Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

You have it exactly backwards. It is far less complex and expensive and resource intensive to build Starlink than to run a new copper or fiber line with associated telecom equipment on both sides to every rural residence in the US, let alone worldwide. Yes, despite the large cost of launching satellites. And it's especially good that we don't have to force everyone to subsidize inefficient monopoly utilities with our tax dollars to get everyone connected. Plus the benefit of mobility is enormous and shouldn't be ignored.

As solar and batteries become cheaper, eventually we can transition to most rural residences being entirely off the grid and self sufficient. This will also be cheaper and less resource intensive than maintaining the electric grid in those rural areas, let alone building it in the first place, and we can all stop paying hidden subsidies for those users.

 help



For just ONE Starlink v3 it weighs around 2000kg and right now falcon heavy has an estimated cost per kg of $1,500 dollars so $3,000,000 per unit to orbit and let's say around $1,000,000 for the v3 itself. According to SpaceX each satellite can support 60 Tbsp downlink (I highly doubt this just purely based on the compute required) but assuming that is true that's 60,000 Gbps or 60,000 customers at 1 Gbps. The biggest issue here is that each satellite only lasts for approx. 7 years in LEO before its deorbited vs a fiber link last basically forever

this.

Except it's no longer only in rural areas, grid connected utilities are now costing more than being off grid in the cities too. Starlink residential 100 Mbps is cheaper ($69/mo AUD) (ignoring hardware and setup costs) than 50 Mbps fixed line internet ($80/mo AUD). Depending on location, home solar + batteries will usually work out cheaper than being on the grid within the battery warranty period too.


The question that comes up then is: how much traffic can Starlink handle until it gets saturated? I'm not sure it can handle even a significant percentage of the users that currently use wired connectivity. And if they see that demand for their services starts overwhelming supply, they will definitely raise the prices...

Starlink has, essentially, fixed capacity per area. Thus, sparsely placed rural users are "cheap" to Starlink.

Densely placed city users would strain the system, but cities also favor wired connectivity.

That makes the two complimentary. Wired makes the most sense in urban formations, satcom makes the most sense in Bum Fuck Nowhere.


_Lots_ of traffic. It's going to end up being the global Internet backbone.

Internet traffic today is estimated to be a few tens of exabytes per day. Even if you assume 100000 Starlink satellites (we're far from that), each satellite would have to handle hundreds of terabytes per day. That's tens of gigabits per second per satellite, assuming traffic is split evenly among them (will never happen in real situations).

Starlink V3 can pump out some seriously impressive speeds and handle thousands of clients. Starlink is both a great leap forward in rocketry and radio technology. I do still think funny how we are going back to the pre war technology tree for a re-visit

That's not even sufficient to handle the needs of a single large city. The limitation is that even with the much larger constellation they hope to deploy there won't be enough satellites visible at once from any given large metro area.

Citation definitely needed.

Grid prices are going to start coming down in some of the most expensive parts of Australia due to SAPS, home generation and storage, and microgrids.

I wouldn’t rule out the grid just yet.


If you find Starlink cheap they just haven't gotten around to the bait and switch in your locality. It'll come.

This is because Australia has high internet prices. Partly because it's huge, but partly because the NBN got stuffed-up by the Liberals because they didn't believe the country should be investing in what they called at the time "a glorified video delivery service", so put the tech back a decade, and the country ended up paying more for a worse rollout.

Your comparison point is also a bit weird to me. If I want a decent speed, my choices are fixed wireless NBN at ~250Mbit (400 in theory, 250 in practice), or Starlink at ~200Mbit, and they cost around the same.

If I were just a few km closer to the city I could get 500Mbit fibre for ~$90 a month.

So while it's definitely not out of the range of other plans, I wouldn't say it's definitively cheaper. And I wonder if the recent price drops are down to people not wanting to have much to do with Elon Musk any more. I know it's worth a few bucks a month to me not to be a customer of his.


Man, I pay $50USD/month for 1Gbps up down in Wisconsin.

Wicked, I wonder what the most juiced option would cost amongst your upgrade options.

“When in Wisconsin.”


Where are you? In the suburbs of Atlanta I paid $80 for AT&T Fiber 1Gbps u/d.

If they're paying Australian Dollars.. probably not Atlanta

Maybe today, but internet over radio cannot defeat physics. There is only so much bandwidth, so much space in the RF spectrum for data. But landline internet is effectively limitless. You can always lay a second, or twentieth, fiber run. A 10cm bundle of fibers can carry more bandwidth than the entire starlink network many times over, with much lower running costs.

The most effective in rural areas is generally a combination. Fiber to a central location and wifi radio out to customers. I am monitoring a property on the west coast connected via such a setup. The last relay is actually solar powered atop an island.


