From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm1-x32c.google.com (mail-wm1-x32c.google.com [IPv6:2a00:1450:4864:20::32c]) (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 B5BB93B29E; Wed, 8 Feb 2023 19:44:13 -0500 (EST) Received: by mail-wm1-x32c.google.com with SMTP id r18so360013wmq.5; Wed, 08 Feb 2023 16:44:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:from:to:cc:subject:date :message-id:reply-to; bh=Zh/PTVEXuWBzBWSsUOTEvhs5yzyFZsV75RQ7KMj74j4=; b=kYTDFqJX0LdzcRIobD6IceiVAR5t5rqSckxOA/B1cPu/dbwfmv1jmmB97pdHfI1Dk2 JU+mHYplTM7nfyFg6bXFjjIx9yHjUuGUjvrUUh1oDyd6NjesMra/8gpfT0Byf0x5YcNr wtTUmtcqkWe8rJNvygCRxygrxI/RHBYbr4qxOYATZ3/10wRnScPf6nT8BisQeqckQeqY qeFESNcPOhNr9Y6pzHKj99dx+Kaxs7/yexJVwkFOlNtdRYzIak9qKiN0REmU8n4Wyb2N VbliXEhZRSdAM6js1n43L+7d0zAre2cxgq5GRSGz1x2N7mFokL3ckvuO5Y0n4Lkpvn11 vNhg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding: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=Zh/PTVEXuWBzBWSsUOTEvhs5yzyFZsV75RQ7KMj74j4=; b=rWQe9YIuzigdwMvYOHPH4FDiNJbCCtud2AJ7OrXQeSJUTDkCrh9DaHB0ou420foFMB Luk9K/tp0mK3BQmibOn6TWALK9bYIIamrUJACLnEjf5PiCLDLGasrQ010n1VBcVivNTN tW2PhXNVNKkKGZOdjfpKmw5wR1dr5o5PEGnQfGgJkL37Cn3g7LDFusJq+YAyew7y+vr5 GTPGKF9Ht/G9KqznwwOEOTajaP5ESXTYt0SriMxjeo0m3xrbSLiZPXVl9o4/0/SdZSVc OHTAJzboTc0teoxjpS+bv0JG7DYCUUqKH22waCSeyWyd5xCAfBFWHI9X65azJuu7+opi whDA== X-Gm-Message-State: AO0yUKUAP5WYxc30pXbQG5+1iPiFMHx25E7bjMTiTNaXFEjmAFcaxGWB /iYLcCdYiEZJ9nk9AmDfaAiocpAe8FjLQ1mISXE= X-Google-Smtp-Source: AK7set9srqUQy4l2kABaeBovVCyU+4ktxUDgzmELP+t4EnEJNqSzRUkH5OAtvzZQ1O2MBRJpaXmVdiae3B0zE7es+4k= X-Received: by 2002:a05:600c:3b0f:b0:3dc:51e5:45d3 with SMTP id m15-20020a05600c3b0f00b003dc51e545d3mr303885wms.138.1675903452081; Wed, 08 Feb 2023 16:44:12 -0800 (PST) MIME-Version: 1.0 References: <49312101.15152.0@wordpress.com> In-Reply-To: From: Dave Taht Date: Wed, 8 Feb 2023 16:43:59 -0800 Message-ID: To: "Luis A. Cornejo" Cc: Sebastian Moeller , Dave Taht via Bloat , libreqos Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Subject: Re: [Bloat] Fwd: [New post] How Good is FWA Wireless? X-BeenThere: bloat@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: General list for discussing Bufferbloat List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 09 Feb 2023 00:44:13 -0000 On Wed, Feb 8, 2023 at 2:41 PM Luis A. Cornejo w= rote: > > As a former T-mobile HSI customer, I can attest the horrible queue manage= ment. The deprioritization was so bad that it became basically unusable, ev= en a ping or DNS lookup just lagged. I latched to a tower near the intersta= te about 2 miles away. I knew when there was an accident at the interstate,= my online experience correlated very well. I ranted over here about how bad just the depriotization could become, and also pointed to a flaw, I believe, in how long devices hold onto packets in that case. I fear that when the network is getting toasty, and comes back online that there is a thundering herd of devices, enormous backlogs of stale syns, and so on that hit it, and what is in place is some sort of timed rate limiter to deal with it, not very well. It is the only explanation I have for seeing pings come back on 1s intervals... https://blog.cerowrt.org/post/flaws_in_flent/ > This was about a year ago, I do have t-mobile for mobile service, I might= try them again since they offer it for $30. Might make an OK backup or sec= ond WAN, specially if I can just use my own modem in 5G SA. > > Meanwhile, I also have Verizon LTE home internet. Which has been a fairly= reliable ~50/3 Mbps. Something changed recently, now I get 100-160 downs 2= 0-30 up. Bloat increases linearly after 100 down, but it is not too bad. Af= ter 22 up bloat increases exponentially. Once you pass the threshold bloat will increase exponentially. >Luckily CAKE at 100/22 seems to mitigate most of the bloat. I haven=E2=80= =99t noticed a slow down. It=E2=80=99s been way more reliable (in terms of = bandwidth, latency and availability) than Starlink for me. I=E2=80=99m fair= ly rural, so that probably has to do with the better experience with Verizo= n, but the tower is like a year old and has fiber backhaul. It sure beats m= y former ADSL in price and performance. As for dealing with the variable rates on both starlink and lte: I also keep encouraging folk to participate in sebastian=C2=B4s team on the cake-autorate stuff: https://github.com/lynxthecat/cake-autorate It may be the only way to get decent latency out of both systems unless they get their act together. > If course the wireline providers are complaining since they are taking cu= stomers, many that will have a good enough experience that might save them = a good bit of money. Business quickly miss their cash cows. Their margins m= ust be very high, access should be cheap, but they love to squeeze a bit to= o much. I kind of hope to level the playing field with a set of comparisons and simulations of various real world behaviors and network measurement tools via the libreqos.io testbed. Right now this is just cake, but I have poured simulations through of pie, fq_pie, wifi, etc, and hope to find a model for lte=C2=B4s uplink behaviors that is a fit. https://payne.taht.net/circuit_queue?id=3D9 Right now, I am re-finding old bugs in cake, and have made a few optimizations... > > Competition works, I hope FWAs eat their lunch. I hope the FWAs move faster, and in particular, the new mikrotik code gets on more routers, as well as cake and fq_codel on more ISP-supplied CPE. But I really am in this to fix latency on everything... T-mobile, or Verizon would sell to a lot more home users if they got their bloat under control. It also really puzzles me that a billing company, with bandwidth caps, gives me no ability to buy more when I run out. > > On Wed, Feb 8, 2023 at 11:07 AM Sebastian Moeller via Bloat wrote: >> >> Hi, >> >> On 8 February 2023 17:13:26 CET, Dave Taht via Bloat wrote: >> >I do find it odd that the carriers deprioritized home users >> >> I would not be amazed if the priority would inversely correlate with exp= ected profit for mobile vs fwa customers and maybe with the availability of= competitive broadband access products.... >> >> (and of course, >> >insane, that they have such lousy queue management still) >> >> That is IMHO bigger issue, even with lower priority competent AQM would = result in pretty usable FWA links even if the advertised rates are not reac= hed, but I am preaching to the choire here ;) >> >> > >> >---------- Forwarded message --------- >> >From: POTs and PANs >> >Date: Wed, Feb 8, 2023 at 4:35 AM >> >Subject: [New post] How Good is FWA Wireless? >> >To: >> > >> > >> >[image: Site logo image] Doug Dawson, CCG Consulting posted: "T-Mobile = got >> >some bad news recently when the the National Advertising Division (NAD)= of >> >BBB National Programs informed T-Mobile that it could not use the words >> >=E2=80=9Cfast=E2=80=9D and =E2=80=9Creliable=E2=80=9D when advertising = for its FWA fixed wireless product >> >that it brands as T-M" POTs and PANs How >> >Good is FWA Wireless? >> > >> > >> >Doug Dawson, CCG Consulting >> > >> >Feb 8 >> > >> >T-Mobile >> >got some bad news recently when the the National Advertising Division (= NAD) >> >of BBB National Programs informed T-Mobile that it could not use the wo= rds >> >=E2=80=9Cfast=E2=80=9D and =E2=80=9Creliable=E2=80=9D when advertising = for its FWA fixed wireless product >> >that it brands as T-Mobile Home Internet. This ruling came as a result = of a >> >complaint from Comcast that T-Mobile is overstating the capabilities of= the >> >FWA product in advertising. >> > >> >Most large carriers belong to the BBB National Programs as a lower cost= way >> >of mitigating advertising disputes than lawsuits. ISPS agree to go alon= g >> >with the rulings issued by the group as a condition of joining. However= , in >> >this case, T-Mobile is appealing the decision. The news wasn=E2=80=99t = all bad for >> >T-Mobile since it was ruled that T-Mobile could continue to advertise t= hat >> >the price of FWA is =E2=80=98locked-in=E2=80=9D since the company hasn= =E2=80=99t raised its rates. >> > >> >Anybody who has looked closely at the performance of FWA wireless from >> >T-Mobile or Verizon would agree with this ruling. The main reason for t= he >> >ruling is that the performance of FWA can vary widely. It=E2=80=99s a b= roadband >> >product that connects to customers from a cell site, and the distance >> >between a customer and the cell site makes a big difference in the spee= d >> >being delivered. I talked to one customer located near to a T-Mobile to= wer >> >who was consistently getting over 200 Mbps download and was really plea= sed >> >with the product. But in this same community, customers only a mile or = so >> >away from that same cell tower were getting speeds closer to 50-100 Mbp= s >> >and were not as happy with the product. A mile further away and speeds = were >> >not good at all, and I talked to a farmer who sent the receiver back. I= n a >> >rural area, a mile isn't very far, and unless there are a lot of towers= , >> >most folks are not getting the advertised fast speeds. >> > >> >The one consistent feedback I=E2=80=99ve gotten in talking to FWA custo= mers is that >> >speeds vary. This is true for all cellular broadband, and cell phone >> >customers are used to seeing a different number of bars of broadband sp= eed >> >over time from the same location such as home or the office. Cellular d= ata >> >speeds vary for a wide variety of reasons like temperature and weather. >> > >> >But the biggest reason for the variability is the overall volume of dat= a >> >being demanded from a given cell site at a given moment. Like most >> >broadband products, cellular broadband is a shared data product where t= he >> >broadband is divvied up among the users at any given time. But unlike >> >landline broadband networks, a cellular company cannot control the numb= er >> >of users at a cell site. Since cell phones are mobile, there is no tell= ing >> >how many people might be demanding a cellular data connection at any gi= ven >> >time. >> > >> >FWA has one more limitation in that the cellular carriers have elected = to >> >give first priority to cell phones over FWA customers. This means that = when >> >a cell site gets busy, the carrier will choke the delivered data speeds= to >> >FWA customers in order to deliver the most speed possible to cellular >> >customers. This makes sense since each big T-Mobile and Verizon have >> >roughly 100 million cellular customers compared to a few million FWA >> >customers. They do not want to make cellular customers unhappy with >> >broadband speeds, and so they throttle FWA when a cell site gets busy. >> > >> >T-Mobile doesn=E2=80=99t hide this, and the throttling is discussed in = the fine >> >print when the product is advertised. But that throttling is part of th= e >> >reason that T-Mobile can=E2=80=99t describe it=E2=80=99s product as rel= iable =E2=80=93 because at >> >busy times it isn=E2=80=99t. >> > >> >The big selling point for FWA is the low price and I=E2=80=99m sure the= price is >> >what attracted urban customers. The speeds are going to be liked in rur= al >> >areas where there are no alternatives, but there is definitely a severe >> >distance limitation - in a rural area a 50 Mbps connection might be a b= ig >> >leap up in performance. But the FWA product is a lot slower than cable >> >company broadband. Households who are heavy broadband users might not l= ike >> >the slower speeds and the variability. This ruling is telling T-Mobile = that >> >it can=E2=80=99t advertise in a way that makes FWA sound like an equiva= lent >> >alternative to cable or fiber broadband, because it isn=E2=80=99t. It= =E2=80=99s going to be >> >interesting to see how T-Mobile adjusts it=E2=80=99s advertising after = this ruling. >> >Comment >> > >> >------------------------------ >> > >> >Unsubscribe >> > >> >to no longer receive posts from POTs and PANs. >> >Change your email settings at manage subscriptions. >> > >> > >> >Trouble clicking? Copy and paste this URL into your browser: >> >https://potsandpansbyccg.com/2023/02/08/how-good-is-fwa-wireless/ >> >Powered by WordPress.com >> > >> >[image: Download on the App Store] >> > >> >[image: >> >Get it on Google Play] >> > >> > >> > >> >> -- >> Sent from my Android device with K-9 Mail. Please excuse my brevity. >> _______________________________________________ >> Bloat mailing list >> Bloat@lists.bufferbloat.net >> https://lists.bufferbloat.net/listinfo/bloat --=20 This song goes out to all the folk that thought Stadia would work: https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-69813666656= 07352320-FXtz Dave T=C3=A4ht CEO, TekLibre, LLC