An interesting side effect of these networks shutting down seems to be a lot of car manufacturers have been cheeping out for years and relying on 2G and 3G networks on even quite modern cars, then blaming the network providers when they stop working after being the network is shut off instead of offering upgrades to the telecommunications modules to a more modern standard.
The newest car I've ever owned was a 2005 model, so at the risk of sounding ridiculous: Why do cars have cellular radios built-in? Just for the GPS, or for actual communication?
Telematics & remote interaction. Want to get remote diagnostics/problem notifications on your car, have emergency 911 service in the event of a crash, be able to unlock it from your phone, preheat it in the winter, precool it in the summer, get notified when your 16-year old child is driving over 70mph, get software updates for the ECUs in your car, enable remote charging notification & management, etc.? Cellular.
Note that the HN crowd is extremely outside the mainstream in its collective view on these kinds of features. The actual mass market for cars embraces most of these even though HN tends to vilify attempts to add connected tech to vehicles.
Just because the actual mass market embraces these things doesn't mean they're good for us all. The skepticism is about how needlessly dependent it makes our individual cars upon a broad shared resources like CDMA and coprorations' servers, in perpetuity.
Tesla drivers locked out of their cars en mass last week because Tesla servers were down? Okay, it's relatively benign, and maybe it has helped people not get locked out of their cars on the whole, versus how often it locked people out of their cars.
But year after year at DEFCON, it is demonstrated that malicious actors can take remote control of your "connected tech" car, while you're driving it. The terrorism potential alone is scary, aside from all of the mildly troublesome tracking and privacy implications.
I'll trade away precooling my car in the summer in favor of exclusively controlling my own vehicle, thanks.
The problem with telematics systems: even if you don't buy the subscription, the telematics system is still active, reporting your location, speed, environmental data, and use of various accessories like windshield wipers etc. to the manufacturer.
A friend used to work for a company that was purchasing that data from automakers to try and enhance weather data for real-time weather reporting.
Let's run down that list of benefits:
911 calls in event of a crash? iOS 16 is going to have that built-in. Ford vehicles (and others, I think VWs too?) can use a bluetooth paired phone to place a crash-detection 911 call.
Remote diagnostics? Only authorized dealers can access, and if you're going to your dealer for non-warranty service, you're a fool. Massachusetts passed an extension to Right To Repair requiring automakers to provide indie garages with access to the same telematics and Subaru responded by remotely disabling all remote diagnostics in any car registered in Massachusetts. That's how petulant they are.
Remote keyfob based starters (optionable on any Ford/GM and a ton of other vehicles) don't need monthly subscriptions to preheat/precool. Idling your car especially in the winter is terrible for the engine and environment anyway; it's much better to just get in and gently drive the car until it has started to warm up. 3/4 of the energy in gasoline is converted into heat.) In the summertime the fastest way to cool a car down is to open up the windows and sunroof, and drive for a minute or two.
Very, VERY few automakers do remote software updates for ECUs. For example, the Chevy Bolt, an electric car, will only update its entertainment software "over the wire" even though Chevy has said they have the capability to do OTA updates for the drivetrain. Ford vehicles can update the infotainment system via WiFi.
Remote charging notification/management? Largely useless except for very specific, limited circumstances. Electric cars have built-in scheduling and charge limits, in many cases location-specific settings can be done. There are also "smart" EVSE adapters that allow for max SoC and charge time settings. If you need to monitor the percentage charge on your electric car regularly, you have too much time on your hands.
Some countries (or states in the usa) demand built in emergency services connectivity for when you crash the car for cars manufactured after some date. Before that date, a lot of cars already had that before it became mandatory.
I know tractors have the same problem, but farmers are paying because they can log into each tractor and make decisions. Things like a field isn't yielding great so they can stop for lunch before getting the next load off...
I personally don't see drivers upgrading their car, but maybe I'm wrong.
This is mostly about drama around Boost Mobile. T-Mobile wanted to shut down their CDMA network on January 1, 2022. Boost (now owned by Dish) doesn't want to upgrade users that have handsets that don't work with VoLTE. Upgrading users involves sending them notices, giving them options of free handsets, shipping those handsets, getting the users to actually activate them, etc.
While T-Mobile gave Boost nearly double the required time, Boost complained that it wasn't enough and regulators don't like the idea of customers losing access. T-Mobile delayed the shutdown for three months to quell the situation.
