Hacker News new | past | comments | ask | show | jobs | submit login
SpaceX Starlink average speedtest is 42.3Mbps down / 9.7 Mbps up (testmy.net)
184 points by punnerud on Aug 27, 2020 | hide | past | favorite | 313 comments



So I guess companies will now fill the entire global sky with thousands of sats just to make up for the fact that the US government ignores rural houses when it comes to digital infrastructure?

I'm sorry but how is 42Mbps worth spamming the entire nightsky with to the point that it prevent astronomers from obtaining useful data? Not even 10Mbps upload?

My inlaws live in a village in France that has 70 people total. It's 25min drive to a bakery, 45min drive to anything resembling a supermarket. I'd say it's rural.

They now have 16Mbps *DSL and are scheduled to get Fiber in 2021. Next village already has it.

But because the American government refuses to invest in infrastructure the rest of the planet just has to accept some American company is going to put 14,000 sats in low orbit?

Besides the progress this resembles I still think it's pretty strange the US government can greenlight this plan even though it has global implications.


But are all villages in France like that? I doubt it.

Even in Norway there are thousands of people without access to sufficiently fast broadband, and some 80k DSL customers who apparently will never get fiber at all. The proposed solution for the latter is to use 4G/5G, which comes with its own issues of course.

I'd say even in well developed countries, Starlink is a nice option to have. As I understand it, the 42Mbps they see now is only going to improve as well.


Eventually everyone is going to get connected, yes.

I mean someone in Norway also had to draw a powerline to these people, to use electricity. Why not draw a fiber link, too? It's just part of infrastructure.

I'm not saying that Starlink isn't useful. I'm just really wondering what the implications for astronomy are going to be, whether we will always see these things at dusk.

Now it's Starlink, but what if 3 or 4 companies in total want to have a network? What if on top of that China and India want their own, just like GPS?

It just seems we're running head first into this without anyone thinking of the implications or governance of it.


Hopefully using satellites like this is a temporary solution for the next decade or two until we have a better solution.

Their limited lifespan is what makes me much less pessimistic than I would be. They have relatively high maintenance costs which means that if utilization is low, it won't make sense to keep replacing them.


I think it'll be nicely self regulated. It's expensive to run, those that are not profitable will not be able to maintain their network and being in low orbit, those satellites will fall into the sky and burn up.


> I'm just really wondering what the implications for astronomy are going to be

More space based telescopes?


The power lines were, to a large extent, built with U.S. money after the war. Many rural areas have seen little infrastructure investments since then.

Anyway, if governments, businesses and individuals care enough then it's easy enough to develop that infrastructure now. The satellites will come down in a few years.

Who volunteers to tell Musk?


You don’t need to draw a power line if you can just tap to a local mountain river


Yes I think all French villages have at least super fast ADSL.

The US is much larger than France of course, but whatever the reason, it's incredible and really upsetting that they can pollute the earth like this, and for the rest of eternity, just to solve a temporary domestic problem.


> The US is much larger than France of course, but whatever the reason, it's incredible and really upsetting that they can pollute the earth like this, and for the rest of eternity, just to solve a temporary domestic problem.

Temporary domestic problem? It's a problem worldwide, the fact that France can get fast ADSL almost everywhere doesn't means it's true worldwide.

Wireless has the advantage of not needing a huge infrastructure locally, which is a huge help where you don't have any. I remember a presentation from Charles Sirois, one of the guy that brought cellular networks to Quebec. He told us how he helped build some cellular networks in some place where they didn't even have phones. Sure I have no doubt it wasn't easy to build theses towers, but it was certainly much easier than bringing cable to every single building. Every business had cellphones afterward, it was a huge boom, it helped develop so many industry.

Now imagine the same but with freaking internet. This is going to be a huge boom for so many third world countries.

But no you are right, France can pay to lay fiber everywhere, theses third world country should just do the same, they are just not trying enough.


Ah yes, third world countries. That's what Musk is after, that great philanthropist.


The satellites deorbit automatically. You are misinformed.


ADSL and super-fast don't go together as adjectives. ADSL is by far now a legacy technology.


You are ignoring millions of people in rural Asia and Africa who don't have internet connection, or a spotty 3g/4g connection for internet. Satellite internet is legitimate use case for these people. Starlink might be too costly for most of them, but some other satellite internet company might cater to them.


Something like 45% of the world's people don't have internet access right now, so that number would be billions, rather than millions.

Also, Starlink doesn't have to be too costly for them. The constellation's operating costs could potentially be fully paid by North American and European users. The satellites would still incidentally fly over the southern hemisphere, and any revenue at all from Africa/South America/Asia would mostly just add to Starlinks bottom line profit. The marginal cost of additional users should be minimal once general revenue targets are met.

I'd like to hear whether the throughput will increase with additional satellites, or whether it requires re-designs. I wouldn't expect the current satellites to be saturated with users by the currently "limited" number of beta test users.

Cost of end-user terminals also will likely have to come down significantly before the worlds poorest would have any access. Although sharing the cost of that among a large group would help, it would also help a lot if the speeds were actually closer to gigabit for sharing the connection.


I'm looking forward to seeing StarLink go up against the Great Firewall. Although Tesla is booming in China right now, so the CCP would probably just retaliate for making service available to Chinese citizens by going after Musk's other company.


I'm French too. France is tiny. It is not about the number of bakery you have but how far your are from the closest medium size city.

Also, the satellite are not geostationary above the US. They rotate above the Earth and will bring the internet everywhere... Bringing internet to everyone in the World is nothing to sneeze at.

(I agree with the last bit of the rant. It sucks that private companies can launch whatever they want in space. That is not specific to the US however.)


On the flip side, why should the majority of us in denser areas in the US subsidize people who want to live in exurbs/rural areas? If you want to live in a place like that, be prepared to pay the cost for the infrastructure.

We already pay enough subsidies for roads, water, power, etc in rural/exurban areas.


I'm healthy but I pay for those who are not.

I grew up fine but I pay for those with special needs.

I have a job during Covid but I pay for those who don't.

I guess in Europe we call this 'society'.


Europeans really barely understand the US. Yes, of course, we understand how investing in communities makes a society work. What Europeans don’t understand is just how fucking huge the US is and how low-density huge swaths of the country are. Farmers might live 500 miles from a metro of 100,000 people. It’s a huge challenge to give everyone fiber without the 1% of hyper rural America doubling costs for everyone else.


This argument shows up in all these discussions and is way way overused. The EU only has ~2-3x the population density of the USA. It's not even a single order of magnitude and people have already given examples of how very remote places in Europe have very reasonable infrastructure. If you already have telephone and/or electricity poles to somewhere running fiber is definitely an easy problem to tackle for what is supposed to be the world's most advanced economy. Capture by telco special interests is a much more likely explanation.


3x population density is a lot, no? That means for every person that can access, say, a fiber line in the US, 3 people could access it in Europe. That would seriously change the economics.

Further, the US's population density is very varied. About a third of the country (the non-coastal West) is very, very sparsely populated, punctuated by only a few semi-large cities like Denver and Phoenix. If you live on the coasts or the Great Lakes, density is a lot like Europe and the services are similar. In between, it's really, really sparsely populated.

https://www.census.gov/dmd/www/pdf/512popdn.pdf


Yes, yes, all of Europe fits into Texas, and even Texas fits into Texas, and that's just one state. We get how mind-bogglingly huge the US is. It's been said over and over again by every single american whenever the subject of your subpar infrastructure comes up.

Europe is just as big. Not as dense, maybe. But your telcos were given hundreds of billions to do something (that should have been the job of the government), and simply pocketed them, with your government doing nothing about it.

So, if that money was properly invested, instead of polluting low orbit, that'd be nice.


I have zero intrest in having the government control my access to information, no thank you

That the biggest thing Europeans do not understand about the US, many of us hate our government, distrust our government, and the only thing we desire from government is for them to leave us alone..

The US is a individualist nation, not a collectivist nation, though many are trying very hard to change that.


You have it backwards. Almost none of the telecom infrastructure in Europe is state owned. As the comment you are replying to explained it's in the US where huge government incentives were paid out to telecom companies for broadband deployment who then proceeded to pocket that money and sit on their hands.


>I have zero intrest in having the government control my access to information, no thank you

They kind of already do. NSA splitters are common knowledge by now, and PRISM was yet another proof that they have access to most of your data. Make no mistake, if the US government wants to find you or what you've been doing, it won't be hard for them.

You're either dealing with Mossad or Not Mossad: https://www.schneier.com/blog/archives/2015/08/mickens_on_se...


You're dismissing the very valid geographic argument completely by being snobby.


If, in Europe, if people decided to pave over every natural area within 100 miles of where you lived because they want to live far away while, you paid to make it cheaper for them to do this, would you still be ok with that?

If this caused all of the water (streams, lakes etc) in the area to be polluted and unsafe to swim in and drink without heavy treatment, would you still be ok with this? I can't think of a single body of water where I live that doesn't have excessive algae growth from fertilizer runoff from the massive number of suburban/exurban lawns. I was exploring in my favorite nature preserve that has a nice size stream (well, one of the few over 10ft that hasn't been culvertized) and one of the park rangers told me to be mindful that there was a septic system a mile upstream leaking E. coli in to the stream. In a city that has a "township" designation literally so they can get tax breaks and grants for roads meant for farmers. There isn't a single farm in this city, nor in any of the other "townships" in the area.

I can go on and on, do you still think subsidizing this life for people is reasonable? It certainly isn't done to this level over in Europe. Why not?


How many of those things are choices? Can you not think of other alternatives to this "problem"?

This seems wholly different to the point of making the analogy laughable.


I love that we still live in a world where different countries can make different choices about how they want to structure their societies.

Kind of telling that North America immigrants were largely people escaping European utopian society. I guess different people have different values?


Exactly, Why did all those European immigrants leave what many now call a more civilized and progressive place.

It makes no sense. I mean it’s not as if Ireland has changed at all since the potato famine when my ancestors came over right?

As you can tell, I do not think highly of the argument you have made. It seems to ignore something important.


I vot for calling it civilization.


Boldly asserting that we already pay "enough" subsidizing other utilities is lazy. Why are "road, water, power, etc" subsidies fine but internet subsidies not? They all fall within the same category: essential services that would be low-return/high-risk if provided by private entities.

In fact, your argument basically boils down to "Why should X pay for Y?" So why should astronomers pay for Starlink?


Oh, I'm not fine with many of the other subsidies as I think they are incredibly excessive at this point.

I should add that I live in the midwest (Cincinnati) so I directly see a lot of rural/exurban areas people are talking about here. There are just miles and miles and miles of suburbs out in the middle of nowhere because people here are obsessed with the idea of them. The amount of money that gets dumped in to those areas while people complain up a storm when the urban areas try to build out their own infrastructure is insane.

I can't tell you how many coworkers I have had that drive 45mins+ just so they can live in some suburb far away from everything. If they want to do that, fine, but don't beg for internet, roads, etc to be paid for at the cost of sucking resources away from people living in urban areas.


Because we need rural areas to supply us with food and other materials.


A massive number of the people that live in rural areas are not providing food and key materials. In fact, the amount of farmland and farmers has been declining for years due to increases in agricultural efficiency (green revolution).

I've got no problem (to a point) subsidizing infrastructure for farmers and other key jobs, but acting like all of these people who insist on living in an incredibly unsustainable way deserve to be subsidized is ridiculous.


Does that matter? Not really. Infrastructure (Internet counts as that) should be available everywhere people live. Who knows, one day you might be in such a situation.


Sure, just expect that if you're going to live that way by choice, pay the costs for it!


What costs? Of course I'll pay for internet/power/water.


Why should the rural people in podunk nowhere subsidize our food?

Ultimately it is in the best interest of cities to keep on good terms with their breadbasket.


Again, people are trying to frame this as an "all people in exurbs and rural areas are farmers" and this straight up is not true in the slightest. The issue here is that massive numbers of people move to the far out of the way places which requires an inordinate amount of resources to support, and there is this expectation that everyone else should pay up to let them live that lifestyle.

Here's a example of these sort of areas on the midwest: https://www.google.com/maps/@38.9232568,-84.4006112,11979m/d...

Almost none of the land in that area is farmland, yet I've had coworkers who got USDA loans for houses in the middle of this area while working white collar jobs.

For the last couple of few years a local developer keeps lobbying to have a multibillion dollar highway built through this area so they can develop more suburbs. http://www.cincyeasternbypass.com/wp-content/uploads/2019/02...

Why does everyone else have to keep paying for this garbage? When you bring up things like toll roads all of these people scream their heads off, because they know it means they wont be getting their subsidies (ironic considering how many people in these areas are highly conservative, anti-subsidy).

This isn't about subsidizing farmers, this is about subsidizing people's choice to live an unsustainable life style.


The economics of farming barely work out at market value. How are they subsidizing others and how can they afford to given those margins?


Because that's what a country is.


Some things are just common goods.


The latest generation of Starlink satellites is invisible to the naked eye once fully raised to full height in a couple of months. The impact on astronomers is real, but very small.

OTOH, the benefit is the ability to participate in the modern economy.

Cost/benefit to the world seems pretty clear to me.


Seems to weigh on the side of not interfering with astronomy, given that we have plenty of alternative options for accessing the internet.


Your comment kind of make me sad... this is such a big issue of perspective. It's something that I often see coming from the US, but any first world country often has that issue, we forget that there's more than us in the world.

> In 2019, Internet access became available to 51 percent of the world's population. When households are taken into account, the index increases, reaching 57.8 percent.

42% of the world don't have internet. We aren't talking about not having competitive internet, literally, no internet at all.

At the bare minimum, you forgot about 42% of the world... that kind of sad.


The 42% of the world that doesn't currently have internet won't be able to afford Starlink anyway, and unless Elon has discovered a way to make satellites 99% cheaper, Starlink will never be the solution for them.


Why? That's not obvious at all. The internet is useful, even essential for many modern economic activities. Modern astronomy is useful for what, exactly? Learning things about objects that are physically impossible to reach?

Ask almost anyone what they care about most, the internet or a book on astronomy, and they'll think you're joking.


But it doesn't significantly interfere with astronomy.


I'm in Italy, 15km from the province capital, 7km from a major turistic village, in a village of 3000 souls. I'm waiting for 2Mbps ADSL since 2007.


Where do you live? (I'm Italian as well). Mobile connections aside, there are some vendors who would reach you through WiFi if you're in line of sight with any of their repeaters. (eolo.it and alternatyva.it came to mind, but there are others). In 2009 I brought a whoppy 4DL/4UL connection (with a really low latency excellent for online gaming) to a house in a rural area where the only available option would have been a 57k analog modem. Now there are much faster solutions.

Also, don't trust the national provider Telecom Italia. Before finding that solution, we pestered them with requests to cover that area, and their canned response has always been "we're working on it", while actually they were waiting for more people to move in that area so that it would have been more convenient for them.


Province of Udine. Yes, Eolo has been my provider for the last 10 years, I was one of the earliest adopters in the province. The service has always been good, but worsening in the last 2 years due to too many contracts been sold without improving the infrastructure. IE, during the lockdown I had to rely on the 4G network (even checking email was not possible through Eolo) and somedays I still have troubles.

If I may, which provider offers you 4DL/4UL and how much does it costs?


The 4UL/4DL was an offer by Alternatyva.it back then. I don't recall the costs, but they were comparable, although a bit higher, to a wired ADSL. They kept it until Telecom Italia finally covered that area with faster ADSL, which if memory serves had been 5 or 6 years later.

I seem to remember we had to settle to a 4/4 setup although the contract allowed a bit more DL speed, due to the distance from the only repeater we had in line of sight. Performance was good during day and very good from evening, with speed tests around or very near the theoretical maximum.


In all your anti-american rambling, you are ignoring that SpaceX isn't the only company doing this.

> Other operators include the London-based OneWeb, which has launched 74 of what it hopes will be a gigantic fleet of 48,000 satellites


It's not anti-American rambling. As I said in my other comment I'm aware of other companies doing the same. Although an American company will be first, with clearance from the American government.

I'm more drawn to the fact that the American government can greenlight this, even though it has global implications.

These things will also impact astronomy in Argentina. Where can they lodge a complaint? They will impact the night sky in Europe, yet we have no say in American government or decision making.

I can understand this has never been a problem because no one has ever been planning to launch 14,000 sats! But within 10 - 15 year maybe 3 or 4 companies might do it. Then we have 50,000 sats in LEO.

Will this be worth it? And if so, who weighs these pro's and cons? Just one government?

Feels like we're running headfirst into a new world where no one has thought of the consequences yet.


I agree that it's irresponsible, but most of your comment is "American infrastructure is horrible, European infrastructure is great. I can't believe American would force this on the rest of the world", even though it's a European company that's planning to put more of these satellites in the sky than every other company combined.


> I'm more drawn to the fact that the American government can greenlight this, even though it has global implications.

They only need to greenlight the launch, and radio-transmissions over the US.

As far as I know, there is no central authority anywhere to stop you from putting/keeping a satellite in a certain orbit. (aside from someone else moving it out of that orbit by force)

Space border laws aren't super-well developed yet.


Oneweb is bankrupt. It may be half-rescued by the British govt as an attempt to build its own GPS system since they got kicked out of Galileo due to Brexit, but it is unlikely to be a player anymore.


Bharti Global didn't plow 1/2 billion into OneWeb so it could just be a GPS system for the UK.


To be fair, the network is currently only at ~5% capacity and average speeds can be expected to increase as the constellation is fleshed out and direct laser linkages are brought online.


My parents live in the mountains of Panama. They have ADSL internet, but it's so bad at times they have to resort to tethering over a weak 3G connection and paying per GB for very slow data. They are waiting anxiously for Starlink to be available outside of the US.

Starlink is not just about the US, it's a global service that will connect people anywhere the local infrastructure is not yet up to speed. Also comparing rural France to rural US is oranges and apples. The US is almost 20x larger than France.


About 70% of the earth is covered by water. If you had a sailboat with an auxiliary motor you could go hang out in the Sargasso Sea and have perfect weather most of the time, eat healthy with fresh caught fish and use silar panels to have enough electricity for laptop and starlink modem.


We are working with a county in West Virginia that is 2 hours by car from Washington DC. 2 hours!

The school officials have decided to do remote schooling.

Fully 50% of the students have no internet at all.

Another 30% have internet that is incapable of remote.

You might say that they're too poor. And that is part of it. But there are plenty of people with lots of money in DC and the surrounding area that would be interested in moving here or having a vacation cabin, but cannot buy internet access for any amount of money. It just doesn't exist.

The communication laws desperately need an upgrade from the 1930s laws on the books that regulate everything like it is a telephone invented in the 1800s.


West Virginia is a bit cherry-picked, it is by far the most systemically poor state in the mid-Atlantic, and it’s got immense infrastructure challenges due to the mountains. I sympathize with those people, but the vast majority of the US is doing better.


Living in Vienna, Austria, I can give at least 4 locations in the inner city which don't receive even this speed of connection even though the city promised to roll out fiber 15ish years ago. Then our telecom became privatized and now we are stuck with 40/10 (on good days) DSL connections because no matter how often I call and ask for them to do something about it, they push a hybrid DSL/LTE Modem on me which isn't significantly faster.


France also has a high population density. Even "rural" areas are still close to major population centers. That is not true in rural Siberia or the Rocky Mountains.


Is it possible to retrofit and run fiber across sewer systems, telephone poles, and electrical towers so we can connect rural areas without much effort digging trenches to get there by using existing infrastructure?


Sad this is the top comment.


Can you elaborate why you find this sad? Do you disagree with the opinion in the comment?


>spamming the entire nightsky with to the point that it prevent astronomers from obtaining useful data

Is simply not true. It's outright false and the commenter that posted it clearly just has an axe to grind against Elon Musk along with a million other people on this website


I think we're going to see some pretty awesome, in the true sense of the word, activism on this one.

Hacked sats. Pirate-style space cleanup sats. It's gonna make an excellent Wired article.


If your inlaws weren't scheduled to get an upgrade to fibre and were instead cut off from the world due to costs of servicing remote areas would this comment be different in any way?

It's a _big_ world.


Imagine being so full of yourself that you forget that literally half of the planet is not connected to the internet in any way whatsoever. The world isn't the States.


Ah but then another company can send telescope satellites to a higher orbit and charge out the ass for NSaaS (night sky as a service)


Maybe it’s not at full capacity? I mean they are just testing it after all. Lots of fine tuning to do.


The US is terrible when it comes to infrastructure. Sweden is similar to France here, expanding fiber internet to all households that want it even if they are rural. Someone mentioned Norway as a reply but I think it's an unfair comparison as Norway will have more trouble building this infrastructure since there is mountains everywhere.


> to the point that it prevent astronomers from obtaining useful data?

Define useful


I am sorry but how it is the role of government to provide everyone with internet?

The American government "refuses" because there is no Constitutional Authority for which it could use to provide such a service, and that is a good thing

People seem to think government should just provide everything to everyone, the government should be very limited in size and scope, and providing internet is WELL outside what a limited government should be doing.

under the federalist system that is the United States you could make an augment that STATE governments could provide such a service but it should not be a Federal program at all


Article I, Section 8, Clause 7 of the United States Constitution, known as the Postal Clause or the Postal Power, empowers Congress "To establish Post Offices and Post Roads."

In an alternate reality, it’s easy to see how this could be interpreted broadly as meaning the government has an impetus to provide for communications infrastructure.


This clause like many others have been abused to the point there the literal words do not mean anything, this is the so call "living document" interpretation of the constitution, one that should be rejected as it sets in place a "standard" that the words mean what ever we choose them to mean

There was an Amendment process put in place expressly to prevent the type of "reinterpretation" you are advocating for, if we the people desire the US Federal Government to provide us with Internet then we the people need to pass an constitutional amendment declaring that power


"... The earth belongs always to the living generation. They may manage it then, and what proceeds from it, as they please, during their usufruct. They are masters too of their own persons, and consequently may govern them as they please. But persons and property make the sum of the objects of government. The constitution and the laws of their predecessors extinguished then in their natural course with those who gave them being. ..." - Thomas Jefferson

https://jeffersonpapers.princeton.edu/selected-documents/tho...

Jefferson was in favor of living people controlling government, rather than trying to divine what dead people would have wanted. The living document approach to understanding the constitution is a way to keep the text of the constitution static while still allowing for each generation to interpret it as they see fit. It's a pretty good compromise.

Also: words always mean whatever we choose them to mean. As my favorite law professor would always remind us, words don't leap off the page and become reality by magic. People read the words, and people take actions. So what you're disparaging as "reinterpreting" I'd argue is just "interpreting," and is a necessary component of all written language.


That is a gross misrepresentation of Jefferson's view

The modern "living document" is nothing like what Jefferson called for, which was all laws, even the constitution be sunsetted after 19 years allowing for the current generation to pass new laws and a new constitution

This is a far cry from "these words mean something else now because we 9 people say they do" which is what the living document of today comes down to.

Jefferson wanted robust debate and the consent of the governed to deiced the rules of their society not 9 people redefining words to fit their personal world view, in fact Jefferson wrote extensively about the problem the power the Supreme court held would have on society


The US spent a bunch of money on rural electrification back in the day. It’s hard to see how this is much different.


There are a few ways to look at this, while your statement is true it in no way refutes my position that it is a proper role of government unless one also believed the Rual Electrification was also proper use of government

the after effects of these government monopolies have been felt by scandal after scandal

Further before these enforced government monopoly on power farms were using a distributed network of Wind Generators, which Ironically are now being put back into place on some of these very same farms but instead of being able to give the farms good power they are being put in place by these came corrupt government run power companies

Had the government not interfered in the market place who knows what out have happened, may renewable energy and distributed power networks would have emerged decades sooner instead of just not emerging as a replacement for Coal power


Speed is variable, capacity can be improved. The bigger question re Starlink is latency, which this test doesn't show.


I saw these Starlink speed test results 2 weeks ago: https://tweakers.net/i/xDBb34kfMAju3YPPMsC3iOc2gXc=/656x/fil...

Seems like pretty solid latency :)


30-100ms is better than I feared, and should be generally reasonable for most uses.


Consistent 30ms would be pretty excellent, and make it useful for many things. Consistent 50ms, similarly. It starts to get a bit more of an issue at 80ms or 100ms, but my worry is more that jitter may be huge, and 30-100ms is a huge jitter window that could limit usefuless not just for games, but also many other things such as voice calls.


A 100ms ping is perfectly playable in everything except twitch-based shooters. In voice calls you will notice it, but it won't get in the way like say a 1-2s delay would (like you get if you phone from one end of the world to the other). It's really a very good result bearing in mind that this can work absolutely anywhere.


Fighting games, too. Most gamers know to avoid wifi, much less a wireless ISP.


I don't think that's still true with 802.11ac. My Wi-Fi adds only 1 or 2 ms latency. There is way more jitter, but it doesn't really matter if the total is always under 10ms.


wifi doesn't add any latency, inhernetly. nor in practice with actual equipment


The shared medium (frequency spectrum) is what can add latency. If a device wants to talk over Wifi but another device is transmitting it has to wait. This introduces (variable) latency, aka jitter.

Here's an anecdotal example for you, in practice with actual equipment:

1) Mac pro via ethernet to router:

    # ping -c 5 -S 192.168.1.88 192.168.1.1
    PING 192.168.1.1 (192.168.1.1) from 192.168.1.88: 56 data bytes
    64 bytes from 192.168.1.1: icmp_seq=0 ttl=64 time=0.413 ms
    64 bytes from 192.168.1.1: icmp_seq=1 ttl=64 time=0.396 ms
    64 bytes from 192.168.1.1: icmp_seq=2 ttl=64 time=0.417 ms
    64 bytes from 192.168.1.1: icmp_seq=3 ttl=64 time=0.553 ms
    64 bytes from 192.168.1.1: icmp_seq=4 ttl=64 time=0.514 ms

    5 packets transmitted, 5 packets received, 0.0% packet loss
    round-trip min/avg/max/stddev = 0.396/0.459/0.553/0.063 ms
2) Same machine via wifi over Unify AP to router:

    # ping -c 5 -S 192.168.1.72 192.168.1.1
    PING 192.168.1.1 (192.168.1.1) from 192.168.1.72: 56 data bytes
    64 bytes from 192.168.1.1: icmp_seq=0 ttl=64 time=2.992 ms
    64 bytes from 192.168.1.1: icmp_seq=1 ttl=64 time=4.136 ms
    64 bytes from 192.168.1.1: icmp_seq=2 ttl=64 time=1.873 ms
    64 bytes from 192.168.1.1: icmp_seq=3 ttl=64 time=2.293 ms
    64 bytes from 192.168.1.1: icmp_seq=4 ttl=64 time=2.552 ms

    5 packets transmitted, 5 packets received, 0.0% packet loss
    round-trip min/avg/max/stddev = 1.873/2.769/4.136/0.774 ms
That's an average of 2.3ms extra latency, or 6x higher.


I speak a lot with customers on satellite phones more or less every day. When they are calling from a BGAN terminal the latency can be a problem. Latency is around 1000-1500ms. But when speaking on a VSAT Ka-band terminal the latency is less of a problem. Data latency is around 600-800ms on VSAT. So I doubt latency of 100ms would be a problem at all. Most VOIP solutions have some mechanisms to reduce the perceived latency.


