When Tarana released the new G2 generation of radios, one of the claims the company made was that the radios are powerful enough for ISPs to bring point-to-multipoint broadband to metropolitan areas.
The new specifications support the premise. The new G2 radio can support up to 512 customers in a sector (2,048 customers in total). Each sector can accept up to 6.4 Gbps of backhaul bandwidth. If all customers are provided with a gigabit product, the oversubscription ratio would be 20:1. That’s about double the ratio for current fiber technology, but many ISPs would find this to be acceptable. If a WISP were to install a Tarana radio on an existing tall urban tower, it should easily be able to see a lot more than 2,000 homes and businesses.
Selling fixed wireless in an urban area is an interesting marketing plan because there are a lot of locations that won’t be reachable from a single radio site. Rural areas have dead zones created for wireless technologies by hills and other impediments. In a city, the dead zones are more pronounced but localized. There is dead zone created behind every tall building. A home next to a taller neighbor might not be able to see a tower directly. This can be solved to some extent by using multiple towers, but tower real estate in many cities is hard to come by. Many towers are already loaded to capacity or are reserving any remaining capacity for the giant cell companies.
But a WISP could never serve everybody in an urban environment. There are nowhere near enough tall towers in a typical city or suburb to serve more than a relatively small percentage of households. A WISP marketing plan will shoot to fill all of the slots on a radio. However, there is a proceeding at the FCC that is considering overriding local restrictions on tower placement, which could be a huge boon to WISPs and cellular FWA providers.
This is an intriguing idea that would bring yet another competitive option to cities. There is already a price war of sorts for gigabit broadband in cities. For example, in highly competitive markets, customers can buy a gigabit from AT&T for $65 per month, with even lower introductory prices for new customers. Cable companies are falling all over themselves with low rates to keep customers from churning. It seems like a WISP that doesn’t have to support a wired network ought to be able to be price-competitive.
Wireless technology also brings a new option for redundancy for businesses. A growing percentage of businesses are willing to pay for two diversely routed ISPs to make sure they can stay connected to broadband. A wireless backup provides a lot more safety as a backup connection since local events like a downed pole might knock out both fiber and the cable company.
It’s an interesting option that brings a real wireless option for many urban homes. The FCC broadband map shows that I have eight choices of broadband at my house, but most of them are wireless carriers that can’t actually serve me with acceptable broadband. I’ve never understood why WISPs claim coverage across whole cities. When I look at detailed Ookla speed tests, most cities show barely any actual use from the WISPs. There are exceptions, but the current FCC maps are highly exaggerated for cities. But the FCC has no incentive to clean up the maps when every home in a city has one or two legitimate options for gigabit broadband.
The Tarana announcement has to worry the big cable companies that are already struggling to hang on to every customer. Losing even just a few thousand customers in every urban market to WISPs presents a new competitive worry.
overlooking a couple of things.
1- price often matters more than anything. a wireless solution can be much cheaper to deploy and so can cost much less…. Tarana is ‘expensive’ but can beat incumbent pricing.
1b- in areas that aren’t getting BEAD or other gov funds, prices are going up.
2- most people really don’t need gigabit or anything close to it. The 20:1 mentioned is based on lets call it ‘advertised price’, not practical use. Typical home uses under 10Mbps even when they have >100Mbps services so that 6.4Gbps is better than 1:1 for actual use. I’m not comparing to fiber which has a better radio, just saying the Tarana client count to typical use is a favorable number.
3- deployment time of wireless is a fraction. Often if you see someone building Tarana in your neighborhood you can probably get hooked up the next week. That’s an advantage over fiber.
4- non-destructive. You’ll struggle to find a community group where fiber is being run that people aren’t upset by the contractors that tear everything up.
5- a few wireless models/designs actually have ‘infinite reuse’ Tarana for example can have the same channels on every single ‘AP’ (BN in Tarana jargon) on every single tower over and over again. G2 has a ~$90k pricetag for a 360 coverage site (A fraction of ‘5G’) and probably another say $900k for infrastructure, fiber runs, or 10-20Gbps microwave link. etc. $200k just to make sure we don’t understate it. And you can repeat that over and over again. It actually scales extremely well. further, the g2 product supports redundnat operation much like a cell service so if a tower goes down, the client may link up to an alternate site.
5b- mmWave products also allow this, but they are shorter range and all line-of-site. technically much more capacity but there’s some balance of tower/site density that has prevented mmWave products from taking over.
6- looping back to funding. Many fiber (and coax, old DSL vendors) ISPs these days simply do not build out without gettting government funding. All those gaps in cities with poor coverage don’t get enough money to justify the new fiber build and so get left behind. It’s one of many faults in the model.
I do run 3 separate mmWave products, and Tarana in 3 and 6Ghz (g1, not g2), as well as some fiber.
“2- most people really don’t need gigabit or anything close to it. The 20:1 mentioned is based on lets call it ‘advertised price’, not practical use. Typical home uses under 10Mbps.”
This blog has for years posted surveys showing the growth of residential bandwidth consumption that would call that number into question.
Statistics are the parents of lies, their eager children. It is very very difficult to not produce bias when looking for data that you want to find.
I don’t think Daniel would disagree that while a typical home shows a 10 Mbps usage, actually capping them to a max of 10 Mbps would very likely be felt in day to day usage. But the point is to push back on the gigabit or bust push.
Daniel and I both operate ISP’s serving thousands of individual people. Our majority plan right now, by far (70%) is 25 Mbps download speed. Our next plan up is 100 Mbps for a 10% increase in price. I would love to get those clients moved to that plan. 99% of up-sell efforts fail. The clients, who are streaming Netflix etc every night, say nothing is broke, why would I pay more.
But, to just put the shoe on the other foot and apply my opening statement to my view of this situation, we are certainly building our network to support 100 Mbps connections as the standard, basic service. At some point I expect we’ll just migrate the 25 Mbps to a 100 Mbps without changing the price. That is to me a rational increase, coming from several years ago. But to say that 1Gbps is where we need to be is crazy. The data 100% doesn’t show that. We sell 200 and 300 Mbps packages also. I can count on one hand how many of those we’ve sold.
capping to 10M would definitely be detrimental, but PtMP oversell works on stats. 100 users locked at 10M each are unhappy while 100 users on a QoE’d 1G are pretty happy. The higher the peak rates on the AP the happier people are even with a larger oversell. Not saying you should oversell like crazy, but if it’s just price and customer satisfaction that matters and not published speeds then oversell isn’t really an important factor.
I can mirror @trendaltoews struggles with upgrades. Sometimes we offer upgrades of just an equipment fee and no montly increase and still get shot down. We just offered an airmax to wave upgrade for $199 that bumps from 30×10 to 300×30 for the same price and get <10% takers.
It's so much about price. The advertised speed is only when they're shopping around, but once that have acceptable service there's just no want to spend more for no 'perceived' gain.
the squeeky wheel is the gamer that needs 1-2Gbps to download their 200GB game update once a week in a reasonable time. That's like 5% of the userbase maybe.