|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
175.45.6.61
I am starting a (cheap-ish) trial of running different ethernet cables between my TP-Link 743 router (working as a wireless client)and my SB Touch. It appears that some companies are coming out with fancy (aka expensive) LAN cables. I guess from a scientific stand point, ethernet cable doesn't matter as long as it is up to spec (maybe it should matter even less in my set-up since signal goes through wireless first). But then again, our hobby has convinced us with some "phenomenons" that are not quite scientific to begin with.
Since I was just using some cheap LAN cable that came with routing equipment, I guess it wouldn't hurt to try some better built cable anyway(though I am not yet convinced enough to get a multi-hundred dollar LAN cable). I eventually bought a $30 Monster cable Cat6 cable (gold plated plugs with some custom plastic casing and thicker than normal construcion. Stranded) and a Elecom Laneed Cat7 cable (about $12)(shielded, gold plated plugs and has a little bit of metal covering on the plug. Feels like a solid core construction. Slightly thicker than generic ones that come with routers).
Initial trial seems to indicate that there is indeed a sound quality difference, with the original cheap lan cable sounding robust but a little homogeneous, the Monster sounding a little relaxed and warm (but possibly a little too much so), and the Elecom sounding cleaner in the treble, bigger soundstage, but slightly less (though tighter) bass. Since the two cables have not been run-in, I'll probably give it a little time before deciding. At this point, my preference goes to the Elecom.
Has anyone given LAN cables a try and what has been your experience? How about experience with the fancy ones from Cardas, Audioquest, Furutech, etc?
Thanks.
Follow Ups:
Thanks for all your interesting and informative feedbacks. I also have a feeling that it is more an RF/EMI issue than things like conductor material and "audiophile" jewelry construction.
In that case, what do you guys think about ethernet isolators? Acoustic Revive sells an isolator called RLi-1. On the other hand, there are medical ethernet isolators out there that look very similar. Anyone tried those?
Have you sorted the usb issue on your tablet?
If not, it seems that the SP1 update actually leaves the old usb drivers in place and in instances can degrade to usb 1.1!
I recently discovered this and updated my Win 7 pcs.
No, I haven't touched it since. Are you saying that there is another auto update of Win 7 or are you putting in new USB drivers? If it is the latter, how is it done?
Thanks.
http://forums.mydigitallife.info/threads/24666-The-Windows-7-SP1-USB-Driver-Bug-(what-it-is-and-how-to-fix-it)
The fix it link works for my 32 bit PCs. Coypying no longer poses hicoughs. The MS 2529073 update fix locks up my mouse on 2 systems!
Check to see what will work on 64 bit system first.
http://meicord.de/index.php?id=91&L=3
I played with this last summer and found yes indeed the ethernet cables could make a diff with the Touch (via 2 x Linksys routers). I didn't dwell on it though, just noted.
You know that you can unplug the ethernet cable from the Touch and still listen for ~20s (buffer) at CD bitrates? That's how I compared/judged, but in the end just went with the shortest ethernet cables to do the job.
So, are you saying that you could play a track and then while the final 15 seconds were playing you could disconnect the Touch and the music would complete the track? And further, it would do so without stopping prematurely if during this period you reconnected and disconnected the cable one or more times? And finally, are you saying that connecting and disconnecting the cable affects the sound quality, i.e. the sound is different according to whether the cable is connected or not at the instant the sound comes out the speaker, even though the music being played out the speaker went down the cable quite a while earlier?
If the answer to all three questions is yes, then you would appear to have pretty solid evidence that the sound was not affected by different bits entering into the Touch, nor was it affected by data transferred into the touch since there wasn't any, but it was affected by the Touch in some other way. That being the case, there is various experimentation that would be possible, i.e. connecting and disconnecting the cable at either end (Touch vs. Router) and seeing if the SQ is different. You could try different cables, ferrite beads, etc... And obviously, different cable dress. Other things you try do would involve cranking up your computers and moving data through your router (or not) during these 15 seconds and seeing if this affects sound quality.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Wow, you have it all organized!At the time it was just an experiment to see how much "the whole networking mess" affected the sound, or actually *if* it did at all. This is why I didn't get too involved with it, the whole thing was getting too complicated with too many variables for me to wrap my head around lol. [I did post about this with the banned or MIA Roger at the time I think he was trying Cat7 cables.]
