 active 1.5Ghz probes

  • The grounding of the probes was bridged and the Ethernet differential signal was generated via a Math function in the scope

 

Result

• I don't want to go too much into the sound differences, as this is always very subjective, but roughly give a tendency. The sound was determined by a direct connection of the network cards with my KEF LS50 W II.

 

Intel X540 T2 10GbE Dual Server NIC

• The 10GbE card is apparently really optimized for 10Gb, but does not produce the best 100Base TX signal. The sound is a bit brighter and shriller.

Supermicro AOC-STG-i2T 10GbE Dual Server NIC

• Similar results to the Intel X540 card. It also has the same chipset.

Intel PRO/1000 PT Dual Gigabit Server NIC with 82571EB chipset

• A somewhat wavy eye patter. A bit less nervous than the 10GbE cards.

The Intel i350 Dual Gigabit Server NIC

• produces the cleanest signal and is also the winner sonically in my system with a very clean, quiet sound. It has the same chipset as the Jcat cards.

Intel i210 Desktop 1 Port Nic

• Poor signal and eye patter. Sound is a bit dull.

TP-LINK TG-3468 1 Port Nic

• The very worst card in the test. Also very dull and undifferentiated on the sound side. No idea what the colleagues find sonic about this one.

 

Summary

 

If you want a very good network card and do not necessarily want to buy audiophile cards with super clocks (although I would like to see measurements of this), then I can highly recommend the Intel i350 T2.

 

Measurements in next poststhe i350 t2 , but i dont know if uses the jcat chip too , can you tell if any better ?

hello , i am looking at ebay and see an fijitsu version of 

 

 

[itomonje]

I am looking for the intel i350 t2 and found a fujitsu version that seems to have better components in the photo, but dont know how to get more info such ss if it uses the same chip as jcat , as the Intel does. any help will do. thanks

[TomJ]

32 minutes ago, itomonje said:

I am looking for the intel i350 t2 and found a fujitsu version that seems to have better components in the photo, but dont know how to get more info such ss if it uses the same chip as jcat , as the Intel does. any help will do. thanks

Could you provide a link to the product specs?

The ones I identified from Fujitsu seems to be OEM versions of the intel card.

 

[itomonje]

item34486211ad:g:bLwAAOSwe1phA6yP

 

NEW Fujitsu Intel Dual Port I350-T2 Gigabit Adapter S26361-F4610-L502

Fujitsu Part Number: S26361-F4610-L502 / S26361-F4610-E502

Intel P/N: I350T2G2P20

Model: I350-T2

 

NEW

Includes:

• 1 x Intel Fujitsu Quad Port I350-T2 Gigabit Adapter S26361-F4610-L502 I350T2G2P20

Overview

The new Intel Ethernet Server Adapter I350 family builds on Intel’s history of excellence in Ethernet products. Intel continues its market leadership with this new generation of PCIe GbE network adapters. Built with the bridgeless Intel Ethernet Controller I350, these adapters represent the next step in the Gigabit Ethernet (GbE) networking evolution for the enterprise and data center by introducing new levels of performance through industry-leading enhancements for both virtualized and iSCSI Unified Networking environments. This new family of adapters also includes new power management technologies such as Energy Efficient Ethernet (EEE) and DMA Coalescing (DMAC).

 

Key Features

• Halogen-free dual- or quad-port Gigabit Ethernet adapters with copper or fiber interface options

• Innovative power management features including Energy Efficient Ethernet (EEE) and DMA Coalescing for increased efficiency and reduced power consumption

• Flexible I/O virtualization for port partitioning and quality of service (QoS) of up to 32 virtual ports

• Scalable iSCSI performance delivering cost-effective SAN connectivity

• High-performing bridgeless design supporting PCI Express Gen 2.1 5GT/s

• Reliable and proven Gigabit Ethernet technology from Intel Corporation

 

Specifications:

 

General

Connectors

• RJ-45 (Copper)
• LC Fiber Optic (Fiber)
• IEEE standard/network topology 
• IEEE 802.3/10BASE-T, 100BASE-TX, 1000BASE-T

Cabling:

• Copper:
  • Category-3 or higher for 10BASE-T operation
  • Category-5 or higher for 100BASE-TX operation
  • Category-5e or higher for 1000BASE-T operation

• Fiber:
  • MMF 62.5/50 um

Technical

• Data rate supported per port: 10/100/1000 Mbps (Copper), 1000 Mbps (Fiber)
• Bus type: PCI Express 2.1 (5 GT/s)
• Bus width: 4-lane PCI Express; operable in x4, x8 and x16 slots
• Interrupt levels: INTA, INTB, INTC, INTD, MSI, MSI-X
• Hardware certifications: FCC B, UL, CE, VCCI, BSMI, CTICK, KCC
• Controller-processor: Intel Ethernet Controller I350

Power consumption (typical)

• Copper:
  •  I350T2V2 4.4 W
  •  I350T4V2 5.0 W
• Fiber:
  •  I350F2 5.5 W
  •  I350F4 6.0 W
• Operating temperature: 0 °C to 55 °C (32 °F to 131 °F)
• Storage temperature: -40 °C to 70 °C (-40 °F to 158 °F)
• Storage humidity: 90% non-condensing relative humidity at 35 °C

Connect Speed LED Indicators

• Not illuminated: 10 Mb/s; green=100 Mb/s; amber=1 Gb/s (Copper)
• Green: 1 Gb/s. Not

Physical Dimensions

• Copper T2 & T4; Fiber F2
• Length: 13.54 cm (5.33 in.)
• Width: 6.89 cm (2.71 in.)
• Low-profile end bracket: 7.92 cm (3.117 in.)

 

[TomJ]

For me this looks like an OEM version of the Intel i350 T2.

Why do you think there are better components used?

[itomonje]

sorry i mitake on itt, the one i was looking at is a generic card with that kind of cooling ,disipator onboard , but dont know if that is better or uses the same chips inside, as the jcat. So if you dont have a better choice for upgrading the sound using this intel card I will do next month

 

[TomJ]

I have just done comparison measurements of Intel i350 T2, T4 and JCAT NET Card XE.

Will publish this next days.

[SuperRoo]

Today I upgraded from motherboard to an Intel i350 T2 NIC. Also, got a deal on an AudioQuest Cinnamon Ethernet cable. Combined they made an obvious improvement. Money well spent.

[TomJ]

To give a preview on the measurements report:

The Intel i350 V2 is one of the best devices I have measured so far regarding jitter and CM noise.

Anyone looking for a network card should seriously consider this one.

[Markus8]

 

On 9/17/2022 at 11:34 AM, Superdad said:

@JohnSwenson is still hard at work on the 3 test boards for EtherREGEN Gen2–and I am loath to distract him from that. Still, we have several times discussed providing at least some of the promised input soon. We really appreciate everyone’s patience in this. 

Hope @Superdad you have had a good vacation time and Gen2 is also progressing nicely. 
And did you also provide @TomJ some feedback so he is able to continue his very passionate work?

I online forums worldwide flooded with the „data is data / everything is perfect, there cannot be any audible difference on Ethernet cables and switches“ and findings or measurements would be a real game changer here.