Despite a recognition that customers are using more data than ever as they cut traditional cable television in favor of streaming, Comcast’s data cap remains stubbornly fixed at 1 TB a month.

The nation’s largest cable operator last week completed a significant speed upgrade in 14 states in its Northeast and Mid-Atlantic service areas from Maine to Virginia. Some plans are getting as much as a 60% speed boost, but Comcast is not budging a megabyte on its fixed data cap that amounts to 1,000 GB of usage per month.

Comcast acknowledges the speed upgrades are designed to meet the exponential increase in demand for high-bandwidth video streaming in households that now average at least ten devices connected to the internet. Many of those devices are now streaming 4K video, which takes double the bandwidth of traditional HD video.

The speed upgrades:

• Performance Starter: was 15 Mbps, now 25 Mbps
• Performance: upgrades from 60 Mbps to 100 Mbps
• Performance Pro: up from 150 Mbps to 200 Mbps
• Blast: A slight upgrade from 250 Mbps to 300 Mbps
• Extreme: This premium plan used to provide 400 Mbps, but it is now 600 Mbps.

Comcast faces significant competition in this part of the country from Verizon’s FiOS fiber to the home network. That may explain why it is also the only significant part of Comcast’s service area that remains exempted from the cable company’s data caps. Verizon has no data cap of consequence, although the company has shut down some customers that were likely using their residential internet connection as a server, running up many terabytes of usage a month.

For now, Comcast’s speed upgrades come with no price hike.

The speed increases are likely to be welcomed by most customers, but Comcast’s pervasive data cap for most of its nationwide footprint is not. In November, that data cap will be tested like never before as Google launches its Stadia cloud-based video game service. Up to six million broadband customers are expected to blow through their provider’s monthly data cap while using the service, which replaces traditional home game consoles. That is because Stadia will consume an enormous amount of bandwidth — as much as 15.75 GB an hour at 4K resolution.

An article published by Vice Media warned video game enthusiasts they could easily face steep overlimit usage penalties on a future bill:

According to data from The NPD Group, America’s estimated 34 million gamers play 22 hours per week on average. Were those gamers to all shift to Stadia as their primary game platform at 4K, they’d burn through 1,386 GB of data monthly. And that’s just the bandwidth consumed by gaming; it doesn’t include music and video streaming or other activities.

The result will be an even higher broadband bill for US consumers who already pay some of the highest prices in the developed world for bandwidth. For many this will be a surprise. Of the 943 gamers surveyed by the company, only 17 percent were certain they had a broadband cap. 21 percent say they weren’t sure one way or the other whether their broadband was metered.

Most providers set their overlimit penalty at $10 per 50 GB of excess usage. Some offer to waive data caps for a monthly additional charge of $50. That makes Google’s $10 video game service much less of a bargain than many initially thought.

When questioned about the impact data caps could have on Stadia, Google vice president Phil Harrison hoped the nation’s ISPs would do the right thing by their customers.

“ISPs have a strong history of staying ahead of consumer trends and if you look at the history of data caps in those small number of markets…the trend over time, when music streaming and download became popular, especially in the early days when it was not necessarily legitimate, data caps moved up,” he said. “Then with the evolution of TV and film streaming, data caps moved up, and we expect that will continue to be the case.”

Except Harrison’s utopian world view is not accurate. In fact, most broadband providers have set data caps and left them unchanged for years, even as those same companies promote frequent speed upgrades. In effect, more and more customers are running over their usage allowances and either paying steep penalties, reducing usage, or agreeing to pay another $50 a month to dispense with the cap altogether.

Vice author Karl Bode reminds readers “broadband caps are complete nonsense.”

“Experts say the real purpose of such limits is to covertly jack up your already expensive broadband bill—and punish customers looking to cut the cord on traditional cable TV services,” Bode added.

Correction: Data cap expressed at 1,000 MB changed to 1,000 GB to reflect the correct allowance.

Frontier residential customers in Nevada could receive a refund and improved service after a court filing from the Nevada Attorney General’s Bureau of Consumer Protection (BCP) found Frontier’s internet services lacking.

Since 2017, BCP has collected scores of complaints about Frontier’s internet service and its performance, mostly regarding slow service, frequent outages, and ongoing billing problems.

The BCP found Frontier liable under NRS Chapter 598 which forbids providers from misleading consumers about internet speed and service performance in marketing and advertising. An Assurance of Discontinuance filed with the court allowed Frontier to settle while avoiding admitting any wrongdoing and agreeing to correct service deficiencies.

The state found Frontier repeatedly did not disclose limitations of broadband service availability and knowingly marketed its DSL service at speeds the company could not provide customers.

According to the court document:

• Frontier is required to “clearly and conspicuously” disclose in its print and broadcast advertising the actual internet speeds available to customers in terms of minimum and maximum speed.
• Customers that sign up for a high-speed plan that Frontier cannot provide may switch to a lower speed plan or discontinue service incurring no penalties or early cancellation fees.
• Existing customers that do not receive at least 90% of the highest speed their current plan advertises will receive a service credit of 50% of the internet charge for each month Frontier did not provide such speed. Credits will begin in 2020 and end three years after the date the court accepts the Assurance.
• Frontier has also agreed to invest at least $1 million to improve internet service in Elko County.

Charter Communications is envisioning building out a rural fixed wireless network on the edges of its existing service areas in rural parts of New York and North Carolina to attract new customers without spending money on extending its hybrid fiber-coax (HFC) network to high-cost areas.

Charter has spent more than a year conducting mobility and fixed wireless tests using small cells in several cities across the country to determine if the technology is commercially viable. The company is focusing on two service scenarios: rural areas within a mile or two of its existing cable footprint and urban and suburban areas already served by Spectrum’s HFC network.

Charter’s rural initiative uses the Citizens Band Radio Service (CBRS) band at 3.5 GHz to provide rural fixed wireless service to areas just out of reach of its cable network. Trials of fixed wireless service are already underway or will be soon in exurban and rural areas near Denver, Tampa, Bakesfield, Calif., Coldwater, Mich., and Lexington, Ky. These first trials were designed to prove the concept of delivering high-speed fixed wireless internet in different areas of the country. In 2020, additional trials are planned for rural parts of New York and North Carolina, with a tentative plan to launch service that same year.

“Results of these trials have been promising as we were seeing speeds that significantly exceed the FCC’s definition of high speed broadband in most circumstances which would allow for video streaming and the use of multiple apps simultaneously,” Charter wrote on its Policy Blog. “We believe fixed wireless access technologies using this mid-band spectrum could offer a cost-effective solution for providing broadband service to homes and businesses in harder to reach rural areas.”

The next step for Charter is a full service trial in rural counties in New York and North Carolina that would offer high-speed wireless broadband to residential customers. Charter began testing its fixed wireless service in Davidson County, N.C. roughly between the communities of Lexington and Salisbury. Each of Charter’s four temporary transmitting locations in Davidson County are licensed to serve a radius of up to 9.3 miles, but most customers are significantly closer to the transmitting sites. Participants get free service for the duration of the trial, a free outdoor antenna and a free combination receiver/router. All equipment remains the property of Charter and is to be returned at the end of the trial.

Charter told attendees at last week’s SCTE/ISBE Cable-Tec Expo in New Orleans that results exceeded performance expectations. Customers are getting in excess of 25/3 Mbps service, and there is enough bandwidth left over for Charter to consider offering a true wireless triple play package of video, internet, and home phone service.

Craig Cowden, Charter’s senior vice president of wireless technology, told attendees Charter envisions CBRS wireless service to extend the Spectrum cable footprint into rural areas just outside of the cable company’s wired footprint, and a good economic case might be possible to offer service to residents that usually fail the company’s Return On Investment test that governs whether Charter will extend wired service into unserved neighborhoods within their franchise area.

But Cowden also sees Charter deploying CBRS in urban and suburban areas to handle wireless traffic for a growing number of its wireless customers. Spectrum Mobile relies on free Wi-Fi networks and an agreement with Verizon Wireless to provide 4G LTE connectivity for its customers. Charter can begin reducing costs by moving mobile traffic off of Verizon’s network and onto Charter’s own mobile network, likely operating on CBRS frequencies.

The CBRS band is suitable for outdoor traffic, but is likely not going to work well when customers go indoors. Charter plans to hand that traffic back to its extensive network of Wi-Fi hotspots, mostly located at businesses using Spectrum’s commercial service, and the customer’s own in-home Wi-Fi.

Charter has been testing its mobile CBRS service from test transmitters in Tampa and Charlotte, N.C., but plans a much more extensive test in New York and Los Angeles utilizing more than 250 cell sites.

In 2017 and 2018, Charter also filed requests for special temporary authority to test 5G service in the 28 GHz millimeter wave band, but those tests appear to be exploratory and there is no indication a commercial deployment effort is forthcoming soon.

