Which company manufactures this Ethernet switch for the other?

Interesting discussion but FWIW, to clarify the photos shown in the first response to this thread:  

• The picture labeled "SA Bonn" is actually the Silent Angel Bonn N8 Pro model, and not the Silent Angel Bonn N8, which is the direct comparison to the English Electric 8Switch.
• The picture labeled "English Electric 8" is actually the Silent Angel Bonn N8, which you can see by looking at the corners of the case, which are square and constructed of sheet steel as in the Bonn N8 and not the rounded machined aluminium enclosure used for the English Electric 8Switch.

This Audiophile Style thread provides a further interesting comparison and value discussion about these switches, and pretty good pictures of the insides of two switches (see the third post by MarkusBarkus).  Notice the picture of the NuPrime Omnia SW-8 Network Switch shows the machined aluminium enclosure with rounded corners, exactly like the case used for the English Electric 8Switch.  Also notice it is the same board used in all three switches. 

Further down in the thread, Superdad, the owner of UpTone Audio, provides an answer to the original question posted in this thread:

"Thunder Data did the actual board themselves.  They market their product themselves as the Silent Angel Bonn N8.  And they sell the same board to NuPrime and English Electric."

You guys can read the rest of the thread and decide for yourselves, but the gist of it is that this is basically a $30 part, as was used in the Zyxel 8-Port Desktop Gigabit Ethernet Switch" (photos), and that it makes "no audible difference."

This article shows no difference in measurements after adding the Bonn N8 switch in the signal chain, and listening tests that correspond with the measurements. 

To wrap up, these manufacturers make a lot of money on these switches under the guise that "audiophile" switches make a significant improvement in sound quality, even going so far as recommending doubling up on their switches for further improvements.  I own two Bonn N8 switches and run both with 5V LPSs.  I have heard no reliably significant difference between having them in the signal chain or not, or between the Bonn N8, the Netgear switch I have here, or my TRENDnet 4-Port Gigabit Switch with SFP Slot that SGC sells for $160 but can be had on Amazon for $35.  YMMV

Mitch, what you wrote exacerbates my greatest audio worry — being exploited in my effort to wisely incorporate high value equipment in my audio system. I read various reviews of switches in ASR, which consistently state that all properly functioning Ethernet switches perform exactly the same. For that reason, I made sure that I could return it. 
 

All that said, last year I purchased a Shunyata Sigma Ethernet cable to connect my current Luxul switch with my Bricasti M21 DAC. This cable has two embedded noise filters. Without doubt, its replacement of the Audioquest Cinnamon cable i had been using provided a easily discernible improvement in the clarity of the sound. I am hardly going to spend thouands of dollars outfitting seven more connections with additional Shunyata Sigma Ethernet cables. Instead, for these lesser connections, I have purchased the English Electric 8Switch with the hope it will proved some improvement for them. 

@jmeyers, how much are you willing to spend on a dedicated network switch? If price is no object then I’d definitely recommend that you look into the Network Acoustics tempus - Precision Ultra Low Noise Ethernet Switch. Rob Osbourne and Rich Trussell are some of the most considerate and attentive folks in the audio gear biz that I’ve ever dealt with. They both took the time to answer every dumb question and email from me no matter how ponderous.

I’m still using their now discontinued 1st gen eno streaming system Ag in my headphone setup, and I can only imagine how great their new stuff sounds.

 In my experience with the English Eight and the SA Bonn switches, neither one had a better impact on the sound in my system than the Cisco SG 250 switch that I’m presently using in sequence with another readily available D-Link switch which I bought on Amazon. 

Let's see, these Ethernet switches support 100/1000 Base-T gigabit, which means the signal travels point-to-point over twisted pairs at speeds of either 100 million bits per second, or 1000 million bits per second.  This is much faster than Redbook Compact Disk needs which is about 1.4 million bits per second.

Please note carefully that Ethernet is not slowed down to match the audio rate! Instead the audio stream is chopped into packets up to 12,000 bits long (Gigabit can have non-standard longer packets if both ends agree!).  This means there is much more silence on an Ethernet Link than signal, punctuated by very short bursts of signal.

Each packet starts with a preamble, then the address of the sender and intended receiver (these are the Media Access Code or MAC addresses and are globally unique).  Finally, there is a cyclic redundancy check which allows corrupted packets to be detected.

The preamble is 56 bits alternating between 0 and 1 which allows the receiver to match the clock rate of the sender. 

When the receiver has to turn the packets into an audio signal, it has to assemble the incoming packets into a memory buffer, and then clock them out at a much slower rate which has nothing to do with the clock rate of the Ethernet!

Ethernet is essentially a broadcast technology, so all a connection box needs to do is listen to any transmission and broadcast it on all other connections.  This is what an Ethernet Repeater does.  An Ethernet Switch is slightly smarter - it looks at the MAC addresses in each packet and works out which physical cable connects to each address.  Then it only has to forward a packet down one cable (OK there is an exception for a broadcast message where the receiving address stands for all addresses).

My conclusion: jitter in an Ethernet switch is inconsequential because the audio stream has to be re-clocked from a memory buffer anyway. Errors in the digital payload are always detectable.  If you can hear a difference between switches, be suspicious of the EMI they are generating which affects your other components.

Pretty amazing the stream goes through all of those channels and circuits and you get music on the other end.