Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
The Internet Space

Starlink Beta Testers Are Impressed With Its Speed (arstechnica.com) 54

Posted by EditorDavid from the signal-in-the-sky dept.
One beta tester of SpaceX's new Starlink satellite internet service is a Reddit user named "wandering-coder". This week they shared their experience online, testing the equpiment in a national forest which gets no cell service from any carrier -- and using it to upload this report: Works beautifully. I did a realtime video call and some tests. My power supply is max 300w, and the drain for the whole system while active was around 116w... It didn't work well with a heavy tree canopy / trees directly in the line of sight, as expected. I would be connected only for about 5 seconds at a time. Make sure you have as clear a view of the sky as possible!
Ars Technica shares more data: As revealed last week, the Starlink beta costs $99 a month plus $499 upfront for the user terminal, mounting tripod, and router... New speed-test data collected by Ookla and published by PCMag last week found average Starlink download speeds of 79.5Mbps and average upload speeds of 13.8Mbps in October, when the service was in a more limited beta. The same data found average download speeds of 24.75Mbps for Viasat's Exede service and 19.84Mbps for HughesNet, both of which offer service from geostationary satellites. Upload speeds for Viasat and HughesNet were 3.25Mbps and 2.64Mbps, respectively. Starlink's low-Earth orbit satellites greatly outperformed the higher-orbit satellites on latency, with Starlink posting a 42ms average. Viasat and HughesNet came in at 643ms and 728ms, respectively, according to PCMag.

SpaceX's invitations to new users last week told them to expect "data speeds vary[ing] from 50Mbps to 150Mbps and latency from 20ms to 40ms over the next several months as we enhance the Starlink system... For latency, we expect to achieve 16ms to 19ms by summer 2021."

One Montana resident posted a speed test result with a 174Mbps download speed, 33Mbps upload speed, and 39ms latency. "Way out in rural Montana where our alternative is to pay by the gig. Starlink will forever change the game," the beta tester wrote on Reddit yesterday. The Starlink Reddit community has several resources for tracking beta progress, such as a list of user speed tests and a list of states where at least some people received beta invitations.
Engadget points out that Friday SpaceX also won approval to sell Starlink access to Canadians.

And TechSpot shares more reactions from beta testers: According to some early speed tests done in different locations around the US, users are getting anywhere from 100 to 203 Mbps on downlink and around 15 to 33 Mbps on uplink. Meanwhile, the latency varies between 20 to 45 milliseconds, which is pretty much in line with SpaceX estimates.

Upload speeds are still lower than the expected 50 to 150 Mbps, and one user in Idaho reported that connection drops every 2 to 3 minutes in games and video calls are common. One possible explanation for this could be that Starlink's constellation currently has around 800 satellites, which is a low number compared to the target of 12,000 that SpaceX wants to build.
"Of particular interest has been performance in adverse weather," writes long-time Slashdot reader Rei: While the network is still being deployed, Pacific Northwest have been reporting rainy-weather download speeds ranging from ~95-140Mbps, upload from 9-18mbps, and ping times from 32-34ms. The upper range is surprisingly not that different from other November clear-weather data (which, while dramatically higher than September reports, is still supposed to improve over the coming year). The tolerance to adverse weather is likely due multiple satellite paths as well as phased array tuning overcoming wind buffeting. SpaceX does plan to add additional higher frequency V-band transmission to future Starlink satellites, but this should suffer more bad-weather attenuation than the current Ka/Ku-band.
He also notes that SpaceX engineers have a nickname for the hardware. "Dishy McFlatface."
This discussion has been archived. No new comments can be posted.

Starlink Beta Testers Are Impressed With Its Speed More

Starlink Beta Testers Are Impressed With Its Speed

Comments Filter:

  • But is it though? (Score:5, Insightful)

    by VicVegas ( 990077 ) on Sunday November 08, 2020 @08:36AM (#60698978) Homepage

    A service with a handful of users had better be fast! Once they start piling on customers, I'm sure all the speed reports won't be so rosy.

    • Re: (Score:2, Funny)

      by Anonymous Coward

      You're right, StarLink was a giant waste of time. We should destroy the satellites, and announce an extra $100B worth of subsidies for telcos.

    • Re:But is it though? (Score:4, Informative)

      by Ritz_Just_Ritz ( 883997 ) on Sunday November 08, 2020 @08:53AM (#60699016)

      And you know this because???

      Frankly, all it really has to do is outperform the incumbent satellite providers to be successful. That's a pretty low bar. Hopefully, this makes it that much more possible for me to continue making a living in IT while not being forced to live near any urban sprawl.

      For people with no access to terrestrial options, this will be a game changer.

      Best,

      • Outperforming current satellite for a few users was never going to be a problem as Starlink uses newer equipment. Scalability is a valid concern.

        • It's also a problem for the remaining "radio quiet zones" on Earth. People in zones where radio traffic is forbidden for military intelligence monitoring, or for radio astronomy, can now activate such receivers and transmitters and interfere with years of data gathering. It does make me wonder if StarLink is able or willing to deny service to such areas.

          • presumably if the satellites are geosynchronous, they might be prohibited from having a satellite near such regions?

            • There is _zero_ chance of that for multiple reasons. "Geo-synchronous" satellites are locked above specific locations only for equatorial orbits, or possibly if they are under continuous thrust. There are some interesting solar-sail based orbits that could provide continuous land-locked coverage, but those do not currently exist. geo-synchronous orbits normally "bobble", ranging from northern to southern latitudes in Earth orbit over a 24 hour cycle unless _very_ carefully aligned in equatorial orbit.

              StarLi

              • More importantly, the shorter signal path for Starlink brings latency down to a reasonable value, and means less traffic funneling through each satellite. A geosynchronous satellite has to have throughput to handle at least an entire country's traffic at once.

                • If I may say, throughput is not latency. Latency is important for Internet traffic of many kinds: and geosynchronous orbit would indeed add roughly 100 msec of unwelcome latency. at the approximately 22,000 miles above the equator for geosynchronous orbit. But it's not the same thing as throughput, let's not confuse them.

                  • Yes, my point is that these are two different measures, both of which cut against geosynchronous satellites and in favor of Starlink's approach. A single Starlink sat only has to handle a few channels of traffic at once, and is so much closer to Earth that it can offer cable-like latency.

              • StarLink is entirely low-earth-orbit, which is _vastly_ cheaper. Geo-synchronous orbit typically costs at least 5 times as much. It involves a _lot_ more fuel, on more powerful, less frequently available launch vehicles.

                I'm not so sure about that. A GEO satellite costs around $500-$600 MM including construction, launch, and insurance to cover an area like the US, whereas Starlink has so far cost around $800 MM (about $1 MM per sat) and still cannot provide continuous coverage over the US. By the time that happens, it'll have cost in the neighborhood of $1.4-$1.5 B.

                Then, you've got to factor in that each SL terminal currently prices for around $500, and that's thought to be the "aspirational" price, not the current cost to

        • Newer equipment and techniques contribute, but Starlink was going to outperform incumbent providers even by using older equipment. The round trip time alone is massively reduced just by the nature of where the satellites are.

          All else being equal, Starlink signed the death certificates for all current satellite Internet providers with the successful launch and deployment of TinTin A and TinTin B back in 2018. The rest of the FEO satellite providers are on borrowed time.

      • Re:But is it though? (Score:4, Informative)

        by SkonkersBeDonkers ( 6780818 ) on Sunday November 08, 2020 @09:12AM (#60699048)

        Hopefully, this makes it that much more possible for me to continue making a living in IT while not being forced to live near any urban sprawl.

        For people with no access to terrestrial options, this will be a game changer.

        Best,

        It's not even just living in (sub)urban sprawl vs the middle of nowhere. There are many places where even being 15 minutes outside town can mean little or no broadband service. Mountainous areas in particular. A friend was looking at buying and building in upstate NY. A property he particularly liked was only about 1/4 mile from the nearest fiber and the teleco told him it would be $5,000 to connect.

        • Many Americans only have 1 choice in high speed data. Soon they will have 2. Competition pushes down prices and improves service.

        • Re: (Score:2)

          by SeaFox ( 739806 )

          Given the cost of the house itself, $5k more to add fiber internet doesn't seem like a huge issue, and it will increase the value of the home immediately for when he sells it later.

      • Re: (Score:2)

        by Zocalo ( 252965 )
        That's a pretty low bar, but I also agree that it is absolutely all that Starlink needs to do in order to be successful. Assuming the production service is reliable enough to have an acceptable SLA in place, has some form of QoS, and can be segregated into private networks similar to MPLS, then I have clients at $dayjob that could pay 10x that and *still* cut their opex costs for remote telemetry and SCADA significantly. Based on preliminary conversations and the early beta-testing numbers it's absolutely

      • Re: (Score:2)

        by cusco ( 717999 )

        We have a vacation house in Peru, 30 miles and 20 years from Cusco. Our two options for connectivity are horrendously expensive high-latency Hughs Satellite or a seriously overtaxed 3G cell tower. Starlink will allow me to semi-retire to Peru, doing occasional contract work to keep from getting bored.

  • Next generation hardware will be better (Score:5, Interesting)

    by Dan East ( 318230 ) on Sunday November 08, 2020 @08:45AM (#60698996) Journal

    It didn't work well with a heavy tree canopy / trees directly in the line of sight, as expected. I would be connected only for about 5 seconds at a time.

    Back in the 90s I had a Garmin handheld GPS receiver - you know, back before smartphones. That thing really didn't work at all in the woods (which, considering I used it for hiking and geocaching, was a problem). Around that time they came out with new (and more expensive) "S" models, which had much greater sensitivity (hence the "S"). They worked perfectly fine in the woods - as do all my modern GPS devices now (I do motorcycle woods racing, and my GoPro GPS tracks perfect in the woods). I expect Starlink receivers to also gain sensitivity over time, as new generations of the chipsets come out and more and more engineering time is devoted to it.

    • Starlink should already be the next-generation hardware. The first generation stuff is the GEO sat internet like HughesNet, Exede etc, which already had this problem with trees.

    • Ehh, I think very little.

      Consumer GPS has been enhanced with better multipath (signal reflecting off buildings) rejection, more satellites, and predictive models (i.e. use an accelerometer to fill in gaps).

      Really only the first one is something Starlink could do. I assume a high-bandwidth signal will make this harder, but I'm not a radio expert.

      Also, consider that Starlink doesn't just receive. It needs to send, too. This is the bit that really really needs a clear line of sight.

  • It will be impossible for current telcos to do anything about it taking their customers away without literally blowing up the satellite or taking over the company. Any attempt to jam or interfere with lawful radio or wireless signals will make the FCC go ballistic [gps.gov].

    • Re: (Score:2)

      by cusco ( 717999 )

      They have until January to bribe Ajit Pai to block Starlink, and any roadblock he can put in place by then can be repealed by the subsequent FCC board. The big telcos made the same mistake that Wall Street and Detroit did, assuming that Musk was just blowing smoke and nothing would come of his proposals. Now it's too late to stop him, so they're desperately grabbing any last subsidy that they can get before the new administration comes in.

  • I have a friend that only has LTE internet where he lives. It's hard for me to believe with all the hype of these satellites they won't even compete with LTE on price. This is only an attractive choice for people who are in the middle of nowhere. Maybe people are excited because this business plan will still work but my friend and I are a bit underwhelmed.

    • I have a friend that only has LTE internet where he lives. It's hard for me to believe with all the hype of these satellites they won't even compete with LTE on price.

      Why should they? People have demonstrated a willingness to pay LTE prices. They will provide larger data caps, higher speeds, and most importantly, greater coverage area than LTE.

      • By buddy has no data cap on his LTE, and it is not that much slower than these sats. Especially considering the price. All I'm saying is that while I thought these sats would be an exciting change for people it is actually just a very boring new business.

        • Your buddy has a grandfathered plan you can no longer buy, then. He is a statistical outlier and can and should be ignored for the purposes of this conversation.

          • Nope he just moved here and set up the service. There is a company setting up dishes and boosters for people in rural areas. It's unlimited and way less than $90 US per month. Maybe you are talking about the US, I am in Canada.

  • The latency was 42ms. Is that 42ms round-trip to the *satellite* or 42ms round-trip to a *server on earth via the satellite*? If it is 42ms to the satellite, then you have 24ms to the satellite, 24ms back to an earth-based server, 24ms back to the satellite, and 24ms back to you. That's 82ms latency.

    • Re: Latency question (Score:4, Informative)

      by topham ( 32406 ) on Sunday November 08, 2020 @11:32AM (#60699460) Homepage

      It's latency, round trip to a Speedtest, or equivalent server.

      Starlink is being targeted for sub 20ms latency, one of the possible target markets is stock trading. The Starlink network can have latency less than ground based fibre in specific scenarios.

    • Comment removed based on user account deletion
      • Packet leaves your router, goes up to satellite, down to ground station, and across the internet. Response comes back from internet, hits ground station, goes up into orbit, and down to your satellite and into your router.

  • This should be great for anybody who lives where there are no trees, rain, clouds, etc.

  • And the real reason why E.Musk is doing this is to provide connectivity for the growing fleet of Tesla vehicles?

    • And the real reason why E.Musk is doing this is to provide connectivity for the growing fleet of Tesla vehicles?

      The real reason why Elon Musk is doing this is getting your ass to Mars is expensive. SpaceX makes good profit on every rocket launch they sell but they know that even at their new vastly lower price points, there are still only so many launches per year that make sense and they want more income than that. Starlink is a utility, the very best revenue stream to have, year after year, decade after decade.

  • The main issue with Starlink will be its capacity issue. I see tens of millions if not hundreds of millions trying to sign up (worldwide), but Starlink will not have the capacity to service them all.

Slashdot Top Deals

Ocean: A body of water occupying about two-thirds of a world made for man -- who has no gills. -- Ambrose Bierce

Close