linear ps for switch and routers

>> Jitter Reduction, ReClocker, Signal Shaper, De-Noiser.

Unfortunately that sounds like quite a lot of poppycock for a network data switch..... Jitter reduction possibly - but of what ? It implies it takes the jitter out of data frames, and what signal is being 'shaped'???  - what a load of rubbish. I suspect if it was more mainstream Advertising Standards Authority might take an interest...

Simon-in-Suffolk posted:

Manu68, these Catalyst switches can bring a better performance of connected streamers... not dissimilar to the effects of fancy Ethernet leads. My theory of why is that these switches use better quality physical layer synchronisation clocks as they support PTP (Ethernet precision time protocol) which works in part at the physical layer. A better more precise clock will produce less intermodulation interference frequencies (think FM) in the  connected device thereby reducing digital noise.

Simon

ordered mine. I'll post whether I find any difference to it. Thanks for the recommendation Simon

What I'm most interested to find out about is whether one of these 2960 switches make any difference in sound quality BEFORE a pair of FMC convertors. Not taking a punt on it myself (yet) but I would want to replace my current switch with at least a 12 port model (currently an 8 port Netgear GS108) and the Cisco's can get pricey, even used. 

charlesphoto posted:

What I'm most interested to find out about is whether one of these 2960 switches make any difference in sound quality BEFORE a pair of FMC convertors. Not taking a punt on it myself (yet) but I would want to replace my current switch with at least a 12 port model (currently an 8 port Netgear GS108) and the Cisco's can get pricey, even used. 

i have the same interrogation. Without the optical bridge (fmc convertors) people find improvement with these pricey switches but with fmc or isolators i didn't found reviews on forum members.

Simon-in-Suffolk posted:

>> Jitter Reduction, ReClocker, Signal Shaper, De-Noiser.

Unfortunately that sounds like quite a lot of poppycock for a network data switch..... Jitter reduction possibly - but of what ? It implies it takes the jitter out of data frames, and what signal is being 'shaped'???  - what a load of rubbish. I suspect if it was more mainstream Advertising Standards Authority might take an interest...

i read just reviews of some people using this switch on the devialet chat. In a lot of forums( devialet chat- whatsbest forum- computer audiophile) the paul pang switch has a good reputation. But some had preferred this aqvox 8 switch. I can just say that.  You have not listen to it, so why say rubbish.  I have also observed that in a majority of topics you are advertising this cisco catalyst 2960 switch. Have you personal financial interest with it?  

For you optical bridge is not a good idea, ethernet isolators not also, high end ethernet cables not so....but this switch is always on the table.  Curious?

SImon is a well respected member here, stating his own opinion. If you don't like it, so be it but don't make these unnecessary innuendos.

C'est aussi simple que ça.

I think S-I-S was suggesting that the marketing speak was rubbish based on the technical claims, not the kit itself.

The Cisco 2960 was mentioned on here some time ago and there were claims of SQ improvements. So for £55 I took a punt on a pre-loved one and have been pleasantly surprised. Yes it improves SQ subtly IMO, and no, it is not expensive.

charlesphoto posted:

What I'm most interested to find out about is whether one of these 2960 switches make any difference in sound quality BEFORE a pair of FMC convertors. Not taking a punt on it myself (yet) but I would want to replace my current switch with at least a 12 port model (currently an 8 port Netgear GS108) and the Cisco's can get pricey, even used. 

Charles, I have an Etalon Isolator post switch. Once I get the Cisco switch, I'll be able to test the configuration you are asking about and I'll report here.

Manu68 posted:

SImon is a well respected member here, stating his own opinion. If you don't like it, so be it but don't make these unnecessary innuendos.

C'est aussi simple que ça.

i have the freedom to think by myself, thanks!  est ce assez claire comme ça?

nigelb posted:

I think S-I-S was suggesting that the marketing speak was rubbish based on the technical claims, not the kit itself.

The Cisco 2960 was mentioned on here some time ago and there were claims of SQ improvements. So for £55 I took a punt on a pre-loved one and have been pleasantly surprised. Yes it improves SQ subtly IMO, and no, it is not expensive.

i will perhaps take it. Have not said it is not an improvement, i don't know yet just.

Hi Keler,

Simon is a long standing member of the community and comes at these matters from a technical standpoint. His recommendation of this Cisco switch from a very large corporation, and available for a very reasonable 2nd hand price should, I would suggest, answer your own suspicions.

My experience is that everything makes a difference, but not necessarily an improvement. On another thread I reported on removing my EMO network isolator. On well recorded track it had a definite positive effect, on less well recorded material it was deleterious, to such an extent that I put it back in.

WRT AC Isolators, John Swenson is pushing another approach, that he links to leakage loops - edited:

Contrary to all "it makes sense to me" thoughts on the subject the way cut down on leakage loop noise with the DAC and pre/power amps is to have the tightest coupling you can get in the AC domain. Since a leakage loop goes through the AC main the higher the impedance in the mains side the higher the noise voltage generated between the boxes. I know the power conditioner is trying to suppress noise on the mains, but the methods frequently used actually increase the impedance between outlets thus increasing the noise from leakage loops.

So to test this hypothesis, try taking the power conditioner out of the system and just use a very simple power strip, no filters, nothing fancy, just outlets connected by wires. This will give a very low impedance between the AC to each power supply, which should cut down on the noise generated by leakage loops. Everything you now have connected to the power conditioner should go into the simple power strip, including all the digital stuff.

 I did this in my system (replaced a $1k power conditioner with a $35 power strip and Topaz isolation transformer) and it made a significant improvement is SQ.

JS recommends the Topaz:

The isolation transformer I'm using is a Topaz model, it is a very special transformer. It has extremely low inter-winding capacitance (.005pf according to the manufacturer). Most transformers have two mechanisms that transfer AC from one side to the other: magnetic and capacitive. The magnetic part is low frequency (it is what the 50/60Hz mains signal uses) and the capacitive is high frequency. The combination means that a "normal" transformer lets a lot of high frequency crud through.

With its extremely low capacitiance the Topaz doesn't pass the high frequency crud on the AV main, just the base line frequency and a couple harmonics. Thus it is a very effective noise filter.

In addition it is a very good surge suppressor as well. Most of the energy in high power surges is contained in high frequency components, which get suppressed by the low capacitance, thus it is quite an effective surge suppressor without needing any other special circuits to achieve this.

This isolation transformer keeps noise and surges from the rest of your house and neighborhood out of your audio system and fully preserves your safety ground.

Yes you are correct about the application, the Topaz plugs into the wall, the power strip plugs into it.

My recommendation is to use a simple power strip with NO filtering or surge suppression, the Topaz does it much better than what will come in almost any power strip. I plug EVERYTHING into thepower strip. This dramatically cuts down on the impedance between boxes, significantly lowering noise generated by leakage loops.

Some people will say "but then the noise injected back into the AC mains can go right into other boxes". Yes it can. BUT recent experiments have been pointing to the leakage loops being a significantly greater detriment to ultimate good sound than the injected noise. Of course different systems are different and this may not be true in all systems, but it is looking like this is a good place to start for many systems.

John S.

Mr Underhill posted:

Hi Keler,

Simon is a long standing member of the community and comes at these matters from a technical standpoint. His recommendation of this Cisco switch from a very large corporation, and available for a very reasonable 2nd hand price should, I would suggest, answer your own suspicions.

My experience is that everything makes a difference, but not necessarily an improvement. On another thread I reported on removing my EMO network isolator. On well recorded track it had a definite positive effect, on less well recorded material it was deleterious, to such an extent that I put it back in.

WRT AC Isolators, John Swenson is pushing another approach, that he links to leakage loops - edited:

Contrary to all "it makes sense to me" thoughts on the subject the way cut down on leakage loop noise with the DAC and pre/power amps is to have the tightest coupling you can get in the AC domain. Since a leakage loop goes through the AC main the higher the impedance in the mains side the higher the noise voltage generated between the boxes. I know the power conditioner is trying to suppress noise on the mains, but the methods frequently used actually increase the impedance between outlets thus increasing the noise from leakage loops.

So to test this hypothesis, try taking the power conditioner out of the system and just use a very simple power strip, no filters, nothing fancy, just outlets connected by wires. This will give a very low impedance between the AC to each power supply, which should cut down on the noise generated by leakage loops. Everything you now have connected to the power conditioner should go into the simple power strip, including all the digital stuff.

 I did this in my system (replaced a $1k power conditioner with a $35 power strip and Topaz isolation transformer) and it made a significant improvement is SQ.

 

JS recommends the Topaz:

The isolation transformer I'm using is a Topaz model, it is a very special transformer. It has extremely low inter-winding capacitance (.005pf according to the manufacturer). Most transformers have two mechanisms that transfer AC from one side to the other: magnetic and capacitive. The magnetic part is low frequency (it is what the 50/60Hz mains signal uses) and the capacitive is high frequency. The combination means that a "normal" transformer lets a lot of high frequency crud through.

With its extremely low capacitiance the Topaz doesn't pass the high frequency crud on the AV main, just the base line frequency and a couple harmonics. Thus it is a very effective noise filter.

In addition it is a very good surge suppressor as well. Most of the energy in high power surges is contained in high frequency components, which get suppressed by the low capacitance, thus it is quite an effective surge suppressor without needing any other special circuits to achieve this.

This isolation transformer keeps noise and surges from the rest of your house and neighborhood out of your audio system and fully preserves your safety ground.

Yes you are correct about the application, the Topaz plugs into the wall, the power strip plugs into it.

My recommendation is to use a simple power strip with NO filtering or surge suppression, the Topaz does it much better than what will come in almost any power strip. I plug EVERYTHING into thepower strip. This dramatically cuts down on the impedance between boxes, significantly lowering noise generated by leakage loops.

Some people will say "but then the noise injected back into the AC mains can go right into other boxes". Yes it can. BUT recent experiments have been pointing to the leakage loops being a significantly greater detriment to ultimate good sound than the injected noise. Of course different systems are different and this may not be true in all systems, but it is looking like this is a good place to start for many systems.

John S.

thanks you for your recommendation of the topaz transformer , i will look and take interest of it.  I appreciate this forum because people are sharing there experiences and helping people when they have technical problems with their system.

I am also trying sometimes to help when i can. The streaming/ ethernet topic is not simple, and relatively new for everyone.  There is no unique truth in the streaming world and this world is progressing very quickly.  

I am just a bit nervous when someone is thinking that he has the only good answer and other answers are not correct.  Ethernet isolators and optical bridges are acclaimed by a majority of users., but the switch question appears more delicate in opinion and user's experience.

To end, aqvox is a serious audio german enterprise and i doubt they will make a rubbish product.

I will follow the advises of a majority of naim members and take ifi powers on my switches.

For a bettter switch, aqvox or cisco catalyst or paul pang, i will take time and read users sharings( who have already optical bridges). 

And of course i will investigate the topaz transformer.  Thanks .

nigelb posted:

I think S-I-S was suggesting that the marketing speak was rubbish based on the technical claims, not the kit itself.

Correct - indeed Nigel. Always suspicious when it appears somebody/ some company has grabbed a string of technical sounding terms together to make a product sound good for those with no or little understanding/knowledge of the subject matter.

Not sure if I am a little off topic but after reading Andrew Everards article about going fibre optic network from router to switch I gave it a try and what a remarkable improvement, Galvanised isolation it's called. All the internet traffic noise is removed. No capacitance or resistance that copper has. For a little over a £ 100 you can not go wrong.

i agree at 100%.  Technical arguments can't explain all. The most important is to try and hear. 

For myself i had an ear tube preamp with naim amp. All dealers and forum members said to me that it is technically a wrong combination. But i tested and it works superbly!

sbilotta posted:

I put an ifi on a Paul Pang switch and there was a clear (even if not dramatic) improvement.

Should be getting an LPS-1 soon and will swap it out and let you know the outcome.

Got the LPS-1 and after some running in I put it on the Paul Pang switch last night (being fed by the ifi) and the improvement was very clear.

Ifi does work well on my Tp-link switch

Bought x2 Audioquest Vodka Ethernet to replace the Cinnamon cables and my world has blossomed. Sounds from what I can remember like  my old  Linn Sondek. Truly I can settle and relax with these. Very Analogue ish. 

