I'm in East Idaho. Currently my dish angles itself to the north. It rarely moves itself north/south, and slightly moves east/west throughout the day. I've read that right now it locks onto a single satellite, although they're adding multi-satellite support later.
My speeds are inconsistent, and interestingly they start slow (around 60 Mbps) but after a couple seconds they'll get to 150-200 Mbps (which is awesome for downloads). Latency is consistently in the low 30ms. I get some downtime every day, so it really is a "beta" like they say. I have a backup WISP.
Setup was literally take dish out of the box, insert into tripod (included), plug in cables, connect to the wireless routers SSID and activate with the starlink app. After that I put the included router into storage and plugged in my Protectli[1] running CentOS. Everything works great. My only complaint is the CGNAT, but given the difficulty associated with procuring IPv4 addresses, it's understandable.
This is OT but can you share a little bit more about what it's like living in East Idaho? I grew up in Boise and now live in Chicago but with the way things have been going lately I've been considering moving back, and there's a real appeal to living in a smaller town in the eastern part of the state (my dad was from Twin Falls). Internet access is always a bit of a concern though, which maybe Starlink ameliorates? Sorry to be so off-topic but you're the first person I've seen on HN from eastern Idaho!
Sure! I love living here. The only thing is the airports are small. You have to drive to SLC or Boise for good selection of flights. Other than that though, I love it. Population density is much nicer. All the stores and restaurants you want, but traffic isn't too bad.
There's a lot of fiber out here, but you'll need to either be near a city center of in a new enough area. The homes that are 10 to 15 years old are underserved and you're mostly stuck with a wireless ISP. But ... Starlink is about to negate that in my opinion!
My family and I spend a lot of time in the northern Idaho handle near Coeur d'Alene. We love it up here. My wife is Asian (matters to the story), we are liberals, but the folks are generally* terrific and kind. I star “generally” because I would say that the average person is much, much nicer than the folks we meet on the west coast, but the not so great people are much more open with racist, hateful, and frankly scary confrontations. We are invited into peoples homes, have made fast friends with many locals, love the pure beauty of the place, but the lows are much lower when they happen.
I agree completely. There are tons of wonderful, tolerant people here. There are a few gross people too that tend to be a lot more gross than on the west coast, but overall it's not too bad.
FWIW I lived and worked in Chicago for 10 years. Never liked it. Expensive, and life just seemed really complicated. I always felt stressed out. If you grew up in a rural or small town lifestyle, it's a big adjustment.
Eventually moved back to my hometown and wouldn't care if I never saw Chicago again.
I know a few people from the Seattle area who do part time work there - unless you enjoy and agree with religious fundamentalists, die-hard MAGA anti maskers or people whose personality revolves around owning 35 guns, you might want to spend some time there first... Boise by comparison is much more secular and liberal than some of those areas.
As you specify people from Seattle, and these people went out of their way to point out that the people are religious, republicans who like guns, I have a feeling they aren't particularly open minded or tolerant people. Religious or secular, blue team or red team, basic politics shouldn't color your ability to get along with people. Fundamentally people are pretty similar, and American culture is quite homogeneous.
Well, particularly in the current situation where masks somehow became a political hill to die on, I think you are underselling the matter some.
It's a bit difficult to feel all that neighborly or friendly with people adamant that they won't even wear a mask to help protect your own health.
And all the topics listed - religion, political affiliation, guns - are all infamous for causing strife. Even the most tolerant person can easily wish to simply minimize the chances of a conflict. Tensions over these matters also only seem to be intensifying, which further exacerbates the matter.
While anti-maskers are definitely more likely to be politically right, due to weird behavior of the last administration, the reality is, the majority of Americans are fine wearing masks in public. Polling i have seen showed 75%+ were fine with masks.
I think that likely shows the problem, rather than anything else.
(I'm going to assume you mean that number nationwide)
75%, or roughly so, nationwide would be one thing if it was evenly distributed. But as you noted, the issue is tied to politics (I think more than you imply, but doesn't matter). We also know political affiliation is not evenly distributed.
So, a number like 75% mask compliance makes it possible, and frankly very likely, that there are areas with substantially lower rates.
Speaking from experience or speculation? I can't stand how HN has started to turn into a political soapbox. The amount of assumptions being propagated makes me understand the social tension the US has.
How did a Starlink post devolve into this chain about rural ideologies?
I would encourage you to listen to, and believe, the very real experiences of people of color, same-sex couples, and transgender or nonbinary people who have spent time in locations like Eastern Idaho.
I don't disagree, but I would caution heavily about stereotyping an entire geographic area based on anecdotal experiences. Different neighborhoods around here can be radically different. One street up is "the Mormon street" where everyone is LDS, but a few streets past that is the polar opposite.
I think when we blanket stereotype people based on geographic location, the error rate is quite high. In my opinion, unacceptably, but then I've always value tolerance, acceptance, and open mindedness (although for people that don't offer that to others I admit little patience, and I don't want to be around them).
> Different neighborhoods around here can be radically different.
Having lived in places with similar ideological demographics, I caution that living in a place means you're going to visit all sorts of neighborhoods there, and it would be (not saying you're recommending this) untenable to recommend that certain people would do just fine if they keep to the neighborhoods where they are welcome.
If a particular neighborhood in an area is hostile to certain people for no good reason, it makes the entire area hostile because all of the people involved can and do operate outside of the geofence.
Eh. They aren't wrong per se, but there are a lot of folks that might look like the fit the picture, but are just from a different cultural background. A little tolerance goes a long way in any direction.
City slickers should live in the city. Moving to a rural area where people want freedom because your 500k a year job lets you live anywhere and then crying my god there is someone not double masking! is not a solution.
Here I am trying to read about satellite internet and you have to pollute the discussion with toxic partisan cliches and irrelevant divisive anecdotes.
I wish you’d quarantine yourself to social networks like Reddit where the discourses are already sufficiently poisoned.
The person asked "what's it like living in east idaho" - and I answered based on some very detailed descriptions provided to me by people who've spent a lot of time there. I consider it an unavoidable fact that unless you are an ideological match with the majority of the persons who already live there, it's going to be a very miserable experience. Some of the social attributes that I described are statistically documented for that region in poll data.
Considering that probably 40% of your hn comments seem to be some variety of partisan axe grinding on American-specific political issues, maybe you might want to take a look in the mirror first.
We actually launched a service that provides you with a public IPv4 address and a /56 IPv6 block over WireGuard. It bypasses CGNAT and provides unblocked ports, including port 25. The cheapest plan starts at $8/month. Here's a link!
We are a managed service meaning you can sign up, download the WireGuard configuration, and forget about it. For a few extra dollars, you get reliability (we use BGP so we can failover to different datacenters), and clean IP addresses that aren't associated with spam or other cloud providers.
If you just want public IPv6 for your LAN, you could get address space for free from TunnelBroker.net and the setup for your router is well documented.
We lease IPv4 /24 and IPv6 /48 blocks from a reputable IP address broker, check it against spam blacklists, and finally announce the clean IP addresses at a datacenter of our choice. We are not tied down to any hosting company. We tested a datacenter failover scenario and switched the BGP route from Chicago to Toronto, while a YouTube video was playing, and it didn't even stutter!
google 'residential proxies for sale' if you really want to do this, the big content sources are getting pretty good at distinguishing between last mile ISP blocks of dhcp customers and IP blocks that are announced in a colocation/datacenter environment.
For most services that distinguish between residential and commercial IPs, your traffic is outbound so CGNAT shouldn't be much of a problem to route normally.
I am curious about your wire-guard setup. Have you been able to automate the setup or is it still a manual process to set up a new customer connection? I have been waiting for wg-dynamic to be ready because I dont feel like using static assignments. It looks like it could be a while until wg-dynamic is ready. What has you experience been?
Yes, it is fully automated. I don't think wg-dynamic will be ready in the next few years, or even remotely cross-platform when compared to the current wireguard implementation and popularity. Our code for choosing an IP address from our pool is really short, just a quick SQL query and a round robin selection.
According to the story it's locked to a designated "cell" with a diameter of something less than 60 miles, so it would be just a nice modern end-table in China. The fact that it depends on accurate GPS and phones home continuously makes the geofence hard to hack.
FWIW, from my limited experience with the firewall there--they do less banning protocols, and more banning specific endpoints people are using to bypass the firewall.
I never tested WireGuard, but, for instance, I had no issue connecting via OpenVPN to personal and work VPNs from China (on standard ports with no effort to hide). If you try and use OpenVPN to connect to any of the well-known VPN providers with it though, it will be blocked.
Even if you find it works initially, as soon as someone catches wind of it being used to bypass the firewall and/or they see a surge of traffic to your services, you're likely to end up blackholed.
Especially for a paid service, it's probably more of a support headache than it's worth.
Unless you're looking to make a stand, I'd probably just steer clear of the whole situation. It's illegal in China to operate an internet services without a license and it could come back to bite you in the ass down the road if you ever intend to do business there or travel there.
Oh 60 is fast around here too. I mostly meant that I've never seen a download start slow and progressively get faster. If anything it usually bursts at the begihning and then throttles back (comcast did that to me the most)
I've observed download starting slow and getting faster on all connection types: dialup, DSL, fibre, mobile wireless from 2g to 4g. It's just how TCP operates. Interesting that your experience on Comcast is different.
It's usually quite common over e.g. 4g. You first have to wake up the radio and attach to a cell tower, which can take 50ms or so. Once you've done that latency for subsequent packets can be much lower (e.g. 5ms).
They provided underlying connectivity to the premises at my prior built-to-rent apartment building in zone 2. The building had its own routers, Ethernet and WiFi infrastructure, and the management of everything was contracted out to an IT consultancy.
200 Mbps to each apartment was offered so I assume they had fibre and not DOCSIS.
I imagine the issue in zone 1 is having to tear up the roads to get cable installed. In new builds you have to do that anyway, so it's good business to get in on the action.
We’ve got that. We were on 200 and it wouldn’t go above 60. Since they finished the upgrade for docsis 3.1 and we switched to 1gbps it’s been what we wanted 200 to be - reliable and invisible. I hate them with a passion, but hate the UK gov more for its incompetent handling of fibre incentives and coordination nationally.
FTTH seems to be fairly prevalent in new builds in the UK now; my house is 5 years old and FTTH is the only option here. This does rather limit your ISP choices somewhat, but you can get pretty decent speeds. BT upgraded me to 950/140 just last week.
FTTH with asynchronous speeds? Thats bullshit. The only reason for that is that they use cable or dsl somewhere in the path from you into their core network
It is done for market segmentation purposes, not because there is a technical requirement for it. Asymmetric vs symmetric is used to separate residential from business customers, and charge the later category higher prices.
If they offer symmetric speeds to home users, they worry that business customers will try to switch to residential plans in order to save money.
> If they offer symmetric speeds to home users, they worry that business customers will try to switch to residential plans in order to save money.
If their key differentiator between the two connection types are upload speeds then they should adjust the pricing and/or clarify (or actually add some) the value add for the business line. That is just unfortunate
As another user commented, it's about market segmentation and not equipment limitations. I can get near-symmetric speeds from other FTTH ISPs but it was significantly more expensive than I was willing to pay, and I don't really need it anyway.
Besides Starlink, I'm curious to hear more about your Protectli. Why do you like it so much? What can it do, that you can't do on your own computer with some scripting and such?
Not OP, but the Protectli is just hardware and you install your own OS. So it is running your own computer with "scripting and such", just a separate low-power, small footprint computer with dual NICs. This is good for managing network for your entire house/office/remote site.
I've got some PC-based firewalls like the Protecli. Mini-ITX, atom-based, 8gb ram. I can often find these for <$50 on auction sites. If it's one I'll be the main "owner" of, I like to run VyOS for firewall and routing. This is the open-source fork of Vyatta, and Ubiquiti's EdgeOS is a commercial sibling (granted, EdgeOS has or at least had some advantages over IPv6 PD). VyOS is debian arm based, so lots of packages like ZeroTier VPN can be added easily. I like VyOS/EdgeOS because of the full CLI/scriptable config.
I recently setup 3 of these for a radio club. These will live in mountaintop tower locations and provide VPN+NAT. Since these might get modified by others, I went with OPNSense. OPNSense is a fork of PFSense with a nice Web UI and community support.
not to quibble too much but I think the bulk of vyos development takes place on x86-64 and the standard build-your-own-install ISO guide is for x86-64. Though it certainly can be compiled for various types of arm CPU.
this is because if you want things like eight or ten 10GbE interfaces, or eight 10GbE and four 100GbE in one server, the only economically viable and fully stable platform for vyos right now is x86-64.
Like the other respondent said it's just a small general purpose computer: you still need to load an OS like Linux. Mine is currently a PFSense firewall mostly out of laziness and a desire for speed. If I had to do it again I'd buy something else because the model I have (FW4B) struggles with gigabit internet speeds and the next model up is too expensive to consider vs just buying a small form factor computer instead.
I love that it runs vanilla Linux/BSDs and is a complete white box. I'm able to configure it exactly the way I want it, and I can even run non-trivial services on it. These days I mostly just port forward as I like to keep things as separate as possible so that when one thing goes down I don't lose everything, but it's great to have options.
Especially while I've got two ISPs, the ability to bond and use both at the same time is super neat.
"no caps" - but watch what happens if you move 200 or 300GB downstream on a LTE based last mile residential connection in a month. you'll either get shut off, throttled to a few Mbps, or billed extra. Look at the fine print in the terms of service. If you have a household with multiple people that watch netflix, download movies, or even want to download a single xbox one or PS4/PS5 game (they can be 120-180GB now), watch out.
T-mobile doesn’t apply a traditional throttle but instead they deprioritize you on the towers. So as long as you're on a tower that doesn't typically get congested (typical in residential areas since people use their home Wi-Fi instead) you're fine.
A whole lot better than the quotas and caps enforced on consumer grade geostationary satellite stuff right now (hughesnet, viasat) that is available in the same price tier of $85-130/month. And also better than 90% of WISPs.
Relative to the helpful details on LTE you posted, though? (Obviously Starlink is incomparably better than the status quo satellite options...)
Why do you expect better than LTE?
And would you say more about WISP behavior? I know WISPs are all over the place, but in my small experience, a technically competent WISP will not look at your usage unless there’s contention impacting other customers.
Like, Starlink has said they won’t service urban areas, to prevent degradation due to contention. So I expect them to use the usual TOS and technical controls... to prevent contention. I don’t see what makes them special here. If they had some special sauce to provide more cumulative bandwidth to subscribers than LTE and WISPs, I’d expect them to open up to urban areas stuck with Comcast, and profit massively.
Replying to your reply, I don’t buy it. Your experience now on Starlink is analogous to being the first WISP subscriber on a newly deployed AP. Time will tell anyway. (And I will still be a grateful Starlink customer at my off-grid cabin, even with the throttling and caps. Starlink is a game changer!)
My perspective is informed by having been in locations previously dependent on geostationary based satellite services, as a comparison.
