With several major new mobile broadband services launched -- or coming soon -- in the U.S., and all promising blazing performance, it can be hard to get a clear picture of what wireless speeds you will actually get when you plug in your new data card or switch on that minty fresh new handheld or netbook.
The official word from AT&T Inc. (NYSE: T), Clearwire LLC (Nasdaq: CLWR), and Verizon Wireless is that their new networks will be faster than the maximum speeds offered today. But not landline-replacement kind of fast.
In fact, the fastest-sounding new network is also the one with the least known about it. Indeed, Verizon doesn’t want to talk numbers at all on its Long Term Evolution (LTE) network until it has done a lot more testing outside the lab. But we can look to other users' experiences, particularly in the case of AT&T’s planned High-Speed Packet Access (HSPA) upgrade, to glean some indication of how these networks will operate in the real-world.
AT&T
AT&T has just unveiled its timetable to update its 3G network with a technology called HSPA 7.2, work that is expected to start this year and be completed by 2011. A spokeswoman for AT&T says that -- as you might expect -- the network upgrade will offer a “theoretical maximum speed of 7.2 Mbit/s” compared to the 3.2 Mbit/s peak rate currently available.
The operator says the new update "will primarily affect downlink speeds." It notes that uplink speeds will be between 500 kbit/s and 1.2 Mbit/s.
We can, however, look to Vodafone Group plc (NYSE: VOD)’s deployment of the 7.2-Mbit/s upgrade in the U.K. to get some general idea of how AT&T’s network might perform. Heavy Reading analyst Gabriel Brown tested the network in June last year and recorded some performance figures.
Brown found that in central London locations where 7.2-Mbit/s service was enabled, service was consistently in the range of 3 to 4 Mbit/s data throughput on the downlink while stationary, with bursts as high as 5.5 Mbit/s. Stationary uplink offered a throughput of around 1 Mbit/s. 3G in motion was not quite so blazing, however: “300 kbit/s is what you can expect on a 3G connection while on the move across London by car or train,” Brown wrote at the time. (See Vodafone's Blazin' 3G Upgrade.)
There's no guarantee that AT&T’s network will perform exactly the same, since each market topology is different and can affect performance, amongst many other variables. Still, the Vodafone experience does offer a starting point for user expectations.
Clearwire
Clearwire is a little more forthcoming about the average speeds of its new Clear mobile WiMax networks. "Average download speeds range from 2 to 6 Mbit/s," a spokesman tells Unstrung. "Upload speeds are typically up to 1 Mbit/s." The company says the network can deliver burst peak rates of "more than 10 Mbit/s," but there's no telling how close you have to live to the cell-tower for that to happen.
Clearwire’s WiMax service is currently available in Atlanta and Portland, Ore. When Clear’s precursor, the Sprint Nextel Corp. (NYSE: S) XOHM service launched in Baltimore, Unstrung saw download speeds between 1 Mbit/s and 3.5 Mbit/s in the various tests around the city, although the Clear networks are said to be somewhat different than the original Sprint offering. (See XOHM's Big WiMax Demo Day.)
Verizon
Verizon is the most tight-lipped of the three about the speeds that will be offered by its LTE network, which is due to start going commercial in 2010. The operator has only said that it has seen between 50- and 60-Mbit/s downloads with the technology in initial tests. A spokesman for the company says that it doesn’t want to set customer expectations until it has done "a lot more testing outside of the lab" with the proto-4G technology.
"The bottom line is that until you load the network you can’t know for certain -- you can make an educated guess, but you can’t know," the spokesman says.
Unstrung has very little hard data on how LTE runs, since the first commercial base station just went live in Sweden. (See Sweden Claims LTE First.) Motorola Inc. (NYSE: MOT) did, however, stream video at 8 Mbit/s on an LTE test network at the CTIA show this year. This may give some indication of the raw speeds but doesn’t yet help to account for how the network will perform when lots of users are roaming on it.
@omniobso -- I didn’t read this as an LTE vs. WiMax , or ClearWire vs VZW, article. More like an attempt to get a handle on what end users might realistically expect.
Broadly, I think the service provided over today's HSPA is phenomenal. Think back five years and 3 Mbit/s downlink was a top end ADSL service. Now it's available on a £10-£15 a month mobile plan.
@John -- Anecdotal evidence backs you up. I know of one independent organization that's been testing the end user performance of Telstra’s HSPA+ network. Feedback was very positive. I would name them, but not sure if they’ve gone public yet.
1. Download technology for both LTE and Wimax is OFDMA.
2. Verizon paid 9+ billion ffor their 22 megs of spectrum
3 the 2.5 gig spectrum that CLWR 'controls is 200 megs wide including all the EBS spectrum. There is no spectrum in lower frequencies that compares.
