Nopiano

Ethernet switch - benefits?

Recommended Posts

I can try from work which is 6 miles away, but that will have to wait until weekend. However it saves me from the methodical approach of 70 X -² until I get home. Hang on.. I wont actually get home.. will I??

Share this post


Link to post
Share on other sites
Super Wammer

Well despite the very sound advice given I have decided to get a cheap and cheerful Ethernet Switch a TP-Link 8 port metal case unit for the princely sum of £16 . This was not for any ideas that it would improve the sound quality but because I have been suffering a bit with drop outs and issues both listening to music and streaming on my TV .

Arrived today and after closing everything down and then connecting the switch to the router and everything else to the switch I booted the system and got nothing no connections . Quicl look at my admin for the router let me see the issue and from then on all today everything has been going through the new Switch . No dropouts even at the normal times of 17.30 to about 18.30 which seems to be a high point for issues . TV worked either streamed from BT , using Fire TV or even through KODI . I have also been working on my main mucis system detailed elsewhere (Owners MiniDSP thread and 2 Channel Main System Issues thread) so from about 11.30 aam been listening to music right through to about 19.00 when I had a conference call with my Brothers and Sisters . Really happy with the results and at this time everything is very stable and just works which is what I want . I have no real honest opinion on if using a switch has improved sound audio quality if it has then it is very subtle and not the night and day claimed in many places .

I know that the next thing will be that the switch is too cheap , the power supply is not clean enough and lots of other theories all of which I am sure make a difference to some. Not to me but I will say that adding the switch does seem at this early stage to have settled and made my network connections to all items that bit better. You have very little (£16) to lose by trying this and I would say even if it just makes things stable as it has my network then it is well worth a try .

If you want to try a fully upgraded and breathed on SwitCh by any numebr of companies then please feel free I have no idea if it will help but I will not be spending anymore than I have now .

  • Like 1
  • Upvote 2

Share this post


Link to post
Share on other sites
Moderator
Posted (edited)
4 hours ago, bencat said:

If you want to try a fully upgraded and breathed on SwitCh by any numebr of companies then please feel free I have no idea if it will help but I will not be spending anymore than I have now .

Hmmm...let me see...£16 outlay for something that clearly does the job vs £1000 for some snake oil. Tough choice there Andrew! :dunno: :nup:

Edited by Tony_J
  • Like 2

Share this post


Link to post
Share on other sites

Share this post


Link to post
Share on other sites
On 05/04/2020 at 20:26, sq225917 said:

His reviews need to be taken with a latge pinch of salt and in respect of the digital path I believe they are irrelevant because what is happening cannot be measured by his equipment. I go on the evedence of my ears and the Cisco 2960G has made a significant difference.  For the £56 I paid it was a steal.

  • Like 2

Share this post


Link to post
Share on other sites
Super Wammer
2 hours ago, ziggy said:

His reviews need to be taken with a latge pinch of salt and in respect of the digital path I believe they are irrelevant because what is happening cannot be measured by his equipment. I go on the evedence of my ears and the Cisco 2960G has made a significant difference.  For the £56 I paid it was a steal.

Can you please indicate which model Cisco you have Cisco 2960G  is the first prefix for a whole family of switches 8 port , 16 Port , 24 Port , 48 Port etc . Is the below what you are talking about .

https://www.ebay.co.uk/itm/CISCO-2960G-WS-C2960G-8TC-L-v02-8-port-gigabit-ethernet-switch/333568853435?epid=1977773809&hash=item4daa40c5bb:g:2MkAAOSwD89bvG~b

Share this post


Link to post
Share on other sites
59 minutes ago, bencat said:

Can you please indicate which model Cisco you have Cisco 2960G  is the first prefix for a whole family of switches 8 port , 16 Port , 24 Port , 48 Port etc . Is the below what you are talking about .

https://www.ebay.co.uk/itm/CISCO-2960G-WS-C2960G-8TC-L-v02-8-port-gigabit-ethernet-switch/333568853435?epid=1977773809&hash=item4daa40c5bb:g:2MkAAOSwD89bvG~b

That is the one I have. I would not have considered more than eight ports.

Share this post


Link to post
Share on other sites
On 05/04/2020 at 20:26, sq225917 said:

This is likely to be correct in 99+% of use cases. 

There's a quick check you can do to see if a switch will improve things for you.

1. Download Ping Plotter  https://www.pingplotter.com

2. Run it and type in the IP address of your NAS / music server

3.  Let it run for a few mins or so.

Here's a result from mine.  You are looking for three things:

a. Is there a number in the PL% column.  If you see a number between 1 - 99%  that could be indicative that you're losing packets of information. Most minor issues will show up as a small loss e.g. 10-15%  (NB. if the number is 100% it means that NAS isn't on, you've typed the wrong IP address, or the NAS is firewalled to stop it responding to these types of tests.) 

b. Are there any red lines in the graph. That means also means there was loss, as above.

c.  Is there much divergence between Avg Min and Cur.  This would indicate jitter.   Typical numbers should be no more than ~ 5 or 6 even on a home network.  If you are seeing frequent  big jumps it could indicate a problem.  Another way to look for this is to look at the graph and make sure it's nice and smooth,  although pay attention to the scale - small variations can look bad on a a zoomed in scale like below.  I had a couple of jumps up to ~ 5ms but still well within target.  I wouldn't really be worried unless it was really jumping about.   If you are on wifi allow a bit more tolerance.

As long as it's mostly stable without erratic jumps or red lines you're fine. A new switch won't help.