For people who have budgeted, procured and installed 'serious' two way geostationary stuff in the past (one example of which would be a 2.4m two port linear compact cassegrain antenna, NJR PLL LNB, a 40W BUC, a Comtech CDM760 modem and a 1U sized Cisco router), for 1:1 SCPC dedicated transponder capacity based services, starlink is atonishingly fast.
I could pull out a check book and spend $45,000 on buying terminal hardware and $30,000 a month in transponder space and not be able to achieve the speeds that starlink can do right now. Even if starlink was only 20 Mbps down and 4 Mbps up, go price what 20x4 service will cost by traditional geostationary right now (hint: start looking at $1200 per Mbps per month and multiply by N number of Mbps).
It is indeed a good theory that I'm seeing unreasonably higher than normal speeds right now and better latency, jitter and packet loss because I'm in a similar situation to being the first customer on a new WISP PtMP AP sector. But I also have a great deal more confidence that spacex's continued paces of launches and satellite deployment will keep up with providing at least a 100 Mbps down x 15 Mbps up service. I do not think that they will let it degrade into a contended-service-hell where customers see a very poor end user experience.
My perspective on starlink is also informed by knowing the price right now for Inmarsat and Iridium based offshore and aviation data services (sub 2 Mbps) and the $ per megabyte costs. There's already starlink aviation terminals in beta, and terminals for maritime and offshore use. It'll be a game changer there. The market for a globe-covering LEO high throughput satellite network is much larger than just the US48 state consumer residential internet/small business last mile internet market.
Not the person you replied to, but off-grid doesn't mean no power. I have an off-grid cabin with enough solar capacity to run my laptop, as well as quite a few other things. I'm looking forward to getting Starlink there myself. It's a fantastic, quiet place to work with no opportunity for interruptions.
photovoltaic power systems are a lot less costly than they used to be - if you're fully off grid and won't be connecting to it, having absolute fidelity to electrical code might not be a high priority. If you have some technical ability and can follow documentation and instructions, big 360W panels are like $0.45/watt now, and the setup process for a charge controller and battery bank isn't excessively difficult.
Typical WISP last mile: Something like a ubiquiti rocket 5ac gen2 on an RF elements 60 degree horn antenna, in a 40 MHz TDD channel somewhere in the 5.x GHz band, aggregate capacity of the entire AP might be around 240 Mbps. Shared between 20 to 30 customers. I'm regularly seeing 250-330 Mbps down on Starlink right now with beta equipment, and very, very few WISPs except those who are doing 60 GHz based PtMP micro-POP setups can match that. It's a real challenge for a WISP to have a few dozen houses all trying to download the latest 180GB Call of Duty update connected to one AP.
I'm not nearly as optimistic about WISPs in the long run, compared to my views 8-10 years ago. Really difficult to reach locations will go to starlink or similar (as a replacement for consumer grade geostationary), other places where the customers per square km density is sufficient will eventually get overbuilt with GPON last mile that provides vastly more throughput and capacity.
The most clueful and forward thinking WISPs I know are all making every effort possible, within whatever capital resources are available to them, to develop in house capacity for doing rural aerial FTTH. Buying bucket trucks, getting training, learning how to splice fiber, design GPON architecture, working with state PUCs for pole access, etc.
WISP and Musk are ruled by the same Shannon, so I'm skeptical that there will be a significant advantage over the long term, but satellite has a big advantage it terms of coverage -- the difference between "can you see that tiny hilltop in the distance over the trees?" and "can you see the sky?"
Same Shannon limit, but very different bands and channel sizes as well. Most WISPs are limited by the unlicensed frequency bands that exist in FCC part 15, and things like 3.5 and 3.65 GHz. One of the things that can go wrong with that is many WISPs in the same area fighting over the same bands from 5100-5900 MHz.
By comparison LTE fixed last mile services in some places (where expensive spectrum is owned by entities like tmobile) have some of the prime tree-penetrating frequencies in the 600 and 700 MHz bands, and 2.5 GHz band. One of the reasons why clearwire was acquired by sprint was for their 2500 band.
I can't because the nearest city to me with 5G is several hours away. The 4G at my house is super spotty. I've used it for Webex calls when my home internet went down and it will mostly handle that, but I would never want to rely on it for my primary connection.
Have you tried T-Mobile? I don't know what they have configured, but they certainly own a lot of midband 2.5mhz spectrum in east idaho. They have 140mhz in Idaho Falls and Twin Falls, and 180mhz in Pocatello, and at least 60mhz everywhere else in east idaho. It looks like they have a decent amount of sites too. I'd have to dig a bit more but if you don't have access to 5g on T-Mobile today, it's not gonna be more than a couple months away.
I have the Tmobile service. I use at least 500 Gb/mo without problems. Others on the reddit forum use over a Tb without problems. The terms of service say Tmobile will deprioritize us when there is congestion, but I haven't experienced it. Also, it appears that Tmobile is limiting the number of subscribers in a given area to prevent degraded service. Overall, I like it (full disclosure: I have a dual Wan setup with xfinity being the other provider)
Internet connection speeds, in context. Latencies faster than milliseconds are not found in home internet connections (or anywhere maybe?) and 30 second latencies in Starlink would have prevented the product from even existing or would imply faulty hardware.
what is clearer about implying the 30 posseses something? the post said 'low 30s', so to me that would imple 30-35, or even 30-33 if one wanted to call 34-36 mid 30s and 37-39 high 30s.
I do think you need the app and a phone for at least 5 minutes. I did the initial power up and setup with that, and after 5 minutes of verifying it worked, replaced the starlink provided router with my own. Anything that is an ordinary 1000BaseT 1500mtu DHCP client will get an address when connected to the PoE injector.
The weird angled router they had out is just a convenience for non technical consumers who want an all in one 802.11ac box. The app on the phone also does the very basic first time setup step of defining an SSID and WPA2-PSK key.
I did the same thing and got the same redirect. I haven't experimented with anything except GET / but it would be interesting to try throwing some params in there and trying other paths.
Yeah, after another week or so to get my initial impressions with the router as-is, I'm going to do some more experiments and also swap out a couple other routers to see what I can do.
Sweet man, would you be interested in teaming up a bit? I'm fairly busy the next couple of weeks but I can find a little time to do some hacking. I'm FreedomBen on Keybase, or if you want to email me freedomben <at> proton mail dot com I can give you a real email address. Totally fine if not though!
It's not the official place where I'm working on this particular project, but if you have any notes or feedback you'd want to track/share through my internet-monitoring project, please feel free! Email is a bit tough, as the volumes right now mean I sometimes see a message quickly, other times after days or weeks :)
Hows the gimble, I read its auto stabilizing is that true? And how well built is the dish, do you think it can handle long periods of constant wind exposure of 15+ knots
I've been contemplating putting one on my boat for use while at anchor. There is constant movement but its horrible
It's not a gimbal, it's a set of stepper motors and gears. It's not designed for constant movement or tracking. The current starlink terminal is a dual beamforming phased array that will align itself to have its flat face oriented towards the area above you that has the highest simultaneous density of satellites at any given point in time. Beta terminals in north WA state, for example, are angled about 10 degrees off flat, looking slightly north.
In fact a current starlink terminal (which has a 6-axis sensor and GPS receiver built into it) will turn itself off if it detects movement. The terminals for things like yachts are not available to the public yet, though I have no doubt they're in the works.
The dish basically locks into an orientation after it links up with the satellites, so it's not really a "gimbal" in that it's not constantly moving around.
The dish is heavy and feels tough; I'd be more worried about your mount than the dish itself with regard to wind; we're having 40 mph gusts today in St. Louis and dishy's working fine.
I'm more worried about hail, though... hopefully we can avoid the golf ball variety this spring.
Thanks for taking the time to post the files. I actually learned a few things and like the setup. Might give it a shot. I haven’t been happy lately with the business behind pfsense, this might push me to configure my own.
I can answer that I've been monitoring the full setup (dish + router) through a Kill-A-Watt for a few days, and it never goes below 94W, and averages a little over 100W.
I don't have numbers for just the starlink stuff since I have other things plugged in to my UPS, but that sounds about right. The power brick is always warm to the touch.
My other ISP also uses PoE to power a wireless dish (line of sight) and uses a little less power but not a ton less.
the power consumed for a WISP PtMP last mile CPE radio is considerably less than starlink. Typical CPE for a ubiquiti, cambium or similar antenna will be 8-12W at the AC wall power side of the PoE injector. Starlink is more like 100W constant.
that's typical on DOCSIS3 cable since they intentionally allocate and bond a much greater number of channels for downstream capacity, to match the usage patterns of hundreds of users in aggregate. The actual amount of RF available for upstream is quite small in a typical configuration.
You can get a modem with 32 down and 8 up channels. 8 up channels can support over 200 Mbps. But Comcast will still limit you to 5-10 Mbps.
Then there's DOCSIS 3.1, which actually supports up to 1 Gbit/s up, but Comcast still only gives you 35 Mbps on their gigabit plan.
IMO it just comes down to Comcast and other cable providers being cheap and not investing in their infrastructure to provide better upload speeds, even though the tech itself is capable of it.
Nearly all cable networks use a low split, cutting off around 42Mhz for upstream bandwidth. Not all bandwidth from 0 to 42Mhz is usable due to external interference, most cable systems are able to get 4 full sized 8Mhz Docsis 3.0 channels into this space and one partial channel of 3 to 4Mhz.
Certain ISPs like Cox have started using OFDMA (Docsis 3.1) upstream channels as it is 50% more efficient than classic Docsis channels and you can operate it closer to spectrum with interference since it can run subchannels at lower modulation
So, how do you use these devices in an SD-WAN active/active configuration? What is your unified VPN solution that lets you use both upstream channels simultaneously?
Yes I SSH all day every day, and it has been totally fine (at least while the service is up. There is daily down time currently). Video calls the same. No issues (when it's up).
That said, when I'm on my own machines or ones that I can install things on, I can't recommend mosh highly enough. I've literally gotten on a plane and had the shell pick up thousands of miles away without missing a beat once the laptop was back online at the hotel.
Not OP but I've setup a continuous reverse SSH from a farm in the South to a Comcast residential service in the Bay area.
SSH works but there's enough latency and other general network variation that makes me think it's not quite good enough an experience to spend a day remotely editting files.
For anything not requiring really low-latency, Starlink absolutely shines. Watching the local news from my childhood farm on the other side of the country via satellite internet feels like the future.
I'm on traditional satellite for one week out of every two - at 600ms RTT. I'm either SSH'ing into hosts over that link, or worse, using a Citrix VDI to access (mostly) SSH terminals at the far end.
It's tolerable but far from enjoyable.
30ms latency would be an utter delight, not just compared to geostationary, but also compared to what we had in the late 20th century in terms of terrestrial connections.
I’m on MS Teams video calls a lot. It’s fine for speed, except sometimes the connection drops for a minute. I assume because a satellite is not overhead at that time. Supposed to improve as more satellites are sent up.
Wow, I'm impressed! I thought it would take more than a decade to implement when it was announced, but it's here and it works. Just amazing.
100/16 Mbps is pretty decent I guess, hopefully it doesn't go down as the number of users goes up. The latency is great imo, 40ms using satellites? I don't think anyone has achieved that before.
A bigger dish would not lower latency but may increase signal strength leading to better throughput. But it's not a simple parabolic antenna. It is the first consumer oriented phased array antenna tracking the satellites as they move across the sky, so that would increase the antenna cost even more.
« A bigger dish would not lower latency but may increase signal strength leading to better throughput. »
A better signal strength would probably lead to better modulations being used, therefore, less transmission and reception time, and a better latency, or am I wrong ? (at least that's the case for Wi-Fi: the better the signal strength, the lower the transmission time so the better the latency).
> Latency in this case is dominated by time-of-flight to the satellite.
Is it though? Starlink orbits at 550 km, time-of-flight from ground to satellite to ground would be only 3.7ms, twice that makes ~20% of the roundtrip latency.
There are more than 2 roundtrips. Any MAC that has to perform time division multiplexing on a shared uplink has to poll all base stations over time to figure out which ones have data to transmit, and how much data is queued. Once the satellite knows how much data the ground station has to transmit, it assigns sufficient timeslots, transmits the assignment and then waits for data to come back. This is very similar to PON networks where upstream is shared, but the difference is sub-1ms latency vs 4-5ms latency. Sadly, this does have unfortunate latency implications for how long web pages take to load and render. Streaming video should, however, work swimmingly.
It’s interesting they’d do TDMA over some flavor of code division multiple access (CDMA).
From what I read it has to do with the fact the antennas are high gain directional antennas and not omnidirectional ones like on cell phones. With cell phones you are kinda walking around in a soup of cell signals all sharing the same spectrum at once... you and hundreds of other people are broadcasting in the same frequencies at the same time and they all tell each other apart because they all use a different “language”; the Wikipedia CDMA article does an excellent job explaining this.
I would think that as more satellites get launched they could use WCDMA and signal from your station could be seen by multiple satellites in orbit much like a cell phone can reach multiple towers.
Writing it out... I bet TDMA is required because the FCC would never grant a block of spectrum where hundreds of thousands of ground stations were using low gain, somewhat omnidirectional antennas to reach a constellation of satellites in space....
It's even more complicated than that. Thanks to MIMO antenna arrays, signals from multiple ground stations can be received and decoded at the same time (MU-MIMO). The advances in radio MACs over the past 20 years is seriously impressive compared to what was considered high tech in the 1990s, and it's all a result of Moore's law making it cheaper to do more math in the same size and power envelope as older semiconductors.
550km is the closest approach. Usually it will be at least sqrt(2) times that (ie at least a slant angle of 45 degrees), maybe 2 times that. Plus the latency from the Gateway to the actual server. And the Gateways can have even greater slant angle to the satellite than the mobile terminals.
I don't think it'll usually be at a slant angle of at least 45°. The beta requires a field of view of 100° after tilting. I can't find the maximum tilt angle, but SpaceX has authorization to transmit only 25° degrees above the horizon, so the maximum slant angle is 65°.
Orbital planes are so close together that it doesn't make much difference. However, now that I've actually calculated it, your sqrt(2) factor seems to be about right for the average distance -- there's too few satellites per plane in the current phase.
In this phase Starlink uses 72 orbital planes, with 22 satellites per plane, so 1440 satellites in total (they're almost there). It orbits at 550km above Earth's surface, so the orbit has radius 6921km, which gives an orbital length of 43486km.
Separation between orbital planes varies depending on your latitude, but assume the worst case, where it is 43486km / 72 / 2 = 302km¹. Thus, the nearest orbital plane is at most 302km / 2 = 151km away from the orbital plane directly overhead. However, since the planes process, on average the nearest orbital plane is only half that, or 76km away from the plane overhead.
Satellites within each plane have a separation of 43486km / 22 = 1976km. Thus, there's always a satellite at most 1976km / 2 = 988km away¹ from any point in each orbital plane, and on average there's a satellite half that away, or 494km.
Adding all this together, the nearest satellite is on average √(550^2 + 76^2 + 494^2) = 743 km away (at the worst latitude).
[EDIT: Actually, that's improper averaging, the correct average is obtained with ∫√(550^2 + x^2 + y^2) dx dy / ∫ dx dy on x=0..151, y=0..988, which yields 777km].
The original plan used 24 planes with 66 satellites, which reduces average distance to 617km. At more favorable latitudes the difference with the current design would be even larger.
[EDIT: This should be 635km.]
¹ This is distance on the surface of the orbital sphere, straight-line distance is a bit less. It probably doesn't make much difference.
Moore’s law was specific to the way IC manufacturing worked. While it does (/did?) help reduce the price of electronic simply because they relied on ICs, the big price decreases you see in products after their early-adopter phase are likely just general economies of scale.
Making small early batches of anything is more expensive per unit than making a ton of them —- regardless of whether it’s cutting edge tech or a plastic chair.
I think in the antenna's case, the primary goal would be to reduce costs for now (while still maintaining reliability), since the things currently cost $1500 per unit to make (SpaceX takes a loss on each new customer initially, for now).
Starlink sats are low earth orbiting (about 250 miles away). The really high latency sats that people used to use were geosynchronous sats that are parked about 30,000 miles away, and the round trip delay between earth, bird, earth.
Its fantastic, can't wait until this is available where I live. Currently paying $200+ a month for 20Mbps from local wireless company.
2 miles North from here there is AT&T fiber and Comcast available, 5 miles South there is Comcast (150Mbps) but I'm in a small community of homes where only options are satellite or fixed wireless.
Yeah, that's how we got most of our Internet access here in Czech Republic, starting ~2005. The monopolistic telecom company was totally incompetent so people banded together, bough some upstream connectivity and the connected themselves to that link via wireless point to point links using commodity WiFi hardware with directional antennas.
Since them most of these networks are still in place, now in the 100 mbit/gbit territory. And while most turned into commercial volcanoes over time, there are still some that work as community organizations to this day.
The latency may well be better than landlines in some cases. For instance, data from an edge cache in your city will be faster than starlink. But data from china? Starlink may win.
I wonder if it could improve gaming/video conferencing with people far away.
Right now, that's not really the case though, as the satellite interlinks aren't operational yet. All your traffic goes through a base station relatively close to you, and continues over "regular" fiber around the world.
I know my Spectrum cable connection gets high 30s range for latency. Starlink is close at low 40s, and as they get more sats up, that latency average may go down.
The cell grouping is interesting. A colleague likes the outdoors, so he's preordered one for his Suzuki Jimny, to mount on it. I wonder if Starlink are considering this use case.
I haven't been able to preorder mine, because we're planning on moving out from the city to a small village next year, but the Starlink website requires a street address.
Our villages are quite primitive, no street names (I think it's cos nobody's thought of it). So, the nearest town where there's street names, is quite far.
I was feeling uneasy about using it as an address, this article sort of cements that concern.
I have 50/50Mbps fiber, but reckon we could still be served by 20 down if needed. Exciting!
Part of the issue is that Starlink cells are going to be very limited in capacity for the foreseeable future: https://www.ispreview.co.uk/index.php/2020/09/analyst-probes.... Cowen suggests that Starlink "should eventually be able to serve 485,000 simultaneous data streams in the USA with 100Mbps speeds or 1.5 million streams with over-subscription." That's in late 2026 or 2027 when Starlink has deployed around 12,000 satellites (they're at about 10% of that now). If a bunch of people decide to bring Starlinks to a popular area, the cell for that are simply won't have the bandwidth to support all those users. Imagine people going to Burning Man with lots of Starlinks or bringing their Starlink when they go on vacation. It's not meant to be a portable WiFi hotspot. I'm guessing that portable use might cost more since Starlink has to assume that you might be taking up capacity in places where bandwidth is more scarce.
One thing I would also point out about Starlink is that they only guarantee that you'll be able to use the $500 dish (plus $50 shipping) for 12 months before being forced to replace it. Starlink is new tech and I'm guessing SpaceX wants to be able to upgrade things without maintaining support for less-efficient, older equipment. I don't expect them to force upgrades on people on a whim, but they do spell out that the $500 dish might not be allowed on their network a year after your purchase. I don't think they want to make customers unhappy, but I think they want to make sure they can upgrade their network without getting sued for not supporting expensive customer-purchased equipment forever.
> they only guarantee that you'll be able to use the $500 dish (plus $50 shipping) for 12 months
I would guess that if they disconnect your dish from the network, they'll give you a new one for free. I know that isn't what their terms and conditions state, but it would be bad business to do anything else.
They'll hope that most users choose to upgrade first (for more speed or other features).
Since the antenna can't be used for anything other than receiving starlink service, it isn't really logical not to consider the cost as "$500 + $100/month forever"
> Imagine people going to Burning Man with lots of Starlinks or bringing their Starlink when they go on vacation.
I deeply hope people aren’t going to those to download OS updates and watch Netflix. Might have some crazies that feel the need to livestream the whole thing, but should be okay if only a handful at a time.
> Imagine people going to Burning Man with lots of Starlinks or bringing their Starlink when they go on vacation.
That is exactly why I would be interested in it. :) Not just Burning Man, but other events. We do special event medical and having some reliable roving internet access like Starlink would be absolutely magical for us.
Can I travel with Starlink, or move it to a different address?
Starlink satellites are scheduled to send internet down to all users within a designated area on the ground. This designated area is referred to as a cell.
Your Starlink is assigned to a single cell. If you move your Starlink outside of its assigned cell, a satellite will not be scheduled to serve your Starlink and you will not receive internet. This is constrained by geometry and is not arbitrary geofencing.
Sounds like some sort of authorization comes down from the satellite, and they don't want to have to push all authorizations from all satellites. Which is odd for a full duplex service. Searching around google suggests a "cell" is roughly a 4.5 mile radius circle, and you probably aren't in the center. So movement would be pretty restricted.
I imagine they'll eventually have more expensive plans available for mobile/roaming applications, but obviously you want a well spread base load to avoid saturating any one area. It sounds from some of the other posts like they also do a bunch of active assignment/handoff stuff as the satellites pass, and that gets a lot more complicated once you're also accounting for moving ground stations.
If only 3 words wasn't a proprietary black box, which demands that you use their service / api and that you are not allowed to reproduce it without their assent. Plus there's a bunch of other drawbacks to w3w if you just search a bit online.
At least the algorithm for plus codes is known and can be reused even if Google decides to drop it in the future.
A system that depends on a functioning proprietary API to resolve coordinates is idiotic. That’s so unreliable that it can’t be used for anything more than an ephemeral exchange. At that point you might as well have a 3 word url shortened link.
I prefer an unambiguous and open way of communicating locations on the surface of the Earth, to a proprietary service from a company with a long history of pulling the plug on products.
Clicking on a map location in Google maps shows a context menu where the first entry is the latitude and longitude, click that and it copies it to the clipboard. No need to use a Google specific encoding.
I'm not sure that using Google Maps to demonstrate copying a single way of encoding coordinates vs using an open standard is a terribly convincing argument.
Having a field that takes lat and lon has all sorts of ways to enter data either incorrectly, or in an unexpected format - if someone has coordinates from some other source they might be typing it in, rather than copy/pasting.
Also, as others have mentioned - plus codes (or Open Location Codes) are an open standard that can be implemented by anyone under an Apache 2.0 license with a whole bunch of example implementations on github[1]
Interesting edge case of no street names (which in retrospect is not an edge case at all given the use case for remote users!). Wonder why they haven’t allowed GPS coordinate input?
they do when you sign up in a geographic location as a "available to ship now" beta customer, after putting in your address, it brings up a view of a satellite/google maps view of your area and asks you to zoom in and click on the map on the precise location where the terminal will be installed.
Aside: The router design is as impractical as it is futuristic. The thing would fall over if you looked at it sideways, and the solitary LED on the front was hard to see unless in a dark room or looking closely, straight at it. Hopefully a 2nd iteration will be better!
I suppose that's a minor nit, given what Starlink delivers, though. I'm curious how practical it might be for on-board aircraft wi-fi. That's a space that could use a leap in bandwidth/tech, as FAA certification makes it difficult to keep equipment current. I'm curious if tracking is hard since the satellites and the "ground station" are both moving around...the aircraft on all axis points.
>Wow, he's right about the shape of the router. It also looks like it forces the ethernet cable in front if you want to be able to see the LED, and the cable itself is pointed downward: https://cdn.hackaday.io/images/8430831598574092393.jpg
I'm not sure the image you link is a fair representation of the router. The one he actually shows is SIGNFICANTLY shorter/smaller/wider and honestly while it's similar it's a different shape.
I have one and it's really light weight and prone to tipping over, it was only used for five minutes during the initial setup. Based on some people who found the serial pins and watched its booloader it's based on a customized openwrt, the circuit board is pretty basic.
it doesn't seem much more complicated to me than the PCB of something like a mikrotik hap ac2 ($65), based on simplicity I'd say it's like a $40-50 router BOM, max.
Acting like the LED not being bright is a /negative/ thing is odd to me. Why would you care if it’s lit up unless you’re investigating an issue and you’re right in front of it anyways? Just seems well designed to me.
I absolutely detest bright LEDs, but in this case, it's a dim LED set back in a recessed hole in the router that is hard to make out at all unless viewing it straight on.
Better this than a blaring blue LED I guess, but it would be nice to just have it easier to see in general. Having a shiny brushed reflective surface to the router (instead of, say, black or pure white) exacerbates the problem.
The router looks 'cool' like the Cybertruck, but could use some refinement physically. The shape also makes it hard to mount on a wall, or even to place it horizontally if the need arises.
Note that one can use any router with the dish. Basically the dish provides you with a NATed private IP which you can assign to the WAN interface of your router of choice and go from there.
It’s a bit more complicated than that if you want reliability. They have two IP ranges on the same physical link, and their patched DHCPD is not standards compliant. If the dish loses its connection, you will be stuck with a useless local IP unless you block DHCP leases in the 192.168 range.
Oh, that's interesting. Is the router then completely just a commodity thing that has no specific Starlink functionality at all?
Edit: Some reddit posts suggests the Starlink App doesn't work if you don't use the router. But also that the app isn't terribly useful outside of the "obstructions view", which is mostly a one-time need during installation.
That's correct. The starlink app only works when using the included router. The app is useful if you like to see numbers and graphs of dropped packets, throughput, etc, but otherwise isn't needed.
> I'm curious how practical it might be for on-board aircraft wi-fi
I suspect it'll work well for this scenario. My understanding is that the cruising speed of a plane is relatively slow compared to all the other motion involved (LEO satellites move fast!). The US Air Force is reportedly working with SpaceX to test Starlink in various conditions including in flight https://www.avweb.com/aviation-news/air-force-testing-starli....
Power consumption could probably be a bit lower, but there are limits. Keep in mind that this is a phased array maintaining a relatively high bandwidth and high SNR link with the satellite. This is complicated microwave electronics design, a whole different ball game than old grandpas bragging about how they made a contact with someone across the ocean with 5W on their homebuilt HF radio. Starlink may have also made a conscious decision to make the user terminals "overpowered" so they can use a less sensitive receiver on the satellite, saving SpaceX weight, power usage, construction costs, and launch costs. The cost of more power usage is paid for by the user with an extra penny per month in their electric bill.
That's assuming that you could get the power consumption down to zero, which is impossible. While a penny was an exaggeration, a 5W power savings over a month is about 50 cents savings on someone's power bill.
For reference, this is an order of magnitude greater than the typical power consumption for existing satellite dishes / 4G modems, which typically operate at around 10W of power consumption.
For reference, most existing satellite dishes are receive only, and 4G modems only have to close link on a tower a few miles away, and at RFs where atmospheric attenuation is much lower.
It's definitely doing... something. I had my Kill-A-Watt in it for a few hours with some usage and lots of idle periods, and between the dish and router, power consumption was never below 94W. Average around 104W, peaks at 124W.
Motors weren't being used during my measurement period either.
The dish gets pretty warm in operation (it was already a warm spring day, so it wasn't trying to de-ice or anything).
The CEO just mentioned first version of the dish cost $3000 to manufacture, and it's currently down to $1500. Major goal is to reduce manufacturing cost, since they're selling the thing for $500 and taking a loss right now.
the closest thing I have ever previously seen to a starlink phased array is the flat panel phased array radar in the nose radome of an air superiority interceptor type aircraft.
It's actually a whole satellite terminal as well as all the electronics for controlling the phased array. It can operate entirely without the router they give you - plug in a computer, or another router (into the white port on the POE injector), and off you go.
100W is actually really low in the class of two way satellite equipment that's capable of 100 x 15 Mbps speeds or greater.
I am not talking about ordinary cheap consumer grade hughesnet or viasat stuff, but if you were to do the power budget for an idirect x3 modem and a traditional geostationary ku-band VSAT setup with a 20W BUC, the actual AC wall power consumed would be quite a lot more. Just the BUC is going to be 200W.
but in real world use, often much, much less. The average screenshot of speed tests I see from unhappy viasat and hughesnet customer is like 2 to 5 Mbps down and 0.9 Mbps up. Maybe at 0300 in the morning.
What I was referring to is the class of VSAT equipment much more costly than a proprietary viasat or hughesnet terminal, which the ordinary consumer will never see, that's capable of doing a dedicated 100/15.
Fair enough, and you seem reasonable based on all your comments. Why is anyone giving any credence to the starlink speed tests? They are maybe 5% of the target user base they'll need to be profitable with this number of satellites, yet again and again, nobody mentions that. The funny thing is that everyone dismisses the common outages as a beta issue, but the speed and latency are never considered a beta feature.
A satellite dish does not transmit any information to the satellite. Satellite TV is a pure broadcast system in the forward direction.
Moreover, comparing a parabolic receiver with a phased array is quite unfair. The amount and complexity of the electronics and processing power required is several orders of magnitude different.
A better comparison might be HughesNet. Not perfect since it's different classes of satellites, but their modem is spec'd at drawing up to 60 watts on their latest gen. Though, in practice, it tends to draw around 15-25 watts, at least according to comments I can find online.
I have Starlink and am in a wooded part of New England. I mostly agree with freedomben's points. All I will add is I still use my cell phone hotspot (Verizon -- only carrier that gets a modicum of service where I am) for Zoom calls or similar. Starlink is definitely fast but too many hiccups for video calls.
Edit: More musings. There is a push to get broadband in our area within the year. If that happens and Starlink's service remains the same, I will cancel Starlink. For me continuous connectivity is much^3 more important than speed.
"Rural" and East Coast is basically the metropolis some other areas of the world. Not surprising you'll have something better than Starlink sooner or later. Lots of other places? Not gonna happen.
I never claimed I was in a "rural" area, just a "wooded" one. Indeed, Starlink will be a great option for many places. I personally hope they improve the coverage with more satellites so I can continue using it.
In my experience, Starlink isn't suitable for many uses right now without some kind of wan bonding(speedify, speedfusion, etc). I currently have two cellular modems + Starlink bonded with some routing rules forcing connections(streaming) directly to Starlink. Since I installed in early March, it has improved significantly and I will probably switch the default routing rules to use Starlink and only use the bonded connection for ssh and similar.
Overall, I am very pleased and looking forward to the continued improvements.
This depends on your current latitude and obstructions - my starlink terminal above 49N is averaging 0.22% packet loss (1/5th of 1%) to stuff in a telecom facility in downtown Seattle. Its overall packet loss over a multi day period is actually better, and jitter is lower, than on a docsis3 cable connection at the same place.
The simultaneous density of satellites and gaps, or lack of gaps at your location will have a big impact on the terminal's availability and packet loss to gateway at the other side of the ground-to-space-to-ground link.
Our Starlink connection (backup WAN) recently switched from a New York ip to one near Chicago. I think we've switched to a downlink station that is geographically closer and we're getting better pings and, possibly, better throughout (needs more testing).
The service is expensive, but it's been a fun way to "participate" with all the exciting stuff SpaceX is working on.
for a very brief period of time my starlink connection was exiting the cgnat to the internet in chicago (the first hop that was a public IP was 1-2ms from various ISPs' looking glasses in chicago), and the latency matched for return trip to LEO/back, down to earth station in the pacific northwest, and then transport circuit to chicago and back. Then after a couple of days it went back to Seattle.
[We are South of the 45th parallel, by the way.] I think service is improving for us; we don't get the regular 5-10 second disconnects that were occurring every 10 or 15 minutes. According to fast.com, our loaded and unloaded latency has improved. Most media consumption (web pages, video, etc) compares favorably to a terrestrial service.
However, Xbox Live (the limit of our online gaming) is still inconsistent. We get disconnected from game servers and from Live itself pretty regularly.
It is definitely much better since mid January with increasing satellite density. Getting the antenna to a really high position on your roof with no tree obstructions around is important. Right now it's averaging about 0.28% loss to Seattle over the past 3 hours. But there are still some very brief periods of 4-5 second hiccups. Based on the monitoring systems I have running, it's more like 0.00% loss for long extended periods of the time now, and then a minute or two with about 10% loss. I haven't tested it a lot for online gaming but realtime voice/video conference type stuff all appears to work normally.
There is presently a very big difference in simultaneous satellite density and lack of gaps at 49N vs 44-45N, however.
I’m in rural NZ and I can’t wait for Starlink to be available here (it says late 2021 for my area)
I have been trying to get the local ISP to install their system for 6 weeks now, their service is just terrible. Installers don’t answer the phone, they make promises they don’t keep, I have the keep hounding them, and the ISP itself has a monopoly position so they are abusive in pricing and customer service because you have no alternative :(
There is nothing more frustrating than having money to pay for a service, the desire to have it, and the company who’s selling just being abusive and totally useless. (Example: before I got involved they were selling my elderly parents 8,000$ in unneeded networking equipment they told them was required)
It really is going to be save the government a fortune trying to get last mile internet to rural NZ, so a win win for everyone.
The only losers are the rural ISPs who everyone complains about. I am sure there are some great ones but the rest are going to disappear as soon as Starlink launches.
If you want to send me an email (address in profile) I would really like to hear how the setup goes when you get it.
When the sat-sat routing happens it will also be a good undersea cable backup Internet option.
I’ve been using Starlink for a few months now. It’s great for the cost.
One annoying thing is their super non-standard and kind of broken network setup if you want to use your own router. They have two subnets with different IP ranges (cgNAT and local) on the same Ethernet link. You have to do some weird manual routes, interface resets, and dhcp client configuration if you want things to work reliably.
The issue you're seeing there, using your own router, is that during a network-unavailability outage, the starlink terminal will continue serving as a DHCP server to whatever you plug into it. When the network is up and you plug a router into it that's a dhcp client, it'll get an IP address in cgnat space and periodically renew it.
When they're doing network maintenance/beta downtime/terminal firmware updates, your router will fail to get an IP address in the cgnat space, but the terminal itself will give the router a new lease in 192.168.100.x (you can ping the antenna at 192.168.100.1), and your router will be happy because it has successfully received a lease and can keep renewing it).
Then when the network unavailability is finished, your router keeps its 192.168.100.whatever address and the terminal doesn't issue it a new cgnat address.
This sort of setup is weird and not normal, so once your default-configuration DHCP client WAN port router is stuck, you need to manually tell it to go get a new lease, or set up your dhcp client in such a way that it always rejects IPs and leases in 192.168.100.x. I can see why they do it, because they want the starlink app on peoples' phones to be able to communicate with the grpc endpoint on the terminal during outages.
But they'll clearly need to do something different with the terminal's firmware and how it treats BYOD router customers, because people will expect to connect a ordinary DHCP client and have it 'just work', and resume working after an outage has cleared.
Using a very short lease time for 192.168.100.0/24 when CGNAT isn't available would be one way SpaceX could deal with that. The DHCP client would then be frequently renewing the lease.
When CGNAT is available the terminal's DHCP server releases 192.168.100.0/24 and issues/proxies the CGNAT lease.
The terminal would be the default route so packets to 192.168.100.1 would still reach it and responses can be correctly routed back to the client using the Starlink monitoring application.
Seems weird. I think they could just always assign local address and route all traffic from CGNAT address to it? It would mask real address from connected device, but as it is CGNAT it isn't very useful anyway.
They've said it's a requirement for the beta. They're limited the number of customers per cell.
If they didn't lock you to a location, folks would say they were going to use it out in the boondocks, and then acutally use it somewhere it's already at the capacity limit they've set.
Not here to disrespect anyone using it or seeking it, just commenting that with many cities worldwide mid way through fiber upgrades I worry this will be used to deprecate in ground investment in optical.
I know a lot of people can legit say "fiber is never coming to me" but the NBN debacle in Australia comes to mind as a politicised capital investment failure (they dropped FTTP for a mixture of copper, optical, wifi GEO sat and coax depending. I got fiber. It's great.)
The UK is another place where national broadband roll-out could freeze on this if the market bifurcated. Or NL with KPN not yet complete on a national fiber deployment.
High Bandwidth and 30ms RTT is awesome. This is stellar tech (sorry)
Lots of US metro areas trapped behind the consequences of the ma bell breakups, RBOC, cable TV and local laws forbidding Google fiber deals. Its the politics which ruined optical not cost really. Starlink shouldn't be such a rational choice for a metro area and it says a lot about capital investment that it is.
Pacific Islands, rural and remote, city fringes with no viable alternatives, this is mega cool. Cities with infrastructure, It's a signal of some major failure this is going to be so popular.
NZ ufb crown holdings, the signal got strong if you were too far away from the big towns it was years off. I can relate to any kiwi wanting this tech. It would suck to be seeing people up the road getting ufb and being off the planning horizon.
Same with my oz mates who aren't on NBN fiber. If you're on the optus GEO sat stuff, this is going to rock.
It’s always been my understanding that Starlink will never be able to offer anything close to what it achieves for rural speeds in a areas where they’d have many densely packed customers (I.e. cities)
A friend here in NZ just installed starlink, has roughly the same experience as others - most of us here have gigabit fibre to the home, he lives just enough out of town that he's replacing marginal VDSL
I really hope StarLink completely replaces CenturyLink DSL. CenturyLink DSL 30mbps DOWN and 5mbps UP, real world speed is 2mbps DOWN and 512mbps UP. Never goes above this speed. CenturyLink DSL is a complete rip off. Costs $45 and they charge for the modem $10. $55/month total I would be willing to pay $99/month if it had at least real world speed of 30mbps DOWN and 5mbps UP ^_^
It's interesting that the post and at least one comment has mentioned intermittent outages. I understand that this is a beta product however can anyone speculate what might be the cause? Is it because there's no yet enough satellites in the constellation or is it that the rotation of the constellation is still be tuned or something else?
The app shows the reasons, typically around 1-2 hours total per day (usually < 1 hour in the four days I've tracked it so far). Most of the time is due to 'obstruction' (when a sat is in field of view of dish, but there's something in the way—in my case a tree covers about 10° of the sky).
After that is 'beta downtime', which is typically 10-20 minutes per day. Then 'no satellites' (see https://satellitemap.space), when there's no satellite within broadcast distance—typically 5-10 minutes per day at my latitude.
Hopefully the latter two metrics go down as time goes on :)
My take is that they're manually scheduling satellite customers/aiming/timeslices and as they get more sophisticated they will allow this.
The ultimate would be a self-contained electric rv with fold out panels - solar for the motors, climate and appliances, and satellite for internet connectivity.
I think an acceptable middle-ground would be the ability to call-it-in change your address. But then there would be people that would get an address in boise, then set it up in NYC.
Metros are not the target market at all, in part because starlink (nor any other satellite internet) cannot serve above a certain density of customers (because there's only so focused you can make radio waves). Even with their planned upgrades and perfectly even distribution of their customers across the continental US, starlink could not serve more than a few percent of households. (Even Musk has said that it is not competition for the big ISPs, but that hasn't stopped a hype machine on the internet for assuming it will be).
That this works on most of the planet. If you can get fibre or good DSL for a reasonable price, take that. But even in the US large parts don't have fibre available, especially in rural locations. Many countries don't have any networking infrastructure. It will be a total game changer once mobile versions (airplanes, ships, yachts) become available.
For some level of density, wires/fibre is usually going to be the better bet. But even though where I live isn't exactly rural, it's definitely not urban or conventional suburban, and cable is still fairly marginal.
The question is: will this stall development of more scalable fibre optic infrastructure and municipal fiber - to support a proprietary single corporation solution?
But on a more serious note, with fiber you can supply a whole street with symmetric gigabit links pretty easily and for what is pretty much a one-time cost of laying the fibers. Upgrading the line to 25 GBit and up is done by only switching the receivers, once that becomes economically viable. Even if the satellites can handle multiple hundred gigabits (they can't), it would be impossible for them to compete with fiber in either bandwidth, price or latency.
When the entire constellation of Starlink satellites is deployed the aggregate bandwidth of all of them will be less than a handful of fiber optic cables.
Starlink is not designed to compete with fiber or copper in areas where you have a high density of internet users over a significant area.
It is for less dense areas, or those few people in areas otherwise well served by fiber or copper that have a house that ended up in a gap between infrastructure build outs.
Fiber is not and probably never will be economical for rural areas. Most of the rural fiber rollouts that exist today are pretty transparent subsidy milking schemes.
I was thinking about buying 40 acres in the middle of nowhere earlier this year. It had no electrical power within a mile, but there was a buried gigabit fiber line running down the dirt road. It probably cost at least hundreds of thousands of dollars to serve fewer than a hundred people. I called the ISP to ask about pricing. It was cheaper to buy a fiber/phone combo than just fiber. Why? “We get more subsidies that way.”
Starlink is clearly a superior option in cases like this.
That this service is designed for customers who don't have other good internet options and are therefore somewhere between willing and _thrilled_ to pay $100 a month for good service.
Exactly. The absolute fastest I can get without Starlink is a flakey WISP with ~20 down (they advertise more, but well, it's a lie). That's why I bought Starlink as soon as I could.
Metro areas aren’t the target. There is a large, underserved market that doesn’t have access to broadband internet, and for those customers this is a breakthrough.
Also while we aren't really used to this being the case with ISP's, lots of products get cheaper from when they are first offered. I think they're pricing a bit higher to start off with to just make sure they don't join the giant heap of bankrupt satellite internet companies.
It’s sad that it is even necessary in the US. Telcos were given massive amounts of money by the government with promises they would do this and then they simply didn’t and said cell phone service was good enough. I would posit that a populace educated on cell phone internet is largely responsible for the rise in intense anti-science stupidity and lack of fact checking we are experiencing. People on phones are too easy to manipulate and sway, the ad giants have mastered it.
Isn't it fantastic? The system of (mostly) free enterprise has created an affordable high speed service product that can be accessed anywhere on the planet.
I lived downtown in a major city and my landlords had an exclusive deal with AT&T so while my neighbors were getting 400 Mb cable I was getting 15 Mb DSL.
You still need fiber to the towers for good connectivity. The amount of infrastructure needed to install even for 4g in remote areas is very expensive compared to these now that spacex can get the satellites up for very cheap as secondary payload with the most of the expense is borne by the primary payloads.
Have you tried using 4G as your only connection? Even carriers that provide “unlimited” transfer typically throttle your connection after a certain amount of usage.
I use 4G as my primary Internet connection in Oslo, Norway. The apartment block where I live has a 100 years old copper line and DSL gives like 10 MBit/s with 100ms ping.
For about 90 USD/month one can get 60 MBit/s connection l both ways with ping bellow 20s. The connection is only throttled after 2 TB of data. But then you need a router and an antenna.
Or one can pay like 15 USD/month extra to upgrade the mobile subscription with 10 MBit/s connection and the throttling limit of 1TB. The ping is about 30.
Have you seen what they do to install just a few blocks of 5G antennas? Ripping up pavement, trenching fiber and power. And that is just in a high density city.
My in laws live In the middle of nowhere. Their option for many years was 3G and eventually 4G. Having to visit them, 4G was just unbearable. Eventually ATT said they could just barely get 1.5Mbps down service.
That was superior to the 4G service they got. Just because it is 4G doesn’t mean it is services well and not overloaded even in rural areas.
Starlink will be the first real viable high speed connection they will have ever been able to get
My dad's house in Maine is at the very end of a DSL connection. He get up to about 1Mbps down with a tailwind--which is barely usable even for non-video. I couldn't really work from the house. (And there's essentially no cell service at all.) And the neighbors further down the road basically have nothing at all other than conventional satellite.)
I doubt they'll price it that way but I'd definitely consider getting it as a backup where I live because my cable Internet is "OK" but definitely a bit glitchy and inconsistent. Especially as I've been pretty much full remote since pre-COVID I'll pay to get more reliable Internet at home.
I mean the question itself is pure ignorance and also shows a kind of lack of care of others. As one other commenter responded "New York City has subways, why do cars exist?" is the kind of question it was.
Because a lot more people these days are moving out to more remote areas and working remotely where it was just not possible in years past. Starlink would open up a lot more parts of the non-urban world to this possibility.
I'm in East Idaho. Currently my dish angles itself to the north. It rarely moves itself north/south, and slightly moves east/west throughout the day. I've read that right now it locks onto a single satellite, although they're adding multi-satellite support later.
My speeds are inconsistent, and interestingly they start slow (around 60 Mbps) but after a couple seconds they'll get to 150-200 Mbps (which is awesome for downloads). Latency is consistently in the low 30ms. I get some downtime every day, so it really is a "beta" like they say. I have a backup WISP.
Setup was literally take dish out of the box, insert into tripod (included), plug in cables, connect to the wireless routers SSID and activate with the starlink app. After that I put the included router into storage and plugged in my Protectli[1] running CentOS. Everything works great. My only complaint is the CGNAT, but given the difficulty associated with procuring IPv4 addresses, it's understandable.
[1] I love this thing. Highly recommend: https://smile.amazon.com/gp/product/B0741F634J/ref=ppx_yo_dt...