I'm struggling to imagine a network path that could induce a 1-2 second delay from one side of the world to the other. Even at just 50% of c-in-vacuum that's only a tenth of a second.


There's a lot of active gear introducing delays between those ends. Pinging www.govt.nz, which is about 17000km from me and as close to the antipode as I can find in a quick search, pings at around 300 to 400ms, so only at about 15 to 20% of c.


Huh. My nearest land antipode (from Boston, MA, USA) is Perth, and I get about 150ms ping to there. Regardless, even 400 ms is a far cry from 2000 ms.


A ping is a round trip, so you have to double the distance.


Even in an FPS unless you are trying to play at a pro level or something 100ms will be barely noticeable. Your reaction time is already probably over 300ms


One thing that may be a game changer: inter-satellite relaying. With the whole network, Starlink client to Starlink client latency might actually drop.


30 is really good actually. At least for what I’m used to.


On my home fiber, pretty much all sites I visit are like 80 ms away, Starlink would be an upgrade for me, latency-wise


anything above 100 ms is starting to get into unusable territory for quite a few applications


Count yourself lucky you don't live in either South America, Africa or Asia, where +100ms latency is more common than not. Fortunately I can count on one hand the applications that don't work with +100ms latency (not counting gaming). Usually it's the fault of the application developers (or rather the infrastructure team) where they put too low timeouts for requests rather than letting the requests go for a bit longer.

Sub 100ms for satellite internet is incredible, hopefully it'll be cheap enough for people to actually get, compared to the current satellite internet we have.

Can't wait for humanity to become a multi-planet species, as then application developers would have to start taking multi-minute latencies into account, and hopefully that'll help me as someone with ~500ms latency to most services.


> Fortunately I can count on one hand the applications that don't work with +100ms latency (not counting gaming).

Pardon my ignorance, but what are some exampes of applications that wouldn’t work with +100 latency?


Depends on where on the +100ms range we're on. Once you start hitting 1s latency, lots of applications (or rather, their servers) have a hard limit on 1s for every request. So when loading data from the backend, you have to continue to retry the request until it gets below 1s and then you will finally get the data.

I think Adobe been one of the worst companies I've dealt with personally, as many of the endpoints have ridiculously low timeouts (for someone with really shitty latency).


Anything that requires real-time interaction between a client and a server and other clients, e.g., gaming, stadia/geforce now, videoconferencing, ...

< 100ms is usually the "minimum", > 150ms is often "unusable", and for a smooth experience you might need < 30ms depending on the application (e.g. depending on the game you might need < 90ms or <60ms or <30ms).


Videoconferencing does not need imperceptible latency.

The other two things on your list are both gaming.


Depends how much you care about people talking over one another. If your call is a presentation/lecture/class with few switches between speakers, latency's no problem.

But if your calls normally have lively discussion where someone different jumps in any time there's a pause, the higher the latency the more likely people will say "meeting in person is much better"

Likewise, with things like remote desktop, 100ms of latency isn't a dealbreaker but it'll certainly leave some of your users saying "things that run locally just feel snappier"


For mobile phone networks, >20ms latency in audio is "unacceptable" from the point-of-view of standards conformance and a client "accepting" the hardware of some vendor.

Up to 100ms is kind of ok-ish barely-sluggish, but over 100ms latency, it becomes extremely annoying to maintain a conversation.

Video conferencing often makes this worse, because it is what people use for meetings, etc. and that involves more than 2 people maintaining a conversation, so latency becomes even more important there.

Otherwise 3-4 people start talking over each other, and none of them notices until they receive what the others are saying. Which is extremely annoying.


Gaming and video conferencing come to mind. Videoconferences need bandwidth for obvious reasons but it's also nice if you don't have any delays in when you says something and when the other side hears it. I've been in some calls lately with very noticable delays. Especially people joining from mobile phones tend to be affected (shit latency, variable bandwidth).


Voice chat is obnoxious with more than a few tens of ms latency.

Around about 100ms things stop appearing instantaneous, beyond the scope that everything starts to suck if you're not a patient person.


> Can't wait for humanity to become a multi-planet species, as then application developers would have to start taking multi-minute latencies into account, and hopefully that'll help me as someone with ~500ms latency to most services.

Stuff that requires realtime (or near realtime) communication simply won't be possible.

What remains is bulk data transfer, here I guess the only viable way is:

1) on both ends in both directions, massive buffers (at least bandwidth x 4)

2) massive FEC (of course it will reduce the net bandwidth, but there's no real other way to avoid lots of retransmissions)

3) sender station transmits the data in blocks, with each object of data having a specified number of blocks

3) receiver station checks all the data blocks for integrity, places it in buffers, and transmits back a list of broken blocks and a list of successfully received blocks

4) sender station receives the list of broken/successful blocks, deletes successful blocks from its buffer and retransmits those marked as broken

5) receiver station waits until all the blocks for an object of data have been successfully transmitted, and delivers the message to the recipient system



Yeah, in short: content-addressable systems are needed if we're ever to send data between planets on a larger scale. Systems like IPFS and alike solves this problem nicely, at least in my tests with high-latency scenarios.


What sort of areas is low latency important in outside of video games?


Finance, stock trading. Theoretically Starlink should eventually have lower latency than the optical cables under the Atlantic.


I wonder when the first CDN's will be put into space when this will become broadly used.


This got me thinking... people _will_ start putting infra into space sooner or later, just for the better latency. Imagine the new availability zones in aws/gcp/azure :)

Could definitely lead to more investment into space, more money to SpaceX / Blue Origin / etc for their new-gen lift vehicles.


Would that not be impossible because of the lack of effective cooling in space? The only way to get rid of heat in space is by radiation, which is a very inefficient process. Assuming temperatures cannot rise above 100°C, every square meter of radiator fin emits at most a kilowatt. And because half the orbit is spent in direct sunlight, a significant part of the surface area will have to be reflective, making it useless for radiating heat.

Also, where'd you get the power? Solar panels will only yield a kilowatt per square meter at most, for half of the orbit. Beam it up from Earth?


That's actually better than what most 3rd world countries got, oh god.


It’s better than what most of the US has (by land not people)


I would guess the same for the UK - certainly for Scotland :-(

Edit: Surprisingly little of the UK is actually built up areas: https://www.sheffield.ac.uk/news/nr/land-cover-atlas-uk-1.74...


That really is surprising considering it’s just an island (or a few islands? I can never remember wether Ireland/isle of man is included.) It’s weird how expensive housing can get when there’s so much empty space.


Part of the island of Ireland, Northern Ireland, is part of the UK the rest being the Republic of Ireland (which everyone not in NI refers to as just Ireland).

The Isle of Man isn't part of the UK, though it is a Crown Dependency.

This is all just as confusing as it sounds...


is that real? how are they beating the speed of light?


Latency to where? One of the promises of starlink is reducing latency over mid distance - say transatlantic or even across the states, due to using vacuum rather than fibres. Will be interesting how HFT use it.

Starlink network is nowhere near complete so I’d expect things to only get better (until customers start piling on)


If I understood it correctly Starlink doesn't send between satellites yet, only sat<->ground. Sat<->sat is a big point of starlink and when they roll that out the latency should go down, especially for starlink<->starlink comms I imagine.


I don't understand how this is meant to lower the latency. Currently:

    ground <--> satellite <--> ground
With satellite links:

    ground <--> satellite <--> satellite <--> ground
How can the latter possibly be faster?

Edit: Thanks for all the responses. I'd been assuming it was a test of Starlink latency only, but if it's Starlink -> ground station -> open internet -> ISP then it would make sense how that would be slower than a pure Starlink connection.


Currently:

ground <--> satellite <--> ground <------------> ground

With satellite links:

ground <--> satellite <--> satellite <--> ground

i.e. the sat-to-sat link should be faster than the ground-to-ground link, on the basis that light transmitted in vacuum goes faster than light transmitted in glass. That's the theory at least.


Plus they're transmitting in a straight line without a lot of switches in between.


More like this:

Currently:

  ground <--> satellite <--> ground <--> ground farther away via legacy infrastructure
With satellite links:

  ground <--> satellite <--> satellite <--> ground farther away directly


Light travels through glass (fiber optics) at 2/3 of the speed in vacuum. So as long as you are skipping some ground links by doing similar length links in space it is faster.

Low altitude orbits mean that the hops up and down can be compensated by faster hops across.

That all of course is not there yet and depends on Starlink implementing the cross sattelite links.


Current satellites are in a much higher orbit


This is a good question, I see three main reasons but I may miss something:

1. Hop count is not one on the ground to go same distance.

2. Speed of light is significantly faster in vacuum.

3. The path is potentially straighter for longer distances.


There are some places where there is no ground, say the middle of the Atlantic. Satellite to satellite is important for that.


Yes that’s a few more years before then, and the more sats the more capacity and the lower the latency.


What is the lifetime of a SpaceX satellite?

Judging from their high launch cadence, it seems satellite-satellite communication was just a means to excite the fanboys and motivate their employees, the same way the peddle the Mars stuff.


Why would HFT even consider using it? They are located as close as possible to the exchange they operate on, not across states or halfway across the world.


Same stock is listed in more than one place, and also e.g. stock movements in NY affect stocks in London. There are undersea cables built for this purpose: https://www.popularmechanics.com/technology/infrastructure/a...


What a waste of human ingenuity. And I thought ad-tech was bad enough


I think you can argue on the whole is a waste, but I do believe it does have some advantages. EG efficient HFTs can reduce bid ask spreads which does save a lot of money for retail traders.


I think the premise was that such investments are wasteful if it's only for a trading arms race.


> EG efficient HFTs can reduce bid ask spreads which does save a lot of money for retail traders.

Berkshire Hathway has $1000+ spreads yet you dont see a lot of people complaining.


Stock price is probably a pretty big factor in that along with volume.


Exactly, that is the point.


But being able to do that 50 milliseconds faster really doesn't.


That is true for much of the finance field.


Actually it's not. There are objective benefits to society that you simply ignore.


As there are objective downsides that we, collectively, are ignoring right now.

Financial capitalism shouldn't be praised, at the very best it's a lesson, we got useful tools out of it and that is all.


I am not praising anything, I am correcting your statement that much of the financial industry is a waste of human ingenuity. That "waste of human ingenuity" enabled us to build the modern world.


It wasn't my statement.


Don't forget using the equivalent of the entire power consumption of Austria to mine bitcoin?


HFT have installed microwave relays between Chicago and New York and between London and Berlin to arbitrage on the 47% fiber optic delay between the exchanges. A LEO satellite relay serves the same purpose. I can see London to New York, New York to Tokyo fiber connections being superceded by LEO satellite.


If you mean Deutsche Börse AG, than that would be Frankfurt am Main and not Berlin which is a slight difference of about 400 km. Frankfurt to London is actually a shorter distance. https://en.wikipedia.org/wiki/List_of_stock_exchanges


Quite surprised why LEO for cross exchange arbitrage was not already done. Microwaves were not super with weather conditions the last time taking an arbitrary interest.


LEO sat constellations that are low enough to beat existing solutions need to be almost global to provide reliable communication anywhere.


Arbitrage between multiple exchanges requires data to go from exchange A to exchange B as fast as possible.

Light in vacuum is faster than light in fiber.


Light in vacuum is only faster if it goes in a straight line from point A to B, not if it zig-zags across 50 satellites.


How could zigzagging through fiber be faster thaz zigzagging through vacuum? You still need to cross nearly the same distance with fiber.


The fiber is a fixed geometry.

The satellite mesh is not, a straight line from point A to point B is not possible most of the time, given the number of satellites available and range of laser communication in space.


https://www.youtube.com/watch?v=QEIUdMiColU is an animation of how it's supposed to work, including latencies

There are enough satelites and low enough routing on the lasers to mean packets will arrive from London to NY faster than even a great-circle fibre.

In reality how much bandwidth is available is a function of money, and HFTs tend to have a lot of money.


Light travels 47% slower in a fiber optic cable than it does in the vacuum of space. [1]

[1] https://youtu.be/giQ8xEWjnBs?t=288 (the whole video is worth watching, but that timestamp answers your specific question.


Too lazy to watch it -- does it take into account the multiple criss-crossing satellite hops?

I watched it -- no it doesn't :) It is comparing fiberoptics latency with straight line light propagation. So the worst case scenario of non-existing inter-satellite communication could easily be worse than fiberoptics.

But I guess if the hedgefunds knew exactly which packet travels in a straight line, they could send one packet via Starlink and others via fiberoptics.


Too lazy to answer


yes


They make trades at one exchange based of prices at another. For that reason there has been a lot of microwave relays set up between New York and Chicago, for example. Starlink could reduce latency from New York to London, another important center of trade.


But if you can access to information from an exchange halfway across the world faster than others, you can definitely trade on that.


And if you can break the speed of light, you don't need a second exchange.


It’s quite easy to “break” the speed of light in fiberoptic glass.


But it's not that useful even if you pour a glass cube over the whole exchange to slow it down.


> It’s quite easy to “break” the speed of light in fiberoptic glass.

I don’t understand this, why? Does light travel faster than the speed of light in fibre optic glass?


"the speed of light in fiberoptic glass" is the thing being broken.


Geographically distributed arbitrage?


arbitrage


They need sat-sat links for that, I think the current generation of starlink doesn't have that so initial customers will not see that kind of benefit, their data will be bounced back to a ground station from the same satellite.


I'd assume latency from endpoint to satellite back to terrestrial base station.

I can get satellite Internet right now with 40Mbps down / 5 up, but the latency is 800 to 1000 ms, so it's slow.


This is easy to have a lot of bandwidth when you only have 10s of users, but when the bandwidth will be shared between 10000 users will it plummet?

I remember when 4G was new and only few user where using it, it was very fast, but when more users arrived on the spectrum, the bandwidth per user decreased and the latency incised.


I have been on EE 4G in central London (Westminster) for several years now while at work. A speed test back when it was new would get me 85/85 Mb, these days, with a much more capable device, I am lucky if I get 15/30 Mb (download is being taxed more). It's not horrific, but it's a far cry from what I was advertised when I started. I guess I'll just upgrade to their 5G plan and start the cycle again sigh


I’m on three. I’ve noticed the speed drop in town over the years. What’s really surprised me has been the lack of an increase since lockdown, even though the City is empty.

Which makes me suspect there’s something else going on rather than just undercapacity.

Chronic underinvestment across networks?


More and more users use 4G at home rather that WiFi. In general, home WiFi seems flaky, stops working in some rooms, cuts out if you go outside, etc.

It's often easier to just use 4G always.


Very interesting point. I've not been into the office since lockdown but I suspect if it was also the case with EE, they've probably shifted a lot of their backbone's bandwidth from the 4G network to 5G for that sweet marketing and word of mouth.


I remember when 4G launched here in Atlanta I could pull down 100Mbps no problem.

I still can (pulled 150mbps down as a test for this comment), but I could then too. Getting 150/30 to nearby servers.

I guess I live next to the tower lol.


Same. EE 4G in CENTRAL LONDON with a SIM-card Home Router and the speeds are horrific — mostly 5Mb, and maybe 15Mb for an hour or two each day.

Used to be consistently 50Mb+ back in the days.


Is 15Mbit really such a horrible speed? For a mobile device I think that’s still great.


They're talking about a home router with a sim card. I'd say 15Mb is pretty poor for home use.

It may be better than some wired home broadband in parts of London though (I used to get 3Mb/0.2Mb in zone 3).


I think they throttled all users to get them to pay for their "double speed" plan upgrade.


I've asked them this outright, several times and they've always denied it. They deny any throttling.


I also remember that. I had a WiMax dongle that I got when they first came out (Internet while commuting, how novel!), and it was very very fast. I was blown away. A few months later, the first WiMax phone came out (yeah...) and the network became completely unusable. Sprint actually ended up giving me a full refund, it failed that badly.


It depends what is limiting the speed... The limits might be due to regulations like "The user terminals may not transmit with more than X watts". If that's the case, more users might lead to a (nearly) linear increase in performance.


There are suggestions that traffic from user A might go up to a satelite, then down to user B, up to another satellite, and then down to the ground station.

If that's the case, more users leads to more efficient paths and potentially a higher peak throughput for each user (assuming most links are idle most of the time, which they are).


Starlink might even be affected by rain and weather conditions like some old wimax internet. I haven't seen any information about this regarding starlink.


This is amazing. I know there will be latency issues, but for majority of web browsing, videos, etc, it shouldn't matter.

We are entering the era of spacenet and the future looks bright.

Imagine having devices as small as phones or set-top boxes that can directly access internet using satellites. It would lead to a revolution in many industries.


Genuine question - why is it a given that there will be latency issues? It looks like the satellites are only ~350 miles, or a little less than 2 light-milliseconds, from Earth's surface.


Preconceptions. Satellite internet have traditionally been slow+laggy. Starlink is not. Yay first principles :)


The issue is with overall latency.

User -> Satellite -> multiple hops between satellites (as they are orbiting earth and the right one has to be selected to relay data to -> earth DNS servers -> multiple hops across internet -> satellite -> user

Additional latency is introduced due to satellites and the grid between the satellites, as well as complications from rotation of satellites.


> User -> Satellite -> multiple hops between satellites (as they are orbiting earth and the right one has to be selected to relay data to -> earth DNS servers -> multiple hops across internet -> satellite -> user

All of that in general is comparable to the costs incurred on the ground from any other kind of network.

> complications from rotation of satellites

As far as a single routed packet is concerned, the satellites are stationary.


Starlink does not at this time use any intersatellite communication links, despite having talked about it.


I'm not sure that's revolutionary outside of very rural areas or in a desert.

Smart phones and set-top boxes etc can already access the internet without satellites, and often with much better latency.

What would satellites add for urban or sub-urban applications?

I do appreciate the extra competition that satellite internet brings. Especially in parts of the world with suppressed competition (like the US..) this could have huge practical effects without having to be a big deal in any technical sense.


Smart phones use 4G networks. And now there is another alternative.

Cellular networks have limited coverage. Imagine tracking a truck traveling across the US, or a ferry, across the English channel. Currently, that's not possible with 4G. But with a satellite driven internet system, it very well could be.

As for urban and sub-urban applications, not much.


I don't think it is possible or planned to use StarLink directly from a mobile phone.

You will need a ground station to communicate that isn't going to fit in a pocket.


