Apple released iOS 5.1 this week, and with it made an interesting change to iPhones that make use of the AT&T network. The new update changes the AT&T 3G logo in the upper left corner of the screen to now read as 4G when connected to HSPA+ networks. Keep in mind that no magic has occurred in the iOS device. No new radios have been added, activated, or even upgraded. Its simply a matter of semantics. AT&T wants their service to compete with other services like Sprint that currently advertise 4G wireless network access.
AT&T can call it whatever they want, 3G is a commonly understood wireless standard at this point but the definition of 4G wireless is still a topic of great debate. Does a network qualify as 4G simply by being able to attain a set base level of performance? Does a 4G network require specific hardware or must it implement certain technology? Or is a 4G network simply any 4th generation version of any wireless service? It depends entirely on whom you ask and what their company has to gain from the answer.
With all of vagary associated with 4G, only one thing really matters to the users– and that’s the speeds at which they access the internet. So, putting aside all of the jargon and getting right down to real world numbers, lets see how Sprint’s 4G wireless compares to AT&T’s “4G” wireless service.
The Sprint devices tested was a Sierra Wireless Overdrive Pro 3G/4G Mobile Hostspot that was running the latest firmware and was sitting in an office window for maximum wireless signal strength (a constant 2-3 bar signal through all tests). For the record, it was in 4G mode and did not revert to 3G mode at any point during the test. Its internet access was shared with the Mac Mini via the built-in wireless hotspot feature. The AT&T device used in the test was an Apple iPhone 4S sharing its wireless (a constant 2-3 bar signal through all tests) via the built-in hotspot functionality of iOS 5.01.
All of these are speed tests run in my office using a Mac Mini i5. All of the tests were run from my single location which had mid-level signal strengths for both the Sprint and AT&T devices. I ran each test 4 times since there were fluctuations in performance from one test to the next. And, if you look closely at the speed test results, you will notice that not all test results were run against the same test node. For these tests, I let Speedtest.net’s tool select its own preferred end point each time I ran the test. This could allow a little room for fluctuation, but in fairness, with any cell phone based internet access fluctuation is a fact of life since towers loads and throughput are in a constant state of flux.
Sprint 4G Wireless:
AT&T HSPA+ 4G Wireless:
And now for the take-away’s. While the Sprint service did seem to have a few spirts where to returned overall better downstream numbers, the AT&T scores were more consistent from one test to the next. Additionally, the AT&T scores returned substantially better numbers on the upstream tests.
Any test of this sort is somewhat antidotal in the end. These scores will vary depending on location given relative to a carrier’s tower as well as the network capacity at that location. But in this case, despite the fact that I personally think that AT&T is stretching its luck by relabeling a 3G phone as 4G, it does seem that at least as far as the wireless performance is concerned the revision to its marketing might actually live up to its implied performance.
One additional data point that some might find interesting. I did a similar test back in 2010 testing another Sprint wireless hotspot against an iPhone 3GS jailbroken to offer hotspot functionality. At the time, Sprint had just announced 4G access in my area and the iPhone was 3G HSPA. Here are the test’s from 2010.
Sprint 4G in 2010:
AT&T 3G HSPA in 2010: