[ih] Internet/Wireless Principle of Levelness

Mon Nov 11 15:25:15 PST 2019

the keyboard of geoff goodfellow via Internet-history
<internet-history at elists.isoc.org> writes:

> Dan & Jack: am curious to know WHAT kind (Cable, DSL, Fiber, ...?) of
> Internet connections y'all have and from WHICH providers?

http://www.dslreports.com/speedtest/results/bufferbloat

is one of the best public resources for measuring bloat that
exists. There's breakdowns by AS number, by ISP, etc, also
if you click in and around.

Pls give the speedtest a shot and see how you do with your ISP. Try both
wired and wireless.

note: dslreports is not as good as the flent tool and has a few
problems:

A) everybody tuning sqm/fq_codel/cake tends to use it to verify their
results, so the dataset is polluted by folk that have "fixed" bloat or
are fixing it.

B) They throw out a lot of outliers. I have longed for them to
show a plot of the outliers alone as I suspect it would look
much like the cosmic background radiation.

C) Each test only runs for 20 sec or less, so the cutoff past 4 sec is
*very* artificial.

D) their "Quality" metric is partially based on packet loss. An A
for bufferbloat and a C for quality means you are throwing away
packets. dslreports apparently supports ECN so that's the only way to
get an A for both at lower bandwidths.

the upload scatterplot worldwide has got tons better in the last few
years, it used to be that the blue vertical cablemodem lines were very
thick and in two groups extending down past a second. Most of that
improvement comes from bandwidths going up and buffersizes staying
constant.

At least my gpon provider (sonic) only has a 60ms uplink buffer.

>
> pretty clear/sure neither of you have satellite (as yours truly dose here
> on The Big Island for backup when the "primary" cable Spectrum
> "service" *reliably
> goes out* -- almost monthly -- so far twice already this month and once
> last month :-/) at which time even a 700-1000 ms latency over the satellite
> link is Most Welcomed! :D

cake has a mode for that. :)

>
> geoff
>
> On Mon, Nov 11, 2019 at 10:50 AM Dan Lynch via Internet-history <
> internet-history at elists.isoc.org> wrote:
>
>> Or you could have just used a kid. 15 or so years ago I had a 12 year old
>> son who bounced back and forth between the home in the Napa Valley and the
>> one in Los Altos. We had T1 service at both places (hot stuff in those
>> days) and he was a gamer,of course.  He could/would not play certain games
>> in Napa because the latency was over 30 ms!  Not so in Los Altos. He knew.
>> And yes, for a twitchy kid 30 ms was everything.

For me too. I'm extremely latency sensitive - playing in a band, 4ms (4
feet) is a comfortable distance. 8ms, far less so.

>> As for TV service up here I have 50 megabit service and it is excellent
>> except for the occasional glitch like Jack described. And it may persist
>> for a few minutes, then goes away for days. I tried calling to complain a
>> few years ago, but nobody home.....  We have won?

Well, sometimes it ain't bloat. 

>>
>> Cheers,
>>
>> Dan
>>
>> Cell 650-776-7313
>>
>> > On Nov 11, 2019, at 11:32 AM, Jack Haverty via Internet-history <
>> internet-history at elists.isoc.org> wrote:
>> >
>> > On 11/11/19 8:31 AM, Dave Taht via Internet-history wrote:
>> >
>> >> And - of course! it's got the "deep buffers" providers require.
>> >
>> > I'm just a User now.  Just last year I helped a friend, another User,
>> > figure out why his "gaming" app, which depends on interactive behavior
>> > across the net, was sometimes unusable.  I was curious, since I also
>> > sometimes see visual and audio artifacts on streaming TV content, making
>> > TV sometimes similarly unusable, even though I have 150+ Mb/sec internet
>> > service.   We Users tend to think "Oh, the net's broken again, they're
>> > probably working on fixing it".
>> >
>> > Using the ancient network management tools, we tracked the cause down to
>> > latency.  The typical latency we measured across the net was 100 msec or
>> > less.  But occasionally it would jump to several seconds and stay there
>> > for a while.   I was surprised to see that zero packets were being lost,
>> > but many were delayed as much as 30 seconds.  Without the ability to dig
>> > inside the boxes, I can only speculate that such behavior at the IP
>> > level was what made the gaming app unusable, and could cause those
>> > artifacts I see in my TV video and audio.
>> >
>> > My friend tried complaining to his ISPs' tech support, but they all said
>> > their service was working fine.  Perhaps that is a consequence of the
>> > "Levelness" that now makes Users' applications involve many different
>> > service and equipment providers?
>> >
>> > Is this latency how Users now see the effects of those "deep buffers"?
>> > Why would providers require a feature that makes their customers
>> > unhappy.....?
>> >
>> > I'm still just being curious about the History of the Internet,
>> > especially how its service evolved -- as seen by the Users.
>> >
>> > /Jack
>> >
>> >
>> > --
>> > Internet-history mailing list
>> > Internet-history at elists.isoc.org
>> > https://elists.isoc.org/mailman/listinfo/internet-history
>>
>> --
>> Internet-history mailing list
>> Internet-history at elists.isoc.org
>> https://elists.isoc.org/mailman/listinfo/internet-history
>>
>>
>
> -- 
> Geoff.Goodfellow at iconia.com
> living as The Truth is True
> http://geoff.livejournal.com
-- 
Internet-history mailing list
Internet-history at elists.isoc.org
https://elists.isoc.org/mailman/listinfo/internet-history