True, for now. There was a time when a cell phone was the size of a brick.

Maybe we will not see such miniaturization, but I can safely assume that a small backpack size device should be light and last long enough for meaningful high speed internet with StarLink.


> There was a time when a cell phone was the size of a brick.

And that was already advanced. Portable phones used to be so unwieldy, car phones used to be a thing!


I live 12km from the centre of Edinburgh in an area that is rural but not "very rural" and all we can get is 22 Mbps down and 1 up.

Edit: I've spent a lot of time trying to find alternatives - none are really practical at this time. Starlink looks pretty exciting to me.

Edit 2: We have fantastic line of sight to Edinburgh and you'd think that would give us more options, apparently not...


Could you expand on which industries will be improved? My normal life is covered entirely by 4G, with 5G now appearing.

I guess you mean things like internet on planes, shipping internet, off-shore rigs? But what things can't they do at the moment because of prohibitorily slow/costly satellite internet?


Logistics - Tracking cargo, without interruption, in real-time, across the world.

Shipping - Same as logistics, but across oceans.

Real-Time weather data, from solar powered sensors, spread across the world.

Real-Time video for tracking endangered wild-life.

I think different people can come up with more ideas.

The difference between satellite and cellular networks is coverage.

Imagine if a nature loving adventurer fell into a ravine and can just video call authorities about his location and condition.


Logistics: already done via gps receivers and low bandwidth spot style satcom. SpaceX hasn't designed for mobile/vehicle mounted ground stations either.

Shipping: already done via AIS

Real Time Weather: already done by satellites with the required features. SpaceX is not building this.

Real Time Video: again, there's no way you're getting animal scale images out of a starlink scale sat.

I swear people have taken the usual Elon hype with this and gone to absolutely absurd lengths. This is not your iphone suddenly works globally. This is not a star trek communicator. This is an alternative to sat internet like Dish network et all. Is your hiker really gonna be hauling around a giant freakin' dish?

It's going to be a premium priced broadband service for rural customers and institutions that can afford it. If you know anything about the market you know DoD is gonna be the customer they focus on.


>Logistics: already done via gps receivers and low bandwidth spot style satcom

Imagine being able to see live video, or AI enhanced video (movement detection, etc). A lot more bandwidth will open up many areas for innovation.

Imagine being able to get real-time data of patients being transported in planes or ground vehicles.

>Real Time Weather: already done by satellites with the required features.

But there is not much data about wind patterns, directions, barometric pressure, etc. Satellites can observe on a macro level. If you can combine that with micro level ground data, perhaps predictions for weather would be more accurate.

>Real Time Video: again, there's no way you're getting animal scale images out of a starlink scale sat.

Perhaps, but periodic photos, would be possible, and that would be enough for many innovations.

>I swear people have taken the usual Elon hype

Even I have the after taste of Hyperloop. However, this is different because there are billions being spent to send actual satellites to space. That is not hype. I would not cast aside Elon's vision so easily, especially if he has committed and spent so much on the idea and is actually building out the infrastructure.

>This is not a star trek communicator.

But Imagine if you could have one.

>Is your hiker really gonna be hauling around a giant freakin' dish?

Would it be too far fetched to think that those can be shrunk down, perhaps to the size of a suitcase or a backpack?

>It's going to be a premium priced broadband service for rural customers and institutions that can afford it. If you know anything about the market you know DoD is gonna be the customer they focus on.

Sure, that would be one revenue source. But many innovations in the world are made possible due to what is available. Would an app store be possible without the iPhone? But, when the iPhone was made, the app store, as a concept was at best, a theoretical idea in Jobs' mind.


> Imagine being able to see live video, or AI enhanced video (movement detection, etc). A lot more bandwidth will open up many areas for innovation.

For what benefit?

> Imagine being able to get real-time data of patients being transported in planes or ground vehicles.

This is a very niche use case.

> But there is not much data about wind patterns, directions, barometric pressure, etc. Satellites can observe on a macro level. If you can combine that with micro level ground data, perhaps predictions for weather would be more accurate.

I'm sorry but this is incredibly ignorant. There's a huge volume of data that's fed into several global weather modeling supercomputers continuously. That's literally where the weather forecasts on your tv come from.

> Perhaps, but periodic photos, would be possible, and that would be enough for many innovations.

You still aren't getting it. There are fundamental physical limits to image resolution based on the size of the optics. You'd have trouble resolving individual creatures even with a telescope the scale of Hubble/KeyHole. Starlink changes nothing about this.

> However, this is different because there are billions being spent to send actual satellites to space.

The hype is people speculating on applications like the above that are entirely implausible and unrealistic. It's on the same scale as thinking the next Tesla is going to be a flying delorian with fusion power.

> But Imagine if you could have one.

But you just aren't fantasizing, you're claiming these fantasies will become real in the short term without understanding the limitations that preclude them.

> Would it be too far fetched to think that those can be shrunk down, perhaps to the size of a suitcase or a backpack?

Yes, it is too far fetched. You cannot shrink a wavelength. In addition, starlink's design requires the transmission/received lobes have a certain maximum beam width. The global capacity of the system depends on this parameter. That means they need phased arrays of a certain minimum size in terms of array area and number of array elements to meet their specs.

> But, when the iPhone was made, the app store, as a concept was at best, a theoretical idea in Jobs' mind.

There were examples of app stores before iPhone. That wasn't a unique new concept.


The latter is pretty well handled by existing devices such as SPOT:

https://www.findmespot.com/en-gb/

Edit: Starlink looks fantastic but maybe not for that particular scenario, at least not any time soon.


Here's one: Teleoperation / autonomy in remote areas. Forestry. Mining.

Not sure how big this, objectively speaking money-wise, but for teleop/autonomy it sure does enable new things when there is true 100% coverage everywhere on earth.


Already being done with existing satellite services.

Starlink is cool and al, but satellite internet isn't new. Both forestry, mining and farming is utilising existing satellite services for monitoring and operations.

Starlink is better than all existing services, but with some drawbacks. Like it will be really expensive to provide service in really remote areas until sat to sat relay is possible as there must be a ground station within 500 km radius of the satellite. This means that to provide service in the Amazonas SpaceX will have to build ground stations in the Amazonas. To provide service in the Pacific and Atlantic SpaceX will have to deploy several floating ground stations in the middle of the oceans.

Again, not taking anything away from Space X. Starlink is really cool and will be a wake up call for the satcom industry, but satcom is not a new thing and Starlink have significant drawbacks for the time being.


> it sure does enable new things when there is true 100% coverage everywhere on earth

But there won't be. Not with Starlink at least. Even at their envisioned full constellation size of 40000 satellites, coverage won't be 100%. The Arctic and Antarctic regions won't have any coverage at all, for example.


Imagine the military setting up an internet base station and releasing remote controlled drones for surveillance or military operations. Having satellite based high speed internet would be a game changer for remote locations.

Imagine being able to conduct a robot controller surgery in Africa, by a doctor in the US. Currently, not possible, but with high speed satellite based internet, it could be.

Imagine an electrical engineer being able to help his colleagues in Africa, with specific issue, over VR or Video calls, enabled by Satellite based internet.

I can see a number of possibilities and opportunities.


> Imagine being able to conduct a robot controller surgery in Africa, by a doctor in the US.

Highly unlikely for many reasons: latency too high, cost of equipment way too high, only very few hospitals are even equipped for remote surgery even in the West, let alone underdeveloped countries...

> Imagine an electrical engineer being able to help his colleagues in Africa

That's already possible even without satellite-based internet and doesn't require it at all.

The only point I agree with is the military - surveillance and remote controlled drones could benefit from greatly from low-latency high-bandwidth satellite internet.


> only very few hospitals are even equipped for remote surgery even in the West, let alone underdeveloped countries...

You're talking about today, which is the lowest potential time scenario. How about the time frame of 15-20 years from now? It'll take years just to fully build out Starlink's constellation, then many more years to fully spread its use. Remote surgery will continue to expand, becoming increasingly common, and the prices will come down. As that process unfolds, you will increasingly see the major robotic-assisted surgery companies donate (or otherwise massively subsidize) equipment to poorer nations, exactly as is done with very expensive drugs now. Also, while today robotic-assisted surgery costs are very high (dominated by companies like Intuitive or Stryker), China and others will eventually fully knock down that door and flood the market with lower cost systems that are 85% good enough and will become very widely used.

It's also more likely that it'll be highly skilled doctors in one less developed nation doing work in another less developed nation. That will help keep the costs down more than the parent's example. US doctors will likely do most of their remote work within the US (other than charitable work), where they can command far higher fees.


>That's already possible even without satellite-based internet and doesn't require it at all.

Maybe in developed countries. In India, most Power Plant construction sites don't even have dial-up internet. There are a lot of projects going on in a lot of remote places, where there is simply no infrastructure available for internet. StarLink could be rented for a while and perhaps used after that.

There is tremendous potential for such a service.


But most of this is already being done with existing satellite services. Starlink is cool and better than existing providers in many respects, but it's not a total game changer.


Restart my production web server from a cruise ship.


Tons of areas have either no 4G at all or the signal is too weak to work fast.


Plantations! No longer will they be isolated little fiefdoms where the company manager is judge, jury and executioner.


I'm not familiar with these remote plantations, where can I read about them?


Palm Oil in remote Kalimantan and Irian Jaya have very little oversight. One of my family members is a plantation manager.


In my experience (multiple clients in rural area) low speed (1.5 Mbps) with low (50 to 80 ms) latency will outperform 40Mbps/800 ms latency connections for video.

Web browsing, of course, isn't really affected by latency.


"Web browsing, of course, isn't really affected by latency."

Web browsing involves a lot of ping-pong traffic pattern that multiplies latency.


When you say video, do you mean video conferencing? I don't see how latency would affect video streaming significantly, other than the initial buffering time. In fact I would say high latency affects general web browsing much more than video streaming.


Web browsing can be super slow if you have a high latency. Between the DNS requests, the TCP handshakes, the TLS handshakes, it takes forever before you start receiving the actual content.


Where are you getting the 800 ms latency though? That's what Geo stationary satellite internet provides due to the physical distance they have to operate at (35786KM), but Starlink and OneWeb and other mega constellations are operating at 550KM which easily gets you below 100ms and likely close to 30 ms without sat to sat comms.


Fantastic. I'm keenly curious on how this will work with national regulations. Having a single global ISP is very attractive on the surface and that's with the huge caveat on monopoly, but as an additional option. I'm just not sure how this will interact with some nations' policy/restrictions. Looking at this aspect very much.


The right way of making sure this doesn't become a monopoly is to treat this satellite infrastructure just as you treat the fiber infrastructure in many countries. Where instead of having the ISP owning the infrastructure, they can rent the usage of it, so you can have many competing companies using it.

So Starlink would be the infrastructure and they need to allow others to use it on the same level as the Starlink ISP uses it.

One example of this already working is in Sweden, where Telia-Sonera is owning the actual infrastructure but they have to allow others to use it as well. So in Sweden you have plenty of ISP choice, one of them being Telia-Sonera, but not only them.


The tech who installed my Starry Internet said that they were going to offer Starlink next year in rural areas. Maybe he was confused but he was reasonably knowledgeable.

We were specifically talking about Elon, Starlink and rural Internet. Obviously Starry is a completely unrelated company.


If countries require permission to use slices of radio spectrum over their territory (I don't know to what extent), then they can control whether a radio service operates there.

Does this usually apply to space-based operations? I would guess there must be minimal coordination, or otherwise anyone space-based could degrade the infrastructure of countries in line of sight.

If such regulation exists, then satellite internet either needs to make sure they have all the permission to use their band (bands?) in particular countries, or... I guess they can just ignore the local laws, blast away, and claim no jurisdiction?


It's coordinated through ITU.


It's going to be very interesting for me. I plan to live on a sailing yacht and travel the world permanently, StarLink will make my working options a lot more flexible. Presently people are forced to see out their contracts over potentially dodgy marina WiFi.

The regulatory situation might get quite complicated I think, for example if you're in international waters, connected to the internet by a global ISP whose infrastructure is also outside of territorial limits then whose regulations do you have to comply with (besides your clients)? What's stopping the likes of Google poaching the concept of Radio Caroline, planting a datacentre somewhere in the North Sea and raising a two-fingered salute to any sort of data protection laws?


>The regulatory situation might get quite complicated I think

I think it'll be OK. Cellphone roaming works just fine for example


True, but cellular infrastructure is definitively in someone's territory. If I stick a phone mast up I'm still subject to things like planning permission even if the signal crosses a border. As far as I'm aware, space is pretty similar to international waters in that nobody is sovereign there.

It could lead to a bit of a "wild west" situation with governments giving themselves extraterritorial powers. In the late 1980s the British government gave itself very draconian powers to conduct armed raids on radio ships in international waters for example, they take a really dim view of information outside of state control even if they're not the state controlling it.


Space X will need a license in each country/regulation to provide their service there. They will probably employ some geofencing or similar to comply.


Hard question. We will see. Take GDPR, only applicable, though extraterritorial, to EU residents. So if on a yaught or extra-EU for 183+ days? Plus multiple other territories doing GDPR similar.

Interesting to watch. And most of the world does not live on yaughts. I think the answer will be a lot of deals. I'm in China and Elon is putting in a lot of energy here, and with Starlink a big potential conflict of interest with Starlink vs Tesla. Not sure how fickle that is. We will see.


Why is a single global isp good? That's basically saying a single company owning and controlling the internet. That sounds dystopic


For many countries like mine this will be a window towards freedom. As long as US stands behind its constitution


Yes, need multiple ISPs so the national restrictions would cancel each other.


The true question is: may I be able to play an online video game in the deep Sahara before the modern human civilization collapses?


50ms ping should be reasonable for most games.

whippersnappers who complain about 20ms ping should watch some quakeworld or quake 3 cpma where this dude rat used to destroy everyone on 300ms.


That's an unfair comparison though. Everyone used to play with high latency. There's a clear advantage to you if your ping is 20 ms and mine is 300 ms.

Whether there's a clear advantage to having 50 vs 20 ms ping I can't answer. I'm quite certain that's an advantage I personally can't make any use of, but a Quake pro might be able to.

Also, today's pro gamers are likely at a higher skill ceiling than the best players were 10 years ago.


> That's an unfair comparison though. Everyone used to play with high latency.

I played a lot of Quake and that was not remotely close to being the case. There were the high ping and low ping players, with not much of a % inbetween. In 1996-1997 the vast majority of people were playing Quake online via dial-up modems. The best players had low pings and were playing over ISDNs or better (a lot of players were on fast university or corporate networks).

> today's pro gamers are likely at a higher skill ceiling than the best players were 10 years ago.

The pro-level players in the later half of the 1990s were as elite as anything that exists today in the FPS space. They were playing like it was a job even when it wasn't, and they had been doing it for a while (Doom LAN competitions were common before that). Quake clans were prolific. They also had professional caliber competitions / tournaments back then, see: the 1997 Red Annihilation Quake tournament in Atlanta, where Dennis Fong won Carmack's Ferrari.


On some games (Rainbow 6: Siege being the one I'm most familiar with) latency actually helps you, as the enemy has an out of date picture of where you are.


Yes, in Natural Selection 2 you can't properly ambush anyone unless you take into account the ping and jump on them before you even see each other.


Depends on what latencies you can tolerate.


And this time it is literally down / up !


We're already doing literally down/up as your traffic goes from your computer on a above-ocean-level altitude, into the ground, eventually probably into the ocean floor and then up into some other building (at least considering transocean traffic)


Has pricing been indicated anywhere?

Edit: See https://www.inverse.com/innovation/spacex-starlink-heres-how... and https://www.zdnet.com/article/spacex-starlink-internet-prepa...

So about $80 pm. That's pretty good imo.


This is only a few Mbps slower than my Fibre-to-the-Cabinet broadband in the UK provides. And I'm living in a town; plenty of rural areas are much slower.


It's actually marginally faster than mine - UK town in the north also.


I'm in a VM cabled area, so the I get ~ 100/10 Mbps and ~ 12 ms ping. This is the second lowest tier that VM offer btw


They put in VM 2 years ago - and missed off the suburb in which I and 2k others live.

I'm not holding my breath... but I had VM before I moved here, and prefer the BT-network service to it.


I assume that this is the only internet connection where gravity plays a part in the up/down speeds!!! :)


I'm eagerly awaiting reports from beta users about the length of downtime during antenna repositioning and handover to another satellite.

Hopefully it's sub-second. I can't think of anything apart from a competitive gamer playing a shooter that might be intolerant of this.


After installation the antenna does not move. That antenna has the capability to connect to two satellites at the same time.

What would change would be the latency which would have a bit of a jump on a handover.


Is there a way to use this on the go, battery powered? I wonder if it would be a good alternative to using 4g or 5g networks. How does it work indoors? Does the antenna/receiver need line of sight?


It requires line of sight and doesn't work indoors.

It also can't be an alternative to cell networks since the antenna is huge and heavy compared to phones, requires LOS and needs to follow the satellites [1].

[1] https://cdn.shopify.com/s/files/1/0173/8204/7844/articles/GM...


The bigger question is what's the total capacity that is shared by a single neighborhood/city. For example - does this mean that a small city can only use 42Mbps down across all receivers?


This is already better than most of Australia's internet


People talk about latency but what about packet loss?


I think as long as the sats don't leave blind spots as they move around this won't be an issue? Or are there other factors? Anyone confirm pls..


Weather could be an issue. Would be interesting to see how Starlink internet fares when there's thunderstorms or heavy rain.


Can someone explain the physical layer aspects of starlink? Is the spectrum same as mobile data spectrum or does it work with different part of spectrum. If it is different part, do we need specific hardware to receive and transmit?

Are there any interference/loses issues that might popup when scaled?

What is stopping spaceX to provide internet outside US?


> Can someone explain the physical layer aspects of starlink? Is the spectrum same as mobile data spectrum or does it work with different part of spectrum. If it is different part, do we need specific hardware to receive and transmit?

Almost certainly has its own slice of spectrum. You need a pizza-box sized antenna set up, so it's not quite person-portable yet, but not too bad for installing on a vehicle.

> Are there any interference/loses issues that might popup when scaled?

Each satellite only has so much bandwidth. So the more people using it in the same area, the worse it's going to get. In theory they'll add more satellites as needed, but it's never going to be much good for towns/cities.

