I’ve recently seen that the cable industry is working towards a solution that will provide what they call “full-duplex,” or, what the rest of the industry calls, “symmetrical” data speeds. Currently cable networks are capable of download speeds of up to a gigabit, but the technology supports relatively tiny upload speeds.
It looks like the solution for full-duplex won’t be cheap. Cisco says that their solution will require having additional fiber to the last active component in the network, which either means more fiber to nodes, or in some cases even past the nodes. It’s also going to require setting aside a few more empty channels from the cable network unless the duplex operates by cannibalizing the downstream data.
This raises the question of how important upload speeds are. Are they important enough for big cable companies to implement an expensive network upgrade? Already today cable companies have built (or plan to build) fiber to larger businesses or to sizable business communities. The DOCSIS cable networks have been unacceptable for business broadband for many years, except perhaps to smaller businesses that use the web in ways similar to home users. Most of the large cable companies have also begun building fiber to greenfield communities rather than extend their coaxial network.
But there is no doubt that upload speeds are vital to businesses. There are huge numbers of businesses served with cable modems today that would benefit from faster upload speeds. The case for symmetrical residential speeds is harder to make.
There are only a few types of residential customers that need fast upload speeds. One is gamers. They don’t necessarily need superfast speeds, but everything I read shows that they love speeds somewhere between 25 – 50 Mbps. There are also lots of folks that work from home that need faster broadband. For example, I know several photographers who send out massive files of pictures and videos to customers and who struggle if they are on slow upload broadband.
In my work I sometimes send fairly large files and attachments. And yet, except for those few times when somebody is on the phone waiting for the files while we talk, I’ve never much cared if it takes a little longer to send files. I’ve always figured that’s how most people feel. One of the services we offer at CCG is conducting consumer surveys and I’ve never reviewed a survey that showed a big consumer demand for fast upload speeds – most survey respondents say that it doesn’t matter.
Many of you probably suppose that once somebody buys fiber broadband that they get blazingly fast upload speeds. But when I look at my small clients that is not the case. I would guess – without sitting and counting – that 60% to 70% of my clients with fiber networks do not offer symmetrical data speeds.
But there is obviously a lot of marketing advantage in offering symmetrical speeds. Verizon FiOS converted all of their products to symmetrical speeds late in 2014. CenturyLink built fiber past a million homes this year and is offering symmetrical data speeds. And of course, Google Fiber made the huge splash a few years ago by offering a symmetrical gigabit product for $70. The first symmetrical data products I can remember were from municipal providers like the ones in Chattanooga and Lafayette.
It will be interesting to watch to see if the cable industry decides to implement full-duplex. By doing so they might be able to wipe out the perceived advantage that fiber has in the marketplace today. I would think today that a lot of consumers would view a 100/100 Mbps product as superior to a 100/25 Mbps product even if they never use the upload capabilities. And perhaps it is that marketing perception that most matters, and maybe that is what will drive the cable companies to make the investment, at least in the markets where they have competition.