3 ms·

Early versions of 5G, known as “non standalone”, are just glorified 4G. With the exception of the radio link, everything else is 4G. This means that the value p

by Foivos 6y ago

Early versions of 5G, known as “non standalone”, are just glorified 4G. With the exception of the radio link, everything else is 4G. This means that the value proposition is that more people per base station (or per surface area) will be able to stream video, download fast etc. When stand alone 5G becomes available, the latency of the network will become comparable to the latency of wired networks, especially if it is combined with mobile edge computing. Then, you will have access to new services that are impossible now, such as game steaming and VR. I agree though that 5G is way more important for the public sector and the industry than the average consumer.

ineedasername 6y ago
I can do game streaming now without an issue @ 1080p. Even latency isn't bad @ 25ms to 30ms ping time. Probably not what hard core twitch gamers want. 5g isn't performing much better yet, Ookla shows ping times about the same, but it is supposed to come down to about 10ms (1ms in theoretically optimal circumstanc. read: never for actual users) Streaming VR would be a valid use though, the resolutions there are much higher. The problem is that data caps will severely limit this. I can play an hour of streaming 1080p gaming before eating through almost 20% of my data cap. I usually tone it down to 720p. Data caps haven't actually improved that much over the years either: Now I pay $20/month with Mint for 8Gb. The supposedly "unlimited" plans from Verizon start at $70 and have an effective cap of 15Gb. Compare this to around 5 years ago where things were around 5Gb, if memory serves (and mint didn't exist). The backbone network capacity that mobile networks tap into isn't going to get any cheaper for providers, even if 5G signals bouncing around their own network have higher bandwidth, so I don't know how much 5G will actually reduce mobile providers' costs per unit of bandwidth if it significantly alters the peering ratio of backbone traffic interchange. Given this, I skeptical that new high-bandwidth applications will be able to get much traction, at least not before 3-5 years of additional product maturity & capital costs to providers for the upgraded hardware have been covered and they can gradually lower costs toward the marginal cost of production.
eru 6y ago
Why the public sector?
- Foivos 6y ago
  From the IoT side, it will be possible to have sensors everywhere in the city. So, you can have services like smart parking, monitoring of pollution at the city block level, etc. With the introduction of slicing, it will be possible to have specific SIM cards as “first class citizens”, with complete priority over everything in the network, while also being secure. These SIMs can become “first class citizens” dynamically. So for example devices that belong to firefighters can become so, during an emergency. This enables ordinary devices to meet the high SLAs needed in such cases. Thus, now firefighters can use the already existing, way more powerful (for example it can do live video) and cheap public network, instead of their current communications solutions.
  - eru 6y ago
    Oh, you meant IoT for infrastructure etc. Yes, that makes a lot of sense. Thanks for explaining! Your example with the firefighters is a good one, no matter whether they are public sector employees like in the US or privatised like in Denmark. That's why I was confused.

by Foivos 6y ago

ineedasername 6y ago

I can do game streaming now without an issue @ 1080p. Even latency isn't bad @ 25ms to 30ms ping time. Probably not what hard core twitch gamers want. 5g isn't performing much better yet, Ookla shows ping times about the same, but it is supposed to come down to about 10ms (1ms in theoretically optimal circumstanc. read: never for actual users) Streaming VR would be a valid use though, the resolutions there are much higher. The problem is that data caps will severely limit this. I can play an hour of streaming 1080p gaming before eating through almost 20% of my data cap. I usually tone it down to 720p. Data caps haven't actually improved that much over the years either: Now I pay $20/month with Mint for 8Gb. The supposedly "unlimited" plans from Verizon start at $70 and have an effective cap of 15Gb. Compare this to around 5 years ago where things were around 5Gb, if memory serves (and mint didn't exist). The backbone network capacity that mobile networks tap into isn't going to get any cheaper for providers, even if 5G signals bouncing around their own network have higher bandwidth, so I don't know how much 5G will actually reduce mobile providers' costs per unit of bandwidth if it significantly alters the peering ratio of backbone traffic interchange. Given this, I skeptical that new high-bandwidth applications will be able to get much traction, at least not before 3-5 years of additional product maturity & capital costs to providers for the upgraded hardware have been covered and they can gradually lower costs toward the marginal cost of production.

eru 6y ago

Why the public sector?

Foivos 6y ago
From the IoT side, it will be possible to have sensors everywhere in the city. So, you can have services like smart parking, monitoring of pollution at the city block level, etc. With the introduction of slicing, it will be possible to have specific SIM cards as “first class citizens”, with complete priority over everything in the network, while also being secure. These SIMs can become “first class citizens” dynamically. So for example devices that belong to firefighters can become so, during an emergency. This enables ordinary devices to meet the high SLAs needed in such cases. Thus, now firefighters can use the already existing, way more powerful (for example it can do live video) and cheap public network, instead of their current communications solutions.
- eru 6y ago
  Oh, you meant IoT for infrastructure etc. Yes, that makes a lot of sense. Thanks for explaining! Your example with the firefighters is a good one, no matter whether they are public sector employees like in the US or privatised like in Denmark. That's why I was confused.