Unlimited gigabits are coming to your phone. Over the past year, all four major US carriers announced quasi-unlimited usage plans, along with plans to boost their 4G LTE networks to previously unheard-of gigabit speeds. One carrier, AT&T, has gotten so high on the new 4G that it's starting to call it 5G.

Gigabit LTE isn't 5G. (Proper 5G will involve different radio technology and spectrum.) But gigabit LTE is faster than our mobile networks have ever been before. So it's time for us to head out on the roads and test to see where we can find it, for you.

Starting today, we're putting drivers on the road headed to 30 major US cities for our eighth annual Fastest Mobile Networks test. FMN is by far the biggest test of its kind done by a journalistic organization. We test all four major US carriers, taking tens of thousands of samples all across the country.

This year, we're using Samsung's Galaxy S8, which has the most advanced modem available: 4x4 MIMO, 256QAM and 3x carrier aggregation on AT&T, T-Mobile, and Verizon, and High Performance User Equipment (HPUE) on Sprint. AT&T dubbed the Galaxy S8 its first "5G Evolution" phone (once again, not actual 5G, but it's the best phone for the network right now.)

Verizon Wireless won in our 2016 tests thanks to its consistent LTE speeds and extensive coverage. But the competition has really ramped up in the past year. T-Mobile's coverage has expanded thanks to its 700MHz wireless spectrum rollout. Sprint's HPUE is improving both speed and coverage. And AT&T, well, "5G Evolution" it basically really good 4G LTE.

We'll once again be using software from our partners at Ookla Speedtest.net. It's the world's largest community of crowdsourced speed tests, and its data is regularly used by major carriers to make their own speed claims.

At PCMag, we believe drive testing proves different points than crowdsourcing does. With our drive tests, we can take measurements across networks at the same time, in the same place, on the same device, eliminating some of the variables introduced by scattered crowdsourced testing. Crowdsourcing, on the other hand, often covers a much wider geographical area and can help tell the difference in performance on different devices and at different times of day.

Our cities are set, but our routes are not. If you want us to drive past a specific place in one of our cities, or to hit your town or region as we drive between cities, Tweet to us @PCMag or ask us in the comments below. We'll swing by! Our final US story will be published in mid-June, and then we'll get ready to test Canada.