I am appreciating this email thread.
It is really hard to “sell” the features ala carte. They need to be bundled
into reference packages. This you know.
> The wisps totally got it with fq codel and cake arriving native for mikeotiks
> entire product line and much of ubnts gear prior to that.
I see that L4S is not really solving everything (I read about issues with
Wi-Fi), although it seems to be a step in the right direction, to be
improved, let's hope.
At least, Nokia is implementing it in its network gear (for mobile
operators), so the bufferbloat problem is somehow acknowledged by
Sorry - I meant ~ 4,5 % of the ISPs, not 2 :)
All the best,
Frank
Frantisek (Frank) Borsik
https://www.linkedin.com/in/frantisekborsik
Signal, Telegram, WhatsApp: +421919416714
iMessage, mobile: +420775230885
Skype: casioa5302ca
frantisek.bor...@gmail.com
On Wed, May 8, 2024 at 9:58 AM
Just to add the latest numbers from our (LibreQoS) ongoing "QoE competitive
landscape research":
Out of 66k plus ISPs worldwide, barely 3k use some QoE middle-box. Preseem
is the market leader, with well over 400 T2 & T3 ISPs (number shared in
their wonderful Fixed Wireless Network Report 2024 Q1
This was a wonderful post, rich!
I note that preseem, paraqum, bequant and libreqos (a bufferbloat.net
backed project) are in the fq codel or cake Middlebox for isps *Qoe) market
and all of us have made a substantial dent in the problem for oh, call it
1000 isps worldwide total between us.
Thanks Frank,
So it is not the universal cure. Not surprising.
I don’t see a show-stopper for pushing adoption.
Gene
--
Eugene Chang
IEEE Life Senior Member
IEEE Communications Society & Signal Processing Society,
Hawaii Chapter Chair
IEEE Life
Here is a current view of it, IIRC:
https://forum.openwrt.org/t/rfc9330-rfc9331-rfc9332-for-lower-latency/180519/12
All the best,
Frank
Frantisek (Frank) Borsik
https://www.linkedin.com/in/frantisekborsik
Signal, Telegram, WhatsApp: +421919416714
iMessage, mobile: +420775230885
Skype:
I thought I saw a reference to an OpenWRT implementation with L4S. How well
does that work?
Gene
--
Eugene Chang
> On May 7, 2024, at 9:46 AM, Dave Taht wrote:
>
> Pete heist, jon morton, and rod grimes published a TON of research as
> to where
Pete heist, jon morton, and rod grimes published a TON of research as
to where l4s went wrong in these github repos:
https://github.com/heistp
The last was:
https://github.com/heistp/l4s-tests?tab=readme-ov-file#key-findings
They were ignored. Me, I had taken one look at it 7+ years ago and
On Tue, May 7, 2024 at 11:11 AM Dave Taht via Starlink <
starlink@lists.bufferbloat.net> wrote:
> The RFC is very plausible but the methods break down in multiple ways,
> particularly with wifi.
>
Dave, can you elaborate more on the failures? Are these being researched or
addressed in the
The RFC is very plausible but the methods break down in multiple ways,
particularly with wifi.
On Tue, May 7, 2024 at 12:10 PM Eugene Y Chang via Starlink
wrote:
>
> Dave,
> Thank you for calling attention to the RFC. I took a quick peek, and I need
> to put more time into reading the whole
Dave,
Thank you for calling attention to the RFC. I took a quick peek, and I need to
put more time into reading the whole doc. It feels very intuitive.
What I like is that it is written for incremental adoption. I will focus on
that in my next pass. It opens the door to be incrementally
It has an RFC at https://datatracker.ietf.org/doc/rfc9330/
I read it as a way to rapidly find the available bandwidth without the TCP
"sawtooth". The paper cites fc_codel and research based on it.
I suspect My Smarter Colleagues know more (;-))
--dave
On 2024-05-07 08:13, David Fernández
Is L4S a solution to bufferbloat? I have read that gamers are happy with it.
Sorry, I read it here, in Spanish:
https://www.adslzone.net/noticias/operadores/retardo-videojuegos-nokia-vodafone
Regards,
David F.
Date: Tue, 7 May 2024 06:50:41 -0400
From: Rich Brown
To: Eugene Y Chang
Cc:
Oh goodness, I wasn't suggesting that we had a total solution, I was
pointing out that the gaming community was missing the point, even with
evidence in their hands.
That suggests we have not made the point to a technically aware part of
our community.
--dave
On 2024-05-06 20:43, Eugene Y
Hi Gene,
> On May 6, 2024, at 8:38 PM, Eugene Y Chang wrote:
>
> It seems like you signed off on this challenge. Don’t do that. Help give me
> the tools to push this to the next level.
Not at all - I'm definitely signed up for this. But I collected all these
points so we can be clear-eyed
Dave,
We just can’t represent that we have the total solution.
We need to show the problem can be reduced.
We need to show that latency is a significant negative phenomena.
Take out one contributor and sic the users to the next contributor.
If we expect to solve the whole problem in one step, we
Rich,
Thanks for the recap in the email.
I have seen all of those bits.
I will help with the marketing magic needed.
We need a team of smart people engaged to help vouch for the technical
integrity.
We need a simple case (call it a special case, if you must) that shows the
problem can be
Thanks! I just posted to:
https://randomneuronsfiring.com/all-the-reasons-that-bufferbloat-isnt-a-problem/
It has mild edits from the original to address a broader audience. Also posted
to the bloat list.
Rich
> On May 6, 2024, at 3:05 PM, Frantisek Borsik
> wrote:
>
> Hey Rich,
>
>
I think that gamer experience doing simple (over-simple) tests with CAKE is a
booby-trap. This discussion suggests that the real performance of their link is
horrid, and that they turn off CAKE to get what they think is full
performance... but isn't.
Hi Gene,
I've been vacillating on whether to send this note, but have decided to pull
the trigger. I apologize in advance for the "Debbie Downer" nature of this
message. I also apologize for any errors, omissions, or over-simplifications of
the "birth of bufferbloat" story and its fixes.
21 matches
Mail list logo