Attached is a recent cloudflare test of verizon, taken about 1.2 miles from
a rural tower. While the bandwidth was low for reasons unknown, the latency
was... quite tolerable. Perhaps they are figuring this out.

I have no idea how atlas calculates latency or loss. Over what interval? In
the starlink case especially, I look at a typical plot, and I hope that if
more people looked at details like that, they would despair of using
averages ever again.

On Wed, Sep 20, 2023 at 8:05 AM Livingood, Jason via Starlink <
starlink@lists.bufferbloat.net> wrote:

> As some of you may recall I recruiting some US-based Starlink users to
> host RIPE Atlas probes as part of my role as a RIPE Atlas Ambassador. I
> just got word that NetForecast conducted a performance study that looked at
> RIPE Atlas data using some of the data from those and other probes. They
> looked at homes across the US with 5G fixed wireless, LEO satellite, and
> DOCSIS.
>
>
>
> They have posted a first report of this study at
> https://www.netforecast.com/wp-content/uploads/FixedWireless_LEO_CableComparisonReport_NFR5148-1.pdf.
> I am looking forward to other researchers looking into the RIPE Atlas data
> – I know a few are working on this.
>
>
>
> Jason
>
>
> _______________________________________________
> Starlink mailing list
> Starlink@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/starlink
>


-- 
Oct 30: https://netdevconf.info/0x17/news/the-maestro-and-the-music-bof.html
Dave Täht CSO, LibreQos