You can beat the physics here though. There are several techniques in signal processing and beam forming that can be used to make insane high-speed transmissions, the days of geostationary broadcast for all radio traffic are over when it comes to advanced radio communication.

Radio technology is truly the closes thing to black magic. I wish there were more places to learn and read outside of an EE degree


> And it's especially good that we don't have to force everyone to subsidize inefficient monopoly utilities with our tax dollars to get everyone connected.

Again*.

In some ways we did subsidize the initial public phone network that put ma bell in the position to take over as an Internet backbone as "the Internet" became a thing. In some ways we're subsidizing starlink like direct grants of taxpayer sourced funding for rural broadband expansion and contracts that subsidize the spacex launches.

I do wonder sometimes if it's actually cheaper to connect a rural farm to the Internet by blasting a satellite into space vs setting up some kind of terrestrial radio based network like lora or microwave. That's not my knowledge area so maybe there are real, unsolvable issues that prevent terrestrial radio as a solution, but I have to assume blasting rockets into orbit is expensive both short term and long term, especially considering space trash.


Isn't this even more of a monopoly utility?

In theory you could have multiple providers but it just doesn't happen much due to market dynamics and incentives.

In this case if I understood it well there's a limit to the amount of satellites we can send into space at those heights and that space is essentially privatized for free uncontested and ESA and China's CNSA already complained about near collision events.

So not only do you get the same market dynamics but practical limitations too and an externalization of costs.


ESA and CNSA were always going to complain, real problems or not.

Why would that be? Having to pull tricks to dodge satellites seems like a valid one.

The externalities need to be weighed into the cost too no?

How much global warming and environmental destruction is caused by launching rockets? A grid is built once and can be maintained for a very long time at a much smaller operating cost. Space stuff is expensive...


As far as electric goes, that's a nice thought but the reality is prices will not go down in such a scenario. I'd rather my bill go to subsidizing rural areas than to pure profit. Nevermind there are benefits helpful to rural areas that grid service can provide versus solar+battery.

Starlink recently hit 10k satellites. I'd hazard a guess that's not anywhere near enough getting everyone in the US, let alone worldwide, online.

While having more satellites sure does help serve more people, there’s a second issue which arises when trying to serve high density areas, where you run into bandwidth limitations. The solution there is not more satellites but either bigger satellites (which can make smaller beams) or more FCC allowance on the spectrum.

Not everyone. But it's enough for rural areas, which are the most expensive and least practical to serve with wires.

I do wonder about what happens when Starlink grows its customer base a lot bigger like many of you are predicting here, since Elon Fucking Musk, the king of over-promising and under-delivering, is at the helm. We might end up yearning for the days of the (slightly more) regulated utilities instead.

Ideally the other couple low orbit satellite based ISPs will get to a decent place by the time of irrecoverable enshittification.

Oh good, so we’ll be in the same place as we are today with a duopoly or similar, where every option sucks. Sounds lovely.

The HN groupthink is to hate on anything Elon adjacent, satellite internet included.

It's not groupthink to believe that the guy sucks and is a threat to humanity. He constantly fights against the type of programs that could have possibly given us satellite internet, the same way we all get to enjoy GPS.

> It's not groupthink to believe that the guy sucks and is a threat to humanity.

Wow, that’s a wild misstatement; that is exactly groupthink nonsense.

You (people) loved him before he went in for Trump.


> You (people) loved him before he went in for Trump.

The inflection point for the public was Musk calling the cave diver, who helped orchestrate the rescue of a dozen trapped kids, a "pedo guy" and then doubling down on it, again, twice in front of his audience of millions.

The inflection point for anyone in tech with two eyes and a brain was Musk insisting his companies produce products that do more than they are, still to this day, capable of.

First was around 2018, the latter was ~2016, although anyone who was familiar with machine learning knew models were not as capable as Musk was insisting they were, and that the hyperloop was a scam.


Before he went in for Trump he created an obviously fake, insanely expensive system that could never work in practice (Hyperloop) just to slow down California rail projects

California rail projects fail because it’s California

Before he went in for Trump he was running a factory with an alarmingly high injury rate, where employees were regularly called the N-word, and union busting. People who liked him then weren't paying attention at all.

For what it's worth, I hated him well before he had anything to do with Trump. Most concretely when he called the cave diver a pedo for not wanting to use his stupid submarine, but I remember thinking that the Hyperloop thing he was proposing was pretty stupid too.

Oh, and when he lied about taking Tesla private so he could quickly boost the price of the stock. That sucks too. He's always sucked.


hopefully that include his business partners , airlines in this case.

And yet this discussion seems to be driven by a load of TSLA maximalists.

Elon is a busted flush. He promises the world, delivers somewhat less, somewhat late, if at all. And then layers it with deeply unpleasant politics.

Not groupthink, a sane reaction. Belated, but sane.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: