Posh switch - another cause of Audiophilia nervosa?

Hi, looking at the 2960, 8 port, you can have a used gigabit version (2960G) for 152 £ or a 100 bit version for 50 £, is the additonal 100 extra good value for money???. Are all 2960 8 port switches, with a build in powersupply, or is there a need for an external PS??

Will use it with 272/555 and 300dr, with a synology NAS and US

Tread carefully, I have not evaluated the gagibit versions from a SQ performance. The PHY sub architecture may well be different on the 1000 BaseT ports compared to the 100BaseT, even though the Naim streamer will be identified as 100BaseT by the switch. This may or may not affect the PHY clock jitter...(certainly 1000BaseT operates differently to 100BaseT at the physical layer), either way it would be unwise to assume a 1000BaseT capable port sounds the same as a 100BaseT port so should be compared and evaluated first.

Switch configuration and functionality, which is not the main motivation here for SQ, will however be the same.

All the 2960 switches have inbuilt powersupllies apart from one of the 8 port PoE variants which can be PoE powered or powered by an external DC supply. I use one of these with its DC supply.

Thanks for info, so to be safe, as I can't compare, is to buy the 100 as this is known stuff, also compare to only pay 50 

Bandwagon jumped on to : yesterday I fleabayed a 2960 8TC-L for £52. Arrives Monday. Will report back. Interestingly, we had some of our oldest friends staying with us yesterday. He's an IT specialist by trade, ex BT and manages various Corporate clients for their IT needs. Like many pro's, he believes that digital signals can't be corrupted because everything can be resampled several times until the data packet is perfect etc, etc, but he DOES believe that switches can make a difference and my multi point Zyxel into which my music system switches is not the best and could be an issue. He still doesn't really believe it will make a difference, but does say " it certainly can't do any harm and the Cisco switches are reliable and well designed. An IEC power switch could also remove the possibility of an SMPS causing noise issues" 

if the benefits are as described above : Scale, Authority, Ease, Flow then these are things I value very highly. Perhaps too often we audiophiles look for more detail as sign of an improvement, but that doesn't always guarantee better music in my experience.

anyway, as HH says, for the money it's worth a try.

kevin J Carden posted:

 

..... An IEC power switch could also remove the possibility of an SMPS causing noise issues

Who said the internal Cisco PS is not a SMPS    ???  Only asking -- honest.   But as the halibut says,  its worth a try at 50 squids a pop.      

Mike, very, very likely I misinterpreted his reasoning, but he approved of not having the SMPS "on a rope" as he put it. I'm another complete numpty when it comes to all this stuff. 

One of the bigger audio chain stores recommends these,they start around $350ish.....and are industrial/commercial andmfg by the same company that makes the Wiremold strip

https://luxul.com/

No probs Kevin,  it's all too easy with this network stuff.       If it sounds better, no matter what the reason, with or without a SMPS, even one on a rope, it's worth the squids. My problem is I've tried a few switches & with the exception of a poor sounding 10/100 switch,  there was little or nothing to choose between the Gigabit types.

Mike-B posted:

kevin J Carden posted:

 

..... An IEC power switch could also remove the possibility of an SMPS causing noise issues

Who said the internal Cisco PS is not a SMPS    ???  Only asking -- honest.   But as the halibut says,  its worth a try at 50 squids a pop.      

The 2960s with inbuilt power supplies, definitely have a screened SMPS built into the device.

As i said a few months ago - i really don't think is anything to do with SMPS 'noise' at this level - but the PHY layer clock. This clock is critical and can be a cheap low precision oscillator or a precision crystal controlled clock with good regulation. This transport clock jitter (approx centred at 30MHz for 100BaseT)  will most likely cause intermod frequencies in connected devices. Most digital equipment doesn't care - however audio does as these alien intermodulated frequencies can modulate PCM sample clocks and delicate analogue stage power supplies. This is similar to the effect of  transport jitter on a SPDIF framing signal.

I also posted a TI engineering white paper on this forum the other week that discussed some of these issues

S

The switch has been running for a few days now, and it's interesting to reflect on the change once the initial 'is it better or isn't it?' phase has passed. 

I'd say that yes, it's definitely better, but not in a Hifi bass and treble way. There is greater drive, and the music seems to flow better. It's just more real, more immersive, more head nodding, with more groove. The system sounds bigger and more solid. It's hard to describe without sounding bonkers, but that's what the difference means to me. 

Another question I've been asking myself: how does it compare with the posh cables? I've recently bought two AQ Vodka cables, worth £500, and now the secondhand Cisco. From memory, I'd say that the switch makes at least as big a difference, but bearing in mind that the switch built on the cables, which came first. I don't know if the difference would have been as big if the switch had come first, and lest anyone ask, I'm not going to try the expreriment. 

What I can say though, is that it's been more than worth the effort, and the system is sounding better than ever. 

I'm particularly grateful to Simon, not just for going on about the benefits, but for giving me the confidence to try it for myself. Thanks!!

