Hacker News new | past | comments | ask | show | jobs | submit login
South Korea now has 11M 5G subscribers (telecomlead.com)
61 points by finphil on Jan 2, 2021 | hide | past | favorite | 68 comments



I remember listening to a Masters In Business podcast [1] with the President of Qualcomm and listening to his explanation of why 5G was going to be a huge game-changer. He perked my interest by mentioning that he thought that _everything_ would have a 5G chip in it one day - garbage cans, mailboxes - but otherwise did not by my judgment give any significant reason that 5G would change something in the market when compared to 4G technology.

Does anyone have a strong reason they think that higher speeds and (I assume) low latencies would unlock something in the economy we're currently unable to achieve? I just don't see it.

[1] https://www.bloomberg.com/news/audio/2020-03-13/cristiano-am...


The real motivation here is that mobile carriers want to take market share from cable companies. Mobile carriers want to be able to provide wireless to your phone that is good enough that you will pay more for an unlimited plan and cancel your cable subscription. 5G doesn't really do that much for garbage cans, they could already have 4G or wifi chips in them if anyone wanted that.


In most EU countries and in the US, major wired network companies are already in the wireless market (T-Mobile, Vodafone, AT&T, etc.), so your take would make sense only if they were to stop laying fiber, and replace it with towers. To achieve this though, it would require a massive investment, especially to circumvent network congestion.


> To achieve this though, it would require a massive investment, especially to circumvent network congestion.

AKA: 5G


Not quite. It would require a massive investment on top of the current 5G deployment.

Replacing fiber, cable, or DSL with 5G in dense urban and suburban areas is definitely not what is currently under development.


The wireless distribution layer is really the only place mobile carriers are meaningfully bandwidth constrained. It's (relatively) easy and cheap to 10x or even 100x bandwidth available at a tower. The hard part is delivering that capacity to the end user. That's where 5G comes in.


There is an argument that it's crossing a magic boudnary to the point wherein it's almost like 'WiFi' as opposed to 'Wireless' in the sense that the bandwidth is there, and there is plenty of it.

If Qualcomm and the Carriers were not stuck in the 1940's with their utterly byzantine business models (remember 10-cents per WAP page on feature phones? They thought there were geniuses and were going to dominate with that Einstein business model) - then - the notion of 'your toaster on 5G' is a very material feasibility.

There is definitely the theoretical possibility for a 'true IoT revolution', but I think that the variety of ways they shoot themselves in the foot will keep any revolution at bay for a while.

Some 'dynamic' carrier in Japan, Korea or wherever could literally do something transformative and the rest wouldn't budge.

It took the creation of the iPhone (innovation + market power) to force a lot of theses parties to shift gears.

Startups, without power, won't be able to budge carriers, they have to be forced.


AR is one case. Maps are great today, but just last week someone delivering packages saw that I was pushing a bike and asked me how to get into the local bike shop. The entrance is in the alley, and at the end of canopy that is not easy to find at all. I believe that image recognition and a very fast connection could help with that problem.

It seems like a gimmick, but how many times have you known the address and not known where the door was? It happens a lot in medical buildings.


I'm all for an Assassin's Creed style augmented reality view of the world! But I don't see why it's something 5G would usher in over 4G (or 3G).


Just got an iPhone 12 in Australia and I see 1100mbit downloads regularly on sub 6 5G. Maybe the network isn’t saturated yet, but that blew me away. That isn’t too far away from mm wave, and I get that inside my apartment. What is Telstra doing differently here to make it work like that? Also worth mentioning, I would consistently get 150mbit down indoors on 4g. Feel lucky to have such a fast network here. I pay ~$30/month for 80gb data


I got an iPhone 12 about a month ago in the US, and noticed my town has 5g. My town is < 1 square mile, and a population of just 600, so I fired up a speedtest and was astounded... I was getting 560kbps, at 6am, on a Saturday.

It is actually very difficult to even send iMessages at those speeds, and forget about sending a photo to someone. If I'm not on my wifi at home, my phone is almost useless as anything but a phone.


Which network are you on?

Verizon's Nationwide 5G is slower than 4G: https://www.pcmag.com/news/heres-why-verizon-iphone-users-mu...

AT&T is in the same boat as Verizon: https://arstechnica.com/information-technology/2020/09/atts-...

T-Mobile is in a similar boat as they have a narrow chunk of 600Mhz dedicated to 5G service, but they own a bunch of spectrum right above 2.4Ghz WiFi (usable with much more powerful radios) that should enable 1100Mbps speeds and higher, but it will take another year or two to see it widely deployed.

Verizon is very wireless spectrum starved right now, AT&T is in a bit better position but still needs more sub-6Ghz spectrum, hence heavy spending to buy 3.5Ghz spectrum: https://www.lightreading.com/ossbss/will-c-band-auction-set-...?


I’m on Verizon, they are the only cell service in town.


Are you sure about this? AT&T has been building sites at a fast clip to fulfiul their FirstNet obligations: https://about.att.com/newsroom/2019/fn_purpose_built_cell_si...

T-Mobile has also been doing a big rural buildout, and even built a map to show coverage on all major carriers: https://www.t-mobile.com/coverage/lte-comparison-map


My sister in law has AT&T, and her boyfriend has T-Mobile. So we are a full spectrum of the American cellular network. We all get full bars of "signal" while outside, but Verizon is the only one (4g or 5g) that will actually connect to the internet to do a speedtest.

Everyone in town just chalks it up to that is the way it goes, but worse than that, we only have one internet provider in our area, and from Christmas to New Year, we had a total of 50 hours without internet.

We live in a tiny tourist town that is a 45 minute express train ride to/from NYC, and while we may be close in travel, we are very distant in our available amenities.


Is the phone saying its connected to 5G?


I thought part of the 5G features of iOS was that you could disable 5G and force the use of LTE. Have you tried doing that?


What good is 1100mbit if you run out of data after a little over 10 minutes?


You have the amazing opportunity to purchase more data at higher rates!


all that speed and still data caps. Congratulations on your ~15 minute data package.


This should bode well with the phone addiction article on HN.


Does anyone know what infra SK telecoms are using for 5G?

In particular, are they using or not using Huawei? What’s their domestic security establishment’s take on it as a vendor?



Did ericsson resolve the issues getting their base station silicon out of intel's 10 nm snafu? Did they switch to TSMC? Or did they get preferential treatment, edging out CPUs making the 10nm shift?

Inquiring minds want to know.



"The country’s 5G networks still lag compared to promises, with the average 5G download speed reaching 690.47 megabits per second (Mbps)."

I assume most of these people are using this speed for cell phones. 86 MB/s sounds really great for a phone, and better than a lot of people in the U.S. have in their homes. This should only get better as well.

Assuming the technology is safe, this will be a major game changer for society.


> Assuming the technology is safe

Whoever invented the 5G conspiracy theories is either having a laugh or is fundamentally disconnected from how radiation works. This pseudo-scientific nonsense has no place on HN.

One of the key design features of 5G is to drop transmit power (instead having more sites, with lower range per site).

Because signal strength drops off with the square of distance to the tower (technically the cube, but because they transmit a flat beam the relevant part of the graph is square), this results in dramatically less transmit radiation vs 4G.

However, all of this is moot, because the radio frequency used isn't a frequency that could ionize anything even if the transmit power weren't negligible.


Every time this is brought up, the only argument that comes up is that "radio frequency used isn't a frequency that could ionize anything even if the transmit power weren't negligible".

This is absolutely true but fails to recognize that radio frequencies have harmful, non-ionizing effects on biological organisms.


Do you have a citation for "harmful" non-ionizing effects? As far as I can tell, the body of literature has no conclusive evidence to suggest that radio waves do anything but heat the tissue that absorbs it.

Rant: Like many things that receive attention on popular and social media, agonizing over 5G distracts from issues that cause real provable widespread harm, such as UV radiation and air pollution. We rarely see articles rise to the top when they cover non-controversial points that would do well to be read by everybody, because how much buzz and discussion can you generate from "wear sunscreen", "don't drive while tired", and "get a prostate exam before you turn 50"?


(not the parent poster)

I don't know for sure, but I'm setting up an experiment this year where I irradiate cells with UV and try to use 5G radiation to jam the BER mechanism, which is one of the DNA error correcting systems. BER uses the charge state of DNA (https://www.pnas.org/content/100/22/12543) to help find thymine dimers, and 5G radiation is in just about the right frequency range to excite nuclear DNA and induce a current in DNA. I would be looking for a higher rate of mutation in the UV + 5G case over UV control, and no effect in 5G / control without UV.

What is interesting about this experiment is that 1) it's easy. 2) you would never detect this using normal procedures to assess mutagenicity, because normal techniques assume that the harm comes from direct mutation effect of the radiation; in this case the harm (if it is there) would come as a secondary effect, by inhibitng proper repair in the presence of a previous insult. Moreover, typical lab experiments (rightly) attempt to reduce outside effects (especially mutagens), so it's less likely it would have been discovered by accident.

Incidentally I'm sick and tired of people saying "it's non-ionizing and therefore not dangerous". This is a very reductive statement that belies a complete lack of understanding of biophysical chemistry.


How can I follow your work?


twitter under same handle as hn. I am slowly ramping engagement on this, you'll have to sift through posts about elixir


Which GHz frequency?


I suspect the person refers to how microwave ovens heat through water molecule resonance, around the 2.4 ghz spectrum


Even then; the human body is pretty good at dealing with excess heat, and conservation of energy says you can’t get more energy deposited than the transmitter puts out - even if you absorb 100% of the signal that’s still only 100mw or so for a phone (vs ~2000w for a microwave oven).


> This is absolutely true but fails to recognize that radio frequencies have harmful, non-ionizing effects on biological organisms.

What harmful effects are you referring to? Outside of intense heat or light, there shouldn't be any potential hazards to biological systems, since non-ionizing radiation by definition cannot remove electrons from atoms.

https://www.cdc.gov/nceh/radiation/nonionizing_radiation.htm...


What mechanism do you propose to create these effects? (To be clear, we are not talking about high-energy cosmic ray interactions, because those don’t happen at these frequencies).


When he said that, I assumed he was talking about China using the technology overhaul to sneak in spying capabilities through Huawei. I forgot people don't know how radio waves are the opposite side of the light spectrum from x-rays.


I think a common misconception is that the concern is about the ionization potential of the radiation. The way I've heard it explained is that microwave frequencies (including, but not limited to, 5G signals) affect the ion channels of cells, preventing normal cellular function and increasing the risk of a variety of disorders, particularly neurological.


The first part of the comment seems to cover this claim already, at least in relation to 5G rollout being bad, without getting into whether it's even a valid claim or not.


> Assuming the technology is safe, this will be a major game changer for society.

Will it really? 4G reaches 70 Mbps, how do you think that'll change society when we now can have a whole ~600 Mpbs? My guess is, people will stream videos faster, but that's about it.


Improvement of boring things leads to interesting things. When we went from 2G to 3G/4G, we went from images to videos. 10x improvement in data transmission may lead to other forms of media becoming more readily accessible. If what is imagined is just “faster video loading”, that is an awful waste of data.


What are these "other forms of media" though? Video streaming didn't suddenly come into existence when we got to 3g, it was already common on wired internet and only became accessible on cellular internet.


Realtime 3D video and virtual environments.


Not exactly sure what "Realtime 3D video" is referring to, maybe this? https://dm.octonus.com/features/realtime-3d-video-capture Seems the "Extra high-quality, large file size" ends up being 87-160 MB per minute, hardly impossible to do with 4G...

Virtual environment been possible to stream on the fly for a long time. Only thing that will change with going from 4G to 5G is that the quality will be higher, hardly a "major game changer for society" as liquidify put it.


yup, 480p videos from 2010 look very dated now when everything is 1080p.


480p doesn't look bad now because you're used to 1080p, it looks bad now because you try to view it on a 4K monitor (or something else, but it's order of magnitude better than what you tried to use for 480p back in the day).

Also, 480p vs 1080p is not a major game changer for society, at least in the society I live in.


4G can go much faster. I get 175Mb/s: https://www.speedtest.net/result/a/6889068582

I wondered what I could possibly do with 600Mb/s, but I noticed that Telenor isn't charging any extra for a 5G connection (my phone doesn't support it). Isn't the main feature that the speed is still high even when there are lots of people using it, rather than my test at midnight in a city centre?


and less towers/infra is needed, which makes it possible to give small towns fast internet that they couldn't do practically with 4g tech.


I am unable to get a decent wired internet connection in the middle of a city center in the U.K. (13Mbps peak via ADSL).

I’ve recently switched to using a 5G connection for home internet and whilst it’s very early days - a slight sacrifice in latency (ADSL isn’t great there either), and a daily modem restart, is a welcome price to pay for 250Mbps down/50Mbps up (at the same price point).


'Cloud computing'? People having remote access to much higher power machines than they do at home. And being able to use them with minimal lag?

Cloud gaming companies are going to benefit a lot from 5G I think. Also Machine Learning on remote GPUs.

I've also read that car manufacturers want to use 5G technology for cars to communicate with each other.


> I've also read that car manufacturers want to use 5G technology for cars to communicate with each other.

That sounds like a marketing statement from some car manufacturer who wants to sound like they on with the times. There are plenty of technologies and protocols they could deploy to allow cars to speak directly with each other (without any 3rd party infrastructure like antennas), no 5G needed here.


All that car stuff came up in some of the various specification papers about 5G.


When you look at phones each as its own, I agree that more bandwidth is not really needed. I certainly never felt that I absolutely needed more that LTE bandwidth when on my phone.

But 5G will improve _total_ bandwidth I assume, which is in fact going to improve the situation for very congested areas.


The bandwidth would be quite useful for music festivals and large gatherings.

I've been to music festivals in the centre of a city (Glasgow, Scotland, UK) where my phone usually works quite well, but at that festival I'd struggle to send a text.

That's when festivals are allowed again, mind you.


Or stream very high resolution video which was unpractical with 4g.


The thing that really interests me is the potential to properly eliminate wires (especially when/if device to device becomes possible).

For example, the Test and Measurement industry (see eevblogs interview with John Kenny - a really good one, seems like a great boss to have and a great engineer to) is apparently really looking forward to being able to having very high throughput links between devices such that they can synthesize and fuse data without having a literal maze of wires - e.g. You have a signal generator and an oscilloscope, you should be able to make a (not very good) SpecAn from it.


Consumer SSD drives only go up to around 500Mbps, and it's far less for SD card storage. So you won't be even able to download a file with 690.47Mbps! (To stream a video in HD, you only need about 5Mbps).


Eh, consumer-quality SSDs these days can hit a good bit more than 500Mbps even with writes. This Inland NVMe drive lists 1,900MB/s write speeds (15.2Gbit/s). ~$60 for 512GB. Even if we take its rated speeds with a grain of salt for sustained writes and cut it in half, that's still >7Gbit.

https://www.microcenter.com/product/600420/inland-premium-51...

That said, almost all phones are not using such an interface or controller for their storage and I do largely agree having a phone with a >1Gbps link is a bit overkill for a handheld mobile devices. Ultimately though once there's a bunch of clients on it your actual throughput in practice will be much slower. The big win is that the overall channel (or channel grouping) can now handle a ton more aggregate throughput meaning higher realized speeds per customer for the same amount of spectrum. Being able to get your slice of data faster means you can quit talking or quit listening sooner and someone else can talk or hear something they care about.


"consumer grade", I meant SATA SSD. Although I acknowledge that NVMe drives are entering the consumer market these days with insane write speeds that you'd only find in the data center before.

Also, good point about why speed is important - it can certainly help serve more customers!


iPhones are using Gbps interfaces to interact with NVMe, that is part of how they perform better than competing phones: https://bgr.com/2016/08/18/iphone-vs-galaxy-note-storage-spe...


Are you sure about those units? Typical SATA SSD can reach ~500 MegaBytes/s, not Megabits/s


yeah, you're right! Seems like I'm even more confused about the specs than the average consumer...


That's the speed at which I'd start expecting my phone to natively allow multiple video streams on screen via different apps, a versatile audio mixer, and just a much better multitasking environment overall. I think/hope we're just getting started in this area on mobile.

(I'm weird and people always ask if this creates focus issues, but personally I run about 6-8 video streams at once on the desktop, typically with only one stream given audio presence at a time. It's been great for extra energy and interest-deepening while working, which helps prevent burnout)


Out of curiosity, what streams do you have up? And how do you work without getting distracted?


I usually stream a lecture, a few favorite TV shows on random playlists, some thriller or action movies, and maybe a favorite Youtuber. The scripts I use for this really randomize things into 200+ item playlists per window, so I can quickly flip through anything I don't want to watch right now.

IMO distraction and healthy prospective activity (stimulation, personal interest, cool new stuff) have become tragically intertwined and confused.

To me, distraction in the negative sense is the byproduct of a lack of what I call the "three C's", comfort, clarity, and confidence. Once I'm sailing through the day productively, singleminded focus is now a problem that can easily turn into productivity exhaustion. The streams help prevent this with entertaining mood support based on my core interests. (I wish everything I had to do at work was based on my core interests!)

BTW I don't generally watch new material in this context. It's usually from my favorites I've already seen, unless I'm ready to take a break.


Frustratingly my Carrier in HK doesn't support 5G even though it runs on a network which has supposedly deployed 5G already. To be honest, it doesn't make any difference to me, except in crowded buildings where there is no reception and perhaps 5G would've helped. When I do have reception, LTE speeds are already more than enough.


I was surprised to see my iPhone 12 hit 1.5 GB per second in Saint Paul Minnesota.

The future is coming...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: