[Bloat] reality2 podcast

[Bloat] reality2 podcast

[Dave Taht]

My nuanced personal tale of the aqm deployment debate, with a
modest starlink WIP update, and a reference to my fav outcomes of the iab
workshop, just went up here:

https://reality2.substack.com/p/improving-the-internets-responsiveness


-- 
Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw

Dave Täht CEO, TekLibre, LLC

Re: [Bloat] reality2 podcast

[Gabe Kassel]

[-- Attachment #1: Type: text/plain, Size: 833 bytes --]

The podcast link appears to go to the wrong episode (83 instead of 85).

https://www.reality2cast.com/85

On Fri, Sep 17, 2021 at 11:27 AM Dave Taht <dave.taht@gmail.com> wrote:

> My nuanced personal tale of the aqm deployment debate, with a
> modest starlink WIP update, and a reference to my fav outcomes of the iab
> workshop, just went up here:
>
> https://reality2.substack.com/p/improving-the-internets-responsiveness
>
>
> --
> Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw
>
> Dave Täht CEO, TekLibre, LLC
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>


-- 
_______

*Gabe Kassel*
Technology Strategist | Office of the CTO

*m* 770.490.0431
*e*  gabe@eero.com

[-- Attachment #2: Type: text/html, Size: 2952 bytes --]

Re: [Bloat] reality2 podcast

[Jonathan Foulkes]

[-- Attachment #1: Type: text/plain, Size: 1531 bytes --]

I’ll note that I defected what seemed to be a bloated line from Doc’s end. Voice quality was nowhere near as good as it could be.

Podcasters should A) use a wired connection for local connections, and B) run a well-tuned AQM on their router.

Running this VoIP quality test would be a good proxy for line latency performance for conference calls (voice only).
https://www.ringcentral.com/support/qos.html <https://www.ringcentral.com/support/qos.html>

Attached are two images of results of that test on two of my lines, both sore the same MOS rating of 4.2.

The first is a cable line with 30MBps of uplink capacity, with an IQrouter Pro.


the second is my backup 10 / 0.7Mbps DSL Line with an IQrouter v3, note how this graph and stats is almost identical to the cable line, some stats are actually better on DSL.



Friends don’t let friends VoIP/Zoom without Cake ;-)

Cheers,

Jonathan

> On Sep 17, 2021, at 2:27 PM, Dave Taht <dave.taht@gmail.com> wrote:
> 
> My nuanced personal tale of the aqm deployment debate, with a
> modest starlink WIP update, and a reference to my fav outcomes of the iab
> workshop, just went up here:
> 
> https://reality2.substack.com/p/improving-the-internets-responsiveness
> 
> 
> -- 
> Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw
> 
> Dave Täht CEO, TekLibre, LLC
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat


[-- Attachment #2.1: Type: text/html, Size: 2991 bytes --]

[-- Attachment #2.2: VoIPTest_WS35Up_line.png --]
[-- Type: image/png, Size: 190658 bytes --]

[-- Attachment #2.3: VoIPTest_DSLline.png --]
[-- Type: image/png, Size: 188624 bytes --]

Re: [Bloat] reality2 podcast

[Jonathan Foulkes]

[-- Attachment #1: Type: text/plain, Size: 1635 bytes --]

I’ll note that I defected what seemed to be a bloated line from Doc’s end. Voice quality was nowhere near as good as it could be.

Podcasters should A) use a wired connection for local connections, and B) run a well-tuned AQM on their router.

Running this VoIP quality test would be a good proxy for line latency performance for conference calls (voice only).
https://www.ringcentral.com/support/qos.html <https://www.ringcentral.com/support/qos.html>

Attached are two images of results of that test on two of my lines, both sore the same MOS rating of 4.2.

The first is a cable line with 30MBps of uplink capacity, with an IQrouter Pro.



the second is my backup 10 / 0.7Mbps DSL Line with an IQrouter v3, note how this graph and stats is almost identical to the cable line, some stats are actually better on DSL.