Dish is notoriously cheap. They're supposed to be building out their own 5G network, but it's looking like they don't want to spend the money to do that. Dish has also been losing the wireless customers it acquired and so they might be souring a bit on the business.
> T-Mobile gave Boost nearly double the required time, Boost complained that it wasn't enough
The CPUC says T-Mobile representatives lied under oath. They claimed that Dish would have until 2023 to migrate customers off of the CDMA network. T-Mobile is facing sanctions because of it:
The hearing was scheduled for 2021-09-20, over two months ago. Any idea what the outcome was?
I tried poking around here[1], but CPUC public discoverability is up there with the worst of enterprise...never mind that a decision/resolution/ruling/general order all sound roughly asymptotic to a non-lawyer.
Thank you for sharing this. This is actual news and I am somewhat dismayed that none of my news feeds picked it up ( I have long given up on dramarama news edition ). I clearly need to made something more custom.
It's kind of amazing to watch them keep pushing out half-baked protocols with the details on what phones actually support being so hard to find. I guess the lesson they learned from rolling out LTE phones with no Voice over LTE support is that nobody really cares, which is why they repeated the same issues with 5G and there's all this stuff about "non-standalone 5G".
Well VoLTE was the first time the world moved to an IP based Voice Calling Protocol. So it was hard for everybody. VoNR will hopefully be a lot easier.
Not mention the actual VoLTE standard came at a slightly later date than LTE itself.
Whitelisting Smartphones by US Carriers is another story.
VoLTE works fine, its the ridiculous rollout strategy that requires every phone to have specific carrier settings in the firmware. This ends up in ridiculous situations where some phones will work on some carriers with VoLTE, but not others (despite both phone + network supporting VoLTE), depending on what firmware you have.
I'm not sure if this was just an Android problem and Google totally dropped the ball, but I cannot understand why there isn't a central database that gets pinged for VoLTE settings dynamically and the phone uses that. Instead, it seems to be need to be shipped in the OS.
iOS is totally fine; but this has totally killed adoption on Android. It's improved recently at least in the UK (last two phones worked out of the box), but previous to that it was a total farce.
It might be to find info about VoLTE support because VoLTE phones started coming out seven years ago.
The Nexus 6 was the first Nexus/Pixel device to support VoLTE (I seem to vaguely recall the 5 was capable of VoLTE, but no carrier enabled it?), and the iPhone 6 was the first iPhone to do so.
I think you'd be very hard pressed to find a 4G phone that doesn't support VoLTE.
GSMarena usually has the most extensive specifications, however, if you're looking to check a specific phone.
How do you keep tabs on all this stuff? Are you reading press releases? I often wonder how some random person on HN seems to always know so much detail about some specific company. I am also trying to figure out the best way to do a deep dive on a company (any company).
There are plenty of sites that publish news about cellular carriers. It's a technical pursuit like any other. You can assign someone to read Microsoft press releases all day, and they won't ever become an expert Windows programmer... You won't be able to become an expert on a company with a crash program, either, years of accumulated knowledge and subject matter experience and interest count for a lot.
> know so much detail about some specific company.
The three major US cellular carriers hold a great deal of power over all of us. It's something people keep informed about. No doubt around here quite a few people's livelihoods are dependent on keeping abreast of cellular technology as well, providing extra motivation.
Smells a lot like the Nextel -> Sprint transition back when I was working for Sprint. (Similarly Verizon's buyout of Centennial Wireless) I quit after the stress involved with companies coming in to move 200+ nextel phones over at the last minute only to discover that the Sprint network would never give them the same coverage. Plus their push to talk features everyone in the oil field relied on wasn't going to work deep in coastal industrial areas like they were used to due to it moving to a much higher frequency.
You basically had to maintain some level of cognitive dissonance to sell these companies on Sprint service because we all knew it would never work.
> Dish is notoriously cheap. They're supposed to be building out their own 5G network, but it's looking like they don't want to spend the money to do that.
Did they ever even build out a 4G network? Some years back I was a regular on Howard Forums[0], and I recall from discussions there that Dish had sat on some big chunks of spectrum for years and never built out a network with it, even back when 4G LTE was the latest and greatest. I guess nothing has changed.
Is that they do not want to pay or is it that the big 3 carriers have bought up wide swaths of spectrum in the "auctions" from the FCC and are just holding them, not using them.
