LTE

LTE and DC-HSPA+

I've been pondering Apple's rather complex implementation of LTE for the past few days.  As has been widely observed, it's chosen not to implement the LTE bands (800MHz and 2600MHz) that the majority of European operators are planning to deploy in 2013 and 2014 (and which Vodafone Germany, Telia and a few others actually do have running now), focusing instead on the 1800MHz band that EE in the UK and T-Mobile in Germany  actually  have running in 2012, along with 2100 for Japan and 850 for Korea all in one 'RoW' handset. This then sits beside two US-focused handsets, reflecting the act that LTE is further ahead in the US than elsewhere (DoCoMo has 5m LTE subs but doesn't carry the iPhone). 

This seems like a deeply pragmatic decision: support the LTE networks that are actually running (EE, DT, SKT Softbank plus the Americans etc) rather than the ones that will be running in 12 months. It also suggests that when Apple refreshes the line this time next year, we might see a different set of variants with a different mix of bands. A model with 800, 1800 and perhaps 2600 would seem very likely in late 2013. 

On the other hand, there will be some disgruntled users in a year's time when the other European operators launch LTE at 800 and iPhone 5 owners discover that their 'LTE' phone can't use these new networks.

Meanwhile the fact that Apple has split the line into no less than three separate models suggests that it might make more. This reinforces speculation that it will make a further model for China, implementing TD-LTE (and possibly even TD-SCDMA) on the right frequencies to support China Mobile, which has 65% of the market (680m users) and on whose network previous iPhones could not run at 3G speeds. This is probably now a commercial rather than technical decision. 

But what does this mean for real consumers? Much like 3G before it, there's a big difference between the headline data rate of LTE and what real users will experience. That '100 megs!' number includes signalling and over head and will never be experienced at a user application level. When you then factor in the amount of spectrum that European operators are actually using, likely congestion and the effect of things like the distance you're standing from the base station, real world LTE speeds in 2013 in Europe will probably be in the 10-15 mbit/sec range. Of course if you stand in the right place on an empty network you'll get rather more. 

Meanwhile, Apple has also included DC-HSPA in the iPhone, one of the latest improvements to 3G. Rather like HSCSD a decade ago, this increases user speeds by letting them use more of the network. Real world speeds? 10-15 mbits/sec. Who's deploying it? Nice chart from Apple:  

Transient

(This list is incomplete. )

Of course, LTE is more efficient than DC-HSPA and over time LTE speeds (with new handsets) will improve substantially. But today, differences in coverage and congestion will create bigger differences.

In other words, rather more iPhone 5 customers will have DC-HSPA than LTE, and they'll probably not notice the difference.

iPhone 5

VAST amounts will be written. But actually, I have a simple set of conclusions: 

  1. It puts Apple back into a tech leadership position with (arguably) the best phone on the market, etc, etc. In 6-9 months the other OEMs will catch up again, and there’ll be a window in which Samsung has arguable better hardware (as the S3 is arguable better than the iPhone 4S). Then the cycle continues.
  2. LTE on most frequencies that are important today, supporting EE, T-Mobile in Germany, Softbank, AT&T, Verizon and Sprint. But nothing new for China (probably too early for TD-LTE, though), and no support (yet) for the 800 and 2600 bands that most European operators are planning to deploy on. Next year, maybe. And nothing about DoCoMo. 
  3. (Corrected) The CDMA model supports international LTE bands, but the US (AT&T) GSM model does not, and the non-US GSM model doesn’t support the LTE bands that have been built in the USA. So Verizon, Sprint and KDDI subs may be able to roam abroad, but AT&T users can’t and Europeans can’t roam in the USA just yet. 
  4. No NFC or anything new from a strategic perspective. Old models dropped in price as before. 

(Details of iPhone LTE bands here)

Effectively, Apple is maintaining the status quo: it has probably the best phone on the market but also the most expensive, and it will continue selling in large numbers at high prices while Android sells in far higher numbers at much lower prices to quite different customers, converting Nokia, RIM and featurephone users to sub-$200 smartphones. 

NFC… well, Apple could technically have done LTE last year, but really didn’t need to and so waited until it was ready. Same for NFC this year, perhaps. 

Also, no iPad Mini, but that doesn’t mean there isn’t one - Apple may want a dedicated event. This would have to be soon to meet a Christmas selling season though.