O2 comes out on top of mobile broadband research survey

[hecatae]

http://consumers.ofcom.org.uk/2011/05/mobile-broadband-speeds-revealed/

and the graph http://bit.ly/kqmD8y


The research, conducted between September and December 2010 in partnership with broadband monitoring specialists Epitiro, found that the average download speed achieved by consumers in Ofcom’s consumer panel survey was 1.5Mbit/s and basic webpages took on average 8.5 seconds to download.

Orange falls so far behind everyone else that it's laughable

[Ben]

I'm absolutely baffled by this. How is it possible that the MBNL participants have such varying results, anyway?

O2? Seriously?

[3GScottishUser]

This kind of confirms my own experience.

For a very long time many have said that 02 were poor for data but as ever the slow horse often ends up sprinting to the finish line. 02 made huge strides when they were chosen by Apple as the launch partner for the iPhone and quickly implimented Edge to satisfy the demand for faster data across the UK. Now Ofcom have published that they have more than caught up with the competition when it comes to 3G and they are first off the blocks providing 900Mhz 3G.

Vodafone are not far behind 02 and have the opportunity to reuse 900 Mhz too.

What about 3? They have the highest bandwidth for 3G in the UK yet come third in a comprenhensive regulator survey. This new independent data will not do 3 any favours with their ambition to be the #1 mobile network for the Internet.

T-Mobile posted a pretty poor result and Orange look like they have a lot of work to do to catch up with the rest. Although they share masts (with 3) they all have different physical transmission systems and backbones, hense the varience in results.

Looking at the published data and keeping in mind that 850Mhz 4G will not arrive before 2014 at the earliest, the networks of choice must be 02 and Vodafone (where potential customers have good coverage) for the best speeds and the prospect of higher and more reliable data speeds using 900Mhz.

(Recently I have been hitting 6.2Kb/s download and around 3Kb/s Upload with Vodafone - that is more than double their published result but it's not uncommon in my area). What will have to be taken into account is that the survey speeds are averages so could be better or worse depending on location. As a guide this survey should be useful for potential customers.

[hecatae]

NOTES FOR EDITORS

1. Over 4.2 million tests were run from September to December 2010 using three data collection methodologies.
- Static test probes installed in 97 locations across the UK, each with connections to all five operators and testing hourly. The objective was to research the performance of 3G/HSPA networks, so locations were chosen where all networks had good 3G coverage and any data collected from 2G connections was discarded.
- Drive testing in dense urban city, urban sprawl, provincial town, rural and semi-rural county locations (performing tests while stationary). The objective was to explore how performance varied both within and between areas with different demographic characteristics.
- A consumer panel of over 1,000 mobile broadband USB modem and datacard users with an installed test application that tested up to 4 times per day. The objective was to research the performance actually delivered to mobile broadband users and therefore included data from 2G and 3G/HPSA connections.
2. The research was commissioned by Ofcom from Epitiro, www.epitiro.com. For more information contact Iain Wood, [email protected], 02920 488 226.
3. Web-page download tests involved downloading the HTML assets only (i.e. the code and text, but not the images) from the home pages of some of the UK’s most popular web sites.
4. Mobile broadband is not generally marketed on the basis of the speed of connection, as headline speeds (for example, ‘up to 7.2Mbit/s’) have little meaning when performance varies so much as a result of different levels of network quality. In June 2009 UK mobile operators published ‘Principles of Good Practice for Marketing Mobile Broadband Services’http://www.mobilebroadbandgroup.com/documents/mbg_mobile_broadband_pr_010609.pdf. One of the principal aims of this code was the provision of clearer information for consumers that actual throughput speeds vary significantly and depend on a range of factors.
5. Ofcom’s UK fixed broadband speeds research can be found here. Ofcom set out its plans for the 4G mobile auction in March 2011.
6. Ofcom’s primary duty under the Communications Act 2003 is to further the interests of UK citizens and consumers. One of Ofcom’s Annual Plan priorities is to ensure communications providers deliver clear information to ensure that consumers can make informed choices.

highlighted in bold the flaws, wahey comic sans

[Ben]

Wow. Why 97 locations? Is that the most places they could find with O2 3G?

:p

[hecatae]

no it's the only places they could find coverage from all 5 mobile networks, fighting for frequency and bandwidth capacity.

test application, so windows, usb modem and datacards, so no mifi units.

Any of these results could be incorrect due to badly written drivers and what antivirus was installed on the windows machines, and if some of these tests were carried out in xp, vista or 7, not a fair test.

out of the 1000 users, was it an equal split of 200 to each network, or did they only find rare o2 users in a suitable location for testing.

Anyone want to contact epitiri for a break down as they carried out the research

[hecatae]

For a very long time many have said that 02 were poor for data but as ever the slow horse often ends up sprinting to the finish line. 02 made huge strides when they were chosen by Apple as the launch partner for the iPhone and quickly implimented Edge to satisfy the demand for faster data across the UK. Now Ofcom have published that they have more than caught up with the competition when it comes to 3G and they are first off the blocks providing 900Mhz 3G.

3GSU, the data is over 6 months old, 900mhz data was not enabled back then, so unless two test locations were Newbury and Slough, I doubt 900mhz 3G applies.

[3GScottishUser]

3GSU, the data is over 6 months old, 900mhz data was not enabled back then, so unless two test locations were Newbury and Slough, I doubt 900mhz 3G applies.

I know that but was making the point re future performance.

Interesting to note that Vodafone had the highest speed recorded at just over 6Kb/s.

Overall I think this survey is a very comprehensive piece of research which the networks will take careful note of.

[hecatae]

full 83 page report here: http://stakeholders.ofcom.org.uk/binaries/research/telecoms-research/bbspeeds2010/Mobile_BB_performance.pdf

reading now:

3.8 All probes were equipped with the same Sierra Wireless AirCard USB 309 modem to connect to the mobile network under test. This is an HSPA+ device capable of supporting download speeds of up to 21.1Mbit/s, and of connecting to all the 3G, HSDPA and HSUPA networks being operated in the UK. The use of a single USB modem for the research is not reflective of the different devices supplied by each MNO to consumers. However it did provide a consistent basis for the measurements of mobile broadband delivery by removing any differences in hardware, and also enables the accurate and consistent measurement of the radio conditions at the point of each individual test.

3.9 The SIM cards deployed with the probes were for the consumer pre-pay (Pay As You Go) mobile broadband services offered by each MNO. The exception was Orange, who at the time of the data collection did not offer a pre-pay mobile broadband service and therefore the SIM card from an Orange Dolphin mobile broadband plan was used. Mobile broadband SIMs were procured on behalf of Epitiro using operators’ own retail channels. During the period of data collection, Epitiro ensured that pre-pay services were topped-up in order to maintain the necessary schedule of testing without exceeding any data caps or fair usage policies.

3.10 At the start of the data collection, the 25 probe rigs were deployed to sites indoors across the UK, with one rig at each site to enable measurement of all five MNOs. During the course of the data collection period, rigs were then relocated three times enabling measurement from 97 unique UK sites in total. Each rig remained at each site for between 2 and 3 weeks before being relocated to the next. The 97 test sites provided measurement of mobile broadband delivery with connections using more than 1,400 unique serving cells. The geographic spread of the selected static probe test sites is shown in Figure 3.1.


Other information missing, 3 uk covered the most locations, 95 out of 97, then vodafone 92 locations, then O2 89 locations

O2 uk enables image compression, and gives no way of removing it.

looks like a lot higher number of users were connecting with 3 uk, figure 3.40 looks weird if it was an equal share

[Ben]

Interesting to note that Vodafone had the highest speed recorded at just over 6Kb/s.
Few typos - unless I've been sucked back in time - we're definitely into mbps, 3GSU, not kbps! :)

[Hands0n]

So hang on a moment. We are being asked to believe that O2 scored top of all the UK 3G providers. This is the same O2 that OFCOM had to reprimand for not honouring the 3G licence agreement for nationwide coverage? The same O2 that was threatened [albeit toothlessly by OFCOM] with a fine for missing its 3G deployment targets by several years?

It is all very well having this test performed from static locations where all five networks were available concurrently. It is a far different story, in reality, using O2's network while moving around the country. Far too often I find that Vodafone and Three are available where O2 is completely absent. And I have yet to be in an area where O2 is the only network apparent. By contrast I find that Three can quite often be where none of the others are.

In terms of this survey 3G900 is completely irrelevant. It did not exist in the field then. Now, however, the primary networks (O2 and Vodafone) have a distinct advantage being holders of 900Mhz licences. That is an accident of history not a strategic investment in the future. However, when the LTE/4G bidding the field will once again be levelled. I won't make any predictions, but I also will not be thinking of O2 as my primary provider.

[Ben]

Other information missing, 3 uk covered the most locations, 95 out of 97, then vodafone 92 locations, then O2 89 locations
My little quip was worryingly close to the mark!

[Hands0n]

Reading the article linked to all I could think of was The Emperor's New Clothes.

[hecatae]

o2 only were in one of the 97 locations, but they were the fastest, so they win.

cheque please