Everyone here whom streams recognises the importance of a robust network set-up - including investing in decent routers/switches, broadband package etc.
Obviously we try out new solutions as they arise, but opinions may vary as to whether they add anything. That’s why demo’ing kit is so important.
It’s more nuanced than that. I feel with these things there are a range of outcomes:
Makes a clear difference. Could be a difference you like or one you dislike. Either way happy days: you’ve either found a system upgrade or saved yourself some money.
Makes no discernible difference. Possibly because you already have a great set-up.
Makes negligible difference. Then only you can decide if the investment is worth it.
With the risk of sounding like a broken record: it would in this case still require more supporting evidence that the device itself is causing the difference, rather than other (environ-)mental factors.
In a technical sense, there are almost no conceivable pathways for a network switch upgrade to cause a ‘clear difference’ in sound output, unless the previously used switch was broken or malfunctioning.
Some switches are inherently electrically noisy (Ethernet transmission can cope with a surprising level of noise on the cables), either through the attached network cables or by injecting RF onto the mains, even though they aren’t defective. This level of noise can disturb analogue electronics attached to it (such as a streamer and DAC or amplifiers on the same mains circuit). In this case replacing that switch with one that is electrically quieter can make a ‘clear’ difference, but I would agree that much more often the difference is marginal - as will be the case will any well designed switch.
The important point here is that it isn’t the digital signal through the switch that’s being directly influenced by the switch, it’s the secondary effect of RF on the analogue circuitry. This is why the results are so variable from system to system - and the same applies to Ethernet cables.
Incidentally, there can also be an interaction between the incoming RF and the firmware of the streamer & DAC.
Maybe borrow something like the Innuos Phoenix net which is at the other end of the price scale to the EE8. A purposely designed, rather than a repurposed / modified switch with a built in linear supply etc so it ticks all the audiophile boxes. Otherwise it’s hard to go wrong with the trusty Cisco 2960.
I would try another EE8, because it shouldn’t sound bad, at max the improvements can be very minor or eventually inexistant.
I suspect that one is faulty.
The Netgear gs108t still seems to be the ‘standard’ reliable, low-noise switch for home environments. At a price of about 10-15% of an EE8 it appears there would be few situations where the latter would noticeably outperform the former to a degree that would be discernable by an average listener.
So for most listening situations, an EE8 could already be regarded as a somewhat ‘excessive’ expense.
My personal null hypothesis for digital cables and switches is that it makes no difference, because digital. So if it was removed because it made no difference, I would not be surprised.
If Naim staff tried and it did make any difference, good or bad, that’s what I would find interesting. In this case it is then maybe a matter of taste
Yes, that’s the well known boogeyman that keeps getting regurgitated without anyone ever having demonstrated it, in the same way as when CD playback was considered possessed by demons, something that nobody talks about anymore after people got used to it.