Charter’s Experimental CBRS Projects (based on filings with the FCC for experimental and permanent licenses)

Lexington, Kentucky

WM9LXR was licensed on March 23, 2018 and a CBRS transmitter capable of reaching up to a radius of 9.3 miles was placed on top of the Fairfield Inn & Suites by Marriott Lexington North at 2100 Hackney Place in Lexington. The license expired Sept. 19, 2018. A new application to operate this transmitter was filed Nov. 16, 2018 expiring June 4, 2019.

Centennial, Colorado

WM9XTL was licensed on June 1, 2018 and a CBRS transmitter capable of reaching up to 15 miles away was erected just northeast of the Centennial Airport along E. Easter Avenue. This transmitter was designed to experiment with mobile CBRS services. The license expired Dec. 5, 2018.

Another experimental license to test CBRS service was sought Nov. 16, 2018 and expired June 4, 2019.

A license to operate WO9XOY was filed on May 10, 2019 to experiment with a private fixed wireless LTE network in the CBRS band for a corporate client from the same transmitter location as above. The license would expire Dec. 2, 2019.

Los Angeles

WM9XXU was licensed on June 22, 2018 to test CBRS mobile service from four transmitting sites around Baird Park, Van Nuys, Baldwin Hills, and West Anaheim Junction areas. The license expired Dec. 22, 2018.

An application to operate WN9XRT was filed with the FCC on Nov. 16, 2018. CBRS transmitters would operate from the same neighborhoods as above to conduct outdoor and indoor fixed wireless mobile testing within 8 miles of the four fixed locations until Dec. 22, 2018.

An application to run WO9XQW on an experimental basis was filed May 31, 2019 to expire Dec. 19, 2019. The license application described the CBRS test project:

Charter will deploy experimental fixed and mobile equipment in various configurations. Depending on the testing scenario, devices will be deployed on existing aerial cable strand, on existing buildings/poles or indoors.

Specifically, Charter will use the following deployment approaches:

1. Strand mount deployment: 118ft. height.
2. Building/pole mount deployment: up to 100ft. height.
3. Indoors: up to 40ft. height (3rd floor indoor).

New York

WM9XXV was licensed on June 22, 2018 to test various CBRS applications from three transmitter sites:

125th Street & Rockaway Blvd. Jamaica
72nd Street Flushing
South Beach, Staten Island

The license expired Dec. 22, 2018.

An application for WN9XRS was filed with the FCC on Nov. 16, 2018 to expire Dec. 23, 2018 to test CBRS services from the three locations noted above. On May 31, 2019, another application was filed to continue testing until Dec. 19, 2019.

Charlotte, North Carolina

A pending application filed Aug. 28, 2019 for WN9XHY, a CBRS transmitter located on S. Caldwell Street next to Spectrum Center was filed on Aug. 28, 2018. Charter sought to cover a radius of just over 9 miles to test fixed and mobile applications with an expiration of March 16, 2019.

An application for WO9XCX was filed on March 15, 2019 set to expire Sept. 29, 2019. This is a CBRS experimental project to test indoor and outdoor fixed and mobile wireless reception from two fixed transmitter locations located at Spectrum Center and the Clanton Park/Roseland neighborhood. An application for an additional experimental license was filed March 15, 2019 with an operational end date of Sep. 28, 2019.

Tampa, Florida

An application for WN9XHZ, a CBRS transmitter covering up to 8 miles from Ybor Heights was sought on Aug. 28, 2018 to expire March 16, 2019. It was to test fixed and mobile CBRS applications.

Keystone, Iowa

A license to operate WN9XIX from a mobile transmitter van was filed Sept. 6, 2018 to expire March 30, 2019. An additional application to operate a similar CBRS test project was filed Sep. 17, 2019 and set to expire March 28, 2020. On Sep. 20, 2019 an application was filed to operate WP9XIC until March 29, 2020. This latter project is designed “to evaluate 5G frequencies and technologies for their use in point-to-multipoint access network capacity (e.g., rate versus range) and data throughput. The proposed operations will advance Charter’s understanding of technology and network potential using mid-band spectrum and will advance the potential deployment of fixed and mobile 5G services.”

Bowling Green (and Lake Wales), Florida

A license application filed Nov. 28, 2018 proposed to test wireless service in the so-called C-Band spectrum now used by satellites to check how well it performs with the potential of interference from licensed satellite TV services. Outdoor-only tests of wireless service within a two-mile radius of fixed transmitter locations in the vicinity of Bowling Green and Lake Wales were underway until the license for WN9XSQ expired June 10, 2019.

An additional license to further test potential C-Band spectrum for interference issues was sought to begin Dec. 12, 2018 and expiring June 10, 2019.

Davidson County, North Carolina

Charter applied for an ongoing license to operate WJ2XZT, a CBRS project consisting of four transmitters each serving a radius of approximately nine miles, to provide fixed wireless service to customers in this part of rural North Carolina. The transmitters are located at three locations:

153 Sigmon Road, Lexington
185 Chestnut Grove Church Road, Lexington
784 Mount Carmel Road, Lexington

Park City, Utah

On July 3, 2019 the company applied for WK2XIP, a new one-year experimental project:

“As part of its efforts to lead the industry in broadband innovation, Charter intends to conduct fixed wireless experiments in the 3550-3700 MHz band. The proposed operations will advance Charter’s understanding of 5G technology and network potential in mid-band spectrum and will advance the potential deployment of 5G fixed and mobile services.

“Charter will conduct the proposed test using antennas at a location in the Park City, Utah area. These experiments will evaluate the 3550-3700 MHz frequencies and 5G technologies for their use in real-time communications in a low-latency environment.

“The tests will utilize fixed transmitters with a 2km or smaller effective radius. The antennas will be mounted on a hydraulic mast attached to a mobile trailer, which will be located at the requested test location. The radios will be pointed towards the side of the mountain, the peak of which is higher than the peak height of the mast. The trailer mast can be raised to 10.4 meters.”

Colorado Springs, Colorado

An experimental license for WO9XXJ was filed July 18, 2019 to test a millimeter wave 5G network in the 37 GHz band. The license expires Jan. 28, 2020.

Verizon and T-Mobile are redlining their up and coming 5G wireless services to target wealthy neighborhoods and business districts while shunning the urban poor.

Dave Burstein examined the coverage maps of both carriers in cities like Manhattan and found a distinction in the service available in wealthy southern Manhattan and what upper Manhattan neighborhoods including Harlem and the mostly Latino Washington Heights are getting. For both companies, 5G is not much of a priority for Brooklyn either.

“I do not think T-Mobile specifically intended to exclude people of color, but that seems to be the practical effect,” Burstein wrote.

Ronan Dunne, executive vice president & group CEO of Verizon Consumer confirmed that Verizon will be targeting 5G service to areas where it makes the most economic sense. He said that more than half of Verizon Wireless customers will continue to get 4G LTE-like speeds, with the rest eventually upgraded to 5G service.

“So we’ve taken a very clear view that we want to have both a coverage strategy and a capability strategy. And a very large majority of the volume of data that we carry on our networks goes to large, dense urban environments,” Dunne told investors recently. “So from a population point of view, it’ll be significantly less than half of the customers [getting 5G]. But from a data traffic point of view, it’s significantly more than half. So when it comes to the ability to use 5G as a significant capacity enhancement, there’s more of an opportunity to leverage that in the urban areas.”

In other words, Verizon plans to target population dense urban areas for 5G service the most, because that is where most of its data-loving customers live and where they network’s cost effectiveness may be the highest. Although the geographic coverage of 5G will seem relatively small, the population density of areas targeted for 5G service is not.

Verizon has been touting its forthcoming nationwide 5G network, but Dunne has hinted to investors that the devil will be in the details. Not every customer will have access to Verizon’s super fast millimeter wave 5G service. In fact, at least half the country will be serviced by existing 4G LTE cell towers upgraded to provide 5G service on lower frequencies capable of reaching far beyond the coverage area offered by millimeter wave service. But that will also mean a much larger number of customers will share the same 5G network connection, potentially dramatically reducing speed and performance. Dunne said the performance of this type of 5G service “will approximate a good 4G service.”

Burstein notes in real terms, this will mean a significant difference in network speed. Verizon’s millimeter wave service will be capable of delivering 1-2 Gbps, while Verizon’s 5G upgrade of its existing 4G cell towers will deliver speeds in the low hundreds of megabits per second, potentially even slower on crowded cell sites.

For over a year, the cable industry has been stalled after deciding to slash investment in broadband while enduring indecision and confusion over the next generation of cable broadband.

At issue is a simmering disagreement — rare for the usually unified cable industry — about the next generation of cable broadband, dubbed DOCSIS 4.0.

Two sides have emerged. Cable giant Comcast has spent years gradually preparing its network for perhaps the last iteration of coaxial copper-delivered cable internet service. It has spent at least five years gradually pushing optical fiber closer to its customers, retiring additional coaxial cable and the amplifiers and other equipment associated with that technology. The result is a company ready to embrace Full Duplex DOCSIS, known as “FDX.”

FDX is designed to allow upload and download traffic to share the same spectrum, letting cable companies put internet service bandwidth to full use with maximum efficiency. Comcast wants FDX to be a central part of DOCSIS 4.0. The company has been working through a long-term plan to offer much faster internet service, including symmetrical broadband — unified upload and download speeds. This would erase the cable industry’s broadband Achilles’ heel: download speeds much faster than upload speeds.

To achieve FDX, cable companies have to push fiber much deeper into their networks, sometimes right up to the edge of neighborhoods. It also means eliminating signal amplifiers that help keep signals robust as they travel across older coaxial cable infrastructure. Engineers call this concept “Node+0” architecture, which means a network with zero amplifiers.

FDX gives the cable industry the opportunity of running a more robust broadband network, easily capable of 10 Gbps with an upgrade path to 25 Gbps later on. The downside is that it can be very expensive to implement, especially if a cable company has under invested in upgrades and not incrementally laid a foundation for FDX. Wall Street may balk at the upgrade costs. The logistics of readying degrading older infrastructure to launch FDX may be so onerous, some cable systems may find it more cost effective to scrap their existing hybrid fiber-coaxial (HFC) networks and switch to a state-of-the-art fiber to the home network instead. That is precisely what Altice USA is doing with its Cablevision/Optimum system in New York, New Jersey, and Connecticut.

Charter Communications, along with many other smaller cable operators, have been pushing an alternative to FDX that is likely to cost much less to implement. Extended Spectrum DOCSIS (ESD) is designed to work over existing cable systems, including those that still rely on amplifiers and aging coaxial cable. Instead of allowing internet traffic to share bandwidth, ESD follows the existing standard by keeping upload traffic on different frequencies than download traffic. It simply extends the amount of bandwidth open to both types of traffic, which will allow cable systems to raise speeds. ESD will dedicate frequencies up to 3 GHz (and higher in some cases) for internet traffic. DOCSIS 3.1, the current standard, only supports internet traffic on frequencies up to around 1.2 GHz. ESD will also allow cable companies to raise upload speeds and should support up to 10 Gbps downloads. But there are some questions about how well ESD will support 25 Gbps speed and the condition of the cable company’s existing coaxial network will matter a lot more than ever before. A substandard network will cause significant speed degradation and could even disrupt service in some cases.

Despite the limitations of ESD, many cable companies consider its low implementation cost a principal reason to support it over FDX.

For much of this year, cable companies have put upgrades on hold as the industry sorts out which direction DOCSIS 4.0 will take. Equipment manufacturers and vendors have resorted to layoffs and cutbacks and have signaled neither Comcast nor other cable companies are big enough to justify different DOCSIS standards supporting FDX or ESD.

Therefore, the cable industry has informally decided DOCSIS 4.0 will need to support both FDX and ESD under a single specification, with next generation cable modems and equipment capable of supporting either technology. At a joint pre-Cable-Tex Expo conference held on Monday, executives from Comcast and Charter appeared to support the new unified approach to DOCSIS 4.0.

John Williams, vice president of outside plant engineering and architecture at Charter Communications, told attendees cable companies need to support both FDX and ESD and stop taking an “either/or” approach.

“In order to do this, we need to look at the synergies and embrace ESD and FDX as the next generation of HFC,” Williams said. “It’s all about scale.”

Charter has been significantly challenged historically because its own legacy cable systems were often behind the times and sometimes dilapidated. Its 2016 acquisition of Time Warner Cable and Bright House Networks only complicated things further, because neither operator had a reputation for using state-of-the-art HFC technology. Costly upgrades have been underway at many Charter-owned cable systems since the merger closed, some still ongoing.

Robert Howald, part of Comcast’s network upgrade team, called the emerging DOCSIS 4.0 standard a “perfect complementary pair” of FDC and ESD. He noted both approaches will allow cable systems to boost speeds to at least 10/10 Gbps, with faster speeds in the future.

Howald pointed out Comcast is already testing FDX technology in Connecticut and Colorado, working out bugs and unexpected technical challenges.

“We feel like we’ve significantly de-risked some of the technology components of FDX,” Howald said. “We felt really good about what we saw in the field.”

What is Full Duplex DOCSIS? This video from CableLabs explains the technology and how it differs from other DOCSIS cable broadband technology. (1:58)