Hacker News Reader

Comment by Numberwang | Hacker News Reader

For those of you in the know here. Is this a realistic project/enterprise he is working on?

Seems both costly to deploy that many satellites and risky in the sense that you can't count on getting a lot of customers. Especially with the new super speedy 5G networks and so on.

View on HN

Hell of a lot of places without 5G, or any, mobile internet. Don't forget aviation and to a lesser extent maritime either. People love their onboard wifi.

Owning the launch platform could mean that he's able to significantly undercut current players, although funnily enough soon-to-be competitor Iridium recently launched a whole bunch of new satellites using SpaceX. Good position to be in when your competitors are dependent on you!

There are certainly challenges, though. Latency is a huge problem with satellite internet, for obvious reasons. I'm not sure what the intended bandwidth is but satellite will always "feel" slower than mobile data of equivalent speed. The fact that it's a mesh network with who knows how many "hops" presumably makes it even worse. So it won't ever compete with mobile, not even 3G probably.

1) Starlink will be in LEO around 800mi up unlike most communications satellites (like HughesNet) which are in geostationary orbit at ~22,200mi. Expect 25-35ms latency as compared to 600+.

2) due to having a lot more birds in the air at lower altitude with newer tech, they should be competitive with 5G networks speed wise as compared to existing satellite internet.

Edit: fixed distance typo for HughesNet

Huh. That's really close.

Your numbers are off though. Firstly it's ~40ms up to the bird if it's right above you, then 40 down to the base station if it is next door to your house. If it's on the other side of the world and has to travel around the whole mesh, add a hundred ms at best. Then from the base station to the destination - which could be anywhere in the world, and back, then back up to the bird, then back down to you.

It's not going to be 40ms, probably not even 400ms, and there is no way it will be competitive with mobile 5G.

It might not even be competitive with the geostationary birds for accessing nearby hosts. 22k miles is ~120ms, so 240ms round trip - because the base station is generally pretty nearby. Say ~20ms both ways for in-country ping and you're looking at 520ms -ish total, which is reasonably consistent with my experience using it from the Solomon Islands.

Huawei's saying 1ms to the tower for 5G!

I make it 4ms [1], not 40ms.

[1] http://www.wolframalpha.com/input/?i=800mi+%2F+speed+of+ligh...

* facepalm

Too late to edit. Misplaced the decimal point. Should have noticed that if 22k is 120, 800 can't possibly be 40. Duh.

Still, the actual latency is going to be very dependent on the distance to the base station. If it's nearby then shit, yeah, that will be within 20ms of 5G. Point conceded @jsjohnst.

It doesn’t make sense to bounce a signal over North America / Europe / etc all around on the mesh. The best course is to send it to the ground almost immediately (aka <2-3x max) for those areas as otherwise you’re wasting bandwidth. But over the Pacific Ocean it makes complete sense as adding a ground station in the middle of the ocean doesn’t make sense and even if you did, you still have latency to nearest server farm.

So yes, my point still stands for the majority use case.

Also, to be clear, I was making following claims:

1) bandwidth on par with initial 5G deployments (aka up to 1gb/sec offering)

2) Latency on par or better than existing LTE networks.

Yeah, my skepticism was definitely misplaced. I guess I didn't internalise just how close that is! Mea culpa.

This could open astounding possibilities if the prices are set within any kind of consumer-affordable level. I'm fairly interested in going on some long sail journeys, for example, but the total lack of connectivity (without a ruinously expensive and hence rare conventional satphone) gives me a lot of pause. What if something happens at work, or worse yet with my family, and i'm literally uncontactable, hundreds of miles offshore somewhere for a week? I know that for some people that's the point, but I wouldn't be able to stop worrying about it.

If your prognosis is correct then not only might it become feasible to be contactable on a boat for a reasonable outlay, you might even have friggin' broadband. For me and I suspect many others that's a very exciting idea.

It’ll never be 4ms in the best case though. Both fiber and radio signals are a lot less than 100% of speed of light.

(Know you likely know this, but not everyone does)

22,000ft -> 22,000 miles.

> Don't forget aviation and to a lesser extent maritime either.

And trains, busses (it might be orders of magnitude cheaper than paying for cell data to provide those local hotspots people use on busses to/from work everyday), cars (I could easily see teslas using it over cell for data in the future), and rural areas where laying fiber is expensive.

