Skip Navigation


Broadband Speed Test Results for June 2016
Monday 04 July 2016 12:22:37 by Andrew Ferguson

The broadband weather is generally better than the real weather for those that have dared to step outdoors this summer, the big changes in June are that the up to 52 Mbps BT Infinity 1 product is making an impact on BT FTTC speeds and Relish has jumped up the table as the Swindon superfast fixed wireless service has been launched.

Technology Comparison From Internet Speed Tests June 2016
Technology Download speed of bottom 10%
Mbps
Median Download
Mbps
Mean Download
Mbps
Median Upload
Mbps
Mean Upload
Mbps
Download speed of top 10%
Mbps
ADSL/ADSL2+ 0.9 5.3 6.6 0.6 0.6 14.7
VDSL2/FTTC 12.5 29.9 31.2 7.6 7.7 51.4
Cable Broadband 8.1 42 50.4 5 6.1 104.3
FTTH/FTTP 20.2 85.1 108 25.8 58.5 230.6
3G/4G Mobile 2 12.7 18 1.8 4.1 39.8
Fixed Wireless 4.3 10.1 18.6 1.8 6.2 41.5
Satellite Statistically too small sample

While satellite broadband is seen as the answer for the USC and maybe even some of the forthcoming USO areas we saw a drop in the number of satellite users testing during June 2016 and this meant the numbers fell below our threshold for showing the results. This might suggest that even with the USC voucher schemes underway many are still deciding to struggle on with below 2 Mbps ADSL type services, or usage allowances are precious so people are reluctant to speed test.

Large Provider Fibre Based Connection Speed Tests June 2016
(Change since previous month)
Provider Download speed of bottom 10%
Mbps
Median Download
Mbps
Mean Download
Mbps
Median Upload
Mbps
Mean Upload
Mbps
Download speed of top 10%
Mbps
BT 14.2 33.5 (+1.3) 34.4 8.5 (=) 8.5 52.4
EE 11 28 (-0.6) 27.1 8.2 (+0.2) 7.3 37.3
Plusnet 12.6 31.7 (-0.7) 32.8 6.2 (-0.2) 7.9 59.4
Sky 10.6 25.2 (-0.4) 25 7.0 (=) 6.7 37.1
TalkTalk 13.7 27.8 (-0.1) 28.9 1.8 (=) 4.5 41
Virgin Media 8.1 42 (-0.7) 50.4 5 (-0.1) 6.1 104.3
Vodafone 14.8 28.7 (-2.7) 31.4 8.7 (-0.1) 8.8 46.1

BT Retail has the big gain this month as it races on with the upgrades from up to 38 Mbps to up to 52 Mbps for Infinity 1 customers, and looking at the speed profile it is clear that the rise is not from more Infinity 2 customers testing since we are now seeing a new speed step at 52 Mbps. Interesting Virgin Media is remaining fairly flat even though it has an on-going speed upgrade programme, but this is tempered by the peak time performance on their network which as we show later has a bigger impact on performance for the cable network.

ADSL/ADSL2+ Connection Speed Tests June 2016
(Change since previous month)
Provider/Area Download speed of bottom 10%
Mbps
Median Download
Mbps
Mean Download
Mbps
Median Upload
Mbps
Mean Upload
Mbps
Download speed of top 10%
Mbps
BT 0.7 4.8 (-0.2) 6.2 0.5 (=) 0.5 15
EE 0.8 4.9 (-0.2) 6.2 0.7 (=) 0.6 14.3
Plusnet 0.8 5.2 (-0.2) 6.4 0.6 (=) 0.6 14.7
Sky 1 5.3 (-0.1) 6.6 0.6 (=) 0.6 14.5
TalkTalk 1.2 5.5 (+0.1) 6.6 0.6 (=) 0.6 14.2
Vodafone 1.1 5.6 (-0.1) 6.7 0.6 0.6 13
Rural ADSL 0.7 3.9 (-0.1) 5.0 0.3 0.4 11.2

Business as usual in the ADSL world and unless something changes dramatically it looks like many are still happy staying with their cheaper but slower ADSL/ADSL2+ services, the roll-outs of faster services are changing the mix of products but whereas dial-up died a very quick death, ADSL services look likely to survive longer.

The 25 Fastest UK Broadband Providers in June 2016
(ordered by median speed)
Smaller providers without enough geographic data samples are not included
Provider Download Speed of bottom 10%
Mbps
Median Download
Mbps
Mean Download
Mbps
Median Upload
Mbps
Mean Upload
Mbps
Download Speed of top 10%
Mbps
Venus 26.7 89.6 129.2 79.9 122 333.2
Hyperoptic 11.8 45.8 83.8 30.2 71.6 195
Virgin Media 8.1 42 50.4 5 6.1 104.3
Gigaclear 12.1 39.6 87.8 31.5 84.4 293
Vodafone Broadband 3.3 23.8 25.5 7.4 7 43.3
Zen Internet 2.5 21.1 29.3 5.5 7.9 67.8
IDNet 5.5 20.5 29.5 5 8.4 65.5
AAISP 4.4 20 30.7 5.2 10.6 68.4
BT 1.8 19.6 23.6 5 5.6 50.8
Daisy Wholesale 1.5 17.4 22.8 6.4 8.3 33
EE Mobile (3G/4G) 2.1 17.2 23.1 1.9 4.7 49.8
XILO Communications 3.3 14.4 27.1 0.8 6.7 74
Relish 4.2 13.6 17.2 1.2 2.1 38.9
KCOM 2.4 11.7 32.3 0.7 9 82.9
Plusnet 1.3 11.5 18.1 1 4 42
Vodafone Mobile 2.5 11.4 16 1.8 3.7 34.7
O2 Mobile 1.7 9.1 14.2 1.4 3.3 34.1
TalkTalk 1.5 8.5 13.8 0.7 1.9 34.6
Sky 1.3 8.3 12.3 0.7 2.6 30.1
Three 1.4 7.9 14.1 1.2 3.8 36.6
EE (ADSL2+/FTTC) 1.1 7.6 14.1 0.8 2.9 34
Entanet 0.7 7.5 21.6 1.1 10.8 49.2
Claranet SOHO 1.3 7.3 21.8 1.4 7.5 69.6
The Cloud 1 7.3 10.4 0.7 5.9 21.8
Eclipse KCOM outside Hull 0.9 6.9 15 0.8 3.4 37.5

The time of day when you use your connection can make a big difference to the performance and we track peak versus off-peak speeds and have decided to share the results for June 2016 with you. Our speed test is almost unique in carrying out two download tests with the tbbx1 test being much more sensitive to congestion issues between the testing device and our servers. Put simply the tbbx1 should reflect the speed of a single file download/video stream and the httpx6 test is what you'd see from multiple downloads at the same time - so for those who have good speed test results but still struggle with video streaming the difference in the two results may help to explain your problem.

Peak and Off-Peak Download Speed Tests Results June 2016
Provider tbbx1 Test httpx6 Test
  7am-3pm 6pm-midnight % difference 7am-3pm 6pm-midnight % difference
BT 24.4 Mbps 24.3 Mbps -0.4% 23.6 Mbps 22.9 Mbps -3.1%
EE 12.7 Mbps 12.7 Mbps = 12.9 Mbps 12.8 Mbps -0.8%
Plusnet 20.6 Mbps 22.1 Mbps +7.2% 18 Mbps 18.2 Mbps +1.1%
Sky 13.6 Mbps 12.6 Mbps -7.9% 13.4 Mbps 12.1 Mbps -10.7%
TalkTalk 14.5 Mbps 13.9 Mbps -4.3% 14.4 Mbps 14 Mbps -2.9%
Virgin Media 42.6 Mbps 34.5 Mbps -23.5% 54.2 Mbps 45.2 Mbps -19.9%

On a good connection you should see the tbbx1 and httpx6 at around the same speeds, but factors such as wireless connectivity, incorrect RWIN settings and congestion can mean the two figures are different, thus is it interesting to see that in addition to the largest difference between peak and off-peak for the two test protocols with Virgin Media there is also a significant difference to the two protocols during the same test period.

Vodafone and other smaller providers are not present not because of any favouritism but there is simply not the same volume of testing to produce meaningful comparisons between the peak and off-peak period. The observant will notice that PlusNet actually seem to get faster at peak times and this is not a one-off blip when looking at the monthly results this is not uncommon and is believed to be a result of the traffic management systems in place at PlusNet.

Comments

Posted by 21again 5 months ago
Having used SatBB for buisiness use I certainly wouldn't use it for home use, too expensive and too many other issues, in rural areas,compared to the cost of BB through copper wires not to mention those able to use fibre if they so wish.
If it was possible for our fore fathers to get copper wires all over the UK then there is no reason that fibre shouldn't also follow over time.
Posted by MCM999 5 months ago
@21again. The copper network developed over many years from small beginnings to the size and extent it is today. So there is no reason why fibre shouldn't be the same, over time, just not all at once and tomorrow as some seem to want.
Posted by comnut 5 months ago
please tell me how you define 'Cable broadband' when Virgin Media is soon to release 300Mbps cable???
Posted by comnut 5 months ago
http://www.ispreview.co.uk/index.php/2016/01/virgin-media-business-uk-launch-300mbps-broadband-package.html
Posted by comnut 5 months ago
I wonder if it will take 'the rest' another five to seven years to catch up with the TOP FOUR... :) :)
NOTE people who think it is 'the whole UK' speed, make up newspaper titles like 'England freezes at minus 5 degrees!'

Well, Newcastle *is* in england... :/
Posted by WWWombat 5 months ago
@21again
The UK's copper line deployment took from 1964 to 1998 to go from near zero to 30m - nearing 1 million lines per year. In 1980 there was still only around half the country with a phone line.
Posted by Somerset 5 months ago
Near zero in 1964?
Posted by 21again 5 months ago
History from the archives:-

http://www.btplc.com/Thegroup/BTsHistory/1912to1968/1912.htm
Posted by ukwoody 5 months ago
@21again. You say those in rural can still use copper for BB. You come here and try it! 0.7M for me with the fastest in the village recording 1.6M occasionally. I have to use a wimax provider and that is utter tosh with speeds varying from 2M to 12M with 30G allowance at £30 per month. Both being self employed we NEED a connection, but we're rural so goodness only knows when or if it will happen. Even Superfast Cymru has now been disbanded so there really is no-one fighting for us in Wales.
Posted by comnut 5 months ago
Of course, 'able to use' does not specify how 'good' it is!!! :/
Posted by andrew (Favicon staff member) 5 months ago
@Comnut The extremely complex definition for cable broadband is all the broadband products provisioned by Virgin Media across its fixed line broadband network, and for the record we are seeing 300 Mbps users no major impact no the stats so far.
Posted by comnut 5 months ago
OF course these 'big figures' are greatly misleading, the top three represented in ONLY 8 to 10 small areas in the WHOLE UK !!
and only TWO are mapped!
http://maps.thinkbroadband.com/#!lat=55.052134424695474&lng=-3.040465187499972&zoom=6&type=terrain&cable-coverage&hyperoptic
Posted by comnut 5 months ago
PLEASE sort it out, TBB.. :)
Therer is only VAGUE details about Gigaclear, and no-one seems to have *any* details about 'Venus' - their website makes you think they are 'business only'..
Posted by comnut 5 months ago
sorry Andrew, but I do HOPE you have read the above link about it!!!

MY close friend in Isleworth has already got a *working* 200Mbps VM, with promise of 300 in a few months.. :)
Posted by andrew (Favicon staff member) 5 months ago
@Comnut No need to read the link well aware 300 Mbps service e.g. http://tbb.st/1467359683783578655 but well under 1% of Virgin Media tests on that sort of connection.

As for misleading, the top 3 Venus, Hyperoptic and Virgin Media cover 50% of the UK. Venus yes they are a business provider, but aggressively selling and expanding in London area.

As for vague Gigaclear expanding rapidly and have BDUK contracts too.
Posted by comnut 5 months ago
well maybe WHEN you add Venus!
Posted by WWWombat 5 months ago
@21again
The 1964 source looks bogus. Let me restate...

There were a lot more lines and phones in 1964 ... but concentrated in businesses. But to talk about widespread copper down every street, you need to consider progress in homes, not businesses

Combining stats, these are the number of households with telephones:
1951 1.5m
1966 4m
1970 7m
1980 14m
1990 19m
2000 23m

Averaging 4m per decade.

For comparison, the VDSL2 deployment ran at 4m per year, and the G.Fast one looks to aim at 3m per year.
Posted by WWWombat 5 months ago
Growth in households doesn't help. 16m households in 1961. Now we're over 25m.
Posted by TheEulerID 5 months ago
@WWW

Another factor that should be plugged into those stats is that many of the households during the 1960s and 1970s would have been serviced with party lines. I don't know when they were finally phased out of the network, but it does means that during that early period the rate of full loops was not as high as it might appear.
Posted by Blackmamba 5 months ago
Hi TheEulerID
The phasing out of shared loop was a gradual process as Cabs were provided making E and D sides flexible plus the introduction of longer lines using smaller gauge pairs dropping from 70cc/40cc to the small gauge. This produced a problem for broadband so fibre had to be used using the same step method as was used in the 1960----1980.
Posted by WWWombat 5 months ago
Against my stats for the number of households with phones, the other way to consider things is the raw number of phones.

One site suggests 5m in 1951, and 10m in 1966.

That shoes how much initial phone usage was weighed towards businesses - ranging from one phone in a small shop, through to a mass on desks in Whitehall on private branch exchanges.
Posted by WWWombat 5 months ago
@euler
True - the use of party lines, and their gradual removal, will tend to skew the data. Growth of the copper access network will have come slightly later than the increase in residential phone ownership.

My parent's line was converted from a party line when I wanted to use a modem on it.
Posted by alexdow 5 months ago
Back about 1978, PO Telephones informed me that the Domestic "Penetration" for Domestic Phone Installations, was 65% for England; and 55% for Scotland.

That was in connection with the potential design of an automated test system for such lines.

We didn't proceed with it.
Posted by Blackmamba 5 months ago
Hi Alex.
In the year. 1978 you quote 65% penetration in England this result you have to split down the preference rating on each line which were cabled via QOS and preference. keys. Eg RES, BUS,E Lines this ties with today's broadband network the only think that has changed is we have switched to ( 0.1).
Posted by bsg017 5 months ago
While generally satisfied, no more, with BT I am somewhat puzzled that my peak download speeds have dropped in recent weeks from 37Mbs to 33Mbs since I signed up again expecting to rise from a 'nominal' 38Mbs to 52Mbs.I know that I have hardware/firmware problems on this old computer but I do not think they have suddenly got worse. Might it be increased contention?
Posted by DrMikeHuntHurtz 5 months ago
@bsg017

If you care about speed then going with BT is not a good idea.
You must be logged in to post comments. Click here to login.