Any dropouts will occur as a result of interference or airtime allocation rather than bandwidth or throughput constraints.
As Wi-Fi is a shared domain, where everyone has access to the same channels in fixed frequency bands, the streamer as a client has to request airtime and for each available time slot of airtime, the Access Point will provide services on a first come first served basis unless you have determined that you wish to fix priority for a particular traffic class or type, in which case that traffic will be granted priority and jump the queue.
More recent revisions of the Wi-Fi standards allow for more efficient seperation of traffic and higher concurrency to overcome the challenge of airtime fair use especially where best effort traffic classes are concerned.
Itâs able to do those things autonomously and can even adapt to the environment based on utilisation behaviour and traffic modelling, in some cases using machine learning to enable those adaptions to the environment of operation.
Most of what weâre dealing with today in a residential setting lacks that intelligence but it will become the main stream within 3-4 years, certainly through 2024 Wi-Fi 6 will become the baseline mass market solution and Wi-Fi 7 brings yet further gains, one of those key gains being significantly reduced link latency which will benefit streamer devices amongst others.
Thank you for the competent explanation and the outlook. It is exciting to see what perspectives WiFi also has for audio components. Do you think that WMM (Wireless Multimedia Extensions - 802.11e) is coming and that this will have a positive impact on sound?
They sound like a diesel car manufacturer!
Used Netperf with the following results.
If I read them correctly, then my figures seem to be quite good.
Itâs your opinion, perhaps your experience, but not a fact. We are so many here and elsewhere to experience noticeable improvements when changing the stock SMPS on standard switches or going up the ladder in prices with audiophile switches. I had for example the Netgear cheap switch before, with its stock SMPS. Changing that stock ps with a quality linear ps was at that time a « Wow ! « moment. Absolutely not meaningful.
The Cisco indeed was also a clear and nice improvement vs the Netgear. But we have to not forget that itâs not a cheap switch, but costing around 600 euros new. Going to an Etheregen / linear ps at 3 times the cost of a new Cisco was a « Wow ! « moment too. It it was subtle I would send it back, save the money, and bought later something else.
I never suggested it was a fact, merely highlighting my own views as you yourself do frequently, such is the nature of an active forum and the basis for reflection and discussion.
What is clear is you donât neccessarily need to spend 1000âs to find improvements you like and are willing to accommodate and keep longer term.
Not everyone has deep pockets or a desire to find the âceilingâ after all, myself included for that matter!
Looks good to me (with my telecoms engineer hat on)
I donât know if you have tried audiophile switches or linear ps on cheap switches. If yes, I respect your point of view. If not, your point of view is not really one, just the same kind of thinking which is spread of so frequently here by persons who are tekies, work in the pc / network professional area and are persuaded that their knowledge applies in the audio domain. Without even trying such products they declare in advance that audiophile switches are just fancy expensive boxes that, at best, will give very subtle improvements.
I have, including EE, Innuos and Melco for the record both in my own environment and under demonstration conditions.
Hopefully that fulfills your stringent criteria for contributions.
But it does. Audio doesnât get a free pass here but there are additional mechanisms at play (discussed many times) which can affect the final result in the audio domain. Itâs good to have people contribute from both sides.
Importantly, and wherever possible, I try to focus on the needs of the OP with consideration for the fact they may be new here, may lack knowledge and personal experience I am happy to share both in my professional capacity as well as a Naim system owner (two in fact) and in sharing insights, try to present them in a straightforward and easy to digest format.
I have tried these things out of my own curiosity and for me personally found what I required and was willing to finance by other means.
That has little to do with my background or worklife and more to do with the enjoyment of my system within my means.
Simple really when you think about it!
You cannot rely upon SpeedTest.
I ran several tests on their SpeedTest browser version and then ran several tests using their SpeedTest App on the same PC.
I got wildly different results.
I repeated the tests several times always with similar results.
These tests were run from the same PC at the same time.
I reported the issue to Ookla and three reminders.
They did not even bother to respond.
The browser reported 205 Mbps and the App reported 97 Mbps!
Such is the difference between line rate, throughput and available bandwidth. Thereâs also the fact itâs best effort so will get to you in a rather large queue of terabytes of other packets journeying across the Internet.
If you run a sufficient number of tests you can at least extrapolate a mean, max and median value rather than expecting a fixed single value to present itself.
Iâd focus more on latency than anything else, bandwidth is unlikely to be an issue certainly where streaming audio is concerned.
That is why I ran multiple tests on both the browser and the App to ensure the results quoted were representative. I was getting very similar results but same variations between the browser and the App.
What concerns me is that Ookla do not care.
It fullfills them completely, yes. Just find it strange that neither of them satisfied your ears. But have nothing to criticise then.
It does partially only. I am not convinced that the particular mechanisms are all well known , because none of the members hear work in the audio science domain. There are only speculations and always the same in general.
I know you have an open mind and have observed clear improvements with your PhoenixNet. My comments were more towards pc/ network tekies in general, who donât work in the audio specialised area.
If I may. Thatâs true, but itâs just as true for the other âfactionâ who, in part, are not shy to conjure up technical âexplanationsâ on the spot, base everything on highly subjective impressions, and have no problem ignoring established cognitive and other science.
It would be good for the forum if both sides could admit to their shortcomings. Thatâs all I have to say.
I understand where you are coming from FR. All Iâm saying is that there is no magic in the audio domain and the same principles apply. I do have a PhoenixNET and find it does benefit my system much more than I expected, but I still wouldnât admit to my work colleagues that I spent 2.5k on an Ethernet switchâŠ
Damn this hobby 
I didnât understood that the OP really needs something. He has questions, yes, but mainly he shared his experience with some Ethernet cables and EE switch vs basic cables and cheap switch as the Tp link. He observed very good results with the Chord switch. His ears.
So does he really need to read that audiophile switches, at best, are just fancy expensive boxes which will have no or eventually just minimal impact on sound ?
Thereâs no magic, but knowledge, I agree. However a knowledge which is grounded only partially on the one you have, or some members have here. If not, why have spent nearly 3k on the PhoenixNet if itâs so simple for you. Just buy the components you need and build it yourself, as you know exactly how this works. 