It's 2017 and advertisers are liars?
No kidding ^^
Re: It's 2017 and advertisers are liars?
Up to 100% of them are tossers.
"need to regularly test their broadband speed...
to check they are getting the service they are paying for"
If you need to regularly run tests to be unhappy with your broadband performance the best advice would be to not run tests.
Re: "need to regularly test their broadband speed...
I signed up for a free SamKnows box.
Its sits on a VLAN on its own connected to my router. Downloads small test files about once an hour. Uploads results to SamKnows, and I get a little report once a month. Tests everything from DNS latency to RTP jitter to "popular website" loading times. It's used to do the government statistics on how fast your broadband is, etc.
Strangely, since having it on my network, have always got the advertised speed. Strange that.
Impact on my usage of the line: Zero.
Plus, I get to keep the box and do what I like with it (it's a small wireless router with custom firmware, but can be wiped back to a normal firmware).
P.S. No. It can't access my local network, or my wireless networks.
Re: "need to regularly test their broadband speed...
Looks like someone needs to write a piece of software that emulates this behaviour on a pc !
If it can fool ISP's to merely 'think' people are watching them, it might actually help improve their broadband connection...
Re: "need to regularly test their broadband speed...
I have a business BT VDSL2+ at home (for the backup servers in my closet) and we never see any variation in speed on it in spite of a very large number of other VSDSL2+ users around and about it is rock solid at 73/19. This is real speed as I can see the sync speed of the Synology NAS backing up the office data as the Synology handily logs this.
I don't really think since vectoring was introduced that cross talk has been that much of an issue.
However, I do think that overall throughput in the backhaul is a huge issue for consumer grade VDSL2+ connections and that BT are guilty of not providing the bandwidth to the DSLAM.
Actually Andrew Fergusson had a lot more to say about that Which? report, and it wasn't very complementary.
People should never, as a matter of principal, believe anything adverts tell them. And that is never more important to remember than if the subject matter is even remotely technical. Advertising is by its nature regional or even national. xDSL performance is by its nature individual. Someone living in London on a modern housing estate can see/read exactly the same broadband advert as someone living in an 18th century cottage at the foot of Ben Nevis. It should be obvious that no advert can convey accurate and detailed xDSL information to both people at the same time. It is and always was a non-starter.
Broadband adverts should never mention connection speed. It simply shouldn't be allowed - other than 'please contact us for a personal estimate'. Unfortunately that doesn't leave much for most of them to brag about. VM is the only one offering something truly different but even VM has locale specific issues.
Now what they could legitimately mention in their adverts is the state of their interconnects and router capacity. Unfortunately that's very technical and also proprietary information that they don't want their competitors to know. I suppose they could also talk about customer service levels but marketing departments don't like selling to the public on the basis of what they do when things go wrong. They'd rather pretend nothing goes wrong.
Problem with regularly testing speed, is that your laptop may at the time only have say, a wireless 4Mbit connection (due to congestion, other devices or whatever), and that speed would be reflected in something like speedof.me - while a direct wired connection would give something much more accurate.
you can skew results by testing from a rubbish device, over a rubbish local wifi connection with a rubbish router.
There are a lot of variables likely to result in the perception of poor performance - this also is a reason to disbelieve any advert, as the best broadband speed will be driven by what your endpoint(s) look like as much as the cables from your property.
You will probably also find on occasion your router believes it can achieve a near advertised speed connection to the exchange.
Rubbish achievable speeds are more often than not related to local bandwidth contention - I notice a perceptible difference in download rates on a Saturday Morning to a Friday night for example.
Based on the above the market is simply broken and we should probably be buying a minimum speed rather than a fabricated maximum with fireworks and flashing lights.
The problem with all speed testers is that they don't actually test the real line speed.
At times I can run all sorts of speed testers from my PC and get rubbish results; however, if I monitor my router's line stat's these will show that throughout the tests my actual line speed remains stable at some figure in the range 32~36Mbps, so clearly the causes of the end-to-end data speed fluctuations I see on the PC speed tests are down to factors other than the FTTC line speed...
It's not always your connection speed
It's not always your connection speed, it's the source of the data. I wonder how many of the surveyed people were specifically testing their speeds or just whinging about download speeds in general? Even on a 100Mb/s connection, I don't often see that in normal use unless there are multiple devices/connections accessing the line at the same time. Speed tests invariably tell me I'm connected at 90-110Mb/s
Re: It's not always your connection speed
It's not always your connection speed, it's the source of the data.
True. In fact the only thing a speed test site tells you is the speed between you and that site - ironically those sites being the only ones in the world that have nothing you want to download. It's like trying to find the average speed of your car by timing the journey from your house to Edinburgh. That's of questionable value if you actually want to drive to Plymouth. And if you live in Glasgow then the test is even less meaningful.
If you run multiple tests based in multiple locations you can begin to get an idea of the state of your connection and your ISP's network. But that is still a bit dubious since speed test servers are likely configured and connected in ways that maximise throughput that a more generic site wouldn't be.
If you run multiple tests, from multiple sites on a reasonably frequent and regular basis they can alert you to issues or degradation which is where I find the most value. One thing Mr. Ferguson's tester does that most others don't is run both a single threaded and multi-threaded test. Ideally the two should perform almost the same and if they don't you know that something isn't quite optimal somewhere.
The widespread understanding in the broadband industry . . .
. . is that saying "up to" means you cannot complain if you're not getting the top tier performance. It'll always be down to your local connection issues and they're in the clear.
Checking the speed you actually get isn't going to change things one iota.
advertised speeds are always rubbish
My new car is advertised as capable of 127MPH but I only got it up to 96MPH and it crashed on a bend. The advert said nothing about speeds being limited on bends. Anyway I turned it off and back on again and it seems ok but some of the glass is broken and I can't see out properly. Could this be a windows problem? Should I do another test to see if it crashes at the same speed again or should I fix the glass first? Maybe the 96MPH is programmed into it somehow. Sounds like a major bug to me! Is there a workaround for this? The fuel consumption also goes down at higher speeds, well, that doesn't sound right to me. I'm getting the car to my destination quicker so there is less time for the fuel to be used. Makes sense to me ...
It's a bunch of bravo-sierra
Advertised "up to 12Mbps". Per the local tech, given my distance from the CO, my line is set at 8Mbps. And on a good day, a speed test comes back at 6Mbps.
And I had to ditch all the telco's gear to get that! Bought my own DSL modem, a real firewall instead of a 'router', etc. And the DSL modem is literally 6 inches from the Dmarc, so there is no "in house" wiring to account for.
And get this... The telco tech suggested they throttle my line speed to 5Mbps. That way I would always run at 100% full speed, rather than the 80% or so that I am getting now.
till we remove the up to
Is it in the interest of the ISP to provide more bandwidth,
they advertise 'upto' and they pass that test.
Now if we paid per bit then I bet service would improve,
as its in the suppliers interest then.
Its called commercial interest, provide the minimum , for the most , for as long as you can gives the most short term profit.
Re: till we remove the up to
Whilst I agree in general with your statement, I must point out that it is not always the case. Back when I was living in the UK I had a Virgin fibre to the home connection advertised as "up to 150 megabits per second", however at off peak times I regularly saw download speeds of over 250 megabits per second.
I wish I still had that connection. Now I'm just outside of Tokyo and have an "up to 100 megabits per second" connection, which is utter crap. Peak or off peak I've never seen it crawl along at anything over 10.
TLDR: Virgin media disproves your otherwise correct statement.
Lies, damned lies and Adverts
BT harps on about their "fastest wireless connection" but again that doesn't take into account the local radio conditions. Any advert that requires small print to contain caveats should be disallowed.
My Honda landbarge is the fastest car in the world!! *
* When crushed down, shaped like a bullet, and fired out of a huge cannon!
Hello broadband sales reps
Thanks for attending this meeting. I look forward to hearing your pitches. Please take a seat in the meeting room, there's tea and coffee if you'd like. Yes, that educational sign on the wall does note "Room atmosphere contains up to 21% O2." Just let me get this door sealed and crack open that nitrogen tank behind the wall.
Well, I'm a little surprised no one else brought an oxygen mask, that sign is pretty clear. Shocking lack of foresight there. Ought to quicken up negotiation a bit though.
I'm lucky. I pay for 45 Mbp/s and I get 55. Then again that tier speed was set when uverse came in on a bonded pair. I have fiber.
I'm surprised so many posters here don't understand why it's advertised as "upto".
The simple explanation is that the signal traversing the connecting cable degrades over distance. The same service is sold nationwide yet everyone's connecting cable is a different length. Longer cables suffer more signal loss and therefore will not achieve the upper end of the upto, shorter cables can achieve the upto limit.
Signals on Fibre also degrade with distance, it's just that the distances required for bb are well within the margins for delivering top speed for the overwhelming majority of end users.
How do you remove the upto part whilst keeping the advertising simple for consumers? The £20 tier will not exceed 40mbs? The £30 tier won't exceed 70mbs? Speeds dependant on survey?
I'm happy with the "upto" marketing and receiving the more accurate estimate when applying.
Upto doesn't tell me anything about the quality of the connection offered by the isp though. An official tier branding from OFCOM would help, tier 1's being bt, Sky etc, tier 2's perhaps plusnet, Zen etc. Consumers won't care though and would just want fastest for cheapest and moan like hell when it doesn't work.
"those expecting speeds of between 10 and 30Mbps"
My service peaks at 1Mb/s (typical average 340 kb/s download, 180kb/s upload) and I'm based in a major commercial town just north of London. I'd be delighted with 10 Mb/s.
The gummint should legislate to require ISPs to charge only the percentage of the fee, relative to the percentage of claimed speed, actually delivered.
So "£30 for 50Mb" becomes £3 for 5Mb actual.
Think the ISPs might change their tune then?
Speed would bother people less if they weren't being charged for something they weren't getting and had no way of getting.
My inlaws are charged for 24mb but barely see 1mb. They can't do anything about it, BT won't do anything about it (rural) yet they are still charged full whack. I'm on 80mb but only get 42mb (I would switch to 40mb but the package i'm on has unlimited this that and the other, no traffic shaping, lower contention etc. etc.). FTTC will fix their issue eventually, FTTDP or G.Fast will fix mine, eventually.
OFCOM should force ISPs to pro rata the charges. ISPs know what speed you'll get and should charge proportionally for the speed you get. Only getting half speed? Then pay half price!
You brits always make the best of the shit you have
When most other countries install broadband they say up to 90mb/s or similar since only getting 83 wont matter that much, but when you are still in the dark ages arguing between the government and (totally not the incumbent telco who has a direct line to government) BT.
The arguments about speed in the UK are a joke. It should be so fast no one is complaining.... and yes that is actually happening in other places... not even EU places, Africa and Asia... UK is not a technology joke, but every little thing needs 5 years of discussion and argument.
Simple solution. You pay pro rata for the speed you get, including line rental - BT OpenReach can only charge full line rental if the line supports at least 95% of the theoretical maximum ADSL 2 speed.
I must be one of the rare UK customers, on plusnet fibre and i get the advertised 70Mbit :) not bad for being up the north east of Scotland :)
fibre was only recently available in my street, been rock steady at 70Mbit sync for 4 months now. Due to overheads etc i easily see 68mbit throughput on steam etc.
14.6mbit when I was dsl2.
Not the only one
I was told I'd get 40Mbps, and I get 40. That's fine.
But I'd be rightfully pissed off if I was getting anywhere near what my previous provider could muster, a mere 0-4. I think in those circumstances (ie: sold 40 but getting no more than 10% of what I pay for) I should have the guaranteed right to pay only 10% of the bill.
I get some factors may affect the ability to get the max advertised speed, but a tenth of the projected speed? Nope, that's blatantly lying and mis-selling!
Wait, let me get this straight. The UK, with Fiber to every home, is not getting the speeds they should?
Hell CenturyLink (fiber to the node, copper to the home) in the US advertises UpTo XXMbps, and in the fine print it states "at least 80% of speed", otherwise they down grade your line to the next tier down (and you pay less) or they fix the problem. Of course, getting them to upgrade the DSLAM to provide the "advertised speed" is a problem, and getting the upgraded DSLAMs registered in their systems with the new speeds are an even bigger problem. If you do not get that 80% of speed, CenturyLink will give you a credit for the number of months that you have not.
Even with Crap-tastic Comcast, you will get the same UpTo XXMbps with the caveat of "depending on available bandwidth". Of course, dealing with Comcast's shitty customer support will get you naught but anger and aggravation. However, you start talking about canceling their service, and BOOM! free 2-3 months of the next tier up speed.
Every one of those UpTo XXMbps should have a caveat that if they do not meet, you should be able to get a refund. Granted I do not know much about the UK's laws, however, money does talk. If you do not like what you have, starting to talk about changing providers should get the lazy corps there off their backsides to bring better service.