With the upcoming microsat 2a & 2b launch let's discuss the SpaceX internet constellation codenamed Starlink.
>what is starlink? A hueg network of LEO satellites for the purpose of providing internet everywhere. The initial constellation consists of 4,425 satellites and a second constellation of an additional 7,518.
>but why? It'll make SpaceX hundreds of billions of dollars and fund future mars projects.
>how speedy are we talking? 1 Gbps per end user.
>but what about the latency? 25ms. LEO! Data travels at 1.0c in a vacuum. 0.7c in fiber-optics.
>Won't this take a fuckton of launches to establish? Yes, 40 sats can fit in a standard f9 fairing. So.... 300 launches.
>But what about the weather? Downlink is in the Ku band and uplink is in the Ka band, so severe weather would have a moderate impact on connectivity.
>How would one connect to the Starlink network? You purchase a pizza-box sized phased array unit and pay a subscription fee.
There wont be an "off the grid" once this thing goes live
Oliver Edwards
how so? I'd say that Starlink will be a benefit for those that live off-the-grid. You can live in the middle of Alaska with some solar panels and still have fast internet.
Landon Johnson
I want to believe, but I don't know if I should. Elon makes a lot of promises, and has a good record so far, but when someone promises you the moon (and Mars, and now this) a bit of healthy skepticism is necessary.
That said, I don't stand to lose anything whether this succeeds or fails, so I'd might as well hope it does. Is there any word on pricing for this yet?
Colton Garcia
Good luck ever going anywhere or doing anything without it being tracked or monitored again, unless you go full primitive.
Brody Myers
They should call it the Kesslernet, in honor of Donald J. Kessler, who predicted that one day there would be so many satellites in LEO that one breaking up would cause others to break up, in a snowball effect that would destroy all satellites in LEO and leave it unusable for decades.
Logan Clark
pricing will probably be like how Steam sets game prices. It will depend on the PPP/etc of each country.
As for promises, this is already much further along than they want you to think: - They're flying working hardware tomorrow to test the basic satellite design - They've filed multiple FCC applications and appeared in congress to work with regulators on various issues - They have already opened a Starlink/satellite specific office in Redmond - Elon has stated that they aim to initially acquire 10% of the ISP market.... keyword: initially. This paired with the somewhat quiet news on "how we'll fund mars colonization" hints that Starlink is SpaceX's ace-in-the-hole for generating huge amounts of revenue.
Leo Richardson
it's low-enough-LEO that air resistance will deorbit alllll of them unless they boost periodically with the onboard Hall effect thrusters. So even if the Chinese blow up a couple it shouldn't be a "big" problem. Kessler syndrome is sort of a meme anyways.
Jaxson Reed
>implying that wasn't already happening with GPS and tracking you
don't be a derp
Ryder Garcia
They're communications satellites and small ones at that. Unless you've got the equipment to talk to them they won't be able to track you. NRO probably has ones that can do it but they're not going to be pointing them at you.
The ability to live off the grid was already dead though assuming you've ever had a car, a phone or any device that connects to the internet. Plate readers and cell towers are everywhere nowadays and the NSA has always monitored everything on the internet.
Isaiah White
one idea is to also put cameras on every starlink sat. 12,000 photos of the earth simultaneously multiple times per day would be amazing for large-scale earth/climate/etc research. Not high enough rez to be too spoopy though, for sure.
Joshua Myers
Info from an old post for comparisons:
Hughesnet uses 2 main sats now; which has increased their bandwidth and speed a great deal, but they still cap data and bandwidth like hungry Jews. The last one launched last year, "EchoStar XIX" operates in geostationary orbit, at a longitude of 97.1 degrees west.
Since SpaceX's Starlink will have 12,000 satellites (fug) as its network they must be very very close to the Earth. Officially between, "1,110km and 1,325km." Compare that to the ISS at 422km and Iridium-104 at 780km. The air travel (bird fly) shortest distance from United States to Europe is 7,895.1 km. Total round trip distance for Starlink communications would be 2,220km to 2,650km.
Geo orbit lag time is 240ms - 279ms for the round trip. Starlink lag time would be less than 10ms round trip.
Kayden Reyes
If you own a smart phone of any kind that already happens.
>tfw i don't own a phone of any kind but never actually go anywhere anyway
Easton Martinez
Hughesnet fucking sucked many years ago when I had it. Lag time was a bitch and they throttled your speed down to dial-up when you used too much too fast then the speed came up like some slowly regenning healthbar.
Kevin Jackson
don't forget, the VLEO (v-band) satellites will operate at an altitude of 335.9 to 345.6 km. 1110-1325 is for the higher orbit ones. So even quicker than that
Joshua Torres
in reality, the hops between sats and the space-ground transfer adds delay.
>"The system’s use of low-Earth orbits will allow it to target latencies of approximately 25-35 ms" is what is actually known from the filings.
Plus... > the 1600 satellites in the Initial Deployment would have a total aggregate capacity of 32 Tbps
noice
Aiden Bell
So where are all the people who usually shit on Elon? I would actually be interested on their take on why this can't be done. Because as far as I can see there is no reason it can't, other than cost. And man I cannot wait for the ISP shrieking to begin when they suddenly realise how far behind the game they are and Elon is raking in billions with a smugface on.
Jaxon Smith
here are some potential reasons why it can't be done: >SpaceX is not able to bring the cost-per satellite down They're designed to last 5-7 years max, so a high unit cost would be a huge issue >Launch cadence is too slow FCC requires that a big chunk of the constellation is operational within X number of years >International pressure China might not want censorship-free internet flying over their country >Not enough capital on hand 300 launches is a lot, and 10k satellites is also a lot >OneWeb beats them to the punch and the good ol' sunk cost fallacy means that Elon cancels starlink >Kessler syndrome wew lad
overall it's a pretty foolproof revenue-generating plan. I don't think it'll fail
Aaron Brooks
Cancellation? Ohh constellation.
Well we'll see I guess
Gabriel Morales
Ok interesting, I guess it really comes down to how cheap the satellites can be made for and what China would do, I mean they can't exactly shoot down 10k satellites, the amount of insanely expensive ASAT missiles would be too much. More likely it would just be made a crime against the state to own a Starlink receiver and the rest of the world gets on with it.
Luis White
>crime for a starlink receiver then people would make homemade phased array units and rip the pizza box code and plop it on an Arduino. The unique decryption/connection key would be an issue though.
I'm probably oversimplifying it however...
Nathaniel Moore
Elon tweeted about the two testsats being onboard the Paz launch; maybe we’ll see deployment of them in the stream?
Ian Cooper
Starlink is impossible and will be blocked on legal grounds. It's essentially attempt to achieve global monopoly and not only will there be massive opposition from inside the US but the international one might as well be even larger. The idea is not bad but it is achievable only through large international effort.
David Clark
>They're designed to last 5-7 years max If it depended on taking longer to make its money back, it just wouldn't be viable.
>>International pressure >China might not want censorship-free internet flying over their country They don't like spysats flying over their country either. They can't do anything about it. Anyway, they can be shut down as they overfly China and other countries where they don't have regulatory approval to operate.
>300 launches is a lot, and 10k satellites is also a lot SpaceX says 800 satellites is enough to provide service, so that's only 20 launches' worth. The other ~12,000 are optional to increase capacity to meet demand.
>>OneWeb beats them to the punch SpaceX is already planning to coexist with OneWeb. Competition isn't necessarily bad for a business. It means people can sign up for internet in a remote location, knowing they can switch and get similar service if their first choice starts to abuse their market position.
Jonathan Collins
>t. increasingly nervous hughesnet executive
Jaxson Robinson
Just like how the GPS sat system was blocked right, Pradman?
Luis Cook
You do realize that Satellite Internet access exists now don't you? The only thing Starlink is bringing to the table that's actually new is Elon Musk's name and the fact that he has a low cost launcher
Kayden Bailey
plus the ability to take over the entire non-city-broadband network.
Literally every (well, most) rural area in the world will be better off with Starlink rather than hughesnet/wired.
Joseph Baker
You don't get it. This'll be affordable low-latency broadband. It's nothing like previous satellite internet offerings.
Jaxson Sanders
I pay about $35 for DSL that gives me 6Mbps down and 1Mbps up. The main problem is connectivity. The lines are literally 50 years old in some places. I think they are close to all 20 year old lines as being the oldest. I live in B.F.E. and for many years I used dial-up to browse Veeky Forums. I tried Hughesnet but it was a horrific waste and went back to dial-up. I was actually able to download more on dial-up than Hughesnet , due to Hughesnet's speed capping scheme and terrible lag.
It looks like Starlink would be about 2 times the lag as I have now and shouldn't even be noticeable. My service contract is about to expire this month too. Starlink can't get here fast enough because service contracts are 3 years long.
Brandon Moore
what's the latency if you ping 192.0.43.10?
Jackson Reed
i hear "12k satellite network in LEO" and my questions are
- how are they planning to deal with failures? with 12k satellites having a MTBF of ~5 years, expect one to fail partly every few minutes and one to experience catastrophic failure every hour to couple of hours - all those number "at best" numbers. is it just going to be a massive armada of what are essentially backup sats?
- all those LEOs will degrade orbit and burn eventually, so i'm guessing periodic reinstallment will be necessary? holy fuck.
Ryan Martinez
actually most of my question already answered in this thread. i'm an idiot, sorry bois
Grayson Thompson
Yeah, the goal is to just saturate the sky with the darn things so that even if a few fail every couple days there will always be at least one in the 40°-40° view that each pizza box has.
Camden Murphy
>MTBF of ~5 years Nope. Planned deorbit prior to failure after 5-7 years, with lifespan set primarily by propellant and technological obsolescence. If they fail on average before the end of their short planned lifespans, it will be a serious shortfall of their goals for the system.
>how are they planning to deal with failures? It's naturally redundant. A failure will just degrade performance until the hole is filled (they'll use hall thrusters, so they can change their own orbits quite a bit to even out the distribution).
