From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-yb1-xb33.google.com (mail-yb1-xb33.google.com [IPv6:2607:f8b0:4864:20::b33]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 90B083B29D; Mon, 14 Nov 2022 17:29:38 -0500 (EST) Received: by mail-yb1-xb33.google.com with SMTP id b131so14525703yba.11; Mon, 14 Nov 2022 14:29:38 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=Kbr6Fk1VP8kBuA/SnL+tNEAx6y/BsynLsrxKHToamvE=; b=l3AZyifeRtp1m3scuHAPyT78uUgyvFWKYR24qn2ryIKb0q24POAmhvyd5yUElciOl5 wgXvElvM9bw2vByupGV6qze0ty3tdlfTOj2PwNyWXF0k7i6RGOIL1Ky5yVgNQCuRsPQ3 0AkWGxUpwlCsspSBSQ9jWPUNleCZoxdhP/nUqP6Rl4zbkxsPAhAHIZLkeG8pvytS/6l8 Ix7v5I+qP06A9oQ6UmV52AI9cwmssWqRe1r23+oL7gdyAddzWYJDFlHTGJFHx6SdmdJA uwMMWXHFLBYIsxoR3jC7E/7nh+fflNHvS5gBF/fCuphsJlOhpoC29LxObRue2DHUw/uZ GNgw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=Kbr6Fk1VP8kBuA/SnL+tNEAx6y/BsynLsrxKHToamvE=; b=a9VzYBvYcDuzkdZJZBywkR5e/YpLIAanY3fBT3bC78jN9ErK6vqSNNOKF/pOPd1yOM 0syMjun5ANTeAQWIs+cyvGDH2u4bHM4w6v9FJ5pAnrTX8mFIQpz0N1yZeupret84foUr BIqWLONJvlUBpAQDcTDV2LU9Lh+yLRunq3JrI9/wyFDF36WrB9Z06CFK+ON5PwOk3zGg hGRtgm1W0oi8PviNLeUx5UZVVltZSiXDYepFa9nIt0YaXWfm+SgfGcimy2WId8ooG+Dz Erb+Lxgn+LFwrvBZMAoQVJdQn7VFXjtHu3c9ZyfR3SlIgKz8UY0DqK30JWL48kv3y8cG QZ1Q== X-Gm-Message-State: ANoB5pmC8a+rOYdiROc98kMlGNIhK5sBwTKqc0DtXKGvE4UioNrRLh2c LUzWo4EdN1LctzzRBpRz0HB/5blMxpsbo6Iz0ic= X-Google-Smtp-Source: AA0mqf43C1/WRA1XtylMntYUhhqevcmTVDCm83oCizbnhT9lcva/Z/rdJ+wuvGdYiabFGpsy0OU5trDwrGQTfRrZ6vs= X-Received: by 2002:a25:e04b:0:b0:6dc:a17d:e4e2 with SMTP id x72-20020a25e04b000000b006dca17de4e2mr14374592ybg.183.1668464977827; Mon, 14 Nov 2022 14:29:37 -0800 (PST) MIME-Version: 1.0 References: <1A0A3366-3907-4F4E-A055-525A3CBDAC2A@gmx.de> In-Reply-To: From: dan Date: Mon, 14 Nov 2022 15:29:29 -0700 Message-ID: To: "MORTON JR., AL" Cc: Sebastian Moeller , =?UTF-8?Q?Robert_Chac=C3=B3n?= , =?UTF-8?Q?Robert_Chac=C3=B3n_via_Bloat?= , libreqos Content-Type: multipart/alternative; boundary="00000000000052dd7f05ed75c676" Subject: Re: [LibreQoS] [Bloat] gamebench quote X-BeenThere: libreqos@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Many ISPs need the kinds of quality shaping cake can do List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 14 Nov 2022 22:29:38 -0000 --00000000000052dd7f05ed75c676 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Al, I do agree with you from a consumer's perspective. There are those of us that are trying to get latency to the top of the heap, which has prerequisites of adequate bandwidth in both directions, but consumers aren't there and big providers are also very much not there. Still slinging 300x20 for reasons unknown unless it's to preserve CPU/Routing performance on aging hardware. All of the cross connects on the internet are full duplex so unless the access tech is limiting things... no good reason for the abysmal uploads in relation to downloads except to preserve underpowered hardware or to prevent customers for using it to push them to different service packages. On Mon, Nov 14, 2022 at 3:18 PM MORTON JR., AL wrote: > Hi Dave, all, > > > > I=E2=80=99m taking a slightly contrarian position here: > > > > When I read, =E2=80=9C...that latency has now clearly overtaken broadband= speed > ...=E2=80=9D, > > IMO, it=E2=80=99s only correct when we include the efforts to increase up= link > speeds, and can therefore minimize the complexity of low-layer uplink > protocols. The continued suppression of uplink speeds in simplified > marketing helps users overlook a critical factor in ISP service. > > > > In other words, all the terrestrial access =E2=80=9Cwants to be fiber=E2= =80=9D. > > > > Example: We stayed at my frugal brother-in-law=E2=80=99s home, where he h= ad > purchased the low-end rate offered by his fiber-based ISP. It was 25Mbps, > symmetrical rates. The bottom line is that this was a perfectly adequate > service offer for zoom conferences and streaming video viewing. My > measurements confirmed the rates, and that packet transfer had few defect= s. > > > > So for me, the 3 pillars are: sufficient capacity for **all** users > involved, symmetrical uplink bit rate, and the upstream access protocol > should converge to zilch*, all leading to few impairments, including > latency. > > > > Al > > > > * this is hard to achieve without the first two... > > > > > > *From:* LibreQoS *On Behalf Of *= Sebastian > Moeller via LibreQoS > *Sent:* Monday, November 14, 2022 3:41 AM > *To:* Robert Chac=C3=B3n ; Robert C= hac=C3=B3n > via Bloat ; dan > *Cc:* libreqos ; bloat < > bloat@lists.bufferbloat.net> > *Subject:* Re: [LibreQoS] [Bloat] gamebench quote > > > > Hi Robert, > > This is surprisingly rich in useful information and quite lean on > marketing (given that this is marketing material having easily accessible > links to source material is unusually open*). Nice. > > Regards > Sebastian > > *) Personally that approach would likely reel me in as customer, if I > lived in your service area ;) > > On 14 November 2022 00:16:57 CET, "Robert Chac=C3=B3n via Bloat" < > bloat@lists.bufferbloat.net> wrote: > > Same here. We compete primarily against Charter Spectrum in the suburbs. > > Gamers and twitch streamers switch to us frequently due to the high uploa= d > bloat and jitter on Spectrum's network. > > We made a page on our website > > that tries to educate prospective clients about latency and bloat. I adde= d > the GameBench quote to it (thanks, Dave!) > > Surprisingly, some people do read it and it helps ease their minds about > "speed" when they're used to being bombarded with messaging about needing > gigabit bandwidth. > > > > > I'm happy enough if TDS and Spectrum take their sweet time figuring out > that latency is important :D > > > > This! That and their inability to really do whole-home WiFi at scale help= s > WISPs offer a true performance answer to cable internet. > > > > On Sun, Nov 13, 2022 at 3:56 PM dan via LibreQoS < > libreqos@lists.bufferbloat.net> wrote: > > This is definitely our focus. We preach low latency first and better > uploads second, downloads are a distant third. We do have to advertise t= he > faster speeds just to get into the conversation because Spectrum for > example and now TDS are all about 1G or 2G services. See my other email > though, Spectrum's latency and jitter is garbage, TDS is actually worse > than that to many services but I don't have a remote login to a TDS site = at > the moment. I'm 15ms more from AWS Oregon on TDS than Spectrum, and I'm > 20ms better than Spectrum on Lumen DIA. > > I'm happy enough if TDS and Spectrum take their sweet time figuring out > that latency is important :D > > > > On Sun, Nov 13, 2022 at 3:52 PM Dave Taht via LibreQoS < > libreqos@lists.bufferbloat.net> wrote: > > > https://www.gamebench.net/network-performance-and-gamer-experience-the-us= -edition/ > > > "Overall, the consistently reinforced takeaway is that latency has now > clearly overtaken broadband speed as the focus area for network > providers seeking to provide =E2=80=93 and guarantee and commercially ben= efit > from =E2=80=93 optimum experience in both online multiplayer and cloud > gaming." > > -- > This song goes out to all the folk that thought Stadia would work: > > https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-698136666= 5607352320-FXtz > > Dave T=C3=A4ht CEO, TekLibre, LLC > _______________________________________________ > LibreQoS mailing list > LibreQoS@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/libreqos > > > _______________________________________________ > LibreQoS mailing list > LibreQoS@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/libreqos > > > > -- > Sent from my Android device with K-9 Mail. Please excuse my brevity. > --00000000000052dd7f05ed75c676 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Al, I do agree with you from a consumer's perspective.= =C2=A0 There are those of us that are trying to get latency to the top of t= he heap, which has prerequisites of adequate bandwidth in both directions, = but consumers aren't there and big providers are also very much not the= re.=C2=A0 Still slinging 300x20 for reasons unknown unless it's to pres= erve CPU/Routing performance on aging hardware.=C2=A0 All of the cross conn= ects on the internet are full duplex so unless the access tech is limiting = things... no good reason for the abysmal uploads in relation to downloads e= xcept to preserve underpowered hardware or to prevent customers for using i= t to push them to different service packages.