Friends don’t let friends VoIP/Zoom without Cake ;-)

Cheers,

Jonathan

> On Sep 17, 2021, at 2:27 PM, Dave Taht <dave.taht@gmail.com <mailto:dave.taht@gmail.com>> wrote:
> 
> My nuanced personal tale of the aqm deployment debate, with a
> modest starlink WIP update, and a reference to my fav outcomes of the iab
> workshop, just went up here:
> 
> https://reality2.substack.com/p/improving-the-internets-responsiveness <https://reality2.substack.com/p/improving-the-internets-responsiveness>
> 
> 
> -- 
> Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw
> 
> Dave Täht CEO, TekLibre, LLC
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat


[-- Attachment #2.1: Type: text/html, Size: 3342 bytes --]

[-- Attachment #2.2: VoIPTest_WS35Up_line.png --]
[-- Type: image/png, Size: 132391 bytes --]

[-- Attachment #2.3: VoIPTest_DSLline.png --]
[-- Type: image/png, Size: 136763 bytes --]

[Bloat] quality podcasting metrics

[Dave Taht]

[-- Attachment #1.1: Type: text/plain, Size: 4060 bytes --]

Thought I'd talk to the podcasting + bufferbloat - packet loss problem with
more folk:

On Thu, Sep 23, 2021 at 8:24 AM Jonathan Foulkes <jf@jonathanfoulkes.com>
wrote:

> I’ll note that I defected what seemed to be a bloated line from Doc’s end.
> Voice quality was nowhere near as good as it could be.
>
> Podcasters should A) use a wired connection for local connections, and B)
> run a well-tuned AQM on their router.
>
> Running this VoIP quality test would be a good proxy for line latency
> performance for conference calls (voice only).
> https://www.ringcentral.com/support/qos.html
>
> Attached are two images of results of that test on two of my lines, both
> sore the same MOS rating of 4.2.
>
> The first is a cable line with 30MBps of uplink capacity, with an IQrouter
> Pro.
>
>
> the second is my backup 10 / 0.7Mbps DSL Line with an IQrouter v3, note
> how this graph and stats is almost identical to the cable line, some stats
> are actually better on DSL.
>
>
> Friends don’t let friends VoIP/Zoom without Cake ;-)
>
>
Jonathan:

Doc has one site with no bloat (pie algorithm) and another with bloat,
(charter), and I can't remember which he was at for this podcast. I also
note that
for PIE, at least, "no bloat" also equals "some packet loss" and I don't
have a grip on how the underlying technologies deal with that nowadays.

On the podcasting front I have been somewhat intentionally using various
forms of underlying wireless tech to make my points, unsubtly, and to learn
more about actual other behavioral problems the tech has. Dogfooding...
taking packet captures...

I completely missed out on the podcasting world until recently, and while
I'm finding it easier to talk than type nowadays, I'm finding that my range
of facial expressions are more suitable for the stage or
vaudeville comedian than videoconferencing, and would prefer to retreat
behind an avatar, or voice a character in an animation! ( I did, for the
first time in a year, get a real haircut and trim my beard way, way back, a
few days ago. I now look a lot closer to my real age, I think)

The first one (modem.show) I did, I was using cake + an iphone +
freeswitch. That had the least glitches. https://www.modem.show/post/s01e07/
The next (signalwire) was iphone 5g + freeswitch - there were some major
glicthes
The one after that (the floss one that took on starlink) was android 5g +
skype, and most of the glitches I recall were actually due to sound
suppression of me from other speakers, which I didn't understand how was
done! til I learned that webrtc, at least, can only handle 3 speakers.
(talkers? people speaking, not physical speakers)
The one you are commenting on had some packet concealment artifacts as
well, and was android wifi + 5g + skype.

Getting to a more organized, real, and instrumented campaign regarding this
and other videoconferencing technologies is on my mind,
and a common language to describe them, is partially why I've been sinking
time into galene.org's algorithms. More feedback from
actual podcasters (cc-d) about the glitches they encounter, on which
underlying technologies, and what they sound or look like, would be helpful.

Perhaps you could have jonathan on your shows one day.


> Cheers,
>
> Jonathan
>
> On Sep 17, 2021, at 2:27 PM, Dave Taht <dave.taht@gmail.com> wrote:
>
> My nuanced personal tale of the aqm deployment debate, with a
> modest starlink WIP update, and a reference to my fav outcomes of the iab
> workshop, just went up here:
>
> https://reality2.substack.com/p/improving-the-internets-responsiveness
>
>
> --
> Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw
>
> Dave Täht CEO, TekLibre, LLC
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>
>
>

-- 
Fixing Starlink's Latencies: https://www.youtube.com/watch?v=c9gLo6Xrwgw

Dave Täht CEO, TekLibre, LLC

[-- Attachment #1.2: Type: text/html, Size: 6235 bytes --]

[-- Attachment #2: VoIPTest_WS35Up_line.png --]
[-- Type: image/png, Size: 132391 bytes --]

[-- Attachment #3: VoIPTest_DSLline.png --]
[-- Type: image/png, Size: 136763 bytes --]