A Must Read 3G Test Report



We’ve mentioned time and again the three most important things to consider for 3G service: location, location, location. Consumers really only have carrier coverage maps to go on when checking service in various locations, though. You can’t easily purchase multiple 3G devices and service plans to run around for testing purposes. OK, maybe you can, but it’s not a cost-effective method. For that reason, we point out 3G test results whenever we find them, provided the method is a smart one. We’ve seen a fairly decent one out of Gizmodo and a reasonable attempt from Engadget, but I just read one that tops them both.

PC World spent time in March and April in 13 different U.S. cities, and they’ve put together one of the best test reports I’ve ever seen. Aside from basic speed tests both up and down, they also attempt to measure the reliability of the network. PC World defines reliability as a the percentage of one-minute tests where the signal was available, wasn’t interrupted, and was faster than dial-up speeds. Over 5,440 individual tests were performed at 283 unique testing locations from parked vehicles in the 13 cities to create the data. With that many data points and a solid method, you can see why I’m so impressed by the effort.

I highly recommend you read the entire report, but here’s a summary, which probably won’t surprise most folks already familiar with 3G services:

  • Verizon Wireless (s VZ) offered the best speeds in the Eastern section of the U.S.
  • AT&T (s T) owned the Central region in terms of speeds.
  • Sprint (s S) cleaned up on the throughput in the West.
  • AT&T was found to be most unreliable network by comparison.
  • Verizon offered a mix of speed and reliability, while Sprint was found to be the most reliable.

PC World also hits a key point that we’ve reiterated many times, although I’ll admit they do a better job of it. While you might have a full-strength wireless signal, you still might not get the best wireless throughput. The backhaul, or wired connection from the local cell to the web, might be congested or too small of a pipe to take advantage of your super signal. The report covers that by looking at what the signal strength indicator meant for their testing. More bars doesn’t always equate faster wireless service, a point well worth remembering.

When considering any wireless service, the only test that matters is one that’s relevant and specific to your needs. As a result, you’re not likely going to read a customized test for you. Your location, the amount of other data customers on the same network in your area, the backhaul infrastructure near you, and other factors all come into play. So my caveat is: Consider these types of reports a general indicator to help you decide what you need and who you should use for your service.



It certainly is location. On my iPhone, I get less than 1Mbps download at my office ‘in the city’ but about twice that at home ‘in the country’. Pretty good either way when you consider what we had just a few years ago.

Kevin Hall

Just to add to and agree with your ‘location’ comment, it becomes pretty important as you move into the lower coverage areas (away from big cities). Looking at benchmarks for a big city somewhat close by may not give any indication which carrier is best for where you are actually using the phone/modem. Each carriers placement and number of cell towers can greatly impact signal and usability of that signal.

Comments are closed.