• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

The Truth About HiFi Network Devices

Or will the performance of these systems evolve to support at least lower latencies

On latencies. I spent a while working in a stock exchange / brokerages as a programmer. We had clients who placed their servers in datacentres in Hawaii so they could trade off the speed of light margins between the west coast US and Japan/Asia markets.

Most of them would be paying a premium (by distance at the speed of light in fibre) as to how close their servers got racked in the data centre with the core stock exchange gateway in it. A 1 hop, 10 nano-second rack slot might cost you a million a month.

But the stock exchange and international markets are "Free, open and fair". Not really.
 
On latencies. I spent a while working in a stock exchange / brokerages as a programmer. We had clients who placed their servers in datacentres in Hawaii so they could trade off the speed of light margins between the west coast US and Japan/Asia markets.

Most of them would be paying a premium (by distance at the speed of light in fibre) as to how close their servers got racked in the data centre with the core stock exchange gateway in it. A 1 hop, 10 nano-second rack slot might cost you a million a month.

But the stock exchange and international markets are "Free, open and fair". Not really.
This is partly why the UK gov invested in the National Timing centre that's set up by the org I work for.

I have a question, as I can grasp the challenge of streaming a single source file to multiple (millions) of end points, but presumably conventional EM technology is live, in the sense that I would receive the TV image at the same perceptible time as my neighbour did? And presumably same for satellite TV, the number of clients has no impact on the receiver timing? (Unless quantum physics shows us that there is some kind of bizarre photon tunnelling / entanglement going on)
 
This is partly why the UK gov invested in the National Timing centre that's set up by the org I work for.

I have a question, as I can grasp the challenge of streaming a single source file to multiple (millions) of end points, but presumably conventional EM technology is live, in the sense that I would receive the TV image at the same perceptible time as my neighbour did? And presumably same for satellite TV, the number of clients has no impact on the receiver timing? (Unless quantum physics shows us that there is some kind of bizarre photon tunnelling / entanglement going on)

Satellite TV and broadcast TV are very different than streaming. When one streams, there is a two-way communications channel. The vast majority of [I don't have a stat on this, so should not say it] Internet video is rate adaptive stream, which means you will change data rates and even skip some frames (P and B) to keep playback smooth. Just because one is streaming over TCP does not mean that you get the whole video file, it will depend on the streaming protocol and codec. When you stream, there are (at least) 3 components: the streaming protocol, the codec, and the container.

Adaptive streamers don't even use one file. They will switch between encodings depending on the current network conditions.
 
I have had AV on Hard Drives for so many years it is hard to figure out exactly when I started. All my CDs and DVDs reside in cake boxes. There is a lot to it. It can drive you nuts thinking about it. I love streaming both Audio and Video as well. The biggest factor I see in all of it is what the source file was to start with. There are so many different digital pressings of music that your copy of an album is most likely not the one that the streaming service is using. Sometimes they have a better mix and sometimes you do. Video tends to be worse yet in the fact that they usually do nothing with the source video but put it on the service. With the state of video streaming, nobody wants to remaster anything. Not to mention that a lot of TV was shot on tape. The fact that some are 480 and that the framerate is not matched is probably more of a problem than the bit rate or your internet hardware Even the newest video shot in 4k can have problems with Dolby or HDR, SDR. When I was a kid my friends thought it was a big deal when I had a 12" BW in my bedroom. I just try and watch and enjoy.
 
Click on the pictures that have electrons in them.....
 
Oh, trust me, digital productions have taken advantage of the platform.

Dynamic range. It's fine for cinema honestly, but it's usually completely un-fine in a domestic setting. Since media distribution via TV went digital the producers have started to expand the dynamic range of both to the extremes in the same way they would do in an actual cinema.

In one scene you can barely see the people it's so dark and the audio is so low and muffled with them whispering you can't hear anything and have to turn the volume up. Suddenly it switches to the next scene in broad dessert daylight or a snow scene scalding your eyes with full white screen and explosions start going off, jets over fly and a bulldozer rumbles past and by the time you've got to the volume control things are walking off shelves. Then it will cut back to muffles in a tunnel.

I even had to commit a cardinal sin and enable the dynamic audio compressor on the TV!
 
I have a question, as I can grasp the challenge of streaming a single source file to multiple (millions) of end points, but presumably conventional EM technology is live, in the sense that I would receive the TV image at the same perceptible time as my neighbour did? And presumably same for satellite TV, the number of clients has no impact on the receiver timing? (Unless quantum physics shows us that there is some kind of bizarre photon tunnelling / entanglement going on)
This is correct. Technically, each antenna and receiver absorb a miniscule amount of the energy*. But 6 million viewers watching a terrestrial broadcaster from a MW transmitter in a city is about as efficient as it gets.

* The first piece of electronics I built was a crystal set which had no batteries and was powered by the energy of the transmission!
 
This is correct. Technically, each antenna and receiver absorb a miniscule amount of the energy*. But 6 million viewers watching a terrestrial broadcaster from a MW transmitter in a city is about as efficient as it gets.

* The first piece of electronics I built was a crystal set which had no batteries and was powered by the energy of the transmission!


Here is my crystal receiver and mechanic record player.
The need for entertainment and news was strong even before people got electrical power in their homes.

IMG_6723.JPG




Bo Thunér , Linköping, Sweden
 
Very nice. My crystal set was quite scruffy, I was only eight at the time and my soldering skills were not well developed.
 
A further point. The Internet is not consequence free. Downloading and streaming 10Mbit/s audio is a criminal amount of waste of electricity and bandwidth. ...4K action camera...Editing and rendering that video will consume about 10 times the processor time and 10 times the electricity. Uploading it will take 10 times the bandwidth and 10 times the electricity...So blame global warming on audiophiles!
I stream Apple Music at regular lossless, hence claim immunity, and point the finger at Class A amplifier owners...
 
Ethernet is a demonstrably effective method of networking computers and other devices. It contains within its structure robust methods of error correction that have only become more effective as network speeds have increased.
True. but most digital audio applications do not use Ethernet. The standard use scenario is either Wi-Fi or mobile.
 
BlueTooth is not a network protocol.
BlueTooth is as much a link layer as ethernet or wifi. It's usually used in a point-to-point mode, but it can operate in scatternets.
 
BlueTooth is as much a link layer as ethernet or wifi. It's usually used in a point-to-point mode, but it can operate in scatternets.
Can you tell us any examples where BT is used as a network connecting multiple devices exchanging data with each other freely? I’ve never seen a system like that.

My limited knowledge is that you can create a piconet but that’s still a master/slave network where one master broadcast packets to a handful of slaves.
 
Can you tell us any examples where BT is used as a network connecting multiple devices exchanging data with each other freely? I’ve never seen a system like that.

My limited knowledge is that you can create a piconet but that’s still a master/slave network where one master broadcast packets to a handful of slaves.
yes, that's a network protocol. A scatternet is a collection of piconets.

IBM's SDLC, for example, only had one address because it was only terminal-to-host, so never needed a host address. But it's clearly a layer 2 link layer protocol and the progenitor of HDLC that lead to 802.2 used today.

Why is a master/slave protocol not a network protocol?
 
yes, that's a network protocol. A scatternet is a collection of piconets.

IBM's SDLC, for example, only had one address because it was only terminal-to-host, so never needed a host address. But it's clearly a layer 2 link layer protocol and the progenitor of HDLC that lead to 802.2 used today.

Why is a master/slave protocol not a network protocol?
That's a great question. Unfortunately, I think it comes down to semantics and definitions. The real world is always a bit of a fudge.

Having used quite a few protocols over the years, things tend to break down into point-to-point protocols which may use, but don't have to use, link-layer addresses. Point-to-point connections can be considered as two-node networks, but can be built using protocols that cannot perform networking. Networking at Level2, usually requires unique end-point addressing which can be scaled beyond 3 nodes, up to a large number. These days, useful Layer2 networks are those that can encapsulate IP packets.

There are while books in this subject. Radia Perlman's "Interconnections" is the best, IMO.
 
yes, that's a network protocol. A scatternet is a collection of piconets.

IBM's SDLC, for example, only had one address because it was only terminal-to-host, so never needed a host address. But it's clearly a layer 2 link layer protocol and the progenitor of HDLC that lead to 802.2 used today.
It seems you do not think of any implementations of a BT network though, which was my question. Otherwise we are talking semantics.
Why is a master/slave protocol not a network protocol?
As slaves only "talk" in response to commands from the master the process is more akin to broadcasting. It is not a network in the sense of Wi-Fi or Ethernet, which is the subject.
 
Back
Top Bottom