1): This hits what I'd expect my connection to be (or close enough anyway):
./betterspeedtest.sh
2020-10-06 12:53:54 Testing against netperf.bufferbloat.net (ipv4) with 5
simultaneous sessions while pinging gstatic.com (60 seconds in each direction)
HE IPv6 space has been tagged as a vpn type service by Netflix, since it has
users all over the world, but it's space is all geolocated in the US. If HE had
geolocated the blocks of each POP to the country the POP resided in, and put
some rules around geolocation of using each POP (IE Canadian
CM AQM was added as an optional feature in DOCSIS 3.0. I don't know of any 3.0
modems where this was actually implemented though.
This feature requires no support from the CMTS/CCAP chassis, so as long as the
modem supports it, it will work, regardless of the supported DOCSIS version of
the
AQM was not enabled, the firmware was a fix for some puma6 latency issues.
We can still get more than 500ms latency when filling the buffers on low
upstream contracts (less than 2mb/s)
It does generally "feel" better, but it isn't AQM when you look at it hard.
-Original Message-
This is my guess.
DSL reports uses many streams from different servers to achieve these speeds.
I’m assuming flent is a single stream, so you’re at the mercy of TCP receive
windows and latency limiting how fast you can go on that single stream.
From: Bloat
Plenty of easy ways to do v4 and v6.
DNS can be built to only have hosts for v4 or v6, and you run two tests, one to
each set of hosts.
Comcast has been doing it forever.
http://speedtest.xfinity.com/ (they use speedtest.net's code, and they have it
skinned).
-Original Message-
From:
