London has decent 5G availability but speeds lag behind Birmingham and Cardiff – research 5G in London is so far lagging behind Birmingham and Cardiff, according to research into the tech's early deployment. According to mobile performance testers Rootmetrics, EE has the fastest median download speeds in the capital of 149.2Mbps, while Vodafone recorded 97.7Mbps. But EE recorded much higher median speeds in …
great scott you want to access networks and such that are not located on EE and Vodafone ?
these pitiful speeds are to resources inside and close to the networks at EE and Vodafone which mean exactly NOTHING try accessing real world resources that actually CONSUME BACKHAUL bandwidth and this will be a very different number I'll bet
until someone actually does decent testing i.e. a good methodology then I will pass, if they show 100Mbps on a train ride / Commute to actual networks then I'll pay them money...
I wonder if the networks have configured different caps depending on mast/cell location, working on the suddumption that in some locations the important criteria is handling large numbers of users doing 'normal' stuff that works with speeds of sub-20Mbps...
Agreed. Backhaul is crucial here, and it's going to be quite tricky to get good backhaul links to all of the new 5g cells.
It'll be interesting to watch the bitching & moaning amongst the early adopters as soon as the channels start to fill.
Is anyone talking about guaranteed contention levels? Guaranteed minimum speeds? And how would they manage it?
VM/BT who have fibre or whatever for the area still means super throttled connections during peak (as many commentards have attested to) how will that be different for 5G where there will be so many more subscribers (as everyone has a phone) with the added bonus of some of them being on the move on buses between tall buildings etc with frequent basestation handovers. What I'm trying to say is those quoted speeds are meaningless today.
I remember when 4G was shiny and new and no one had it, I was an early adopter and regularly got speed tests of 50-60Mbps. Now everyone has it I'm lucky to get 10 as every cell is so overloaded. Be interesting to see how 5G holds up after a year or two. It'll probably be a congested mess and the operators will be banging on about 6G as the answer! All this is assuming you can get coverage, there are still big gaps in 4G and even 3G coverage even inside major cities.
.......but what I need is big numbers for my laptop connection at my nice tidy desk.
*
Others have pointed out here that the bandwidth delivered is the SLOWEST SPEED IN THE PATH between my user device and my chosen server.
*
For example, my 801.11ac WiFi adaptor may not be able to deliver ANY OF THE SPEEDS mentioned in the article (149.2Mbps 97.7Mbps 185.7Mbps 112.2Mbps 163.1Mbps 113.6Mbps all mentioned). And my cheap laptop may be even slower if it only implements 801.11n.
*
And that's just my laptop......never mind any of the other links between me and, say, Google.
*
Can someone explain why I should pay ANY ATTENTION AT ALL to the bandwidth numbers quoted in this article?
801.11ac should be able to deliver 433Mbps+
801.11n should be able to deliver ~300Mbps assuming you don't have any of the well known wifi limitations (i.e. 2.4GHz only, single antenna limiting you to <150Mbps).
Everything else assumes good configuration practices (i.e. avoiding interference which is the real killer in the 2.4GHz band, ensuring line of sight to access points to avoid depending on signal bounce, not using terrible client devices that cause all wifi devices on same band to drop to lower speeds, within acceptable range of AP to get advertised speed, sufficient AP's to maintain an acceptable client-to-AP ratio to meet required bandwidth and sufficient upstream bandwidth for client:AP ratio etc)
For 5G in public spaces with contention, physical obstructions, distance and limitations for tower locations? If they can manage ~20Mbps sustained with bursting to >100Mbps I suspect >90% of users would be satisfied.
Aye, t' floor bauds. Kept t' puddle from draining away...
One wonders what will happen when more than a few 5G devices are on these superfast cells. Available spectrum would seem to dictate that they'll go south with alarming rapidity with popularity. Then us'll be back to vibratin' t' damp string, mekin' our own ones to go wi' t' zeroes for 25 hours a day down t' pit while chewing on a gobful of cold gravel as dad chops us up wi' 'is knife and dances on our graves singin' 'allelujah.
Don't worry folks! We'll soon have broadband all over the country, so that Tory diversionary plot to shift the not-spots onto 5G will be foiled.
Mind you, Jezza hasn't said how fast the Free National Broadband will run at, so don't be surprised if we're all levelled to 2Mbps to achieve the goal.
/s <- but actually I'm not sure...
It will be the broadest of bands, outward looking and open for business. We'll have a referendum about the definition of broad, where the party will back a 100mB/s¹ rate. Broadband for the many, not the few! Boris has consistently missed his targets to get broadband done and misled the public about the benefits. We will take a new direction.
I'd say you couldn't make it up, but I just did :-)
¹ yes, that's millibytes. Ed Millibyte will be broadband minister and the GPO will return to its core function of not giving a shit
Speeds of 100Mbps to a mobile are more than enough for 99.9% of users. Even if your streaming a 4K movie on your phone - which is overkill as it would be near impossible to tell the difference between 4K and HD on a phone screen. You will still have plenty of bandwidth left for any apps running in the background. So I don't see a problem with those numbers.
Although as others have pointed out, they may go down as more people get 5G capable handsets and the backhaul bandwidth gets saturated.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