This is the big problem, the FCC is far to lax with allowing the big wireless carriers to buy and hold spectrum for years (in some cases decades) never using it for anything just so no one else uses it either
In general, no they aren't required. However, if we're talking 5% of your customer base, you might not want to lose them.
In Boost's case, they might legally have to. Dish bought Boost as part of T-Mobile's deal to buy Sprint. I don't remember all the specifics of the agreement or what was committed to in terms of non-VoLTE Boost customers. T-Mobile had to give Dish 6 months notice before shutting down the CDMA network and they gave 14-months (now 17 months with the 3-month delay). I don't know if Dish committed to upgrading those customers as part of the deal (I know that it's their responsibility, but I don't know if they're required to upgrade them or if they can just wave goodbye to customers that don't want to migrate).
Ultimately, I think Dish is at least informally required. Dish is likely going to be looking for goodwill from regulators in the coming years. They have major build-out requirements coming due soon and they don't seem to have made much progress - and they have licenses worth tens of billions at stake (worth more than the company's market cap). Angering regulators at this point isn't a great move for Dish.
It seemed strange that T-Mobile would have a CDMA network, until you remembered that it came from Sprint.
This seems like good news. I'm glad that starting with LTE all the carriers started agreeing on a single platform with SIM cards. Even though there are little nuances between each carrier, the fact that you can buy a single phone and have it work with the four (three now) major carriers is a big deal.
Not really. There have always been smartphones that supported both GSM and CDMA. The three major carriers not only use different LTE bands, but also Verizon and shortly AT&T (in February) have adopted a whitelist model where only select handsets can be activated on their networks. It's a much less open marketplace than it was five years ago.
That's why I left AT&T. They shut off my daughter's phone, without warning, in the middle of the night because it wasn't whitelisted. Their whitelist is utterly arbitrary, and even brand new phones are not on the list if they happen to be an international version. They explicitly state that certain models of the Samsung Galaxy S20 are not whitelisted.
As for my daughter, she was under the impression that she had until February of 2022 to replace her phone, but it was disconnected five months earlier than anticipated. I'm now with T-Mobile and satisfied. For my location it has better coverage than AT&T did, is significantly faster, and I have unlimited data.
By select handsets, I assume any big name handset, e.g. iPhone, Pixel, Galaxy, would be allowed, but a no-name handset wouldn't be allowed?
Isn't this similar to the pre-iPhone days, when if you brought an overseas GSM phone you had to go through some hoops to get it activated with Cingular/T-Mobile? I remember getting handsets from China and Europe and the reseller would have to circumvent something to get it activated on the networks.
Generally, but with a lot of exceptions and weirdness. For example, my OnePlus 6 is not supported, but if I modify my IMEI and APN those "impossible" features like VoLTE suddenly start working. Before they stopped automatic activation, I could put my SIM out of a supported phone, put it in an unsupported one, and those features would work for between fifteen minutes and a couple of hours.
On the other hand, some handsets that are blatantly missing LTE bands are on the whitelist, such as the Chinese Oneplus 8T.
Keep in mind that supported and features are not the same. Even if it technically can work doesn't mean that the provider wants to support it. Sometimes it's a simple as simply not having put the device through the testing process yet, or it simply hasn't been documented. And until it has, they don't want to risk any service costs so they simply don't whitelist it when the user base is insignificant enough.
That argument sounds good until you remember T-Mobile, a company with less than half the annual revenue of AT&T, has absolutely no problem servicing any device with the proper frequency bands.
>when if you brought an overseas GSM phone you had to go through some hoops to get it activated with Cingular/T-Mobile?
To be frank: No. You popped your SIM in and the phone worked immediately on the network. If it didn't work you manually entered APN settings in.
Some years later AT&T started maintaining a whitelist of smartphone IMEI numbers so they could charge more for data plans than a featurephone would pay. T-Mobile did not do this. AT&T went as far as maintaining whitelisting for phones that would be allowed on wifi calling or voLTE, T-Mobile phones just needed to carry the profiles.
No 'circumvention' other than carrier/subsidy unlock was ever required to use an international phone save for the very rare times international phones had bands software disabled (and this is quite rare).
This doesn't really go into how it's legal, but the situation around VoLTE has been difficult. It hasn't just been every phone working perfectly. Beyond that, it's possible to buy a phone that sort-of works.
For example, let's say that you're an AT&T customer and AT&T has 700MHz, 850MHz, 1900MHz, 1700MHz, and 2300MHz spectrum. You buy some foreign phone that supports 1900MHz. Great, now you're running a device that won't have coverage everywhere that AT&T has coverage. You're running a device that won't be using the full AT&T network and your speeds will be really slow compared to devices that fully support the network. You go somewhere with friends and you're like, "AT&T never works when I leave <city-name>!" Except that AT&T isn't the problem - you're the problem. You complain that your device is always slow and how AT&T is crap - but it's your device that's the problem.
Semi-compatible devices are just terrible for network operators. Let's say that AT&T wants to move that 1900MHz spectrum to 5G - but now you have a device that will stop working. Do they need to reach out to you? Do they need to offer you an upgrade? Sure, you and 90% of people would be like, "I knew what I was getting in for and it's my problem." But omg the 10% of people that will yell and scream and the meetings you'll need to have with regulators and the delays in your plans and the free devices you'll offer them as part of a settlement and all the mailings and admin.
Offering tech support isn't cheap and it's more expensive when you're dealing with random crap from users. Now you have some rep on the phone who doesn't totally know what they're talking about trying to figure out why they're getting bad coverage on their kid's device while their device has good coverage. They don't want to hear, "well, you shouldn't have bought some POS off Amazon that isn't even meant for the US market." They're going to say that it's a compatible phone - compatible being a black/white issue, not something that's only partly compatible and will technically get coverage 80% of the time, but even then won't be able to use most of the network.
Part of it is a cultural thing. Americans are willing to accept that companies are evil and don't allow something. You'll go online and say, "how is this legal" and then go back to your life and the company has spent no money on you. But once something is allowed, there's no concept of "we'll allow this, but you're on your own if anything goes wrong and we'll charge your account $100 if you call us for support." Once you've been accepted as a paying customer, you (and by you, I mean at least 10% of the people who will do that) will demand the same premium service for your dollar.
"AT&T only allows certain phones." Ok, companies are crap, but we're all used to that. If they allow you to put a crappy device on their network, there's so much bad word of mouth, regulator crap, support complaints, etc. I'm not saying the world should be this way, but it certainly is.
So many restrictions that exist are there because the company loses almost nothing from the restriction while the pain in the ass they receive from not having that restriction simply isn't worth it. Keep that in mind when thinking about why companies do things that seem to have no benefit. A bunch of people were likely a complete pain in the ass over something that was the customer's fault, but that the company ultimately had to deal with.
And I think the big area is foreign phones. Phones made for the US that have the right network support for the carrier are good to have. Someone grabs a Galaxy A32 meant for non-US markets (rather than the A32 5G) and it doesn't support bands 12/13/17 which are the main coverage bands for T-Mobile/Verizon/AT&T. It does support bands 2 and 5 and AT&T and Verizon probably have ok coverage in cities with band 2, but in-building coverage will be worse and speeds will be slow. I think they have some band 5 LTE, but they weren't necessarily anticipating that being one of the bands customers would really be relying on. And now there's an angry customer, spreading their anger to their friends, tying up support resources and costing you money, using your network inefficiently, and demanding that the carrier fix the situation and the only way you can fix the situation is basically by giving them a free phone.
In terms of how it's legal, spectrum licenses didn't come with open access requirements. The one exception was Verizon's 700MHz licenses. Except that Verizon kinda redefined open-access to mean that they wouldn't lock their phones - not that you could use any phone you wanted on their network. Google had pushed for the open access requirements, but a decade later Google didn't care and regulators didn't care.
Is it really so hard for them to spend 30 seconds explaining this to users? It's not exactly rocket science, right? They could literally send a text when they notice a new incompatible device like this appearing on a line (and maybe periodically, like once every year or something):
"Hi, this is a free SMS from AT&T. We noticed you're trying to use a Galaxy A32 phone on our network. Unfortunately this phone doesn't support some of our bands (12/13/17), so you might experience poor or no coverage in some situations. We recommend you upgrade your phone, otherwise you might run into issues. If you have any questions, please call our representatives at 1-800-288-2020."
You'd think they could automatically annotate these accounts with a little note for the next customer service rep so they're aware of this possible issue if the customer calls. Then users won't be confused as to why their coverage is poor, but they can still use their devices if they really want to.
Why is it so hard for companies to communicate honestly and openly? Why do they prefer to let others stay in the dark and make problems 10x harder for everyone instead? Isn't it kind of unreasonable to expect customers to be "understanding" of problems when they don't put in an honest effort into helping them understand the problems?
Now you have to have a database of a million phones from Shenzhen that you need to keep track of whether it has your bands. Who's gonna pay for their QA and know what phone manufacturers are doing what with their phones?
Indeed, this has been an issue since the 3G era. I live in New Zealand. 10 years ago one of the largest local carriers switched from CDMA to WCDMA-only using the exotic North American 850MHz band whereas all other carriers use the more common 900MHz. They also both have 2100MHz base stations in busier urban areas.
This caused endless headaches for users and tech support alike because the market was flooded with phones that might have 850MHz support. There is no way to tell from the model or serial number - even Samsung and Nokia often failed to properly indicate that in their specs. And because both networks were also on 2100MHz, a phone may appear to work fine for months, but you will find out the hard way when reception completely disappears in a slightly remote area. There was also no GSM to fall back to since they built it from the ground up with no 2G layer.
Legend has it that one intrepid phone reseller would routinely take his stock on a boat to a spot 15km away from the nearest shore where no 2100MHz signal could reach, and only there could one be sure that the reception you are getting is on the 850MHz band.
Not really. Nobody's swearing under penalty of perjury here, 100.000% accuracy is very much not a requirement. I think a best-effort list would go an insanely long way, and carriers already have large IMEI->model mappings for a lot of phones all over the world. And there are already gazillions of random websites with databases of model->band mappings out there too. And in the absolute worst case where they see a device they have zero clue about they could at least give a slightly different message like "we don't know if this is supported, but please be aware [...]" or whatever, and annotate their profile for the rep accordingly. Again: perfection is not required here.
Because if it were that clear, people would not use them or stick with them, or would get upset and complain to regulators. Market monopoly through obfuscation and inertia.
I literally have no idea if I’ve ever been this customer and have no way of finding out from anyone who’s ever sold me a phone. They didn’t tell me information like this when I bought a random handset from Cingular at their store, or when I bought an iPhone directly from Apple.
Not only are phones cross network compatible, with eSIM support, you can provision new service within minutes on any supported carrier, no physical SIM required.
I just activated a new phone with T-Mobile and they basically insisted on a physical SIM card. When I asked why they said they can do an eSIM but then I wouldn't get 5G.
Is that true? If not, then why are they pushing physical SIM cards so hard?
Furthermore, what they're specifically referring to here is Sprint's 3G network. Sprint's 4G LTE network is currently planned to be de-commissioned on June 30 2022, but that will almost certainly get pushed as well.
Why did they buy Sprint if it was not for their national CDMA network? Just to snag some subscribers? Were they assuming it would all depreciate away and they'd be left with the tower sites anyway?
The big asset is spectrum allocation, right? Anyone can put buildings on the ground, just get some metal and a backhoe, but only one chosen winner gets the right to use a certain part of the RF spectrum.
Are there any updates on their GSM network? There are rumors that it will be also sunsetted, but I hope it will stay on. GSM is still useful because of its broad compatibility.
1st gen IoT devices used 2G, even when 3G was a thing. Things like emergency response in cars (eCall in Europe) and other types of infrastructure level "call home" technologies that were intended to work for decades but now must be upgraded before their 2G networks go dark.
There's a bunch of hard to access industrial sensors that have worked fine for 20 years but now crews have to be sent out to service and replace them because the network is going offline. Yes, that does sound like an expensive pain in the ass, you're right.
I think it was foolish for manufacturers to design systems that are supposed to last decades while relying entirely on the technical implementation of a service provider without signing a contract for the entire lifetime for the device.
I've seen it here with systems that rely on the voltage of phone lines to operate. They simply don't work with most digital telephony without additional hardware.
The consumer and business service industry doesn't stand still for industrial and IoT applications and time and time again do companies building these things assume this time things will be different. Whether it's 2G networking or Windows XP, failure to plan and make the necessary deals for the entire lifetime of the device is what's really causing these devices to fail.
If you can't make sure the necessary technology is available for the entire lifetime of your device (which can easily be 20 years or more for a car!) at least make your interface upgradable and well-documented.
There are more devices that can use 2G but not 3G or LTE than those that can use 3G but not 2G or LTE. If you're going to leave a little slice of spectrum on something, you might as well catch the most users.
I have this nasty feeling my 3G prepaid phone is going to stop working one day, and I'll somehow be expected to have known in advance.
