from Hacker News

Mobile LTE Coverage Map

by interweb on 9/28/21, 7:39 PM with 79 comments

  • by joecool1029 on 9/29/21, 12:34 AM

    >The coverage map was created using data submitted voluntarily by the four mobile carriers using certain standardized propagation model assumptions

    Eh? Why link or even talk about this. Here's 3rd party coverage collected by members over at Cellmapper. It's not perfect or complete but it's real data without 'assumptions' being made: https://www.cellmapper.net

    At least there's a site for Canada that has exactly where every single tower is located: https://www.ertyu.org/steven_nikkel/cancellsites.html

  • by Maxburn on 9/28/21, 8:24 PM

    The map is only a piece of the story too. Depriotitization is a thing you will encounter, mostly on MVNO's. I call it four bars of no data. Just moved from Mint back to AT&T Prepaid because of this situation. Your results will vary wildly regionally, and time of day.
  • by fastaguy88 on 9/28/21, 11:04 PM

    Interesting map, but entirely fictional on US Rt2 between East Glacier and West Glacier for ATT. There is no ATT coverage for 60+ miles, yet the map suggests complete coverage. Correct for Verizon and T-Mobile.
  • by mdip on 9/29/21, 1:19 PM

    For those using these maps to determine the service you can expect from an MVNO operating on these providers, beware.

    I've been using Google Fi and tried a few others with Visible being the one that I found to be the most difficult to sort out. Visible uses Verizon's network (IIRC, they're owned by Verizon). In the thumb of Michigan, there's a large spot at the tip that Verizon post-paid customers have service (via the Extended Network[0]) but Visible does not. In fact, Lexington and Port Huron, with about a 5 mile buffer, were the only places service functioned similarly to home (albeit LTE rather than 5G) where coverage is solid. And it's not a matter of "there was really poor/slow/spotty service", there was simply nothing from just north of Lexington up M-25 to Port Austin and for much of M-59 from Armada to Port Austin[1].

    Over the summer that I had this service, my mental model of the map would have the entire thumb empty with a few bursts of service over a some of the more populated areas, much like T-Mobile indicates. And that's curious -- the T-Mobile map looks a whole lot more like I'd expect for most of the service. Even the post-paid Verizon/AT&T service isn't great in a lot of places -- effectively or actually no service, but according to AT&T and Verizon's maps, I should be working almost everywhere.

    [0] Which, if I understand things correctly, is Verizon buying service from someone else.

    [1] We hit the dark sky park in Port Austin using Waze which caused me to pay closer attention, one route up, one route back; on M-25, I'd pick up service briefly enough to get a routing update but it sat "looking for service" with the circle/slash (No) symbol matching it. On M-59, it was dead except for Sandusky.

  • by topspin on 9/28/21, 7:57 PM

    This is "mobile broadband" only. Not wired broadband, the thing everyone is really interested in.
  • by cmg on 9/28/21, 9:47 PM

    How trustworthy / realistic is this data, though?

    > The coverage map was created using data submitted voluntarily by the four mobile carriers using certain standardized propagation model assumptions or parameters that were established by the FCC as part of the Broadband Data Collection.

    > Please note: The map depicts the coverage a customer can expect to receive when outdoors and stationary. It is not meant to reflect where service is available when a user is indoors or in a moving vehicle.

    > Because the coverage map is based on propagation modeling, a user’s actual, on-the-ground experience may vary due to factors such as the end-user device used to connect to the network, cell site capacity, and terrain.

  • by jsjohnst on 9/28/21, 11:23 PM

    I’m so tired of maps that are grossly inaccurate like this. Since Christmas last year I’ve been traveling across the US working remotely from my RV. I’m very meticulous about checking over a dozen sources and still have been bitten by campgrounds without any (let alone useable) cell coverage on any carrier, despite having a high gain directional antenna. I spot checked a bunch of the places I’m aware of and just like all the other maps out there, this is just flat misleading and inaccurate.
  • by sparker72678 on 9/29/21, 2:14 AM

    Coverage maps like these are no longer useful. I have "4 Bars" with extremely slow data rates constantly. My average speeds are about 1/10th what they were 7ish years ago (1.5 Mbps down vs 15 when networks were less loaded, "LTE" networks) as device density and usage has gone up.

    We need data throughput maps, maybe even by time of day.

    Having "coverage" doesn't mean you can actually do anything with it.

  • by JohnJamesRambo on 9/29/21, 5:39 AM

    I question what the point of making this map was, when it isn’t even remotely accurate in so many places. I guess it makes the FCC feel like they did something and the carriers can continue saying how awesome they are and how they have brought broadband to the citizens as promised.
  • by ece on 9/29/21, 12:38 AM

    It's funny that US Cellular is listed here, and Dish/Boost isn't. Neither is Sprint, but it's LTE network is still relatively separate and something most T-Mobile customers can't access yet. Pretty useless map. Maybe comparing this with how much spectrum these companies have in an area might give more incentive to build where there's demand.
  • by shadowtree on 9/28/21, 11:25 PM

    Interesting how much wider Verizon covers the west vs. AT&T.

    I am on AT&T and motorcycle quite a bit through the backcountry and use a real GPS navi for that reason. Trusting your phone is useless.

    Such a change coming from Europe where I stay connected on any pass through the Alps.

  • by LeoPanthera on 9/28/21, 10:16 PM

    This map shows no coverage for T-Mobile at my house...

    which I can confirm is true!

    It also shows blanket coverage at my house by both Verizon and AT&T, which is definitely not true.

  • by vlovich123 on 9/28/21, 8:45 PM

    This should really be normalized as MB/capita. Basic data/voice coverage in the vast wilderness is important for safety, but speed is really more relevant in more populated regions. Geographic maps like this aren't useful, especially without more powerful filtering tools (e.g. specifying a minimum speed & showing that coverage).
  • by sushisource on 9/28/21, 11:20 PM

    Apparently the government doesn't test on Firefox
  • by zachberger on 9/28/21, 7:57 PM

    "Just" as in August 6th.
  • by Mup_TIpekpaceH on 9/29/21, 8:03 AM

    wow
  • by ojagodzinski on 9/28/21, 10:15 PM

    *for US.