Quote:
Originally Posted by
Toruk 
I am unable to find an absolute definition on the Ofcom website, however from the information I found I believe Ofcom may define 2G networks as offering theoretical speeds of up to 115 kbit/s via GPRS (sometimes referred to as 2.5G), 3G networks as offering theoretical speeds from 512 kbit/s to 7.2 Mbit/s via HSPA (sometimes referred to as 3.5G), 3.75G networks as offering theoretical speeds of up to 42 Mbit/s via HSPA+ and 4G networks as offering theoretical speeds of up to 100 Mbit/s via LTE.
UMTS is as low as 384 kb/s but CDMA2000 is even lower but you luckily never had to deal with that.
But lets take the 512 kb/s speed as the minimum and pair that with the 42 mb/s speed of HSPA+ even though it can go as high, currently, to 168 kb/s.
If we call 512 kb/s a unit of 1 then increment by 1 for each 512 kb/s that gives 42 Mb/s a value of 84. That's a huge jump to call such an expansive change all the same generation when marketing it to a customer. From a highly technical, backend perspective there is a need to know whether the air interface and access method are of the same generation and type, but the customer doesn't know about this stuff (no should they) so why are we so hell bent in marketing this obscure nomenclature to the customer.
And what happens to LTE when it gets over 100 mb/s? Do they call it 5G? What will they call LTE Advance? 6G?
edit: Here's something I wrote awhile back on the subject:
Using 3GPP and 3GPP2 standards the slowest theoretical maximum speed of a '3G' is CDMA2000 1X with a whopping speed of 153 Kbit/s.
If we compare that to the old ITU definition of '3G' you find that HSPA+ has a current theoretical maximum speed of 168 Mbit/s (172,032 Kbit/s) and LTE has 299.6 Mbit/s (306,790 Kbit/s).
That's a performance increase of 112,439% for HSPA+ and 200,516% for LTE that we are told all belong in the same generational category as a marketing term presented to the average consumer. Does that really make sense?
Customers don't care about what underlying air interfaces and channel access methods these technologies use; they only care about what offers more or less performance. The variance to far too large to be useful. While things might get better as we move into more LTE connected devices and the disparity in performance differences between LTE networks and devices grow we're still going to be faced with the same issue of the device either saying LTE or 4G. Now
One solution for marketing that the average customer could understand would be to start with a base-level theoretical maximum speed that uses the small common factor between the tower and handset. For example, let's refer to 10 Mbit/s as 1x. For 14.4 Mbps HSPA+ the handset would show 1.4x, for 21.1Mbps HSPA+ it would round to 2x, for 42Mbps DC-HSDPA it would show 4x, and for 73Mbps LTE it would show 7x.