Can I get a 5g non-smart phone? I don't want internet, email, or anything other that text messages and phone calls. I'm already too attached to the internet, I DON'T want it in my pocket at all times.
Nearly all major market have plan or are in the discussion of shutting down 3G. Some are even discussing 2G as well. In 2018 the industry sort of expect everyone to shut down 3G and leave 2G for legacy support. Now it seems there are discussions they might as well shut down both 2G and 3G in the future to cut down operation cost.
Given we expect 6G to land in 2030 I dont think that is too bad. But legacy devices like IoT will need lots of upgrade.
Alcatel Go Flip 4 prob. I haven't used it but it's the only featurephone I can think of with US band support and it's under $100.
EDIT: Also uh, this submission is about CDMA shutdown. T-Mobile's 2G GSM is still going to be around for at least another year. If you have an active SIM in something real old, it'll keep working until then. If you need a replacement SIM you'll likely be forced into an upgrade (as the newer SIM cards in their stores lack the 2G applet)
It seems that a lot of companies have delayed their 3G sunset replacement plans until the last minute and then supply chain shortages pushed them over the deadline.
Not surprising - Boost Mobile's primary market is poor folks generally and especially inner city minorities.
"Getting a new phone" or "getting a fancy newly released phone" is less viable financially for this market so they are going to lag on abandoning older standards like CDMA because they are sticking to cheap phones in their budget.
>Boost Mobile's primary market is poor folks generally and especially inner city minorities.
Any support for that claim? My spouse and I used to be on Virgin Mobile, which closed and transferred us to Boost. We are certainly not in the category you mention, and I never got the feeling that either Virgin Mobile or Boost were just for poor, "especially inner city" people.
> I never got the feeling that either Virgin Mobile or Boost were just for poor, "especially inner city" people.
Virgin wasn't. Boost was. That's why they needed a second prepaid brand.
Tons of evidence to this effect, you just have to go looking for it:
"Boost’s strategy was to go after the young adult market in inner cities. The company tailored its advertising to this group by working with hiphop artists and action sports stars. It also used the slogan “Where you at?” in its marketing messages."
T-Mobile is of largely the same ilk. Call up support and you'll be greeted by a rep with: "Yo. What's up?" Probably why T-Mobile always retained its namesake prepaid brand as well. In the old days pre-paid cell service was overwhelmingly for the poor, who might be late to pay. Respectable folks signed a 2-year contract and paid any price their provider told them they owed.
I ended up on Boost Mobile in the early days by a complete fluke. It was lucky though, as the service was cheap, phones were heavily subsidized, coverage of the iDEN network was excellent everywhere I tried it (Sprint coverage was always horrible everywhere) with just some problems of network congestion.
That's a hard question to give a definitive answer to. In many ways, the answer is "no". However, as with most things in tech, there are always edge cases.
Sprint's network (which T-Mobile bought and is integrating into the T-Mobile network) never went fully VoLTE (voice over LTE) so if you had a phone that didn't support CDMA voice and you were on the Sprint network experience and you traveled to one of the rare areas where Sprint had coverage and T-Mobile didn't and there was no Sprint VoLTE available...
The thing about wireless is that there's always some vocal person who is getting better service from something that is worse for 99.9% of people. Of course, 99.9% means that you have one person out of every thousand and if you're a wireless carrier with over 100M customers, that means 100,000 people might be annoyed by progress.
In America where there's a huge rural population that expects all the amenities of big cities while the nearest neighbor is a quarter mile away, that can create some issues.
Plus, it's not just the 3 national carriers. There are also smaller carriers like US Cellular whose VoLTE support is newer, Shentel (who T-Mobile has more recently bought and whose network will get integrated probably in 2022/2023), etc. Those smaller carriers may not have upgraded all of their network yet. If you're European, the idea that we chopped up our licensing into hundreds of tiny license areas might sound weird (734 cellular market areas for the initial licenses, different sizes/shapes for subsequent licensing). VoLTE interoperability has been a bit of a nightmare and so CDMA has offered a certain amount of fallback for roaming (though that situation has mostly gotten cleared up, partly because T-Mobile didn't really have legacy compatibility to fall back on).
Finally, is it worth supporting? Ultimately, the support comes for free given that Qualcomm's X60 modem was going to support it whether Apple wanted that support or not. I think the iPhone 14 will support CDMA as well simply because Qualcomm's X65 modem will support it. I'm guessing that CDMA support will be dropped with the iPhone 15 when Apple moves to their own, in-house modems.
