That was all, really.
Which? magazine's claim that the UK has "worse 4G than Peru", widely reported by the national media this week, has reopened an highly charged industry debate about the reliability of network data collection. The debate can be crudely summed as “crowd vs non crowd”, but actually goes deeper: does enough "Big Data" from a large …
That was all, really.
Llamas are larger than frogs
Llamas are larger than frogs
And I'll swear the Llamas know exactly where people go to take photos and photobomb them deliberately, I've got a very similar image at home on the wall from when the wife and I went to Machu Pichu...
If true, it would be quite allaming.
All the networks claim the village has coverage, but most can only get it outdoors - and then only if you're lucky.
Vodafone told me I would need to pay for a SureSignal (and that's only 3G) so that they can use my internet to allow them to charge me for using their "service"...
Don't know about 4G in Peru but if I compare the speeds I get in Blighty with those I get in various locations in France, Switzerland and Germany I'm inclined to believe that Peru may well be faster. (My experience is not covering the whole of these countries, obviously, but I am travelling widely in those countries and I do have "anecdotal evidence" for scores of locations in F, D and CH (and a little in Spain).) So perhaps our "world-class" infrastructure is getting a bit long in the tooth now?
(Incidentally, there's a similar picture with old copper-based net access: I get consistently higher speeds on the continent than in the UK. Fibre is different though, here the UK seems to be in front, at least sometimes. YMMV of course.)
Having just returned from two weeks in FrogLand and avoiding the cities wherever possible, my experience is that data speeds are pretty slow when compared to the UK.
Mind you, I live next to a Voda not spot but naturally they refuse to ack that it exists so I use 3 as my network.
Ironically, the best speeds I got on the whole trip were out in the middle of the Carmargue.
> Having just returned from two weeks in FrogLand and avoiding the cities wherever possible, my experience is that data speeds are pretty slow when compared to the UK.
Indeed. The other gentleman must have been confined to populated nuclei. Mobile service (not just data) is pretty appalling in France, especially for roaming terminals (it is always a guess whether a roaming phone will connect, stay connected, and make and receive calls or SMS).
> Ironically, the best speeds I got on the whole trip were out in the middle of the Carmargue.
Flat open space by the sea, and depending where and when you were exactly, you might have benefited from a network sized to serve crowds of holiday makers being at low capacity.
Its concerning how OpenSignal’s CTO, Dr William Webb uses 4G and LTE interchangeably where as we all know most 4G in the UK is actually only HSPA which is why we only get 20Mbps compared to the 100Mbp people get in USA and Sweden.
"......as we all know most 4G in the UK is actually only HSPA" - not true at all. Granted 4G and LTE are interchangeable terms, but HSPA is a 3G (UMTS) technology, and is distinct from 4G, so 4G is NOT the same as HSPA! There is though some sharing of frequency bands between 3G and 4G bearers. Probably best to get your facts straight before commenting. It's possible to get peak speeds of more than 60Mbps downlink in the UK dependent upon handset type, network operator, tariff plan & location. In certain parts of the country Dual Carrier, and now even Triple Carrier 4G LTE (sometimes known as 4G+) has been rolled-out improving capacity and peak speeds considerably. FYI, the current AVERAGE downlink speed that can be experienced in major US cities are generally between 15 and 30Mbps, which is broadly in line with averages in UK (depending on your network provider and location).
wow this is Not the USA where they use Fake 4G (3G+ DC-HSPA+) 4G is LTE for the rest of the world
3g is 3g (UMTS to DC-HSPA+, phones show it as 3G, H and H+)
4g is 4g (LTE is 4G, and 4G+ for LTE-A but most phones show 4G for both)
its like in USA where they are going to be calling LTE-U as fake 5G
2g 3g and 4g is a New network that (In the USA the network operators managed to get away with renaming 3g+ as 4G, the G is Generation Not speed)
300mbps 4G is readily available on EE in the UK if you can find a quiet mast in an urban location: https://twitter.com/PedroClarke1/status/853930150299398144 The maximum theoretical speed for that service is over 400mbps but the site being center of a city means it's rarely all that quiet. I regularly observe speeds of over 200mbps on Vodafone too. These are all obtained using 2x2 MIMO and 64QAM so with new devices able to better use what's deployed, the available speeds will improve even further, leading us to at least one network in the UK being able to do gigabit class 4G.
The Opensignal report does not break down the 4G results by operator which is sad considering one of the UK's networks really lets the side down
I get better 4g coverage in Turkey than i do here in lincolnshire
4G? I don't even get 3G! I can see the local Vodaphone mast on Mynydd Rhiw from my house which is all of three miles away and, for some reason, it doesn't even carry data.
I think that the nearest proper mast is over in Harlech and that could well be in a completely different time zone as far as we are concerned. Certainly the indication on the 3 Network maps suggest so.
All this crap about 99% of the population being covered is nonsense. Nought percent is covered here.
Both EE and CTIL (Vodafone & O2 joint venture) have plans to upgrade your Mynydd Rhiw mast to 3G/4G in 2017, but neither appear to have submitted planning applications yet. I imagine the challenge is getting decent backhaul as there is no fibre and probably no line of sight for a microwave link.
You also need to look at who has access to the tech in Peru, perhaps it isn't as widespread in rural locations.
In Peru the only 4G is in the Lima metropolitan area, and Cuzco city (where the tourists go for Macchu Picchu)
There isn't any 4G anywhere else at all.
There's barely any 2G across much of the rest of the country.
In fact, most of the country has no mobile signal at all.
Their data is based on the tiny number of people who are rich enough to not worry about the data cost of the app. That means it is simply wrong in every country where data charges are high.
Last time this story came out, I went onto open signals site and checked..
It was easy to make that analysis, but nobody seemed to do the same. Always worrynig when something so easy to check is copied without any thought. Worse, it seems authorities listen to the same flawed analysis.
It's self selected and by definition often is in areas with coverage. So what ever it reports the reality is worse.
The "scientific" drive testing can also be hopelessly unrealistic and like operator maps paints a better picture than reality because the samples are not big enough and usually outdoors.
So take the worst data of all three and assume reality is worse. Time of day and day of week is also a major issue.
Every summer I visit a location (a county with a population density of less than one person per square mile) that has almost no cell coverage. Not "no LTE", no nothing. There are a couple spots on the property where you can pick up a signal, but there are no guarantees it will work if you try to actually use it to make a call, and forget trying to browse the web.
Because of that, if you leave cellular enabled all day your battery drains pretty quickly, so it is reasonable to just turn it off entirely and rely on wifi (where available) for wifi calling and internet access. If everyone did that, crowdsourced data wouldn't show any coverage problems there!
I'm English but moved to Italy some years ago. Vodafone gives me pretty good 3G/4G coverage everywhere including rural areas. When I came back to uk recently i had a 'normal' 2 days of coverage in London but when I got on a train at headed to the South-west I was genuinely surprised at how poor the coverage was for most of the journey and also travelling the area afterwards.
It really surprised me as the two countries have similar population, area and density and I normally assume italian investment to be a bit behind UK.
I upgraded to a phone that had 4G about 6 months ago and in London O2 coverage is pretty disappointing. I rarely activate 4G now. At least on 3G I tend to get full HSDPA+ or whatever it is.
With 4G enabled my phone seems to spend more time without any signal at all, unable to receive calls and SMS.
I don't know what actual speed my '4G' is (I'm on EE in the Midlands) but it seems hell of a lot slower than when I used to have a GPRS enabled phone. All I can think of is that web page bloat, even for seemingly text only pages, has got so out of control page load times are comparable to when I used to use GPRS on a 'feature' phone. It has got to such a point that I rarely even fire up the web browser on my phone when out and about because the page load times are so long and If I didn't know better I'd say the adverts were prioritised before the actual content data.
Welcome to the result of the constant procrastination by government, BT and ofcom.
I have 4g here in ho chi Minh with download around 75mb/s
The UK talks alot but really is always a catch up country in terms of internet.
Can a new company lay fibre in the sewer in the UK ... No way, that would be accepting market forces and being efficient. Much better to make it political and wait for Asian countries to leave you behind.
Honestly, coverage in Vietnam pisses on the uk. It's standard to have 24mb in your house and true 4g on mobile. Keep that imperial arrogance uk, the rest of the world is actually getting on with it!
Having got myself a spiffy dual sim phone a while back, I got testing various providers. My conclusion is time connected to an LTE signal over say a 3G signal is only part of the picture.
Still being on the old Three One plan, I have unlimited everything but speeds had slowed down quite a lot. In fact their LTE speed in my area is comparable to their HSPDA 3G speeds. 82% of the time I was connected to LTE and, 13% of the time HSPAP and 5% HSPA. Download speeds about 35mbps.
EE was connected to LTE much less but when it was connected, the speeds were much faster. Also the latency was much much better than on three. Using apps which utilised many simultaneous connections got information and responded so much faster.
02 I connected to LTE most of the time, latency for each connection was quick but speeds varied a lot.
While LTE has benefits of quick connects and faster speeds, it's only part of the equation. It reminds me of people shouting how they have 120mbit cable broadband but video stutters because of jitter and latency, where those on 12mbit adsl had no issues. Is being shunted down to 3G/HSPDA a problem if it's faster than the LTE connection from the same provider in the same area?
LTE coverage is shouted about in the same way top speeds are, ignoring the quality aspect.
"Webb explains that OpenSignal is confident if just 100 app users are gathering data. The firm derives its confidence intervals using CLT (Central Limit Theory) “which argues that the mean of any collection of random variables (subject to minor technical constraints our data can readily be shown to satisfy) will tend toward a Gaussian distribution,” the firm explains. “Further it says that the variance of this mean will equal the variance of the underlying population divided by the sample size. Hence, if the CLT holds, we can derive the variance of our mean from the variance of the underlying data set. It is important then to show that the CLT holds for our data and for the sampling sizes we normally use.”"
You cannot use the Central Limit Theorem if your underlying sample is biased. That's the end of that whole paragraph's chuntering. And it tends towards a normal distribution, not a Gaussian distribution. Muppet.
"And it tends towards a normal distribution, not a Gaussian distribution" -- DavCrav
Erm, what do you understand by Gaussian distribution? I always thought it was another name for the Normal.
As for CLT, isn't the whole point that samples taken ("with replacement") at random from any distribution (with finite variance) will have their means approximately normally distributed? I think this is true even if the population being sampled is skewed - or even non-normal - providing it is large enough.
And while getting there on the tube I saw and had 4G, once I got out of Liverpool St Station, around that area (which is always quiet at a weekend) and then around Green Park and Piccadilly, I struggled to get anything above HSDPA, and even then the signal was low.
I spend a lot of time in Portugal and, whilst generally more expensive, pretty much every connectivity option is better over there than in the U.K.
Wider LTE coverage with better overall speeds and a major deployment of FTTP in pretty much every major city and now into smaller ones and also some rural areas.
All those things that, according to BT (in my eyes one of the most evil companies in the world), are unrealistic, unaffordable and "nobody wants or needs".
"It's a great example. The data tells you there is 100 per cent LTE coverage for where the people actually are. If they stray out of range, then it falls below 100 per cent," he said.
I'm not sure an app of any form was required to find that out.
I'd hate to be the first person to stray out of range after the Base management had decided to use LTE for emergency communications!
I travel a lot and I've noticed something similar in Eastern Europe and Asia. In a lot of areas there the fixed-line infrastructure simply isn't up to high speed internet, and so as a result they've invested much more heavily in mobile internet instead as this is a hell of a lot cheaper and easier than upgrading the fixed infrastructure.
It's much more common in a lot of those areas for people to have a home 4G router for example, instead of (V)DSL/Fibre.
So in that sense, it's understandable.
Spent about 10 days in the eastern end of the Pyrenees last October. Amazed to find the broadband in the house supplied over a telephone line from across the mountain provided a perfect netlfix experience totally free of any buffering unlike my home fibre. Also on my travels both in France and Spain across the mountains that Google maps never lost the data signal at any time and this was really rural, unlike the M4 where I regularly lose the data signal!
I live within 25 miles of the biggest city in Scotland and close to a main trunk road.
Data coverage [and coverage to even make a phone call] varies between reasonable [when walking halfway up a hill] and none at all - the Ofcom coverage checker proves it - and that's on all networks apart from O2, where they confess to having 2G coverage.
I have an EE boost box to provide coverage indoors, because otherwise there isn't any.. which only works properly because fibre has just arrived....... 400k up is not enough on an ADSL line to make it work properly.
EE are going to be planting a lot of masts to provide emergency service coverage up here..
Statistics based on population coverage are a fib.
DTraceunder the GPL
Biting the hand that feeds IT © 1998–2018