Here's an example on a wired connection:

image.thumb.png.cb37f8123e5cdd8e44aac17e31deb67f.png

Here's one on wifi - as you can see it's pretty stable at ~ 9.8ms, but a small blip of latency right up to 210ms.  A one off like this isn't really a concern, but if it was doing a lot of this then try either a wired link instead or a different wifi channel. 

image.thumb.png.f193ec23ad2d69db2cf62d698cd23ebd.png

Share this post


Link to post
Share on other sites
Super Wammer
16 hours ago, jamster said:

This is likely to be correct in 99+% of use cases. 

There's a quick check you can do to see if a switch will improve things for you.

Crumbs, a lot of that is rather lost on me, but I get the general drift.  Thanks for sharing. 

I take it that you don’t believe there is any other sort of unwanted ‘noise’ that a switch might reduce or eliminate, yet not be evident in the ping plotter test?  I use the term noise widely, as it might be some type of interference, either from within the home from, say, a Dect phone charger, or externally from, say, an industrial plant.

Share this post


Link to post
Share on other sites
Crumbs, a lot of that is rather lost on me, but I get the general drift.  Thanks for sharing. 
I take it that you don’t believe there is any other sort of unwanted ‘noise’ that a switch might reduce or eliminate, yet not be evident in the ping plotter test?  I use the term noise widely, as it might be some type of interference, either from within the home from, say, a Dect phone charger, or externally from, say, an industrial plant.
That's right.

The way data is transmitted on ethernet means that you would be very likely to see the effects of excessive noise as packet loss or jitter.

The quality of a switch is unlikely to make so much of a difference here, even the basic £15 hobbies are pretty well built. They have economies of scale which are many multiples of the boutique hifi companies.

Don't get me wrong, a faulty switch could cause problems, which could (edge case) affect audio quality, but you'd see it through stuttering / dropouts, or reduced bitrate (on systems that do transcoding / adaptive bitrate - not many do and certainly not audiophile ones).

Also some switches have shitty firmware which can stop some things working properly eg. Upnp / dnla but again this is quite rare, and again, it would be pretty obvious that it wasn't working.





Sent from my SM-N960F using Tapatalk

  • Like 1

Share this post


Link to post
Share on other sites
On 07/04/2020 at 20:28, jamster said:

This is likely to be correct in 99+% of use cases. 

There's a quick check you can do to see if a switch will improve things for you.

1. Download Ping Plotter  https://www.pingplotter.com

2. Run it and type in the IP address of your NAS / music server

3.  Let it run for a few mins or so.

Here's a result from mine.  You are looking for three things:

a. Is there a number in the PL% column.  If you see a number between 1 - 99%  that could be indicative that you're losing packets of information. Most minor issues will show up as a small loss e.g. 10-15%  (NB. if the number is 100% it means that NAS isn't on, you've typed the wrong IP address, or the NAS is firewalled to stop it responding to these types of tests.) 

b. Are there any red lines in the graph. That means also means there was loss, as above.

c.  Is there much divergence between Avg Min and Cur.  This would indicate jitter.   Typical numbers should be no more than ~ 5 or 6 even on a home network.  If you are seeing frequent  big jumps it could indicate a problem.  Another way to look for this is to look at the graph and make sure it's nice and smooth,  although pay attention to the scale - small variations can look bad on a a zoomed in scale like below.  I had a couple of jumps up to ~ 5ms but still well within target.  I wouldn't really be worried unless it was really jumping about.   If you are on wifi allow a bit more tolerance.

As long as it's mostly stable without erratic jumps or red lines you're fine. A new switch won't help.

Here's an example on a wired connection:

image.thumb.png.cb37f8123e5cdd8e44aac17e31deb67f.png

Here's one on wifi - as you can see it's pretty stable at ~ 9.8ms, but a small blip of latency right up to 210ms.  A one off like this isn't really a concern, but if it was doing a lot of this then try either a wired link instead or a different wifi channel. 

image.thumb.png.f193ec23ad2d69db2cf62d698cd23ebd.png

Thank you for the software link and instructions. It's been extremely useful.

I have been trying various ethernet options between my router and streamer. 

Namely: 

1. Cat6 cable (original setup)

2. Cat8 cable - cheap

3. Cat8 cable - expensive

4. Fiber optic cable - to isolate noise(?) : 

https://www.audiostream.com/content/electrically-isolate-your-networked-audio

https://andreweverard.com/2015/06/08/high-resolution-audio-now-with-added-fibre/

To me, they all sounded different. No idea why.

However, last week, just to check and baseline my experiments, I tried using wireless after ordering the proprietary wireless dongle for my streamer. 

Long story short, in my setup, this proved to be the best sounding option with the most transparent, balanced and engaging sound of all.

A real shock, after using a wired connection for the past 10+ years.

However, after using the PingPlotter to track the stats to the streamer in wireless mode, I could see the average was over 15ms. On closer inspection, this was due to a regular spike in the metrics, every 30 secs or so, where the latency rode up to beyond 300ms, similar to the spike on your wireless graph.

What's going on? Is the streamer or router wifi connection continuously refreshing/reconnecting?

I tried changing the wireless channel, but to no effect.

It turns out to be a background process running on my Windows 10 PC. 

This article explains how to disable this process : https://hitech-us.com/articles/entry/292/Ping-spikes-on-Wi-Fi-every-10-seconds-Solved

(I also tried  enabling/disabling the OneDrive settings explained later on in the article. In my setup, it made no difference.)

The wireless stats have dropped down to an average 2.4ms, with no spikes beyond 59ms and no packet loss. 

So, my psychoneurosis around the connection to the streamer has been cured. 

No more looking at uber expensive ethernet cables, lan isolators, audio switches, router power supplies etc. 

Just listening to music.

  • Like 1

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Recently Browsing   0 members

    No registered users viewing this page.