Verizon is shutting down their CDMA network at the end of 2022. By the time the iPhone 15 launches in fall 2023, CDMA will be quite dead. But legacy things stick around for a long time because even when it's not used by 99.9% of people, there's always some that will vocally complain about change and regulators don't like the idea of customer equipment being rendered useless.
CDMA is the superior protocol afaict, unfortunately they wanted more in licensing fees at the beginning which stunted it. In any case a phone can support both in software.
> By the time the iPhone 15 launches in fall 2023, CDMA will be quite dead.
Perhaps in the US, but CDMA is used elsewhere too. In several countries in the recent past, fixed wireless CDMA operators (intended for home phone lines) started operating mobile handsets. Those places may not be on the upgrade path to LTE anytime soon, because of expense.
This is one of those "but actually" things that always happens (and why I can't give a straight answer and why I should have hedged that statement as well). The question was about the US and Apple and I responded about the US and Apple, but I should have know I'd get a "but actually" if I didn't hedge everything.
I don't know what country/network you're talking about, but it looks like Uzbekistan might be the only country that's valid for.
Beeline Armenia doesn't have a shutdown date, they also have GSM/UMTS/LTE service and their CDMA network is at 450MHz which the iPhone has never supported. MPT has an LTE service. That leaves Perfectum Mobile in Uzbekistan which looks like it's still heavy in CDMA. Uzmobile also seems to run CDMA at 450MHz, but again the iPhone never supported that. Going through Perfectum's website and looking at their suitable models feels like growing Verizon and Sprint's website from years past.
Yea, I should have hedged further and said "in the US and for Apple's purposes." Really though, I can only find Uzbekistan as an exception. And I could see Uzbekistan lingering on CDMA. Are there more?
I'm not even saying that old tech doesn't linger. Part of it is that CDMA wasn't popular in most of the world. GSM support will stick around longer because it was so popular. Most places that had CDMA networks have replaced them. CDMA kinda became a dead-end technology as Qualcomm's UMB evolution didn't happen - once Verizon chose LTE, there wasn't much of a point.
Honestly, it would have been cool if you'd added information about countries and networks where CDMA might stick around. I'd love interesting information - I literally went around looking for the information not included in your comment. Without that information, it just feels like "I know you were talking about Apple and the US, but actually..."
I'm having trouble finding things with searching tonight. But I think I know where to look to get the pointers I need... I'll see if I can dig up some examples tomorrow. Although, maybe I'll find that everyone has moved to LTE.
I was dealing with phone numbers for several years and there had been several networks which were CDMA for home phones only in otherwise GSM nations that started doing mobile phones. I suspect I can find some of the ones I managed to get added to libphonenumber when I was contributing to it. But, I don't know the frequencies, maybe they were at 450 and not iPhone accessible.
Well, I still can't find the references I'm looking for. So I'll accept that CDMA will be dead, and we can both be surprised if it survives a little bit longer.
CDMA2000 network (Au KDDI) is going to close but still operating in Japan, but iPhone 8 and successor for Japan model dropped support. iPhone works for LTE only on Au.
This likely mirrors what happened when we moved from analog to digital. CDMA versions of phones like the Motorola v60 came with pull out antennas and were still capable of latching on to analog signals. It would drain the battery pretty quickly, but it worked.
Most people, most of the time, were riding digital signals and their phones never saw analog, but yet it was still there. A few things came into play:
1) The CDMA chipset in the phone had analog capability, so it was basically "free" to utilize.
2) Analog signal still worked well in rural/mountainous areas where it carried better.
3) It was still in active use with regional carriers they had roaming agreements with.
My guess is that similar conditions are why it is still in play in today's phones.
Lots of people buy their phones in the USA through relatives and then ship them to their own country. That might have prompted Apple to support stuff that they no longer need in the US?
Apple uses a Qualcomm baseband chip. They share the capabilities with other manufacturers, but Apple is working on their own baseband chip which they expect to ship in 2023.
The chipset the qcom modem they are using probably supports it. Would depend on the firmware and sometimes a grounded pin if it is on or not. Much like FM radio in many of these devices is 'there' but not turned on/wired up. For the new chipset that Apple is making I seriously doubt they would support that standard, esp if the big telcos are turning it off. But it would depend on what sort of contract they have with the telco.