4. It is not 'rocket science' to comprend that a biigger pipe will carry more digital information than a small pipe. 20megs versus 1-5 megs is significant in terms of capacity.
5. Upload speeds that suppert legacy systems; by definition cannot be as efficient as ones that don't.
6. LTE will carry a significantly larger IPR load than Wimax making it intrinsicly more expensive.
7. I'll pit Cisco, Intel, Google, Huwei, Motorola and other Wimax supporters against the Alca-lu, copper twisted pair obsolete infrastructure ANY DAY OF THE WEEK.
"The Next G Network is also the fastest cellular network in the world, delivering theoretical network speeds of up to 21Mbit/s,[2] and is planning to increase the speed up to 42Mbit/s in 2009." HSPA ...
"Telstra is excited to announce the launch of the new Turbo 21 Modem^ that provides typical download speeds of 550kbps to 8Mbps in selected CBD, metro and regional areas^. Available now exclusively to business customers, it will unleash the power of the world’s fastest national mobile broadband network for Australian businesses in these areas."
Dan, great article - it brings up a variety of points and angles for consideration.
My point of view is: It's not about LTE vs WiMAX speeds. It's about best experience using available spectrum with full mobility and access to a plethora of devices and laptops. HSPA and HSPA+ networks will deliver a great multi-Mb experience for years to come - today. Why wait for LTE? Look at Telstra...people are experiencing 2-3Mb and sometimes more. AT&T will be able to replicate this here in the US much like Vodafone and T-Mobile will be able to do the same in Europe and US for years to come.
Verizon wants access to the global device market and to simplify its device/service offerings, thus the path to LTE makes good sense. As for spectrum, all of the operators need more spectrum - period (video is a network destroyer). Where will they get it from? Yes, Clearwire is a good target as is Sprint. I would be surprised if both brand names are still around 12 months from now. Suitable buyers include: T-Mobile, A&T, VZ (market entry by: China Mobile, Vodafone).
I think you're spot on, Dan -- if CLWR were to be sold it wouldn't be the first time a Craig McCaw spectrum-acquisition strategy paid off, would it? But I see a sticking point in the Comcast and Google investment... they may not approve any such deal, think it will be something to watch if CLWR can prove the business case.
Gabriel, I understand your desire for everywhere connectivity, but I wonder how much that caveat applies to the general populace. For every globe-trotting mobile professional there must be many more folks who stay close to home and might see greater value in faster speeds with local mobility. Not sure there is enough data yet to prove this idea, but it's a flip side to your coin that bears consideration.
Asked Dick Lynch that back at MWC in Barcelona, he said he was "very comfortable" with their spectrum position, but then what did you really expect him to say?
I think the spectrum postion is interesting, but it is what it is, the only way AT&T and Verizon add more in the future is by buying it in. I sometimes wonder if that the eventual outcome of Clearwire, to be bought by one of the big two? I think its a possibility, I don't know how likely a possibility yet, Clearwire still has show they can make it work across more markets.
I actually think the upload speeds of these technologies is as interesting as the downloads. If they really want cameras and netbooks and all kinds of other gizmos on there I think the upload speeds become much more important.
They have some more recent data, but I can't find a link.
For my money, coverage is at least as important as high data rates. Knowing you can get a connection everywhere at, say, 1 Mbit/s is more valuable to me than being able to get, say, 5 Mbit/s every now and then.
Verizon's Dick Lynch has said, publicly, that advanced radio technologies like WiMax and LTE are essentially equivalent. So what will make for higher end-user speeds? The amount of available spectrum.
If you REALLY want to see Verizon and AT&T get tight-lipped, start asking them about how much spectrum they have available for mobile wireless data use. Even the so-called "beachfront property" 700 MHz chunks won at auction don't really offer a whole lot of wireless pipe. And when you don't have enough overall spectrum depth, you have to use smaller channels per antenna -- which reduces the speed. See this handy chart Motorola has put together to get an idea on some "real world" speeds.
The betting line -- WiMax will stay fastest because Clearwire has gobs of spectrum, but likely available in fewer markets. LTE when it launches will be in the 7 Mbps range if Verizon is lucky; and AT&T's HSPA+ upgrades... well, this company said a year ago it was going to 20 Mbps this year. Now it's down to 7.2. How low can you go and keep the iPhone all to yourself?
The blogs and comments are the opinions only of the writers and do not reflect the views of Light Reading. They are no substitute for your own research and should not be relied upon for trading or any other purpose.
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Managing Risk with Analytics - by IBM 7/28/2010
Digg
Del.icio.us
Reddit