Slashdot Mirror
Eight Major 3G & 4G Networks Tested Nationwide
adeelarshad82 writes "Building on last year's efforts, PCMag once again hit the road on a 6,000 mile trip to test out eight 3G and 4G networks to determine which ones were the fastest (and slowest) in 21 different cities. With 10 stops in each city for at least 15 minutes each, the team used custom speed test software on 16 different handsets which ran HTTP upload and download tests every 25 seconds to 3 minutes. The test results were broken down by city as well as region. As expected, Verizon's 4G led the pack. It performed the best in Dallas, where it averaged 15.75 Mbsp and also hit the highest download speed of 37.66 Mbsp. On the other hand, Sprint's 4G results were disappointing; in some cities even AT&T provided better download speeds. Beyond the 4G, T-Mobile's HSPA+ offered blazing fast speeds as well, going as high up as 15.93 Mbsp in Detroit while averaging the best in Dallas at 6.44 Mbps. Amongst the 3G networks, AT&T mostly outperformed all others."
We will now proceed to the obligatory 579 posts as follows -
...followed by anecdotes about cell coverage, speed and pricing in Korea.
"This is bunk. In [Insert City], [carrier A] sucks donkey balls. [Carrier B] is much better!"
"Are you joking? [Carrier B] STINKS here in [City C]. I love [Carrier A]! "
"I wish I could just buy a phone that makes calls!"
Mablespoons. It's a measurement of data in liquid form.
Beyond the 4G, T-Mobile's HSPA+ offered blazing fast speeds as well, going as high up as 15.93 Mbsp in Detroit
Of course the downside is - you have to live in Detroit.
#DeleteChrome
When Verizon markets their network as "4G", I expect it to blow away other 3G networks. It's good that it does, and it's also good for PCMag to verify that it does. As a counterexample, Sprint's 4G network wasn't really much faster than AT&T's 3G network.
These new standards like LTE are much, MUCH faster than existing 3G standards. So it makes sense for them to be called something new for marketing sake and for consumer understanding. However ITU-T decided that to be "4G" you had to be much faster than could currently be done.
Well the companies decided to just ignore that, and call the current stuff 4G and I don't blame them at all. I mean with LTE on Verizion you are talking a new frequency band, new encoding, much faster speeds, and you don't want to call that a new generation of wireless?
Standards organizations need to be reasonable with what can be met. Nobody is saying that the current wireless technologies are where we should stop but they are what we can deliver NOW and they are a big step up. Targets shouldn't be set so far advanced. Set that for 5G, or whatever.
10 times what wired speed?
I have 10Gbps gear here at work, and even at the house 1Gbps. My FIOS connection is 25/25 and if I was willing to pay would go up to 150/150.
You sir, are far behind the times.
Uh, even the ITU relented and says that LTE and HSPA+ are 4G technologies. When LTE Advanced becomes available sometime around the end of the decade it might be called 4.5G or perhaps 5G for marketing purposes.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
True. Latency be damned, when it downloads its fast.
But what I found interesting in the charts is that they more or vindicated AT&T as far from the worst carrier, and
usually second only to T-Mobile in the 3G arena. From the grouse level on the web you would be lead to believe that
AT&T were the slowest and offered no connectivity at all in most place.
With an independent assessment, will any of these carriers change their advertising to avoid false advertising claims.
Sig Battery depleted. Reverting to safe mode.
I have AT&T and I have driven all over the U.S. for my job with my 3G phone. I have yet to drive down a freeway anywhere and continuously receive a 128 kbps stream. Again, 1 mbps is all I ask.
I once took an excursion to Reddit, and later HN. Unlimited up/down voting sucks when dealing with a hive-mind.
Theoretically, on a completely empty highway I can achieve speeds in excess of 200mph in a sufficiently powerful car. In practice, most highways will be congested, police officers will be present, the weather may be bad, etc., and I will never be able to go above an average of 60 (in rush hours of many cities, even less). To that effect my compact car is all I really need, and I am more interested in routes to and from work that will avoid congested areas instead of freeways that offer a theoretically higher speed.
Likewise, this article isn't about raw speed, it's about what is practically possible. This is an excellent guide for those interested in data plans.
Latency with LTE is drastically reduced than with 3G.
,the 4G LTE standard was drafted to allow it to be more directly shoved into packets. In the 3G world, there's actually a lot of processing time spent turning them into IP packets, and sometimes you actually have to wait for more data/control signals form the 3G side before you finish crafting your packets and send them off.
This is because
Geez, guys. The whole POINT of wireless is that you can use it anywhere, rather than tethered to a personal access point.
Yet the wireless companies, during the upgrades from analog to digital and voice to voice-plus-data, have abandoned the space between the cities in favor of serving only the concentrated populations wandering around in urban areas. You aren't limited to your hardwired tether. But you ARE tethered to your "coverage area". And even within that, some areas are drastically degraded compared to others.
How about some testing of service ON THE ROAD and otherwise out in the boonies, rather than going cross-country yet measuring only in one big city after another.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Nah, we just had to plot out a trip that took no more than 22 days, where both cars started in Detroit (that's where we got the cars) and covered the maximum amount of large cities, where each city had as many 3G/4G choices as possible, and with no more than about 350 miles of driving per day. It was gated by things like car availability and publication deadlines.
I'm Sascha Segan. Who are you?
No, it's because ...
1) Last year we used USB dongles and it was a bit of an operational nightmare. The drivers kept conflicting and we had to rebuild the laptops a few times, Windows would freak out from time to time for no apparent reason, the tests all had to be initiated by hand, and there are differences in USB dongle performance anyway so it doesn't totally insulate you against that.
2) Last year I got a lot of criticism around "who uses USB dongles? Everybody's on smartphones!" ... so I thought we'd use those popular smartphone devices. I tried to make the devices as similar as possible.
3) Sensorly was willing to work with us to create a great field test app that I think tested a bunch of connectivity aspects really wekk.
I'm Sascha Segan. Who are you?