> What is stopping spaceX to provide internet outside US?

Ground stations and regulations. Once they have satellite-satellite communications running the ground stations become less of an issue (though there's a latency penalty if your signal has to go a long way round the satellites before getting to the ground).


Well, I would add that both satellite and ground receiver have phased arrays antennae. These allow emitting multiple focused beams simultaneously.

Since the angular resolution of these beams is limited, they each cover a fixed ground area (let's say 20km across). If more than one customer sits in that area, they will have to share airtime with the others.

Now, it's probably easier for the satellite to perform (time-division or other) multiplexing, since it knows how much data each node is going to receive.

I guess it's a lot more involved when it comes to sharing upload bandwidth with another client you can't see.

Anyway, this makes it easy to scale in rural areas, and pretty much a no-go for cities.

Now, a question of my own: do SpaceX satellites already have HW for sat-to-sat communication?


FYI On this site if you press "I do not accept" on the privacy pop-up it it redirects you to another page.


And hitting the back button sends you to the linked page.


This is already better than anything I can get (rural NZ). Any info on service tiers and pricing?


Has anyone tested bufferbloat yet? [1] If so, please link your results for Starlink.

[1] - http://www.dslreports.com/speedtest


My mom can't get broadband even though she has a house in Wrocław, a quite large city in Poland. She is using 4G internet sticks.


As good as my residential ISP, though I'm using a 30/10Mbps because it suits my current needs. Why pay more?


That sounds pretty good for version 0.1. More sats would make it better, as would better sats. A promising start.


If this provides unfiltered internet to regimes which dislike it, will we see them blowing satellites out of orbit?


No, they'll simply deny licensing of spectrum over their land area, and SpaceX will comply.

This will never be some tool to counteract totalitarian governments.


I am interested in that. Are there any international treaties that force SpaceX to comply? If SpaceX has no business in totalitarian country X, do they have any way to sue SpaceX for offering service? I imagine a black market will establish to get access to starlink even if SpaceX doesn't officially sell access. Can you expand on this topic?


https://www.itu.int/en/history/Pages/ConstitutionAndConventi...

If SpaceX violates ITU policy they risk their licenses globally.

A black market for this service is not possible. The equipment necessary to confirm SpaceX is transmitting into an area, or to radiolocate the ground stations themselves costs a few grand. Even low income nations can afford to monitor and enforce spectrum requirements if they desire to.

But more fundamentally, this is not a goal of SpaceX or Elon. The projection of "hero" onto him and this project is out of control. Elon is doing this to make money, plain and simple.


It's much easier to simply block payments to Starlink.

Can't use a service that you have no method of paying for...


How many rural locations share the sky which a major metropolitan area which will clog up the satellites?


People always complain about Latency. But Latency to where?

And is latency in Asia or the Caribbean or anywhere else outside of North America that bad? Nah.

In 2020, latency from almost every continent is good enough for 99.99% of use cases.

Good enough that I can sit here in Bangkok and use Vim all day remotely into servers in Texas without any difficulty.

Starlink will be awesome for the majority of the world. Even "gamers" (if the pricing is good).


I don't think putting quotes around the word gamers helps you get your point across particularly effectively - it sounds like you don't really think it's important enough of a usecase to be worthy of much consideration.

Many years ago, before life got so busy, I used to play FPS games online over various forms of dialup and then (somewhat) faster links. Bandwidth was fine, if you were patient about downloads, but latency caused real problems, even on "good" days when you could get close to 100ms on some of the better links. In short, 100ms latency (taken from another thread) is definitely enough to cause problems for many games, and would also make things like stadia/geforce now/playstation now not viable. This doesn't mean it's not useful tech - I can think of many useful applications, but I don't think we can just handwave away all the cases where 100ms of latency might, actually, cause an issue.


Starlink aims to achieve sub 20ms latency. This is OK for gaming, especially with modern shooters.


While latency is more important than bandwidth for most games, jitter is even more important. Lower latency is obviously better, but under what duration? If it fluctuates between 10ms and 50ms, while only being 20ms for short periods, many games will have problems with this, rather than a 50ms constant latency where the game servers can start accounting for it without issues.


Do you have a source? A link in a different thread paints a somewhat different picture. If they can achieve stable ~20ms latency though, that would indeed be excellent.


Going from 5-10ms to 30-90ms* might take some getting used to. That's on top of geographic latency.

Chrome's devtools will let you simulate this. It's not pretty but it is better than nothing.

* Yes yes yes, Elon promises better than that, but 30-90ms is what it's currently tested at.


Yep, except for FPS & RTS games involving fast hands and quick minds. Anything more than 75 ms is gonna hurt your reflexes.


Exactly.

Elon confirmed the target is <20ms and <10ms over time.


what's the current status of the constellation? are there moments where the network is down because there are no satellites overhead or is the thing already big enough that some part of the world are covered all the time?


So much better than the best wired connection I can currently get here in Aus...

Awesome!


May be silly question, but why is the up so low compared to down?


They must have provisioned less spectrum to uplink. Usually people want more downlink, less uplink. Symmetric services exist, but are usually offered as a "pro" or enterprise solution.


I actually have DSL and it is similar, not speed, but asymmetry.

12 Mbps down and 768kbps up, almost 20 to 1 ratio.

Most ‘consumers’ download more than they upload.


anyone know, what happens if some of these satellites fail and are uncontrollable (which is bound to happen at some point)


wondering how many down-link stations they have. this will be v. interesting when it comes to latency vs current solutions.


Have they said anything about pricing yet? This would seem like a decent solution for roaming data during travel too.


Nothing specific, I think they are still figuring that out, but Gwynne Shotwell, the president of Spacex, is hinting a bit I think with:

> Shotwell said millions of people in the U.S. pay $80 per month to get “crappy service.” She didn’t say whether Starlink will cost more or less than $80 per month but suggested that would be a segment of the public the company would target as well as rural areas that currently have no connectivity.


Yeah, the last mile is huge. My parents have paid between $100 and $200 per month for the last decade to get metered Internet that's worse than this.


I prefer a clean sky and earth telescopes to work instead of 42.3Mbps down / 9.7 Mbps up.

see: https://www.nature.com/articles/d41586-020-02480-5


Yeah it sucks, but I think it is kind of inevitable that someone was going to do this, and even Iridium had similar problems. Now that we have at least 3 companies pouring in serious money, it seems likely that the economics will keep pushing for it.

My hope is that this will lower the cost of access to space with another magnitude, which should allow a serious scale up of the amount of astronomy done from space.


I would also be glad, if you could just look at the sky and don't be disturbed by satellites and air and light pollution. I think, SpaceX and others will eventually manage to make the satellites less or mostly not disturbing. But lets face it, SpaceX actually enables better professional astronomy because launching a telescope into space is now much more affordable. Also all those remote locations, where normal earth based astronomy still kind of works will have a usable internet connection to transfer the results e.g. for computation and further instructions. Also scientists will be able to e.g. video-talk home which should make these long trips into nowhere more socially acceptable in their respective families. I can imagine, that would make some of that research more surmountable.


I don't think SpaceX would break Earth satellites. Isn't it trivial to remove a light streak from a picture? Satellites were a problem before SpaceX and I'm sure it's 1 click in some software astronomers use. The surface area of a streak of light must be less than 1% of the total surface area so virtually no information is lost. If you use multiple exposures it's even less. Seriously, an astronomer who can't deal with this problem should just give back their PhD. The idea that humanity should be prevented from extending internet access to the entire planet because an astronomer finds it too inconvenient to click a button is example of entitled behaviour on a scale I've never encountered in my life.


I prefer an orbital ring already across the entire sky.


I'm sure you do, Chief.


Last time I mentioned latency all the fanboys squealed about how latency isn’t a problem any more with low orbit satellites.

Whatever.


Just as a heads up, you'll be downvoted, not because you're mentioning latency, but because your comment is low effort and basically complaining about discussions in a place where discussions are kind of the point.

Is there any new information about latency since before that you want to expand on?


Cause with satellite is ALL about latency. It’s THE topic. Most comments here will reference it. It’s a critical issue. It doesn’t go away.


If you check out the comment I made earlier, I'm very aware of that (https://news.ycombinator.com/item?id=24291548)

But it's more interesting if we discuss the latency itself, rather than complaining about previous discussions around the latency.

As it seems right now, the latency is more than alright, at least for people who are used to really shitty latency. Sure, it's not gonna replace your home connection if you live in-or-near a city, but for the rest of us, it's much better than what was offered before.


You have a point.

At 550k off the deck, my house to Los Angeles isn't that much closer via satellites link that terrestrial fibre, maybe a few hundred kilometres.


But by the existing evidence, the latency seems to be fine: https://www.reddit.com/r/Starlink/comments/i8n1s5/here_is_a_...


Starlink blah blah blah.

Wait till the service is massively over subscribed and get back to me.

With any luck Starlink operators (Elon Musk?) won't let that happen, but it does tend to be the way of things.

Additions, the unintended consequences haven't revealed themselves fully yet.

One thing I personally dislike about electricity, cable internet, and fibre optic, is what they do to the visual appeal of the built environs. Could be underground, but only areas missed that boat.

Not saying the trade-offs, if any, won't be worth it though.




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

Search: