Starlink has bufferbloat. Bad.
 help / color / mirror / Atom feed
From: Sauli Kiviranta <smksauli@gmail.com>
To: J Pan <Pan@uvic.ca>
Cc: Dave Taht via Starlink <starlink@lists.bufferbloat.net>
Subject: [Starlink] Re: Testing 4K live video over Starlink with added +10% packet loss
Date: Wed, 10 Sep 2025 00:26:53 +0300	[thread overview]
Message-ID: <CAJbqEYDjb02t0D8422oxmXRcMesrCaK0nu4JrtDivZxdAY=Xug@mail.gmail.com> (raw)
In-Reply-To: <CAHn=e4h74EwyutsDEfRD7wqfrOL_5DKnitFAHcYAJjzaBTgf8Q@mail.gmail.com>

You can see the bottom left corner that its pushing about 12Mbps, even when
the Fast.com can only "see" mere 2Mbps (they already cheat a bit), that's
what the RTSP source was configured to push from the Axis camera (
https://www.axis.com/products/axis-v5938 ).

In this test the end to end latency was 2.5sec (P99), pretty nice for a
100ms RTT on such a lossy scenario.

Best regards,
Suali

On Wed, Sep 10, 2025 at 12:13 AM J Pan <Pan@uvic.ca> wrote:

> that's really cool. most tcp cannot handle 10% packet loss. what's
> video datarate/latency?
> --
> J Pan, UVic CSc, ECS566, 250-472-5796 (NO VM), Pan@UVic.CA,
> Web.UVic.CA/~pan
>
> On Tue, Sep 9, 2025 at 2:05 PM Sauli Kiviranta <smksauli@gmail.com> wrote:
> >
> > This is all just TCP, that was kind of thew idea. Not yet with FEC
> though, working on that, just rusty good old TCP.
> >
> > Best regards,
> > Sauli
> >
> > On Wed, Sep 10, 2025 at 12:02 AM J Pan <Pan@uvic.ca> wrote:
> >>
> >> fast.com measures tcp performance and you used udp to get the best out
> >> of a lossy link? spacex uses multiple starlink dishes on a launch
> >> vehicle and pick whatever made through too
> >> --
> >> J Pan, UVic CSc, ECS566, 250-472-5796 (NO VM), Pan@UVic.CA,
> Web.UVic.CA/~pan
> >>
> >> On Tue, Sep 9, 2025 at 1:50 PM Sauli Kiviranta via Starlink
> >> <starlink@lists.bufferbloat.net> wrote:
> >> >
> >> > We start to get somewhere that is what I would say cool level. Our
> target
> >> > is Ulrich Spiedel storm conditions measured over the Gabriella
> cyclone with
> >> > Starlink.
> >> >
> >> > https://x.com/sksauli/status/1964349734304301268
> >> >
> >> > We have also now test running just accelerating all data, so we can
> get
> >> > same effects for better utilization rate of the link.
> >> >
> >> > Anyone know what kind of bit error rates or packet loss levels are in
> the
> >> > streams for SpaceX launches? Usually they look very very stable.
> >> >
> >> > Anyone want to test this? Send me email sauli.kiviranta@xrtc.io
> >> >
> >> > Best regards,
> >> > Sauli
> >> > _______________________________________________
> >> > Starlink mailing list -- starlink@lists.bufferbloat.net
> >> > To unsubscribe send an email to starlink-leave@lists.bufferbloat.net
>

  reply	other threads:[~2025-09-09 21:28 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-09 20:49 [Starlink] " Sauli Kiviranta
2025-09-09 21:02 ` [Starlink] " J Pan
2025-09-09 21:05   ` Sauli Kiviranta
2025-09-09 21:13     ` J Pan
2025-09-09 21:26       ` Sauli Kiviranta [this message]
2025-09-10  8:11       ` Sauli Kiviranta
2025-09-10 15:53         ` J Pan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/starlink.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAJbqEYDjb02t0D8422oxmXRcMesrCaK0nu4JrtDivZxdAY=Xug@mail.gmail.com' \
    --to=smksauli@gmail.com \
    --cc=Pan@uvic.ca \
    --cc=starlink@lists.bufferbloat.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox