3 called a bunch of journos over to the Groucho this morning for a breakfast briefing about mobile broadband speeds, why 3 takes the right approach in advertising these, and why Vodafone (who 3 has hauled off to the ASA and GSMA on this point) does not.
A moot point, as it turns out, because Vodafone has apparently said it will stop advertising 7.2Mbps nationally when that speed's only available in 18 London postcodes and some airports (although it is investing to extend that reach quite a bit).
Anyway, the meeting did provide a useful primer as to why, as with fixed-line broadband, you should never expect to get those 7.2Mbps or 3.6Mbps headline speeds with mobile broadband. First off, here's those maximum line rates vs the maximum possible user rates that can be attained after the basic overhead is subtracted (all in Mbps obviously) - and remember, this is assuming you're right next to the base station in perfect environmental conditions:
14.4 / 12.91
7.2 / 6.69
3.6 / 3.42
1.8 / 1.71
Then come the other factors, each chipping away at that data rate:
- Non-mobile-broadband 3G traffic (voice, video, SMS etc)
- Applications being used (browsing, streaming, downloading etc)
- Signal quality (interference, distance from base station, signal strength, trees etc)
- Technology being used (device capability, network capability, network configuration)
- Backhaul capacity on the operator's side
- Number of mobile broadband users in a cell (although this can be mitigated by splitting the cell, using second and third frequencies, or adding more base stations)
Another interesting point that was made - and I must get the graph of this - is that, once you're around 500m from the nearest base station, it doesn't really matter much whether you're on a 3.6Mbps or 7.2Mbps connection. You'll get much of a muchness because, of course, the higher the bandwidth the steeper the falloff in propogation over distance.
3's take on this is that you shouldn't worry so much about high numbers (most mobile broadband applications only require a downlink of 1Mbps or less) and focus rather on density and reach of coverage - 3's speciality as it has the most comprehensive 3G network in the UK (especially as its network-sharing agreement with T-Mobile comes to fruition).
All very impressive, but we're not just streaming from YouTube, we're increasingly sending videos to YouTube, and try doing that without some high-number standard like HSUPA. 3 says it's rolling that out from July, which is good, but it will take a while to get significant coverage on that.
All fair enough. I can't see the point in worrying about or advertising high numbers and standard names - what consumer really knows or cares what those means? It just needs to work properly. Funnily enough, it was Vodafone that first started going on about "mobile broadband", in an attempt to get away from alphabetic jargon. Let's hope they, and everyone else, stay away.