All I really determined was that the network layout (all wired with 2 routers in the path) could cause slight differences in the sound depending on the physical layout (with the same logical layout) i.e. changing cable lengths/types and router position along those lengths.
Of course I had to leave an ethernet cable in for a short time before I could "hear" it i.e. the SBT ethernet buffer had to clear first. But for your first question, the answer is "yes" in that the buffer will play out. It seems quite forgiving of making/breaking the ethernet connection, as is proper I believe, in it's desire to keep the buffer full.
Edit: I have to correct myself here. I checked my audio notebook and I did make one significant physical/topological change after the "experiment" last summer. I originally had 3 routers, but only 2 in the path to the SBT. I removed the non-path router and replaced it with a high quality switch; everything "unimportant" (TV/BDPs/AVR/PS3) goes into this switch, which can be turned off. The Linksys router the SBT is directly connected to only has the SBT and the switch connected to it, besides the cable to the router near the computer.
Bottom line is when I'm using the SBT for music, it and the PC are the only live items on the network.
Edits: 04/14/12
"Wow, you have it all organized!"
These are all things that someone must do if they want to get to the bottom of why "bits aren't just bits".
Alternatively, one can select and set up one's system in such a matter that one is unconcerned about this level of detail. It's pretty clear that any wires or electronic components in the vicinity of audio equipment can affect sound quality, something that I've noted for a long time. Of course, since the advent of computers and other digital devices the EMI situation has gotten much worse than back in the day of tube based phonographs, tape machines and radios. Most radio hams, as I was 50 years ago, had direct experience with high power radio transmitters causing obvious interference with phonographs and radios, but there were also subtle effects that weren't so obvious, and these could happen with lower power signals.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
^ Yes. I am a ham too. This is my 41st year of being an audio dweeb...
P.S. I corrected myself in my last post above, I did apparently get some benefit from the "directionless" screwing around after all.
As ususal. It's all about noise/EMI/RFI.
Cables act like antennas.
Shielded cables build nasty groundloops into the devices causing interferences/intermodulations and noise inside your audio equipment.
Every measure that'll reduce or change the pollution spectrum will therefore show an impact.
The USB connection comes with similar problems. It's not even isolated to reduce some common mode noise.
The ethernet of course gathers the mess of all clients involved.
One of the main issues is the basically not existing quality grounding
in normal houses/apartments.
Since all cables are built differently (filter and antenna characteristcs differ), you'll experience more or less respectively different impact.
From the feedback I gathered over time, I have to say that it doesn't have to be "audiophile" grade/priced cables to go after. There seem to be quite good cables out there at ridiculous low price, performing better
then those audiophile cables.
If audio-interfaces would have a better isolation against those typical distortions, we wouldn't have that discussion.
Obviously vendors/manufacturer need to accept first that those distortions will have an impact. That's IMO one of the main problem.
You'll see the very same and stupid argumentation on the ethernet issue
as we've seen it in the early days of async USB audio.
"It's all digital, it's all async --- there can't be any impact"
Enjoy.
--------------------------------------------------------------------------------------------
::: Squeezebox Touch Toolbox 3.0 and more ::: by soundcheck
you are repeating the same kind of stuff that just scratch the surface.
> > The ethernet of course gathers the mess of all clients involved < <
I wonder if this is true.
To the best of my knowledge the standard requires an Ethernet card to be galvanic isolated.
The Well Tempered Computer
. . . the standard requires an Ethernet card to be galvanic isolated.
For perfectly proper safety reasons at 50 or 60 Hz mains frequencies, yes they do but AFAIK they have little to say about the KHz & MHz range.
How much "galvanic isolation" is provided at these frequencies by a couple of generic NICs connected by fifteen metres of unshielded cabling and (as SC notes) who knows what by way of ground is a moot point. I'd have thought that a reasonably decent, locally grounded switching hub (i.e. one that regenerates data) with a short, not-so-antenna-like connection to the target device, though certainly not a cure-all, less likely to degrade local conditions less than the usual PnP (Plug and Pollute) rat's nest.
