Australia’s 4G network is faster than 5G: study
thenewdaily.com.au
thenewdaily.com.au
- 5G: 792mbps
- 4G: 950mbps
https://www.opensignal.com/blog/2019/07/08/5g-boosts-the-max...
I have seen gigabit speeds on 4GX - once, when it was new and the spectrum was hardly used.
Since the reports are quoting real world maximums, I'm sure that number is true - but doesn't reflect your average experience.
They only started providing wireline services very recently, one of the few "western" markets they were not a full telco player in.
[0] https://en.wikipedia.org/wiki/List_of_Australian_mobile_virt...
Edit: Seems my info was bad. Disappointing! Latest stats page 9 @ http://clients3.weblink.com.au/pdf/HTA/02127798.pdf and page 31 @ https://www.acma.gov.au/-/media/Research-and-Analysis/Report... They are indeed third (last!), though currently the fastest growing. Their MNVO base = 350K customers, total = just under 6M customers. Guess it shows how long I've been outta dodge :)
_Maybe_ they have a good share of multinational contracts from their parent company, that is the only area I can see this statement being true.
Vodafone are doing some work with NB-IoT but the general market for this are metro utility companies - the coverage doesn't exist out of these areas to be taken seriously for a nationwide fleet.
I would put good money on the number of payment card terminals and power meters alone on Telstra being bigger than Vodafone AU's entire M2M business.
Telstra's mobile footprint is vastly superior to the other two MNOs in terms of geographic coverage.
Anyway, this doesn't really highlight how good our 4G is, it highlights how poor our interconnected fibre network is. It's a bloody multi-generational disaster, a train wreck salvaged. And the price of it all lands on the consumers.
For example, I'm on cable, and life is good. When I get migrated (by force) to fibre, I will have less bandwidth and pay more. It's also wildly known at the moment to have constant drop outs. YAY the future.
And remote areas are put on Satellite, they were never going to run fibre to the home.
Regardless, it's not really relevant to an article about 4G vs 5G - they are separate fibre networks I believe?
That speed is just double the maximum line speed of my FTTN NBN connection at the same property (but I keep it anyway as household data usage on 4G is uneconomical).
At a constant 118 MBit/s I calculate that you'd go through over 50 GB per hour, so the usefulness is limited.
I suspect that part of the reason that 4G is so good in Australia is that data allowances are relatively small which limits congestion.
The less bandwidth and paying more are only due to stupid pricing models from the infrastructure provider (NBN), for anybody who isn't familiar with the system. There's no technical reason for it in 60% of the country. They could upgrade anybody in the FTTP footprint to gigabit this afternoon and anybody in the HFC footprint to at least 250 or 300Mbps but are limiting everybody mostly to 100Mbps (but most people are on lower speeds) just because they want to gouge providers for bandwidth. The artificial limitations created by these pricing models also mean that a lot of providers skimp on buying the necessary bandwidth, so for most providers (there are only a few good ones) there are unnecessary slow-downs at peak times as the artificially limited bandwidth gets congested. It's madness!
The rest is unfortunately served by satellite, FTTN or FTTC that aren't able to deliver the speeds, or for the fixed wireless footprint it is also limited, but to a large degree only because the LNP government shelved the planned backhaul upgrades to save some money, so there are all these wireless towers that are served by low-speed microwave backhaul. The plan had been to upgrade any tower to fibre backhaul once the microwave backhaul reached a certain level of utilisation, before it was fully congested. That now doesn't happen, so those users just have to deal with speeds dropping significantly at peak times.
I'm fortunate that the connection is 100/40/unlimited for $60 AUD, but it still annoys me that I have _no_ options to obtain a static IP whatsoever.
4g here is also far better than the shithouse fake NBN foisted on us by our far-right government, which loathes and fears any technology more recent than coal & witch burning.
I’m not saying the NBN isn’t a trash fire of an infrastructure project (it is), or that our 4G is bad (it isn’t), but at least for me I get rock solid 100/40 on HFC, which is for sure better than my NBN service in inner-North Melbourne
40 odd km from me there are estates wired up with NBN's original FTTP. Yet in my location, all that's available is the hopeless Sky Muster satellite service, which tends to go offline for days at a time, and has trouble with weird contingencies like rain. It was originally slated as a last-ditch option for people in truly remote areas, but the hapless NBN Co has gradually extended its use to anywhere they can't be arsed to serve properly.
[Edit: as a side note, my location shows the corporate fakery behind the 'mixed tech' model. Using a cheap 4g modem from ebay, and without even so much as a fixed antenna, I can get 60Mbps down with 4g. Yet NBNCo doesn't make the fixed wireless option available here, instead foisting on us the 'outback solution' of satellite with its 600ms ping times, pitiful data caps & unreliability]
Each technology has their own hardware faults though, and I've witnessed both. But as a user of the service, I'm actually surprised how decent HFC has been working.
My parents of FTTN get about 50/20,which is what I expect a lot of users on these lines would only be able to access. But the sheer cost of using a NBN over 4G makes it a lot easier to sell, data sims are not cheap
I no longer have a FTTC connection to test, but I don't think it's 3ms low.
Using AussieBB. It's around 20ms from Sydney to Melbourne on our routes
---------
Whereas you can watch the moment of television where all respect I had for Malcolm Turnbull crashed into the floor (fast forward to the 17 minute mark and watch for about one minute):
https://www.abc.net.au/news/2013-08-12/nbn-debate/4882232
"Anthony said that fibre to the premises can deliver one gigabit per second, 1000 megs, and you're quite right, it can. Do you know what it would cost to have a guaranteed one gig' service? At least $20,000 a month."
So basically Turnbull used a deceptive truth (the cost of a high speed service with dedicated guaranteed backhaul) to blatantly and intentionally lie about the cost of gigabit services that would be sold to 99.9% of NBN-connected customers. The fact that he knew enough to concoct this deception made me realise that he wasn't just as bad as the rest of them, he's worse. He used his knowledge and intelligence to cleverly and carefully craft a statement that's blatantly false yet defensible as technically true if he were caught out. Nobody is that clever by accident.
> Please don't use Hacker News for political or ideological battle
HN is a great place for considered discussion. Let's keep it that way.
> our far-right government, which loathes and fears any technology more recent than coal & witch burning.
Being stuck inside it is not interesting, it's horrifying and depressing.
I moved into an apartment that was built with NBN Fibre to every apartment. I moved here and paid a premium for it because I wanted to get a head-start on the FTTH rollout we were promised.
Instead of the promised ability to move just about anywhere and get the same connection quality and bandwidth options, we now have a roulette wheel where most of the options are mediocre at best.
Worse still, we paid even more for this bungled mix of copper based tech than the original FTTH rollout.
We are one of the most urbanized countries in the world, more so than many who have rolled out fiber.
Big projects, govenment & commercial alike often fail. The real NBN certainly might have. But to make up out of a complete vacuum that it was some sort of deliberate con is just another barmy conspiracy theory.
Lets see, Kevin Rudd said it would cost $15 billion. The total cost at the moment seems to be $51 billion. So your conspiracy theory is that the crappier NBN we got which required less work cost $36 billion more than the one you are saying was budgeted for?
> Kevin Rudd said it would cost $15 billion.
The first 90% fibre plan was $43b, later revised down to $37.4b of which it was supposed to be $30b of investment from the Federal government. [0]
About $12b of that was in payments to Telstra ($11b) and Optus ($800m) in return for access to lead-in and last-mile conduits/pits, and decommissioning of HFC networks.
That original business plan was revised later after a KPMG study to recommend ~93% fibre, ~4% fixed wireless (LTE variant), and the remaining ~2-3% on two dedicated satellites (new launches).
(Edited to add: Note that this plan included an eventual ROI, where the government would get it's money back - and hence it be zero/low cost. The current plan has no hope of that happening short of selling the entire thing off to Telstra.)
FTTH installations were budgeted at $1500/residence to install, which by the time the Abbott government was coming into power was actually trending closer to $1200 as the workforce was trained up in deploying the gear and became more experienced in dealing with any odd situations.
There is/was a fantastic video of a talk given by someone from NBNCo (I believe Mike Quigly?) at a University where they go over the architectural design decisions, and how the 93% number was arrived at. The video also went over in detail how the fixed-wireless process was to work.
Arguments that remediating conduit and installing new copper is either cheaper or less work to install aren't backed up by evidence, with NBNCo's current cost for FTTC being somewhere over $3000/residence in 2019[1].
Additionally, because there's now FTTP, FTTN, FTTC, HFC, Fixed-Wireless and Satellite - the whole network is harder to manage and maintain. You need a bigger, more widely trained workforce, you have more equipment to maintain, and more vendors to deal with.
[0] https://en.wikipedia.org/wiki/History_of_the_National_Broadb...
[1] https://www.itnews.com.au/news/nbn-cos-actual-cost-per-premi...
I did have to go through a few modems to find one that worked reliably. Oddly enough, an old Huawei B315s has turned out to be by far the best for my situation. I've tried fancier models that multiplex radio bands, but they were slower and less reliable.
Weather seems to have little to no effect. What does sometimes cause a problem is tourist influxes to the nearest town. The occasional big weekend event can bring speeds down to 12Mbps or so. Still, I'm not complaining and consider ubiquitous internet availability a kind of modern miracle.
The nbn from its conception was a half-baked, last minute idea made up by Kevin Rudd. If any public network was needed, it should have been wireless. The nbn discounted any possible future that resulted in wireless being a viable way of delivering the internet to everyone. Wireless is also necessary because things aren’t stationary. Cars can’t use fiber internet, for example. Fiber to every home is ridiculously expensive, with a large percentage of homes not even using any more bandwidth than is required to stream Netflix. Fiber also excludes rural areas, and any lower density areas in general.
A public wireless network would be utilised much more, as even a lower speed wireless network that is available everywhere and cheaper opens up more possibilities than a limited high speed network. It isn’t just speed that is important. We also have the issue of data caps. High speed connections that have 20gb per month data caps are pretty useless, yet a large percentage of the population can’t afford any more than that. The nbn, whether fiber or not, would always just result in a product a large percentage of normal people couldn’t even utilise.
The liberal party made the nbn worse, but it was a terrible idea to begin with. All the political parties are corrupt.
I don't believe the rural NBN customers on Wireless Satellite technology would support this premise.
Even customers in suburban shopping centres struggle with their cellular connection - hence Free Wifi being so ubiquitous.
But regardless, our conservative government relies on minerals and education rather than investing in Australia's ability to compete in a digital world.
So here's hoping we can keep up with developed nations who are making use of the high-bandwidth infrastructure made available to them today.
The left proposed an NBN based on 'fiber to the home' technology.
The right threw out that plan and replaced it with 'fiber to the node and copper to the home' technology.
Now it is hard to say if the NBN proposed by the left would have been a disaster as it was only rolled out to a few place. But being fiber based it would at least have been future proofed.
However, now that the most of the money has been spent and the NBN copper system is nearing completion, we are seeing that the NBN as it now stands is a total disaster.
It is already obsolete before it is complete.
Can people not see and pros/cons based on suburbs that have fttp/ftth?
You can do the same with fttn based on some suburbs - where people who have paid for 100Mbps end up only getting 20Mbps during peak hour
Now when I go back to my parent's house I get 10+ Mbps 4G on Verizon and Tmobile. It's awesome.
(The downside is that you have to live in Beatty, Nevada. ;))
I've noticed these kinds of discrepancies every time I leave a major city. I don't trust the G anymore. It's all either really good, or really terrible from mile to mile regardless of how it's branded.
I worked at a Telco that installed extremely old hardware, but because it was theoretically capable of achieving speeds that were similar to the low-end of 3G, they branded and marketed it as 3G.
Marketing loved it.
When you pulled out your phone it clearly said "2G" on the screen, but you were paying for 3G.
This report seems to be lacking a reality check. (Is it practically possible to get ~600 Mbps from LTE? Exotic bundles like LTE MIMO don't really count unless they're actually offered as part of regular plans.)
But not implemented. ALL current "5G" networks are 5G non-standalone in which 4G does all registiration with the basestation and most of the work. 5G is just a bit of extra data channel on top.
(Hence why my later part of the comment references “nanosecond region”)
Thanks for calling it out!
Edit:
To say my original post another way, about the fastest I’ve ever seen for a ping time is ten microseconds over a one meter fiber optic cable. That’s 3,000x slower than the speed of light. Fiber optic cable is far more efficient than a Point-To-Multipoint radio which is in a noisy environment (aka cellular), so to think you’ll get within 2 orders of magnitude using 5G vs the best fiber speeds I’ve seen seems crazy to me.
My point is his comparison of 14ms against his expectation of 1ms. That is not realistic, given the use-case being a speed test to [semi-random] provider.
There is a physical limit in play in these scenarios - being the speed of light in a vacuum and light refraction index (fiber). Simply assuming that due to the fact that 4G is "becoming" 5G you should see an order of magnitude less latency (1ms vs 14ms) is in my view, from a physics perspective, silly.
Even including the RAN (Radio Access Network), the majority of said latency to and from the throughput server endpoint is most likely NOT the result of delay on the RAN, but rather the physical distance (proximity) from the mobile station (MS, smartphone/handset).
This article provides a good overview of how the Telstra deployment works: https://cellularinsights.com/telstras-gigabit-class-lte-netw...
Sprint use to offer 4G in the United States that was Wi-Max based, but now almost all 4G is LTE. In Australia there was a Wi-Max provider (Vivi Wireless) but on a different frequency than the US and only via hot-spots (no phone support, that I know of).
Is 4G for all these reports LTE? What exactly is 5G in all these contexts? Are we talking about the same protocol, or another collection of protocols expected to have x speed at y range?
Exposing protocol details to consumers when those details do not correlate with performance is likely to just cause confusion.
When it comes to discussing these two distinct clusters of spectrum, the difference is so stark that we might as well be talking about different protocols.
5G and 5G pro?...
Recently moving back to Australia after a year living in Finland, I really miss unlimited 300Mbit/sec 4G for 25€/month (~$40 AUD).
It's well and good that our mobile networks are as good as the Finns (anecdotally), speed and reliability wise (whoops, as long as you dodge Telstra outages). Sure there are a couple of 'unlimited' plans popping up in Australia now, but it's more like 'trickle at 1.5Mbit/sec when you use up your allowance'. Like a lot of things, I think Australia would do well to copy-paste what the Finns are doing. Education, telecoms, energy, well-being, ... for our post-coal economy we'd do well to learn it and scale it.
I moved back from the Netherlands a few years ago. I'm awaiting the NBN at my place in the next few weeks hopefully. Then I will be able to get the same connection that I had in Amsterdam in 2011. This is after the government has spent $50bil+ on the NBN - go Australia!
Good luck with the NBN lottery! In inner-city Brisbane I'm lucky to have NBN and non-NBN fibre options. But I would cancel fixed fibre in a heartbeat for unlimited 5G if it meant only taking a bandwidth hit.
It's fast, but it's not cheap compared to landline. Unless you use a absurdly high frequency that probably won't work in most of Australia, 5G won't be much faster for the individual user. However MIMO does up the cell capacity, so it will probably be cheaper. That's not a small thing.
It was INSANELY fast. Presumably because I was one of the few folks on it and Verizon had built out all the towers and bandwidth was ready for growth.
As time passed the performance suffered, granted 4G seems plenty for me most of the time.
Basically, as an early adopter you were a 4G user visiting the 3G web. But now it's the 4G web.
Is all this telecom positioning to setup a replacement for landline internet service?
I wonder what it would look like for my TV to be connected and sending my viewing habits back to the manufacturer without a method for me to turn it off. 5G makes this idea much easier.
Phone companies can get contracts with all these device manufacturers and make money that way.
If you have cable, satellite, or streaming this already happens. The only way to watch TV without being watched is to use an antenna or local storage.
Pi-hole[1] might still help at the moment, until the TVs start coming with their own SIM cards.
In any case, I wouldn't use a TV as anything more than an extra large monitor for Raspberry Pi running something like Kodi[2].
[2] https://kodi.tv/
There's no way this can be cheaper, especially in the US where carriers are currently charging $10-20 per device per month. I can imagine a dystopian future where device vendors think "x% of people can't figure out how to connect to their own WiFi, so let's ship a 5G radio with Verizon eSIM instead and make up the extra cost with more privacy-invading ads" but I'm going to ragequit the Internet if that happens.
Seems disgustingly plausible, tbh.
Comcast, Verizon, AT&T, and T-Mobile/Sprint are all working on vertically integrating to provide phone service, 5G, and streaming content options.
Comcast is now in the cell phone business, and owns NBC and Universal. T-Mobile (soon to be merged with Sprint) bought Layer3 and is now in the TV business.
AT&T owns Warner Brothers and HBO. Verizon will likely follow suit.
No coax cables. No modems. Just streaming content over 5G, via oligopolistic walled gardens.
It will be interesting to see what kind of price-bundling options they might offer, if any, to combine this fixed wireless subscription with a mobile phone subscription.
Sure, you could buy a 4G/5G router, but for average consumers that may be a thing of the past. At least, that seems to be the direction that every major player in every relevant industry is gearing up for.
Medium (wireless vs copper vs fiber) defines who is faster.
I can only think of one conversation I've had in which 5G was even mentioned, but boy, I know people here in rural Canada would like affordable data plans, and I don't see that 5G is going to help at all.
5G has some positives and negatives over 4G. For example, 5G frequencies interfere with the detection of water vapor for weather forecasts, are hyper local so lower latency but need to have towers in every neighborhood, can handle more devices connected (so all the hardware reporting back to manufacturers and others can phone home), and has the potential for faster speeds.
In theory, 5G is faster than 4G. In practice, I'm not likely to see the faster and I don't need it for my day to day activity.
I can name GPRS, EDGE, UMTS, HSPA, 1xRTT, EVDO, WiMAX, and LTE off the top of my head and tell you what “G” they are associated with (1, 2, 3, 3.5, 2, 3, 4, 4 respectively).
5G is a vaporous term with no real definition right now other than “whatever the fuck the ITU has decided to call 5G”.
EDIT: I guess they’re calling the air interface “5G NR (New Radio)”. How memorable. Guess with everyone finally getting on the same page with LTE in the current generation they felt no need to come up with a real name.
If we don't look at the bigger picture we can't look at the wider impacts, the ethics, or any of that. This is something I learned in engineering ethics. It was a required class at my university for engineers for a reason.
Providers love to boast about max speeds, but the reality is every device under a base station shares the total capacity, more devices means competition and slower speeds.
Carriers started releasing what they called "4G" networks, but those networks did not meet the standards required to call it 4G. So they came up with LTE because it's a long-term evolution of 3G service, evolving until it meets the standards for 4G.
http://www.itu.int/net/pressoffice/press_releases/2010/48.as...
[0] https://www.lifehacker.com.au/2019/06/mobile-broadband-vs-ho...
Yes, the original NBN plan was definitely worth it. Even the current NBN plan is still in many cases going to be better.
The reason <choose wireless technology> is "faster" than <choose NBN connection option> is because there's a lower density of people using it for higher bandwidth applications.
Do you remember the 3G networks in Australia before the iPhone 3G launch? It was fantastic and fast, and there were some unlimited plans but few phones that could make use of it. Then the iPhone comes along, everyone suddenly every mobile network grinds to a halt as they don't have a high enough tower density and/or not enough tower backhaul.
It took Vodafone and Optus a long long time to fully recover from that.
If carriers don't carefully manage demand, they're going to run into major congestion issues - their networks just can't handle a significant shift of data from wired to wireless.
I guess there may be cheap plans like this that you could use. It's generally operated by MVNOs [1], you could look for those.
[1] https://en.wikipedia.org/wiki/Mobile_virtual_network_operato...
Telstra 5G 2019: 3.5GHz
Telstra 5G planned future: 26 - 28GHz
The estimated density for a 4G network is: 4,000 devices / km^2.
The estimated density for a 5G network is: 1,000,000 devices / km^2.
The density means we can scale up machine <-> machine interactions significantly, rather than focus on human <-> machine interactions that we have today.
I dont know what their 5G is bit I suspect it is not 5G