Will you at some stage revert for a short period to confirm that you are hearing a difference, or are you sufficiently convinced already?

Harry - I'm not going to bother. Even if there is no difference, the fact that I think there is, and that the music is more enjoyable, is good enough. In this instance, the heart is ruling the head. I changed my speaker cables, and didn't bother to revert to the old ones. The same for the din to XLR, and the same for the Vodkas, so I'll do the same for the switch. 

Anyway, with my head injury, every time I put my head on one side everything spins and I feel awful. Installing the new switch entailed lying on the floor on my side and fumbling around under the sideboard, and I have no desire whatsoever to repeat the process!!

Good to hear its been a success & hope it continues HH,  dare I mention "burn in"   

If I had the space in my NAS cabinet I would definitely go for one,  for 50 squids, what's to lose.  My problem is I am space limited with my existing switch location & whilst I'm always open to every/anything that adds to better SQ,  I'm not prepared to change furniture and/or system layout,   but I'm always on the lookout for a mini "Catalyst" eq performance switch.  

I have a Cisco SG-300 28 port switch as the main distribution point, but drop fibre to a dedicated media converter and Netgear GS-105, both running off a bespoke linear power supply that only connects to my NDS. One day I may drop a cat-6 cable down as well to do an A/B comparison, but for now just enjoy the music. My rationale was to have the simplest possible 'clean' network dedicated to the HiFi, galvanically isolated from all the other consumer grade stuff. 

How does your nas connect to the network?

Direct to main Cisco switch.

I wonder if it would be better to connect it to the Netgear, to isolate it from everything else, which seems to be best practice. And then you could try a 2960 in place of the Netgear? Just a thought. 

Welcome thought... there's certainly no universally correct answer here, however I'm more inclined to keep the NAS away as I have no idea how clean it's power supply is. It's a QNAP TS-453S Pro, with 4 1Gb SSDs. I'm pretty convinced that any audible issues result from poor power supplies coupled through copper cables.

We have a 253A, with its standard power supply. I always assume that because the stereo has its own mains that the power supply issue can be set aside. The nas lives in the dining room, with the switch, on the other side of the wall from the stereo in the sitting room. So the wire from nas to switch is 75cm long, and the one from switch to 272, through the wall, is only 150cm. The network is, therefore, compact.

DaveBk posted:

Welcome thought... there's certainly no universally correct answer here, however I'm more inclined to keep the NAS away as I have no idea how clean it's power supply is. It's a QNAP TS-453S Pro, with 4 1Gb SSDs. I'm pretty convinced that any audible issues result from poor power supplies coupled through copper cables.

Yes, I'm sure issues occur from electrical noise with cheaper devices, but when you go to higher quality switches and/or fibre you still hear the benefit of reduced Ethernet transport clock jitter... Clock jitter is no stranger to audio SQ related issues... 

Ok so a little late to chime in but +1 to real switches and network equipment - at first I had the N-172 hooked up to a 4 port "dumb" netgear switch - but since have brought my professional life into the house and upgraded my whole network to be a full stack of Cisco Meraki equipment - MX65 router/gateway - 2x MS220-8P - MR36 AP's

I have also done some bandwidth control and VLAN isolation of the music system from the rest of the network and have POE off on the ports connected to non POE Equiptment

Since the swap I have better connectivity with tidal, slightly fuller sound - more engaging and addictive, faster response with the naim app and browsing the NAS overall it is an alarming improvement - and the significant other is happy with the faster wifi!

Win win situation 

Hi, I am not sure how you are achieving your egress traffic shaping ... but Tidal, and other cloud services can suffer with Naim streamers if you are dropping TCP packets because of the web latency and the limited current Naim streamer buffers... so would be interested to hear.. or perhaps you are just shaping the ingress, or using some sort of PfR/CoS to give the highest priority to web streaming, or home audio VLAN, and effectively shaping everything else?

Simon-in-Suffolk posted:

DaveBk posted:

Welcome thought... there's certainly no universally correct answer here, however I'm more inclined to keep the NAS away as I have no idea how clean it's power supply is. It's a QNAP TS-453S Pro, with 4 1Gb SSDs. I'm pretty convinced that any audible issues result from poor power supplies coupled through copper cables.

Yes, I'm sure issues occur from electrical noise with cheaper devices, but when you go to higher quality switches and/or fibre you still hear the benefit of reduced Ethernet transport clock jitter... Clock jitter is no stranger to audio SQ related issues... 

Yes, I agree in principle, but I would be interested to understand the relative magnitude of the effects. Just how much better is the clock accuracy on a professional switch vs. a simple unmanaged consumer item? How does this compare in terms of audible impact against the detrimental influence of a noisy SMPS?  Very difficult to compare, so I'm happy with my choices at present.

Every time TCP is mentioned it reminds me of childhood. I don't recall the TCP packet, merely the bottle. 

Is it possible connect a NAS directly to a streamer, therefore eliminating the need for a switch.