On Mon, Nov 14, 2022 at 3:18 P= M MORTON JR., AL <acmorton@att.com> wrote:

Hi Dave, all,

=C2=A0

I=E2=80=99m taking a slightly contrarian position here:

=C2=A0

When I read, =E2=80=9C...that latency has now clearly overta= ken broadband speed ...=E2=80=9D,

IMO, it=E2=80=99s only correct when we include the efforts t= o increase uplink speeds, and can therefore minimize the complexity of low-layer uplink protocols. The continued suppression of uplink speeds = in simplified marketing helps users overlook a critical factor in ISP servi= ce.

=C2=A0

In other words, all the terrestrial access =E2=80=9Cwants to= be fiber=E2=80=9D.

=C2=A0

Example: We stayed at my frugal brother-in-law=E2=80=99s hom= e, where he had purchased the low-end rate offered by his fiber-based ISP. It was 25Mbps, symmetrical rates. The bottom line is that this was a = perfectly adequate service offer for zoom conferences and streaming video v= iewing. My measurements confirmed the rates, and that packet transfer had f= ew defects.

=C2=A0

So for me, the 3 pillars are: sufficient capacity for *al= l* users involved, symmetrical uplink bit rate, and the upstream access protocol should converge to zilch*, all leading to few= impairments, including latency.

=C2=A0

Al

=C2=A0

* this is hard to achieve without the first two...=

=C2=A0

=C2=A0

From: LibreQoS <libreq= os-bounces@lists.bufferbloat.net> On Behalf Of Sebastian Moeller via LibreQoS
Sent: Monday, November 14, 2022 3:41 AM
To: Robert Chac=C3=B3n <robert.chacon@jackrabbitwireless.com&g= t;; Robert Chac=C3=B3n via Bloat <bloat@lists.bufferbloat.net>; dan <dandenson@gmail.com>
Cc: libreqos <libreqos@lists.bufferbloat.net>; bloat <bloat@lists.buffe= rbloat.net>
Subject: Re: [LibreQoS] [Bloat] gamebench quote=

=C2=A0

Hi Robert,

This is surprisingly rich in useful information and quite lean on marketing= (given that this is marketing material having easily accessible links to s= ource material is unusually open*). Nice.

Regards
Sebastian

*) Personally that approach would likely reel me in as customer, if I lived= in your service area ;)

On 14 November 2022 00:16:57 CET, "Robert= Chac=C3=B3n via Bloat" <bloat@lists.bufferbloat.net> wrote:=

Same here. We compete primarily against Charte= r Spectrum in the suburbs.

Gamers and twitch streamers switch to us frequ= ently due to the high upload bloat and jitter on Spectrum's network.=

We made a page on our website that tries to educate prospective clients about lat= ency and bloat. I added the GameBench quote to it (thanks, Dave!)=

Surprisingly, some people do read it and it he= lps ease their minds about "speed" when they're used to being= bombarded with messaging about needing gigabit bandwidth.

=C2=A0

> I'm happy enough if TDS and Spectrum = take their sweet time figuring out that latency is important :D

=C2=A0

This! That and their inability to really do wh= ole-home WiFi at scale helps WISPs offer a true performance answer to cable= internet.

=C2=A0

On Sun, Nov 13, 2022 at 3:56 PM dan via LibreQ= oS <= libreqos@lists.bufferbloat.net> wrote:

This is definitely our focus.=C2=A0 We preach = low latency first and better uploads second, downloads are a distant third.= =C2=A0 We do have to advertise the faster speeds just to get into the conversation because Spectrum for example and now TDS are all about 1G or = 2G services.=C2=A0 See my other email though, Spectrum's latency and ji= tter is garbage, TDS is actually worse than that to many services but I don= 't have a remote login to a TDS site at the moment.=C2=A0 I'm 15ms more from AWS Oregon on TDS than Spectrum, and = I'm 20ms better than Spectrum on Lumen DIA.=C2=A0=C2=A0

I'm happy enough if TDS and Spectrum take their sweet time figuring out= that latency is important :D

=C2=A0

On Sun, Nov 13, 2022 at 3:52 PM Dave Taht via = LibreQoS <libreqos@lists.bufferbloat.net> wrote:

https://www.gamebench.net/network= -performance-and-gamer-experience-the-us-edition/

"Overall, the consistently reinforced takeaway is that latency has now=
clearly overtaken broadband speed as the focus area for network
providers seeking to provide =E2=80=93 and guarantee and commercially benef= it
from =E2=80=93 optimum experience in both online multiplayer and cloud
gaming."

--
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-a= ctivity-6981366665607352320-FXtz
Dave T=C3=A4ht CEO, TekLibre, LLC
_______________________________________________
LibreQoS mailing list
LibreQo= S@lists.bufferbloat.net
https://lists.bufferbl= oat.net/listinfo/libreqos

______________________________________________= _
LibreQoS mailing list
LibreQo= S@lists.bufferbloat.net
https://lists.bufferbl= oat.net/listinfo/libreqos


--
Sent from my Android device with K-9 Mail. Please excuse my brevity.=

--00000000000052dd7f05ed75c676--