StarLink will be mostly in LEO, so latency will be competitive with terrestrial services.

Depends on the terrestrial service. When I used to have FIOS, many places on the web were sub-10ms. HFC ISPs however are typically in the range Starlink will be in though (25-35ms) which is already better than many cellular data networks by a factor of ~2.

Thinking at the scale of Musk's ambitions real quick, this makes a lot of sense.

At Tesla: Think about every Tesla car worldwide being able to phone home with self-driving learning data. Being able to remote operate them in the 0.001% of cases where human intervention is neccessary, no matter whether in the Australian outback or the Alps (without "no network").

At SpaceX: Being able to communicate with every Mars-bound spacecraft 24/7 by laser transmission, without having to rely on NASA's expensive Deep Space Network.

The antennas for this are the size of pizza boxes - which puts a question over the car connectivity. That's a lot of flat sky-facing surface to hide.

Putting it on the roof (or the hood) in a color that matches the car would work well enough.

They will be to start. But think about how big the first briefcase cell phones were. They’ll shrink over time.

Even in the developed world, coverage of high speed wireless broadband (e.g. 3G and up) isn't complete. In the less developed world the coverage is even worse[1].

I don't know the costs to set up and run the network, but I do feel like there is a large possible customer base.

[1] https://opensignal.com/reports/2016/08/global-state-of-the-m...

Well populated and tech heavy parts of the UK have spotty and unreliable 4g which is baffling to me. If Elon can deliver a low latency satellite network even just for peering it will be a big step potentially.

Even parts of London have spots where even 3G coverage is poor.

Can confirm that there's a radius of about 600m around my flat (SE3) which has spotty 3G coverage on Three with the inner ~100m being as good as zero (and my flat has had nailed-on zero 3G signal since 2010.)

Yeah fully agree. Going down the M4 from London to Bristol i will constantly drop between 2g 3g and 4g. Really disappointing situation for one of the richest areas of the world

I think there was a recent report from the FCC that said only ~60% of Americans have access to 25+ Mbps internet. So, at least in theory, SpaceX has an enormous market.

I was at Motorola when Iridium went up, and witness first hand the changing market dynamics that were happening even before their first satellite was deployed. Was thinking then that if they mad the satellites generic data switches instead of a flying phone switch, they would have had a much bigger market.

So for Musk's network, it really depends on if 80% of the potential customers (people with money to spend on the service) get saturated with ground-level 3G - 5G by the time the network is operational or not.

Something else to consider, if the system allows for land-based carriers to tap into it and re-broadcast it on 5G (with different pricing). Imagine if there is no coverage in your area, and anyone can purchase a box to put on their roof that pulls down the satellite Internet and makes it available to everyone's 3G - 5G phones in the area. You lease the box from someone, and you get a commission on every client connection.

It would be most interesting indeed if the satellite network allowed operating your own ISP for extremely low cost. Making 3-5G available to phones and peering to existing networks maybe? (I'm not that familiar with Mobile ISP Politics, unlike BGP Politics)

An "ISP in a box" is something I've toyed with, but on the basis of connectivity options ranging from a leased line, a microwave link and high gain wifi, but that stills ranges from expensive to "just" allowing you to extend out from the borders of nearby services. Having a cheap satellite option so that people can start providing that in rural locations where they are effectively an island of coverage would certainly be very interesting.

For a while the Trump administration considered building a government 5G network to prevent Chinese companies from doing it first. American telecoms are dragging their feet so much that they are starting to pose a threat to national security. They say building a nationwide network could take three years, so I don't see much threat to SpaceX now.

https://www.wsj.com/articles/u-s-memo-urges-big-push-on-5g-w...

it avoids all of the annoying issues that cartels spent a lot of time setting up on a state-by-state basis in the US (which results in limited choice).

Effectively you could connect to this anywhere in the world without the approval of local (e.g. Chinese citizens could get an unfiltered internet and the only "tell" is that they have a satellite dish). That makes the potential of this reasonably neat but it will depend on bottoming out the price and having enough capacity.

The equipment will no doubt be illegal to own in places like China. So it would get smuggled in or produced locally and sold on the black market. So far so good. But even if you install it surreptitiously I am pretty sure it will not be a LPI signal. So the government has a fleet of vans driving around sniffing for the spurious emissions and away you go to the gulag.

And governments not interfering with its operation.

Reply on news.ycombinator.com