As ever, YMMV.
Dave
"For perfectly proper safety reasons at 50 or 60 Hz mains frequencies, yes they do [have isolation] but AFAIK they have little to say about the KHz & MHz range."
Ok, I didn't recall either so I dug out a schematic of an interface and it looks like common mode stuff on the data lines see a 75 ohm resistor in series with a 1000pF cap to case ground. Now in this particular case my case is really isolated from the works but I have no idea how carefully the PC or Audio mfg's do theirs. The shield, if it exists goes directly to the case.
Soooo what I would do to try and address RF concerns is clamp a Z-bead over the LAN cable near the plug. The wanted signals are all differential and won't see it but the common mode, longitudinal stuff will and the effective impedance should reduce any detrimental effects.
Really a USB cable, especially an unshielded one, should not affect the sound...
Rick
In the light of Section 2 of the link below, I sited a Netgear GS105 hub close to my audio PC and connected the two with a home-made, 30 cm lead and stock RJ45 connectors (not a longer but fancy lead as SC suggests). It made a small but noticeable difference.
I chose the Netgear because it's small, inexpensive and metal-boxed but you can easily, as I did, test the idea with any old hub you have kicking around. As I understand it, the SB doesn't need gigabit speeds though my system, where data are transferred in bursts, starts playing noticeably faster with the faster LAN.
I haven't tried fancy cables.
HTH
Dave
All I've found is that they have to be long enough to reach, and made up with the wires in the correct order. You can get away with some mixed up pairs with 10BaseT, but not 100.
There's 7 layers in the OSI reference model for networks. Application, Presentation, Session, Transport, Network, Data Link, and Physical. It's enormously unlikely that subtle differences in the cabling will make an audible difference once the data has passed through all those preceding and subsequent layers in its journey from box a to box b.
"Layers" exist only in the imagination of network architects and standards committee weenies. (Been there, done that.) In the real world there is hardware in boxes and wires and some of these boxes run software which manipulates the hardware, causing it to do things that most programmers never imagined, because most programmers are clueless as to how hardware works, believing that the hardware magically manipulates bits and not electrons. At the very least, details of Ethernet cabling effect the electromagnetic environment, as these cables act as antennas to radiate interference as well as to receive it. The engineers who designed the Ethernet signalling were well aware of these issues and specified cabling, including shielding where needed.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
of the things that affects audio significantly on my setup is the Linksys Cable Router (Wireless disabled) and Thompson Modem.
Switching off stops a small degree of treble brightness (digitirise?)
Is the audio data going through the Linksys? If so, then you might try wiring the Ethernet source and sink directly by an Ethernet cable. That way you could then power the Linksys up and down while playing music to see if the effect is via the Ethernet cables or some other way. (If the Linksys is providing DHCP service this may not work if you aren't using static IP addresses.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
NO
If you get a chance, try playing around with the Linksys and the Ethernet cables connected to it and seeing what/when affects the brightness. If you have several categories of Ethernet cables, try them, including shielded ones.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It appears to be mains feedback. Both PC and modem/router are connected to filtered ups output
It might be possible to ascertain which audio components are being affected and how they are being affected by mixing and matching audio components and cables, but doing so might be a nightmare, depending on the cable routing. (I use random rat's nest wiring which is probably better for SQ than neat parallel cabling, but it is pretty awful when time comes to make equipment changes due to problems with cable ends handing up, etc. It would be nice to have more space.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I'd rather avoid the n factorial combinations and just switch the thing off when I need to.I guess I can always do away with the UPS (which annoys me with bleeps when the power is out) and just do a normal or protected mode restart when this happens. If something goes wrong, I can always restore. I shall then power components thru different isolating transformers.
But this too, appears too much trouble at present. I have always done this for my audio rigs, but not computer rigs.
Edits: 04/14/12
It would seem that you are focused on personal satisfaction, rather than love for truth.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Afraid that life is too short.
You may have noticed that I choose to do certain things but not others. My audio rig (including PC) is my reference, and I give this the works.
Otherwise, my work and other PCs are just things which I use.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: