Lacking any way to get comprehensive data from the dslreports folk, what I've been doing has been taking a screenshot of the thing every so often, and sticking it in a presentation, and then comparing over time.
If you look at the third slide here, with data from 2018: http://www.taht.net/~d/sch_cake_ieee_lanman2018%20(2).pdf And then get a current report, things are really looking up! http://www.dslreports.com/speedtest/results/bufferbloat?up=1 My problem is that I feel that the dslreports data is now seriously over-polluted by folk that use the site to tune up their routers with sqm/fq_codel/cake whatever, so I keep wondering what the real state O the world is. It is pretty obvious that upload speeds have been migrating upwards (without changing the buffer size), at least. I'm pretty sure docsis 3.1 modems have not deployed to a huge extent, so we're not seeing pie all that much. About the only way I can think of to actually track the fq_codel and cake deployments would be to start tracking ECE and ECN marks from apple gear. -- Dave Täht CTO, TekLibre, LLC http://www.teklibre.com Tel: 1-831-205-9740 _______________________________________________ Bloat mailing list [email protected] https://lists.bufferbloat.net/listinfo/bloat