Sedgey Rirrf posted:

Ifi does work well on my Tp-link switch

Bought x2 Audioquest Vodka Ethernet to replace the Cinnamon cables and my world has blossomed. Sounds from what I can remember like  my old  Linn Sondek. Truly I can settle and relax with these. Very Analogue ish. 

Have you compare them with diamonds?

Great question, I have had them a week so I am listening closely as they burn in incrementally. Audio T orders them in specially plus you need two cables so it's a lot of money, but who knows one day      You have got me wondering now...

Sedgey Rirrf posted:

Ifi does work well on my Tp-link switch

Bought x2 Audioquest Vodka Ethernet to replace the Cinnamon cables and my world has blossomed. Sounds from what I can remember like  my old  Linn Sondek. Truly I can settle and relax with these. Very Analogue ish. 

good news. I am waiting ifi power for my switch and a second one for my last switch on the optical bridge. It will be cool...

sbilotta posted:

sbilotta posted:

I put an ifi on a Paul Pang switch and there was a clear (even if not dramatic) improvement.

Should be getting an LPS-1 soon and will swap it out and let you know the outcome.

Got the LPS-1 and after some running in I put it on the Paul Pang switch last night (being fed by the ifi) and the improvement was very clear.

so lps 1 is an improvement over ifi power on your switch?    do you have an optical bridge or ethernet isolator?  

Sedgey Rirrf posted:

Great question, I have had them a week so I am listening closely as they burn in incrementally. Audio T orders them in specially plus you need two cables so it's a lot of money, but who knows one day      You have got me wondering now...

i have an optical bridge. After i bought an audioquest diamond to replace the meicord on my last switch before the nds and could detect no improvement, or perhaps very little....audioquest vodka is enough i think .

Sedgey Rirrf posted:

Not sure if I am a little off topic but after reading Andrew Everards article about going fibre optic network from router to switch I gave it a try and what a remarkable improvement, Galvanised isolation it's called. All the internet traffic noise is removed. No capacitance or resistance that copper has. For a little over a £ 100 you can not go wrong.

Hi, for your info all wired Ethernet should be galvanically isolated by design. It's often achieved by little transformers in the NICs. Using a BaseT to BaseX concertvers will/should have Galvanic isolators in their BaseT NICs just like any compliant Ethernet device. There will be no difference to 'internet traffic noise'. That is the same whether you use fibre or twisted pair for the physical link.

What fibre can offer is, in some circumstances, a removal of high frequency or radio frequency common mode noise conducting along the Ethernet cables. This may occur because there is a electrically noisy or poor quality network component / host physically attached to your network. This is the same sort of effect of a noisey device plugged into your mains. It might be worth trying to hunt out out the device that is adding the common mode electrical noise on your Ethernet, you never know it could if sufficiently be bad cause issues elsewhere. You might find  it's removal or replacement gives a better performance boost rather than relying on sticking plasters.

One point to consider  however is that the main role of BaseT to BaseX converters is to extend range not reducing common mode noise. You need to ensure the noise generated from the converter itself is not undermining to some extent what you are trying to achieve, a potential risk with cheap consumer devices. Again I would try if you can at using devices with fibre SFPs and run the fibres between them and if you ensure the SFP host is a quality low noise device like a quality switch and then you should be confident of being in a good place.

Simon

Keler Pierre posted:

sbilotta posted:

sbilotta posted:

I put an ifi on a Paul Pang switch and there was a clear (even if not dramatic) improvement.

Should be getting an LPS-1 soon and will swap it out and let you know the outcome.

Got the LPS-1 and after some running in I put it on the Paul Pang switch last night (being fed by the ifi) and the improvement was very clear.

so lps 1 is an improvement over ifi power on your switch?    do you have an optical bridge or ethernet isolator?  

Yes, LPS-1 is definitely better than the IFI.

I had an optical bridge but I didn't feel it bettered anything so I took it off, but it is also true that I have Acoustic Reveive Lan Isolator (which does work !) so probably that is already doing a good job and made the optical bridge less (or not) worthwhile.

for me too, LPS-1 is definitely better than ifi, same experience