So when the satellites run out of propellant, they are just deorbited and replaced? Are they gonna do it in waves? (first in, first out)
Easton Bennett
yes
Leo Peterson
>25ms. LEO! Data travels at 1.0c in a vacuum. 0.7c in fiber-optics. bull-fucking-shit will you get 25 ms with satellite transfer
Julian Brown
you're dumb. It's an extra couple hundred miles up and down, but in return you go line-of-sight to your destination at 1.0c.
The FCC application states:
>"Broadband services: The system will be able to provide broadband service at speeds of up to 1 Gbps per end user. The system’s use of low-Earth orbits will allow it to target latencies of approximately 25-35 ms."
Xavier Ramirez
They'll be more or less obsolete in 5 years when they run out of propellant anyways
Wyatt Reed
basic LEO stationkeeping with Hall effect thrusters doesn't take much propellant.... The deorbiting is more due to FCC regs and wanting to demonstrate the ability to keep the sat #'s high
Jacob Wilson
With the numbers that came out initially the problem was bandwidth. People were circle-jerking over how he could capture a significant fraction of the global market and make bank, but the bandwidth of the constellation was insufficient to provide that many users with fast speeds. I think there is a disconnect between what is possible (i.e. affordable and not shit internet in places where there is no wired connection) and what people have come to think (destroying Comcast and making trillions).
Asher Clark
What happened to Google or Facebook whoever launching high atmosphere weather balloons with receivers on them?
Also, how come Veeky Forums only talks about satellite constellations and sat telephony when Musk's name comes up? You losers are as bad as Reddit...
Jace Long
How about you lurk moar before posting?
Gabriel Garcia
Starlink test satellites deployed! One step closer to a full constellation.
>the application said it's true so it must be correct
Levi Walker
The math checks out though. Why would they lie on an FCC doc? OneWeb claims to be able to achieve similar speeds. It’s not some spacex ubertechnology that makes it possible.
Luis Jenkins
because they want to convince the government to approve the technology
Kevin Robinson
ah, I see. you're just a conspiratard moron.
Ryder Ross
You act as if this has never happened before it has on a few occasions, and fusion is too big a technology to embrace at a moments notice Need more substance to work off of before one can fully believe
Xavier Gray
>fusion wrong thread buddy? this is about starlink.
Jason Foster
yes it is please disregard, wrong thread
John Torres
>tfw I once started a lit thread on k and everyone made fun of me
Matthew Green
that's fucking hilarious
Christian Anderson
>mfw once started a Veeky Forums thread on /k/ and nobody noticed
Dylan Peterson
>mfw people start multiple /x/ threads on Veeky Forums every day
SpaceX should probably plan how to outcompete broadband providers, not Oneweb or Samsung. People who live in bumfuck nowhere aren't known for their high financial prowess and broadband is cheap as fuck and has no data caps (it still does in 3rd world countries, such as the US). Nobody's going to buy a heavy-ass phased array antenna when you can get fiber installed in 15 minutes for free.
Aaron Garcia
Nigga you high? It cost my brother $15,000 to have fiber run the three miles out to his middle of nowhere House. He’s love to just buy a $200 pizza box instead.
Benjamin Scott
RF Network Engineering Faggot here with specialty in RF design and deployment (Currently working with a WISP)
All the physics checks out on this thing, but it would seem a lot more potent and effective to deploy other wireless ground based solutions.
I mean, consider the cost per unit area. It would be so immense that, unless they personally use it for something useful (IE Tesla fleet always connected everywhere and Added reliability of their drone capture ships) and sell subscriptions for end users as a benefit + somewhat offsetting cost, then it wouldn't make sense. What is there to gain from connecting a region like Greenland that currently has 0.03 humans per square km?
To extrapolate - it's likely they're trying to sell it as high-bandwidth internet for the masses everywhere so they can gather investors to fund the project so the system can be inevitably used for a more reliable experience for the other Musk products. But even then I think there's better ways of doing it. Like what's being done currently is utilize the "IoT" bands of RF propagation..
I'm a big fan of Musk and what he's doing. I really would love to be apart of this project, desu.
Dominic Anderson
t. retard
Aaron Hill
They have mentioned working with local isps and serving as the backbone instead of a direct connection. It could be done that way with cell towers. I also think the coverage wont be as dense in the higher latitudes, so places like greenland won't be a huge waste. Really, alot of satelites won't be communicating directly since theyll be over oceans.
Dylan Turner
>I mean, consider the cost >I have no idea what the costs are, but I want to seem thoughtful
>cost per unit area >I didn't even notice that if the costs are remotely reasonable, since this thing covers the entire Earth, the cost per unit area must be ridiculously low
>Tesla fleet always connected everywhere Maybe they can pull off a mobile ground station, but making it work while the ground station is moving unpredictably is another level of technological challenge, and they've only announced support for stationary ground stations.
The main advantage here for SpaceX is to bypass state and local regulations. They only need to deal with national governments, backbone interface points, and individual customers. On top of that, no part of their investment is at the mercy of any particular national government. When regulations are a bigger obstacle than technical considerations, this is orders of magnitude better where it counts most.
