Hacker News new | past | comments | ask | show | jobs | submit login
How Qualcomm shook down the cell phone industry for almost 20 years (arstechnica.com)
528 points by headalgorithm 52 days ago | hide | past | web | favorite | 300 comments

As a former worker at Blackberry(RIM) cellular chip division (hello fellow RCT-ees), this article explains so much it physically hurts me.

RIM pushed hard to develop its own chip 3G/4G chips because Qualcomm absolutely strangled the company with its licensing cost/terms. They pretty much demanded a separate cellular R&D team (QCT) that duplicates half the effort. All CDMA based phones must contain chips from QCT because Qualcomm holds CDMA patents and they won't license them to anyone. Finally, Qualcomm was, while very technically competent, will give you no assistance whatsoever if you deviate from what they want you to do. Basically 'my way or the highway.'

Edit: As an aside, my heart goes out to Intel's defunct Infineon-based 5G modem team.

I know one person who works in that team. Said he was driving to lunch and heard the news that Apple and Qualcomm had reached a deal. So he buys a few Qualcomm shares (he sold them at peak!), gets back to work and then this e-mail shows up saying Intel is getting out of the 5G modem business. I'm paraphrasing here: "The sight of a whole floor of heads popping up from their cubes at the same time... what an experience!"

So everyone there is hoping that Apple buys them out.

My bet is that Apple buys Intel's modem parents and hires most of the staff from Intel that was doing modem chips. Apple already has a team that is likely working on this.


Why would Intel sell that though?

Selling toast isn't easy when you let the customer borrow your bread and butter.

To put it another way, Apple took their ball and went home. No soup for them!

Intel has a track record of selling off assets that end up being valuable in the future.

Basically Intel is trying to give away their market share like IBM did.

FYI - IBM handed Microsoft and Intel a big chunk of market share because who would ever want a personal computer...

> IBM handed Microsoft and Intel a big chunk of market share because who would ever want a personal computer...

I believe it was because they needed a PC in a hurry. I have heard that Ken Olsen ridiculed the IBM PC for its crude design, but it was the beginning of the end for DEC.

> I believe it was because they needed a PC in a hurry.

That's a much better explanation. They made the entire thing in a year, and 90% of it was made by outside vendors. I think both of these things, long-term, contributed to its success.

>Qualcomm's first weapon against competitors: patent licensing terms requiring customers to pay a royalty on every phone sold—not just phones that contained Qualcomm's wireless chips. This gave Qualcomm an inherent advantage in competition with other chipmakers. If another chipmaker tried to undercut Qualcomm's chips on price, Qualcomm could easily afford to cut the price of its own chips, knowing that the customer would still be paying Qualcomm a hefty patent licensing fee on every phone.

And that was just the start. Fantastic article. Qualcomm, man, just evil business practices that worked for a decade and may continue depending on appeals.

And if Ericsson, Huawei, Samsung, or ZTE had offer a modern modem, it would be the same.

I don't think Qualcomm strangled RIM with their 3.4% (peak!) licensing fee on the selling price of the phone. If RIMs profit margin was 3.5% they had bigger problems.

Here is the thing. If Apple sold a 2G network based iPhone with a non Qualcomm IP modem, and a 4G iPhone at 2x the price with a Qualcomm modem, which one would make more money for Apple?

Why should the price of the phone matter to Qualcomm? Did they have anything to do with the CPU, screen, memory or battery?

It doesn't and it also doesn't matter. There are lots of license agreement, in tech or not that takes a percentage of your selling price.

And if it wasn't a percentage, it will be a fixed price for everyone, while it may sound fair, all of a sudden everyone selling a phone would be paying in total ~$50 to Patents license. That might not matter to higher end phones, for the lower spectrum that is a huge increase in Total Cost.

I 100% agree with you but those were among the abusive terms

> Did they have anything to do with the CPU, screen, memory or battery?

Sure, without QCOM's IP, other non-wireless parts of the phone won't add/contribute as much value or drive market demand for the phone.

Ehh, the original cellphone patents had already expired at that point. QCOM’s IP had arguable direct value only representing improvements to existing tech.

Their business model existed largely via the mismatch between cellphone companies and wireless carriers.

Yea, my point exactly. All the recent IP covers improvements since 2G. Apple can try to sell a 2G network connected iPhone based on expired IP without the Qualcomm tax.

Free market pricing is based on demand of the feature not the cost to make it. Qualcomm should be pricing the part at the point where the phone without the qualcomm part is equal value per dollar.

If a monopolist (or dominant oligopolist) charges too much, it would encourage competitors, at least if they don't have a stranglehold through patents. There may be a sweet spot where the "free market" allows them to remain a monopolist.

Patents don’t last forever, and there are a dozen ways to solve each problem.

Apple is sitting on 200b in cash. What would be great for consumers if they spent some of their cash on 6g networks before they turn into a sovereign wealth fund.

Apple is not a carrier and they don’t own spectrum. So, they would need to buy one expand 6g networks.

Anyway, such verticals integration is unlikely to be good for consumers.

Qualcomm doesn’t own spectrum either. Apple should be investing in tech, not in cash. It’s crazy to think of tech companies sitting on that much cash. At this rate, apple will have 1T in the bank in 15 years.

But they're adding a fixed amount of value regardless of the phone, and so they should get a fixed sum per phone. This business where they get a percentage of the retail price is ridiculous.

Consider the 64 GB iPhone XS selling for $1000 vs the 512 GB one for $1349. Is Qualcomm adding any extra value to the second phone as compared to the first one? Obviously not, so why should they capture any of the extra purchase price?

At least for the license with Apple, it was capped at the first $400 of the phone. I.e. it was a flat rate for virtually all iPhones.[0]

For the industry at large it was $500 (now $400) [1]

https://fudzilla.com/news/mobile/47691-apple-qualcomm-licens... [0]

https://www.reuters.com/article/uk-qualcomm-licensing/qualco... [1]

I believe there was no flat rate. I think during the first FTC trial, it was revealed that Apple paid on average $7.50 per device, or ~$230 (Foxconn's wholesale price for iPhones) x 3.25 = $7.50

If they just charged a flat rate then no low end phones would use their tech as it would eat more heavily into their profit.

On the other hand, consider a video call on the iPhone XS with v.s. a $50 phone with a low resolution screen. The value of a high bandwidth internet connection is higher when the screen allows you to perceive the difference.

That doesn't seem like a reasonable way to price things. It's like Apple charging companies a percentage of all revenue made using their Macbooks

Reasonable way to price things when neither the seller nor buyer are under duress is to for them to an agreement. It doesn’t matter if it’s % of revenue or flat, if they both came to an agreement, then it’s reasonable.

Exactly. Or Apple charging a percentage of all in-app purchases.

> Consider the 64 GB iPhone XS selling for $1000 vs the 512 GB one for $1349.

Sure, first, QCOM licensing fee only applies to the first $400, which is in turn also based on the wholesale, not retail, price from Apple's contract manufacturers.

second, without QCOM, Apple wouldn't be able to price the iPhone XS at $1000 (or the iPhone 7 at $550), much less sell nearly as many units. Would you even consider paying $550 for the 128GB iPod Touch, which is essentially at feature parity with the iPhone 7, but without QCOM's IP?

What a ridiculous argument. Without the OLED screen Apple wouldn't be able to charge $1000 for an iPhone either. And I bet those $1000 that not one of apple's customers, not a single one, bought an iPhone instead of an iPod touch because it had a Qualcomm chip in it. Even in later years half of iPhones include an Intel chip not a Qualcomm one; it has made zero difference. 99.9% of customers don't even know Qualcomm is a thing a won't care if you told them.

> 99.9% of customers don't even know Qualcomm is a thing a won't care if you told them.

Sure, QCOM is not a marketing/consumer electronic company they way Apple is. Your 99% of consumers buy the iPhones over the iPod because of QCOM's wireless functionality, not because of their branding.

And how many of those 99% of consumers do you think would pay $1000 for a OLED iPhone XS without QCOM's patented wireless features that allow to make phones call or data? Or how many would pay half the price? or even a third? The iPhone 7 without OLED, the most popular selling iPhone in many countries, still costs at least twice as much as comparable iPod Touch and sells multiples of the unit sales of the iPod touch.

Yes, because the value of the 512gb without the qualcomm modem is lower than the 64GB version.

Free market pricing is based on demand of the feature not the cost to make it. Qualcomm should be pricing the part at the point where the phone without the qualcomm part is equal value per dollar.

I’m pro patent and copyright law. But they are, by strict definition, government sanctioned monopolies. There ain’t nothing free market about this licensing situation.

It’s even worse in this case. For Qualcomm’s patented IP to be a part of the standard they agreed to license it on FRAND. They didn’t.

They did, it is just whether it was fair or not was up to debate.

Yea. For sure. Apple should be encouraged to spend their 200b in cash on an alternative instead of lawyers.

I’m sure they would love to. The billions they’ve paid in licensing fees is more than it would cost to actually make themselves.

This is the problem with mandatory standards and patents that cover those standards.

If it cost apple less to build improvement into 5G that would have been licensed to everyone they would have done so.

Standards require all participants to license at a FRAND rate. If apple developed 5G IP they would have effectively received a rebate based on the relative impact of their contribution.

Qualcomm actually has great execution. You realize 5G is going to put a 1 gigabit network in the palm of your hand? I can’t even get that out of cable.

> Standards require all participants to license at a FRAND rate.

Clearly you didn’t read the article. There’s an entire section about Qualcomm not issuing FRAND licenses.

@aey used to work at Qualcomm according to their resume. So, take that for whatever you think it’s worth.

Yep, worked at qcom. I think they should charge what the market could bear, like apple charges consumers :)

They haven’t lost a FRAND case yet.

In the ideal world the phone would be like a PC, and you could buy an iPod touch from apple and swap in the qualcomm 5g modem or the intel one.

Apple pays qcom $14 bucks for the modem. In the PC world I think qualcomm would be able to price theirs closer to 50 compared to intels.

Meh, 1Gbps over cable is nothing, really. Switches with 10g uplink ports cost almost nothing now. Problem is that most software and hardware on user’s side can’t deal with that speed :) 5G smartphones probably won’t be able to give you 1Ggps either.

I just moved, and I have access to 1gbps fiber at my home for $50 a month vs a severely crippled 100mbps from comcast for $120 a month two blocks away. So land internet is not free, and 1gbps is rare.

Qcom charges Apple $14 bucks for the modem. That doesn't sound crippling or unfair or unreasonable. I think if consumers had the option to buy an ipod touch for 500, then add the modem like you do with a PC no one would be buying a fully integrated device from Apple for 1000.

I worked at Qualcomm, so take this with a grain of salt. But I find it laughable that Apple is complaining when they can sell the exact same thing at 2x the price by just adding a $14 dollar qualcomm modem.

I suspect that the fact that those were the only choices is a result of Qualcomm’s abusive licensing.

They have other problems, that's for sure.

But if you need to give Qualcomm money for the 9000 Bold (in house stack, no qualcomm at all, GSM, and sold around the world), because you need to license it to buy the modem for the 9670 Bold (CDMA, sold in select market, and CDMA use was minimal elsewhere), then out of the 99 problems you've got, Qualcomm is one of them.

RIM sold more 9000 than 9670. In its heyday, the GSM stuff is everywhere, but needed to keep the CDMA for NA and a half-hearted attempt to enter China.

>I don't think Qualcomm strangled RIM with their 3.4% (peak!) licensing fee on the selling price of the phone. If RIMs profit margin was 3.5% they had bigger problems.

That entirely depends on what Qualcomm was charging RIM's competitors.

Im inclined to think that there is a third choice.

Yea. Apple can build its own. There is nothing wrong with them spending their 200b+ in cash on integrated modem tech.

Yup. And I remember how quickly all the press panned subsequent attempts, saying BlackBerry overcharged.

This is one big piece of that puzzle. I'd still kill to have a vertical slider for a phone...

Did they really 'strangle the company'?

Blackberry had a meteroic rise and they collapsed because of iPhone and (later) Android.

After reading about it particularly "Losing the Signal: The Untold Story Behind the Extraordinary Rise and Spectacular Fall of BlackBerry", I think guardiangod is correct, but also Verizon and, to a much lesser extent, the whole stock dating scandal. If Verizon hadn't been so gun-ho for an iPhone killer, Blackberry would probably of got to something like the Passport a whole lot faster. They needed to stay in their lane for a bit longer. Their market cap actually hit its max a year after the iPhone.

I wouldn't call RIM / Blackberry a meteroic rise. They start in the mid 90's and built up the products in a pretty smart way with each following the next until the total shift of the Storm.

The Passport was a phenomenal device, and nothing like it has been built since. Large square screen and big touch sensitive keyboard, but yet still super manageable due to its squat height... if there was a modern Android version I'd go for it in a heartbeat. Not great for watching videos, but it would be super for everything else.

The problem with the Passport was that it didn't have the Google store. 95% of developers couldn't be bothered to submit their applications to Amazon.

Not by default it didn't, but if you had any sort of technical know-how it was fairly easy to sideload it on.

> Not by default it didn't, but if you had any sort of technical know-how it was fairly easy to sideload it on.

And it was a super outdated Android runtime that ran like dogshit compared to the native BB10 apps.

Also, if you wanted any sort of Google Play services stuff, you had to have an app on a laptop/desktop to sign all your apk's the same and remove a specific check... and you'd have to do it all over again every time you wanted to update an apk with play services.

Not ideal.

I wonder if it's still worth using, despite the outdated OS.

If you don't need many apps and are okay with a web browser that hasn't received updates since 2015 or so, then sure. BB10 can run Android apps with its embedded Android runtime, but it's limited to 4.2 Jellybean (from 2012), so a frustrating experience. The BlackBerry App store is still running, but it's pretty much dead, with zero new development. All apps are free now.

It would make for a great email + basic usage phone, but not for media consumption. And the camera's not great.

I used a BlackBerry Classic that also ran BB10 until December 2017, and I still pick it up from time to time and wish I could still use it. The Classic is unique in that it has three independent input methods, all of which work super well:

1. Touch screen: the BB10 OS is actually designed for touch, with fluid and intuitive gestures. Swipe up to go home, swipe up and to the right to go to the notifications hub (reverse the gesture to just peek), swipe left to pull off a screen layer to go back, swipe down from the top to view the settings pane (standard place for settings in every app).

2. Optical trackpad and hard nav buttons, just like on old BlackBerries. You can fully control the phone without touching the screen (great for when you're wearing thick gloves during winter). Works super well for a mouse pointer in the browser. The trackpad even works reasonably well in Android apps, which weren't designed for it (presumably, it piggy-backed on the accessibility focus stuff).

3. Hardware keyboard: typing experience is great—not necessarily faster, but super precise, and no pop-up keyboard getting in the way. Every app is designed with keyboard shortcuts in mind, and the OS makes discovery super easy, so stuff like [C]omposing an email, scrolling to the [T]op of the page, or scrolling to the [N]ext page in a PDF are single click operations. Modern BlackBerries also have great keyboards, but since they run Android, they have none of the shortcuts (except app shortcuts).

I worked at RIM during the rise/collapse and would indicate the issues were mostly structural, cultural, and competitive.

BlackBerry was a small company up against the #1 product of the era (iPhone) and a 'free' competitor. Both companies are basically #1 and #2 brands in the world. They have very deep experience in all things, channels, money etc..

We had no 'mapping team' really, Google had 100 people.

iOS is from MacOS. BB had this ancient thing.

We were just a tiny company that got clobbered, but frankly, BB was culturally more like 'motorola' than an advanced, well run entity. BB could not compete.

RIM started in 1982, not in the 1990's. The 'rise' was quite meteoric when they decided to put email on your hip, and added a phone.

I still don't see how in any way Qualcomm was responsible for or contributed to Blackberry's downfall. No doubt that Blackberry sent tons of licensing money to them, but that didn't seem to hold back their crazy growth rate. Neither did it seem to contribute to their downfall.

I can't get into details, but basically I agree with you. The Storm...was a product of many compromises. But it's only the start of a series of decisions that...

I dearly wish someone would explain the whole idea of why Blackberry built the Playbook. It seems like a product that was so forced with such a confusing development story.

Not sure how much I can share without giving myself away but...

As an active participant in Playbook's entire development from (near) start to end, there were multiple reasons. They weren't good reasons (most of my colleagues had, eh, concerns which got override,) but they somewhat made sense.

What happened was that multiple teams (internal and external) screwed up, so much so that the thing's internal was completely redesigned half way through development ie. Switched to a completely different SoC .

As another poster noted, their power supplies were great. I still have a couple left around my house. They had to be great partly because of a design oversight....

To give BB credit, the playbook was a '1st gen' device for that category, it's really only when compared to Apple, an amazing company with so much to leverage, that the Playbook was bad.

If they could have taken more time, it might have been better.

But the Playbook product did effectively expose the lack of depth, talent and relevant experience at RIM. I'm ashamed to say.

I worked with a VAR when that shipped. They gave us some to try and encourage sales. It was a terribly slow device with poor battery life compared to iPad.

I still use the well made power adapter (Micro USB) that came with it to charge devices to this day. Kinda sad seeing that blackberry logo and knowing the tablet was horrible.

I've heard that the Playbook was actually in development before Apple's iPad was announced, and after that it got rushed out. I don't remember where I read this (it was a couple of years ago), so I don't know how accurate that is.

My wife got one of the last Blackberries for work a few years back. She had never had one so wanted to try.

Build quality was very poor. Overall experience was very poor.

She ran back to iPhone as soon as she could.

RIM died because their products were poor compared to the competition, it's as simple as that.

That's a good question. Yes, margins are low for most phones, most of the time. And cheaper modem chipsets would help with that. But if everyone had cheaper chipsets... then the phones would be a little cheaper. But I don't think that would have changed the competitive landscape at that time.

There was a fundamental change in phone capability and interaction that really shifted the market, and made new winners and losers.

I was a die hard Blackberry (my business phone, using a private iPhone) user, the last one perhaps in the large corporation I've worked for.

Then the Passport launched, LOVED the format, best format ever squared and large screen, beautifully fitting everywhere though being larger, loved the TOUCH keyboard, loved the unified messaging but had NO F* network connection ever. Just could't be called or call others. A second Passport the same. Then one got dropped and it totally broke internally, the "steel frame" didn't help in any way, I assume it contributed due to its weight and momentum.

That's like saying all of the US's problem is Trump.

Reality is often a bit more nuanced than that.

Patent licensing fees were a big part of the equation, which led to inability to lower costs to compete.

>Patent licensing fees were a big part of the equation, which led to inability to lower costs to compete.

Yes. OP asserted that too. Like OP, you haven't actually given a reason why. Here's some counter-arguments:

1) BB grew exponentially all the while paying Qualcomm licensing fees - why didn't it affect them then?

2) EVERYONE was paying the same licensing fees to Qualcomm. You couldn't build a smartphone connected to the cellular network and not pay those fees. So are you saying that it affected BB more than it affected Apple, Palm, MS, Samsung, etc. ?

BB phones, <= v7, are split between an internal gear without Qualcomm, and the CDMA capable stuff is with Qualcomm. So they would have to pay Qualcomm even for the phone with internal gear.

As to #1, around BBOS v5 and v6, when they were pretty much top dog, all the cellphone accounts that wants to use a BB pays a service-access fee, or a BlackBerry Tax that goes to the company. Enterprise companies that want their emails and everything else also pays for the BlackBerry Enterprise Server. They were practically the only game in town.

They didn't have problem then, because they were rolling in cash; just like how they didn't have a problem basically buying up Philips st expanding, or keep coming up with phone models even though there's no way that there's enough demand to actually absorb the production.

#2 on top of needing to deal with Qualcomm licensing fees, which often applied (as the article stated) on non-qualcomm phones, they basically freaked out after losing that lawsuit to the patent troll, so they end up overpaying for licenses (not sure who's the licenser in this case) to avoid getting hit with more patent trolls. The contracts for those licenses were quite hard to get out of. By the time they were making the BlackBerry Priv, they don't have the SAF to juice their income. Patent licensing, Qualcomm and others, were disproportionally high, compared to other Android providers. They can't renegotiate some of them due to the contract wording. Being 1% of the market means Qualcomm hardly cared to give support, because Qualcomm is far more likely to help Samsung and other android phone, that pushed out far more numbers than BlackBerry did at the end.

It ends up being that they end up licensing their brand, because had they manufactured their own phone, they'd have to deal with the patent licensing contracts that they can't get out from.

There's probably a reason why OP won't be too forthright in describing why. I'll leave that as an exercise to the reader.

To add, the SAF is not just a pure BlackBerry tax. Without getting too deep into it, there's stuff that needs to be set up with the carrier, and that's why BlackBerry push was a true push back in the day, without the phone burning through both battery and data. And when this gear malfunctions, you'd know, because you're not getting push at all.

They sipped battery and data, and once that gear is setup, the SAF was very profitable.

Even to this day, email is not the same level of instant.

I'd say Intel should just spin out it's 5G modem team and let a neutral (!Apple) vendor develop it.

Samsung should start selling 5G modems in NA as well. I know this stuff isn't going to work in the short term, but this is the sort of choices we need.

Semiaccurate has a rather interesting take here. TL;DR is that modems are very hard to make. Intel's modems were much slower and used much more power for last generation. They ran into issues and don't seem to even have a shippable 5g chip.



Semiaccurate also said Intel's 10nm is dead. And Nokia was going bankrupt.

So please, please, stop bringing up SemiAccurate on HN.

Intel's 10nm was a failure. Cannonlake was a mostly broken chip. It's taken them a year and a half from shipping Cannonlake to starting to ship Ice Lake. The 10nm process Ice Lake is fabbed with is not the same process Cannonlake was fabbed with; Intel has disclosed a few of the changes that were made in order to produce a viable chip, but we won't know the full extent of the changes until TechInsights gets their hands on an Ice Lake chip.

Icelake was never meant for the 10nm used in Cannonlake, it was a 10nm+ part. 2nd the shrink is there, at least according to what just announced ( or picture shown ) in Computex by counting how many die on Wafers. 3rd They are shipping it, we don't know yield but we know they are shipping it at least in decent quantities.

And 4th, 10nm is not dead, Intel is not skipping it for 7nm. Nokia is not dead either. And it doesn't matter how anyone spin it, SemiAccruate is like tabloid of Tech Report, and to ensure the quality of news I personally believe it should serve no role on HN.

> Qualcomm holds CDMA patents and they won't license them to anyone

I thought chip-makers weren't required to license QCOM's patents -- QCOM like all other wireless SEP holders only collect licenses at system-level.

> They pretty much demanded a separate cellular R&D team (QCT) that duplicates half the effort.

They could have formed a joint venture with other manufacturers to split the development costs.

Qualcomm charges the same 4-5% patents price to license to EVERY manufacturer. You have no idea what you are talking about. They did a 3%(domestic sales) - 7%(exports) deal in China because it was a poor country in the early 2000's and they demanded it but Chinese makers later switched back to the 4-5% deal.

After reading through the comments quarreling about whether Qualcomm and Apple are acting within their rights to charge monopoly-sized rent where they can, and do, it's hard for me to fathom the depths of confusion Americans are reduced to whenever these topics come up.

First and foremost: when you are as big as Qualcomm and Apple are in their respective spheres, you don't need rights. You have POWER, which is far, far better, especially when you can use it to extract money to buy more power with.

Rights are for little people. Qualcomm has been at this for decades. How long can you get away with just parking your car in the wrong place?

Second, when you talk about abuse of monopoly power, it makes no sense to look at it from the standpoint of the monopolist. Everything is stacked in their favor. The important question is, what can the little guy do? Not other companies, even though they complain most effectively (like Apple or Intel complaining about Qualcomm), or even small ones that can't. They would all love to be monopolies too, and sometimes are. What matters is, what is the experience of somebody who got an iPhone for her quinceaña, or of the farmer stuck with a busted John Deere tractor?

Each is obliged to pay a huge premium -- for software, or repairs -- because the 400-pound gorilla can force the matter. The fact that somebody else has an Android phone or an IH tractor doesn't help either of them. Each is stuck with what Apple or Deere is willing to let them have.

The law doesn't need to define a monopoly: only a monopoly can exercise monopoly power, so it suffices to look for that happening. Any monopoly not visibly abusing its monopoly power has nothing to worry about, because nobody can tell.

Anti-trust law doesn't make it illegal to have a monopoly. It just says that if you choose to make one -- and every monopoly chose it -- different rules apply. The law defines lots of rules, but the DOJ has chosen to ignore almost all because of a lately fashionable notion promoted by Chicago-school economists that nothing matters except high prices. Under the law there is a hell of a lot of monopoly abuse nobody at DOJ will even look at, that have all kinds of deleterious results. Anybody who wants to do anything about any of it gets mobbed by defenders of monopolists' rights.

And that's where we are today. Qualcomm got in trouble by making things unpleasant for other 400-pound gorillas, and we get to watch. Offenses against us are entirely ignored, and now people insist that's normal and right, law or no law.

Offenses against us are entirely ignored...

Meanwhile everybody else is saying that US antitrust law is concerned entirely with consumer harm and not with competition. Although in this case the consumer harm seems to be that a >$750 phone costs ~$20 more than it might otherwise.

No, nobody thinks the enforcement is about consumers. Obviously it is on behalf of Apple and, maybe, Intel. But case law says they have to cite costs to consumers, so that is how the complaint reads.

A zillion other companies do the same things, and much worse for "consumers", but are less openly greedy toward other big companies, so are ignored.

That said, Qualcomm's bad behavior has harmed the rest of us far, far more than just costing us a little extra for each phone. That Intel and Samsung have not been able to enter the market means that Qualcomm gets to dictate not just prices, but what you can buy. As the only implementer of many standards, they get to say what will be in standards, so the standards are ruinously complex to keep anybody smaller from being able to implement them. They get to dictate what RF engineers are paid. And so on. Monopoly power is poison wherever it operates.

The strangle hold is extremely bad for security, too. All of those Qualcomm chips are designed as black boxes. Nobody knows what kind of horrible flaws are inside them because trying to get at the code is like trying to rob the federal reserve.

We need open source 3/4/5G chips, but the R & D cost would be astronomical. If you ever think that TCP/IP is complicated check out the mobile system specs some time! They span 1000s of pages... It's amazing to me how much complexity is there. Those phone engineers deserve way more money, IMO.

Where can one take a look at those specs? Are there RFC like things available?

Here's the landing point I use: https://www.3gpp.org/specifications/specification-numbering

In particular, serie 36 is for the LTE RAN (radio access network) while series 38 is for the NR (~5G) RAN. But other series are relevant, for the core and services parts among (many) other things.

All the specs are publicly accessible, and also the change requests (CR) submitted for discussion at the standardization meetings, as well as the minutes. From a spec page you can get to its history and the related CRs. Now of course, there are side discussions but the process is still very open in practice. But it's really huge, don't really expect to make sense of it with a little casual looking around. It takes some real time investment and commitment.

Another point to keep in mind: the cellular industry has one of the most demanding certification process for consumer grade devices. And operators often add on top of this. You don't want buggy devices messing up a shared medium, particularly when it's a scarce resource in many places (high density urban area at peak time). If designing a cellular modem is hard, making it go successfully through certification and also work well in the field and pass the tiers 1 operators field trials (drive tests in particularly harsh locations) and interoperability tests is even harder.

A lot of people miss this part. It takes millions in test equipment just to get ready to enter certification, and that's not the main cost. Even if it's painful, it's also important to make sure things work reasonably well in such complex, multi-parties systems. Which is why certification is legally required to operate in many places (e.g.: Europe requires GCF certification). There are open implementation of LTE, but none certified and legally usable in the field because of this.

Thanks for the great info.

There you go! https://www.3gpp.org/

At the beginning of my corporate career I was working on implementing LTE layer1 (PHY) in internal testing tool. It sometimes took us weeks to understand a single page from the specification - those were fun times ;)

This article is a good read, adding a lot of context to my understanding of the Qualcomm/Apple spat and it's merits. You can see in the history there were several potential market opportunities that companies were unable to act on due to Qualcomm's behavior.

Part of this comes from the willingness of EE standards bodies to standardize on stuff that is covered by patents.

I was going to say "part of the blame" but it was Qualcomm's choice to take advantage of this. Rambus is another bad guy in this regard.

This +1. Imagine if ECMA script or the ISO C++ standard were covered by patents.

Well there is that one company that keeps claiming that they own C++ and keeps trying to sue people for using it.

Don’t you mean java?

Oh? Please expand.

Sorry for the late response. A company named SCO claims they own the language. They obviously don't (possibly just a specific compiler), but they still try (or tried for a while) to charge licensing fees. I'm having a hard time finding references to this now, but here are two links that talk a bit about it:



You cant compare the R&D spent in Wireless and Video Codec to an EMCA Script or C++ Standard.

Wasn't C# standardized by both ECMA and ISO while having Microsoft patents?

Is Qualcomm a ruthless monopolist? Yep. But what people forget is that EVERYBODY was knifing one another in that space when CDMA originally came out. The only difference is that Qualcomm is the ruthless monopolist that won.

In addition, while I have no love lost for Qualcomm, what you do have to remember is that CDMA was a "bet the farm" moment. If CDMA could be delivered, it was significantly better than what existed at the time.

However, there was no promise that a power amplifier could be delivered with the specifications demanded by CDMA--Qualcomm bet that they could. Rockwell had a worldwide monopoly on CDMA power amplifiers for years because they were delivering that amp, and the yield was abysmal.

The original Qualcomm stuff was a significant engineering achievement, and, while they probably don't deserve to continue to abuse their position at this point, they certainly deserved some time to profit from their work.

What other corporation requires paying to license their patent portfolio in order to be allowed buy their products? I don't have to pay for a license from McDonald's to eat their burger. I'm not an expert in this domain, so there might be more relevant example.

Update: fixed up my wording.

Um, wasn't that the case for MP3 until the patents ran out?

You basically couldn't buy MP3 encoding hardware because nobody could negotiate a rate that didn't suck.

You do have to agree to the McDonald's app terms and conditions so you can get its benefits, like skipping the line. You pay twice: first, with money; then, with data.

You're paying to skip the line using data, but you're not paying for the meal with data. If you could only purchase the meal using the app, then you could make the argument that some part of the meal price could be paid from the data itself. I don't know if there are vending machines that will only sell if you have the relevant app installed first, but I think those come closer.

However, unless the terms of service of any of those approaches penalized me for buying a meal at Burger King, then we aren't quite at the same level of scrutiny as opposed to what was reported to have happened in this article.

> and the yield was abysmal.

Nit: abysmal means very great/limitless... I suspect you meant dismal here.



    1.    extremely bad; appalling.
    "the quality of her work is abysmal"
    synonyms: very bad, dreadful, awful, terrible, frightful, atrocious, disgraceful, deplorable, shameful, woeful, hopeless, lamentable, laughable, substandard, poor, inadequate, inferior, unsatisfactory

You might want to go check a dictionary.

"abysmal" can mean very great or limitless DOWNWARD (a la "abyssal")--but that's a secondary definition.

However, the primary definition appears to be: "immeasurably low or wretched : extremely poor or bad"--which is the sense in which I used the word.

I did, but apparently my local WordNet definitions in 'dict' failed me:

      adj 1: very great; limitless; "abysmal misery"; "abysmal
      2: resembling an abyss in depth; so deep as to be unmeasurable;
         "the abyssal depths of the ocean" [syn: {abysmal}, {abyssal},

I recently worked with a technical VP at a chip foundry in South Korea. One day Quallcom came up in discussion and he said "Qualcomm are like gangsters in the industry".

I know this is completely tangential, but just wondering if Apple ever came up in discussion.

Apple is the gangster in the hood.

What Apple did to/with that glass foundry that went under was straight-up gangster.

And because of that the little glass (actually sapphire) company´s execs are in court for fraud and the little glass company Apple raised from obscurity (Gorilla) is doing pretty well - Right. Gangster

I might be missing your sarcasm (as they are not 'little'), but are you suggesting that Apple raised the 168 year old, S&P 500 Corning Glass from obscurity?

I find great irony in Qualcomm exercising a monopoly position to take a % of total phone price from Apple... since Apple uses its monopoly position to take a % of all app revenue.

What Apple thinks would be fair for Qualcomm chip license fees would also be reasonable for App Store fees.

Apple doesn’t get a percent cut of Android app sales.

I don't think this "blunte" guy really quite understands what people in the cell phone industry mean when they say that Qualcomm had a stranglehold on everything.

They mean everything. Every thin dime spent anywhere you care to spend it. There really exists nothing else quite like Qualcomm around out there. We tend to throw the monopoly word around so loosely that we start to equate things like Apple and Qualcomm. It really does make it difficult to try to generate support to dig out from under abusive companies like Qualcomm when they have an army of people out there advancing the narrative that they are no different than any other company.

Meanwhile real engineers are losing their livelihoods by the thousands when companies succumb to the Qualcomm chokehold. Then a lot of those engineers have to go work for Qualcomm. Etc etc, on and on. It's sad.

And not just Qualcomm, Ericsson, ZTE, Nokia get a percentage cut of iPhone as well.

Maybe not, but they get a cut of each Android phone.

> since Apple uses its monopoly position to take a % of all app revenue

Wait. What are you talking about? They own a platform, others want to use that platform to make money, so Apple is charging fees to use their platform. Where is the monopoly part? Developers have other avenues to sell their software on (Android)

iOS and Android can be classified as a Duopoly though.

Duopoly is when the two heterogenous companies coöperate to form a monopoly. That was Wintel.

A cartel is when a small number of homogenous companies share the market while ostensibly competing with each other, but coöperate to erect barriers to anyone else getting into the market.

If Apple and Google block other companies from getting into the mobile device business, they would be a cartel.

Did you start using that spelling from reading the New Yorker?

Which spelling?

Probably the diæresis:


The "æ" is a whole nother thing.

I was taught about the diaeresis way back in the 1970s in high school English. But I didn’t embrace it until the late 80s when I started writing with a Macintosh, and read “The Mac is Not a Typewriter,” a great little book.

I've always liked typography and graphic design, and I enjoy anachronisms like ligatures, alternate letterforms, drop capitals, visual justification, latin phrases, and so on ad nauseum.

ad nause-am.

I am particularly fond of the looped st and sc ligatures.

Apple only takes payment for apps sold through their own platform. Only if, on the other hand, they also demanded a cut of revenue from sales on competing platforms, you could talk about a situation similar to Qualcomm.

Apple never agreed to FRAND and allow its IP to be part of a standard. The App Store is not required to sell a phone.

This article makes it sound like Apple and Intel and poor victims of Qualcomm's bullying. That is hardly the case. Both are much bigger than Qualcomm, and have been bullies in their own way. Funny to see them cry wolf

Try working at say LG, HTC, BlackBerry, anyone that wasn't top dog in their ecosystem.

You pay Qualcomm a lot of money just to get access to the modems, but they'll ignore you for bug fixes, firmware, etc...

The word Qualcomm drop (update in the Qualcomm code) sent a shiver down my spine. They'd fix one bug and introduce two others, that is if the bug big enough that they couldn't simply ignore you in the first place.

This is more or less totally false more than any other hardware vendor Qualcomm delivers working phones to its chipset customers so that they can copy and modify the designs and so that they succeed it's the failure of Intel to provide a well-integrated working product to customers that made them flop. Qualcomm provides demo phones and schematics for every generation of chipset.

Big, rich victims.

The monopolistic position of qualcomm has ~~not only~~ perhaps driven up prices, but of even greater concern, has created a single point of failure for phone security.

I have concerns that, even as full device encryption has evolved and many of us have moved communication to more secure media (Signal, Riot, etc), the very baseband chip of the phone is still acting like a bug in our homes.

> The monopolistic position of qualcomm has not only driven up prices,

What is your evidence of such claim?

I think that a monopolistic shake down of this nature is always connected to price increases, but you're right that I don't have specific evidence.

And in any case, my comment is not about prices, but about security.

>The monopolistic position of qualcomm has not only driven up prices

There has been no evidence as such. Even in court.

Fine, I added a "perhaps" - my real point was about spying, not prices.

And it is not a single point of failure either. Huawei, Samsung, Mediatek, all have their own modem, as well as Spreatrum ( It has a new name I can't even remember ) , and Xiaomi and Soon, Apple.

I know a lot of people hate Qualcomm, but reading the 230 pages of the report I cant help but felt sorry for them.

> I know a lot of people hate Qualcomm, but reading the 230 pages of the report I cant help but felt sorry for them.

They had a special opportunity to take a lead role in protecting the American public from being victimized by criminal behavior within the NSA and they chose to side with the criminals instead of the victims.

What's to feel sorry for?

>They had a special opportunity to take a lead role in protecting the American public from being victimized by criminal behavior within the NSA and they chose to side with the criminals instead of the victims.

And where is that from?

By safeguarding against Smurf Suite. And by blowing the whistle every time gangsters show up and want to use a baseband chip to abuse people.

They had an opportunity in June 2013 to say, "we're aware of the former intelligence chief James Clapper's acknowledgement that his March 2013 testimony to the United States Senate Select Committee on Intelligence was erroneous ; we are taking every action possible to thwart efforts by the NSA to collect information about our users as described in his recent letter and as disclosed last month by Edward Snowden."

At a time when many people (including some members of Congress) were hoping that Clapper was going to be prosecuted for perjury and the entire regime of illegal spying was going to be provable undone, Qualcomm, a potentially huge contributor to safety of the public, did and said nearly nothing.

for a counter view on this from yesterdays WSJ https://www.wsj.com/articles/a-courts-dangerous-antitrust-ov... this is by the FTC commissioner (the agency that sued Q)

Very interesting I had not seen this.

Note there are five ftc commissioners. Christine Wilson is one of three republicans. She is unfortunately the one who seems the most trumpian. She misleadingly quotes a filing her own agency made in this op-ed. Because of that, and a few other turns of phrase in here that appear frequently in Qualcomm PR (such as the factually incorrect statement that this case was filed during the Obama administration--it was actually filed under Trump) I'm willing to bet a lot of money Qualcomm wrote this and gave it to her and she signed off on it.

it was filed when the last administration was on the way out but still had dem majority. also see the dissent written by one of the commissioner at that time. https://www.ftc.gov/system/files/documents/cases/170117qualc... i feel the ars article is a hit job as is the PR from Q.

I just looked at the FTC government archives from 1919-2009 and there is a gap after 2008 with missing volumes 146-148.


Does anyone know why this is the case?

I just double-checked and you are right! Filed Jan. 17, 2019. Apologies. For some reason I thought it was filed a few days after Trump became president, but I was mis-remembering.

And congress will do nothing about IP, specifically patents as it relates to this story.

I wish... patent reform on extension patents in particular in addition to a dual sourcing requirement from the FDA would let a more natural market take care of a lot of what's wrong with medications.

In general, at this point, I'd love to see the entire patent system dismantled rather than keep what we have.

It seems it is pretty much a one sided opinion against Qualcomm, without taking a look into actual details. I read the 233 page report, and I have to admit I am surprised at how some of the wordings and conclusion the judge came to be. ( And I was very surprised how Ars's Author read the report and did not question anything about it )

And in defence of Qualcomm, I would post a few links from people that had the same thought as I did. [1] Is the Qualcomm information, I think it provides a angle just how much Qualcomm has spent on R&D. While I don't normally consider Cnet as Quality source of news, Shara Tibken [2] has been following the Apple and Qualcomm Trial and has been in the court for every trial between the two, reporting on the "exact" wordings some of these people gave in court. There were lots of details, and Qualcomm clearly has some very decent explanation in every accusation against them. And somehow none of these made it into many other mainstream media.

Patrick Moorhead wrote about this [4][5] and I agree with him on every thing ( Apart from the US politics which I am not well versed in ). Things Such as the "common interest agreement” Apple has with the FTC, and Huawei's involvement. Along with completely ignoring the defendant's testimony. It really opens my eyes as to how US law system works compared to UK or Germany.

And I would just like to remind everyone, Hating Qualcomm, Qualcomm doing wrong, and Qualcomm is a monopoly are 3 different things.

[1] https://www.qualcomm.com/ftc

[2] https://twitter.com/sharatibken

[3] https://www.forbes.com/sites/patrickmoorhead/2019/05/28/ftc-...


Sounds like we should just abolish patents.

So why did Apple decide to make its own SOCs but not its own modem chips?

I would expect making CPUs and GPUs to be a lot more difficult than modems, no?

No, handoff design is more difficult than picking and tuning ALU designs from the 70 years of CPU design history, then generating the control circuitry.

A lack of design-for-handoff OF ANY SORT is why Intel failed with 802.16. Qualcomm invented soft handoff which made frequency reuse 3x more efficient, and it made CDMA possible. People are pretty ignorant that due to shadow fading you have to keep 3 or more MACIDs active on different base stations and use high speed feedback control to hand off roughly every 1.6 seconds. I statistically analyzed drive testing logs for 1x-EV-DO at Qualcomm to produce one of the world's first handoff markov models, as part of OFDMA handoff design for Qualcomm.

A lot of practical experimentation goes into cellular system design in the most challenging handOff locales on Earth - San Diego and Hong Kong.

It seems that there is a lot of focus on Qualcomm's strong-arm monopoly tactics. But that seems to me to work exactly as designed - patents are intended to literally grant a temporary monopoly to reward the holder for the expensive research work that otherwise wouldn't have happened. If Qualcomm won't license someone a patent for whatever reason, that is up to them. Clearly if customers are deciding to pay the price, it must be worth it. In fact, Apple itself did after realizing Intel can't deliver the 5G chipsets on time.

The article points out where QCOM actually violated a contract, though. It's related to FRAND patents [1] - they committed to licensing a certain set of patents this way so they can be included in a standard, something very valuable to Qualcomm, and then later reneged on that commitment. That I can totally side with. But all the other bellyaching about QCOM the big bad monopolist for other reasons is missing the point IMO.

[1] https://en.wikipedia.org/wiki/Reasonable_and_non-discriminat...

I agree. If they alone in the world can make good modems well they can charge a lot for it. 3% seems ok with the value I get from the the modem compared to the whole phone. If not the market will price them out. The issue is not with clients but other suppliers (or lack theeeof). The issue is patents. And enforcing frand policies. Again if no other team in the world can deliver what they do why should they charge less ? If they violated agreements it's not even an systemic issue with patents. Am I missing something ?

The issue with patents is, no other team in the world is even allowed to try.

You are theoretically under frand terms, although not free of charge: https://en.m.wikipedia.org/wiki/Reasonable_and_non-discrimin...

CPU design and RF stuff are both very difficult; I wouldn't assume one way or the other. I wonder if knowledge of CPU design is more widely dispersed (e.g. you can hire ex-Intel, ex-AMD, ex-IBM people) than cellular radios where the only (working) ones seem to be made by Qualcomm.

One difference is there is a hella lot more engineers that know how to design and tape out digital designs than mixed signal RF ones. Qualcomm and a few other companies employ most of them.

> cellular radios where the only (working) ones seem to be made by Qualcomm.

Seems strange, given that plenty of manufacturers are making the radios on the other side of the cellular connection. Can't, say, Ericsson take the radios in their cell towers or picocells, redo their tape-outs for phone scale, and end up with a workable mobile radio?

> Can't, say, Ericsson take the radios in their cell towers or picocells, redo their tape-outs for phone scale, and end up with a workable mobile radio?

Generally speaking, no.

Sometimes in the base stations, you'll find FPGA and such, which are super expensive.

The even bigger issue is that you're implementing the other side of the over-the-air protocol, and a whole bunch more stuff too (signaling in the carrier's network). Whereas the modem in a phone is designed to talk to just a single system.

Also, the protocols are designed to be asymmetric. The cell station is fixed, but has relatively unlimited size and power constraints compared to the phone. They also have to talk to multiple phones simultaneously, and manage their access to the network.

> Ericsson take the radios in their cell towers or picocells, redo their tape-outs for phone scale, and end up with a workable mobile radio?

Ericsson used to make cellular chips for mobile devices. They decided to leave that industry rather abruptly, leaving us high and dry on one product I worked on.

What does "redo their tape-outs for phone scale" mean exactly? Can Intel redo their tape-outs to make a Xeon work in a phone?

I suspect those high-power radios are not doing the majority of the RF in silicon.

> Can't, say, Ericsson take the radios in their cell towers or picocells, redo their tape-outs for phone scale, and end up with a workable mobile radio?

Workable maybe but not power efficient, which is crucial in mobile. That takes years to figure out and is the reason why ARM is so utterly dominant on mobile - there simply is not any Intel x86 offering even close to the power envelope of ARM, even after years of ARM dominance Intel hasn't managed to get anything meaningful.

> there simply is not any Intel x86 offering even close to the power envelope of ARM

They are very close. Intel has many x86 chips consuming 3-4W: https://en.wikipedia.org/wiki/List_of_Intel_Core_M_microproc... Some older Atoms were under 1W.

> the reason why ARM is so utterly dominant on mobile

The economy. When you sell complex chips for $10, you have to sell a lot of them to return R&D costs. That's why they shut down cheap Atoms introducing Core M series: technically the two are close in many respects, but Atoms were sold for $10-20, core M's for $200-300.

Intel could probably do it if they had the luxury of redesigning their instruction set. They're stuck with supporting an overly CISC-based instruction set whose roots date back to the early 1970s. ARM didn't have that problem and designed a much more modern RISC-like instruction set which requires a lot less power.

This argument has held progressively less weight since 1995, when Intel released the Pentium Pro and started the precedent of deciding x86 CISC instructions into the micro-ops which are actually executed. ARM is a respectable architecture and Apple has shipped some very competitive chips but it’s not like Intel’s engineers have been in a coma for forty years.

All the moving pieces in Intel x86-to-RISC decoding (instruction decoder, μop cache, Microcode Sequencer ROM...) use up a non-trivial amount of silicon and power.

Yes, it’s more than zero but you have to look at that as part of the entire chip budget. For example, this USENIX paper estimates 3-10% on power:


That’s not nothing but usually when people talk about this the rhetoric assume it’s much greater and that e.g. ARM doesn’t have similar issues supporting its older instructions, albeit at smaller scale. If you look at the results, and a couple decades where everyone else was struggling to match X86 on either performance or non-embedded power efficiency, it clearly wasn’t holding them back that much. Even Intel’s huge moon-shot clean architecture failed to outperform despite starting with considerable experience and no legacy baggage.

I'm not sure. If that were the case why wouldn't intel expose a native, better fit to uops instruction set in addition to the legacy, difficult to decode one and let the apps use the newer ones (or the subset of existing ones which do map well)?

My guess is (1) inertia, and (2) not wanting to commit to a specific instruction set because they tweak the internal μop set with every release.

And (3), this would add complexity, because, even though one is much simpler, now you need two completely separate decoding pipelines and mechanisms for switching between them.

Can't argue with 1. For 2 it's still an instruction set separate from uops so you might not be as sensitive to changes you still get a level of indirection. For 3 .. it depends. You might gain in power if you use the newer one more and you might as well make the older one simpler to achieve 90% of speed maybe. But maybe the decoding of instructions that counts is not that expensive compared to OOO branch predictors and 512 bits ALUs

Micro-ops aren't RISC. Instructions like VFMADD132PS perform a dozen of math operations, combined with RAM access, yet decode into a single micro-op.

Note that I wasn’t arguing that Intel had switched from one textbook architecture to another: only that it isn’t really an accurate way to discuss modern chips after decades of large teams of smart people have been borrowing each other’s ideas.

As to your specific example, ARM has instructions which do complex operations as well. Does that mean it’s not a RISC CPU, or just that some engineers made a pragmatic decision to support things which are done heavily like AES or SHA?

> it isn’t really an accurate way to discuss modern chips

I think it’s still mostly accurate, CISC/RISC is about public API i.e. instruction set. What’s inside a core is implementation details, very interesting ones and can be important for performance, but still.

> ARM has instructions which do complex operations as well

True, but I don’t think they combine these complex math operations with RAM access, in a single instruction?

Intel has made a number of clean sheet RISC architectures. They abandoned it all for marketing 5GHz pipedreams.

Apple phones do not use ARM-designed processors. They just implement the same instruction set. If it’s possible for Apple to develop a mobile processor independently of ARM, why shouldn’t it be possible for Intel?

> If it’s possible for Apple to develop a mobile processor independently of ARM, why shouldn’t it be possible for Intel?

Apple holds an ARM architecture license (https://en.wikipedia.org/wiki/ARM_architecture#Architectural...) and to this day the Apple CPUs are ARM (https://en.wikipedia.org/wiki/Apple_A12X).

Read my comment again. You are restating exactly what I said.

Apple designs processors that implement the arm64 instruction set. They do NOT use processors designed by ARM (the company).

It is. They made ARM chips for years and still hold a license.

> Can't, say, Ericsson take the radios in their cell towers or picocells, redo their tape-outs for phone scale, and end up with a workable mobile radio?

They can, and they did. Not exactly of course, the two products are fundamentally different, but the organization owns enough IP and good engineers to make a viable product. From a technology standpoint it seems reasonable.

The problem is that the competition is tough and you need to stay on your toes with much more rapid iteration than on the "other leg" where you sell expensive stuff to telcos. And the margins on the consumer side are smaller. So, historically, it has proven very difficult to host the same two "legs" in the same organization. The one with bigger more reliable business tends to push the other out. It's been the same with much of the computer industry too.

Heh, there are various levels of 'working'. Think Windows 9x kernel vs NT kernel. Both kernels 'work'.

You can license the primary CPU cores from ARM. There is no equivalent in the modem space (that I'm aware of).

Samsung, Huawei and Intel all make working 4G modems. They work fine outside the US, which suggests that there is some US-specific problem at work.

And you can also buy the company that designed PowerPC chips better than IBM and Motorola could back in the day and have them kickstart designing custom ARM chips.


They're opening an office in San Diego to do just that.


They bought a company that already was a significant player building ARM CPUs, that’s how. There aren’t many (any?) significant baseband companies that are as easily acquirable for Apple to buy.

Apple bought SiByte who were a lot of ex-DEC VLSI design engineers.

So, "Apple" designing a CPU is technically correct, but its a design team from elsewhere.

Apple has acquired a lot of chip design talent, but its an Apple design not one bought from elsewhere.

I think you mean P.A. Semi.

Uh, yeah. You are correct.

Sorry. I forgot which of the ex-DEC design groups got bought when. It's been a while.

AFAIK Apple designs and manufactures its SOCs but keeps using ARM designed CPUs. Not sure about the GPUs though.

Edit: Protomyth below is correct. Apple licenses the architecture from ARM but designs its custom chips.


The chips themselves are manufactured by TSMC.

I am pretty sure the A-series is not an ARM supplied core these days.

Yep, they were using ARM cores up until the A5 (iPhone 4S, ARM Cortex-A9), then a custom core for the A6, and then... have people forgotten how the A7 came out of nowhere with its custom 64-bit core and completely blew away the rest of the industry with its performance?

Source: https://en.wikipedia.org/wiki/Apple-designed_processors

You are correct. I've edited my comment.

From everything I hear from him, Judge Koh is consistently one of the better judges I have seen in some time.

I don't think so at least if you look at the background of this journalist he writes 100% hit pieces on the tech world. 100% hit pieces.

I hate this entire situation and the article. The feds literally give Qualcomm an IP monopoly and then act all surprised when they exercise those powers. Then the fed expands its own powers to 'deal with it', and the cycle continues. Ugh.

I was missing the three most important words in this ongoing battle: 1) baseband, 2) backdoors, 3) Huawei.

America is rife with de facto monopolies and duopolies. That is our version of capitalism which champions competition but its all talk with little walk. That the government would pick on Qualcomm is a mystery to me. Its lobbyists must not have paid off the right people.

Whilst it seems good to break up this market, it also was rather interesting to see Apple whinge about monopolist market behaviour. Pot calling the kettle black; how very hypocritical.

In which market does Apple have a monopoly?

App Store, in their ios ecosystem.

Jeep doesn’t allow me to install my own apps on Uconnect. I guess they are a monopoly. Sirius, Apple, and Android are 3rd party products that are allowed on Uconnect but not my own.

> Jeep doesn’t allow me to install my own apps on Uconnect. I guess they are a monopoly.

Why wouldn't they be? Does someone else compete with them for distributing apps on Uconnect? Can you install Ford Sync apps on your Jeep?

Can I install android apps on my iPhone? Automakers like BMW have cut out people like Dinan when they locked down their firmware and refused to work with them.

Alpine and Pioneer probably make replacements for Uconnect (iPhone) but it is not even close to the integration I get with my Jeep and Uconnect.

Apple has the majority of the US market and elite market around the world.

Boy GM is really raking it in from UConnect. Everyone's complaining about their monopolistic practices.

Last checking GM's share price it was barely a composition of their revenue.

Does their profit or lack of it change the point? Does it only matter if someone is making money?

So they have a monopoly on selling their own product, essentially, I imagine would be the counter-argument?

Perhaps you should take a moment to review what a monopoly is.

I don't think things are as clear cut as you make them out to be.

Apple says nobody can run apps on their phone without handing them a 30% cut. No side-loading, no other stores, nothing. There is very obviously a market for apps that Apple controls completely. Just because there are other app stores for other phones does not mean Apple doesn't have control over the entire market for iPhone apps. Markets can be narrowly defined in anti-trust law. Apple is fleecing iPhone users for that 30%.

I don't think there's a good parallel case for this. DRM'd Keurig cups are the closest thing I could think of, but that never went to court that I know of. The case will be interesting: can you create a secondary market that you exert total control over? Can I bar anyone else from selling apps for my phone?

Personally, I don't think Apple should be able to do what they are doing and I hope they lose. It would be a win for consumer rights and free market principles.

>Apple says nobody can run apps on their phone without handing them a 30% cut.

False. You can pay for an app/service via the internet (side-pay). The app can be free on the app store. This is what Netflix does. Is it a bit of a hassle? Sure. This is the only part that has an optics problem even though it probably meets any legal thresholds. I do think Apple will get some pressure to make this process easier and that change will happen. Once that is done, there will be even less of an argument here.

From the consumer perspective, if one doesn't like the app store, buy a different brand of phone. There is no monopoly here from a legal or functional perspective.

Has Apple's so called monopoly reduced consumer app selection or pricing? Absolutely not. High quality software has never been more affordable and never has the selection been so incredible. It will be rather easy for Apple to argue that the app store is a net positive for consumers.

For the vast number of non HN users, the app store is a huge win. The whole morass of app selection, installation, update, and developer trust has been solved by the app store. If you are looking for harm to anyone other than the minuscule number of users that care about side loading, it just isn't there. Has everyone forgotten what software purchase, installation, and maintenance looked like a decade ago? What a mess. Fleecing is not the right word for something that has made consumers lives simpler in so many ways.

Some readers of HN might not like Apple's approach, but that doesn't make it illegal. There is just not a way to show substantial (any) harm to consumers. As I said, I think they will probably improve the side-pay options to improve their PR optics and that will be end of it.

> From the consumer perspective, if one doesn't like the app store, buy a different brand of phone. There is no monopoly here from a legal or functional perspective.

That is not how monopolies are defined. Apple says all apps for the iPhone must be signed by them and they must collect 30% of sales and in-app-purchases. The fact that they run an "App Store" means there is a market for iPhone apps. They control the market for iPhone apps. I don't know how to make that any clearer. The fact that Android exists does not change anything about the market for iPhone apps.

However, side-pay will be a very interesting defense. I haven't really considered that enough to speak on it.

I understand that Apple defines the narrow market for iPhone apps. That doesn't mean anything illegal is happening. Antitrust law doesn't say that the existence of a monopoly is illegal. There has to be harm, or exploitation, or price fixing, etc., for the courts to take action. A monopoly may just be the result of great execution by a company and it may not be in violation of any antitrust laws. Again, controlling a particular market may or may not be illegal.

Also, while it is true that Apple is controlling the marketplace for iOS apps, it does not control the market for smart phone apps in general. There is a good chance that that is sufficient for the courts to find no monopoly. As many of others have pointed out, there are many, many examples of narrowly defined monopolies. With a different definition, there is no monopoly.

Of course, none of us know what the courts and I get your concern. My point is that the simple control of a marketplace (created by Apple) for a specific product is not a sufficient condition for court action. The benefits to consumers could easily outweigh any concerns about control. Time (years) will tell.

Here is an interesting essay on press (public) perception of monopolies vs antitrust laws:


This is a tangent, but FYI, there is a DRM'd Keurig case, and it is still going! https://www.pbwt.com/antitrust-update-blog/in-long-awaited-o...

The judge essentially paused the case for four years while he ruled on a motion (which is ridiculous) so it's been quite slow. But I'm still watching it...

Thank you for this link, I'll start watching it too

> Markets can be narrowly defined in anti-trust law

I doubt one has ever been defined as narrowly as a single product. The differences between an iPhone and an Android phone are functionally so minimal as to be irrelevant.

>I doubt one has ever been defined as narrowly as a single product.

Actually, QCOM's antitrust case is exactly that: QCOM's dominance in one or two particular markets (or product), in this case, the CDMA and "premium" (aka, LTE) modems.

The market is LTE modems, not Qualcomm LTE modems.

If you define the market to be a single company’s offerings, when there are other companies that make similar products (Android and the Android App Store) then the concept of monopoly has become a tautology.

The problem some people are having goes back to the concept of the walled garden in the first place and general purpose compute devices.

Apple does not have a responsibility to provide general purpose computing smartphones that can run any software you want. If Congress wants to pass a law saying that they do, then fine, but it seems quite to stretch to claim that anti-trust law is designed for that.

Well no. In this case, it's "premium" LTE market.

So in that case, Microsoft with the XBox, Sony, and Nintendo are “monopolies” since you can’t sell a digital or physical game without paying them.

Except that isn't what is happening. Side loading is allowed on iOS and is how Netflix is currently monetizing their iOS users.

No, netflix isn't side loading apps. They are asking you sign up for the service on their site, not through the app to avoid having to pay Apple the 30%. A user still must get the app through the app store. There is no way to get an app on the phone that is not signed by apple..

You can now get developer certificates/provisioning profiles without paying the $100 you used to, just by having an Apple ID. You can replace the signature of an app given to you in binary form or build one given to you in source form via Xcode, which is also free.

Sorry I misspoke. I should have said only that iOS allows side loading. I thought Netflix was utilizing this.

I believe OP is referring to the recent ruling allowing an antitrust lawsuit against Apple's app store to proceed.

Not necessarily declared a monopoly yet, but also not declared not one.


Yes. My point was that they are exhibiting the same type of behaviour that they're accusing Qualcomm of, and that's why I said that's the pot calling the kettle black; they are more than happy to embrace the same self serving behaviour when it serves them.

The app store paid out $34bn to devs last year, so a very simple and probably naive calculation of 34/.7*.3 tells me they pocketed around ~$14.5bn in revenue from a total of $48.5bn. (source: https://finance.yahoo.com/news/developer-apos-34-billion-ear... )

That really ought to be big enough a pot to warrant my drawing the parallel.


- They have absolute control and will shut down things whenever they want, such as parental control apps - https://www.macworld.com/article/3391361/apple-privacy-paren... .

- They claim you may install alternative apps, yet you may not set them as default handlers to replace Apple's native ones- https://www.theverge.com/2019/5/29/18644045/apple-defends-ap...

- They're playing hardball with Spotify, who are also using the word "monopoly": https://www.theverge.com/2019/5/6/18530894/apple-music-monop...

You realize those “parental control apps” gave a third party complete access to your child’s phone and were able to intercept all of your cbild’s communication.

Do you think parents were aware of that? Are you okay with that? If you want to control your child’s phone, you are free to use MDM software where you control provisioning and the profiles.

That’s like complaining that IKEA has a monopoly within IKEA stores.

> That’s like complaining that IKEA has a monopoly within IKEA stores.

I feel like the fact that only Apple makes iOS devices is what's confusing people.

Suppose that Tesla is the only company that makes electric cars and also the only company that has any electric car charging ports. (This is a hypothetical; in practice other companies offer both electric cars and charging ports.)

They don't have a monopoly on cars -- you can go buy a Volkswagen diesel if you like, and "electric cars" isn't necessarily a separate market than "cars" at the time of purchase. But once you have an electric car, they do have a monopoly on charging ports, which is a separate market because you can't run your electric car on petroleum.

This is easier to see if you add another electric car company but not another electric car charging company. So you go buy your Nissan Leaf but you still have to use a Tesla charging port with it. Clearly a monopoly on charging ports. But still the exact same charging port monopoly when they're the only company making electric cars.

You can have a monopoly on iOS app stores even if you're the only company that makes iOS devices. Having fewer competitors in the other market doesn't make you less of a monopoly. Even the fact that they compete there at all -- if Amazon was the only company with an iOS app store, they would have a monopoly on iOS app distribution even if they don't even make iOS devices.

The problem with your argument is that electricity is fungible and it's wide available. When you buy you Nissan Leaf, you don't need Tesla's charging ports, you can plug it on any power outlet. You might miss the fast charging, but that's it.

When you go buy a phone, one of the factors you have to decide is "what app store do I want". If you go with Android, you might miss a few apps, but so what ? Most of them are also fungible, just use a similar one.

If it wasn't like that, we could say that Wall Mart is a monopoly because only they sell Great Value branded products.

> The problem with your argument is that electricity is fungible and it's wide available. When you buy you Nissan Leaf, you don't need Tesla's charging ports, you can plug it on any power outlet. You might miss the fast charging, but that's it.

You're explaining why Tesla doesn't have a charging port monopoly in practice. The hypothetical is that they do. You can't use any other power outlet in the same way that you can't install an Android app on an iPhone.

> When you go buy a phone, one of the factors you have to decide is "what app store do I want".

Which is what creates the app store monopoly. There wouldn't be one if you could make that decision independently, which is the whole point. It's classic tying.

Moreover, the "customer" of the app store is as much the developer as the user, given that they're the ones paying all the fees. But they don't get to choose between platforms any more than Walmart chooses to operate in Florida instead of California, because they have to reach their customers in both regions, not just one or the other.

By contrast, Nike can choose Target instead of Walmart because they both operate in both regions, so they can still reach substantially all of their customers through one store if the other is being unreasonable, since it's much easier for the customer to switch stores for a desired product when that doesn't require switching locations/platforms at the same time.

But the market isn’t iOS devices, the market is smart phones. And in that market Apple does not have a monopoly.

If you start declaring that the market is the specific store or product, then literally every single product and service is a monopoly begging for state intervention. Can I sue Netflix because as the monopoly holder of the Netflix network they refuse to carry my home made videos? Is Spotify abusing their monopoly over Spotify by refusing to carry my karaoke? Should I be able to sue them in order to bend them to my will?

Of course not. That’s patently ridiculous.

The market is what products people actually move to and from in response to Apple pricw changes; if there's a segment of buyers and price range in which Apple has pricing power—price changes do not cause such movement—Apple is an antitrust monopoly. Popular descriptive “markets” like “smartphones” often include many isolated submarkets and are not single markets in which competition exists.

A major purpose of brnad marketing is to create such pocket markets.

So your point is that there are no high priced smart phones other than Apple, therefore it’s a monopoly?

Aside from being extremely silly, under this formulation Rolls Royce might be a monopoly, it lacks factual backing. There are plenty of android offerings that overlap Apple offerings in price, including the Pixel 3 and most of the Galaxy S10 series. If your point is that someone willing to spend $1k on a smart phone has no other choice other than to buy Apple, the $1k+ Galaxy S10+ would like to have a word with you.

If your point instead is that any brand that builds up a “pocket market” via good products and marketing is an abusive monopoly that just be stopped, then you’re signing up basically every top company in the every market segment for stringent antitrust enforcement, which is so broad a definition as to be useless. You can’t sue Nike as a monopoly because they’ve built up their own fan base and “pocket market”.

> So your point is that there are no high priced smart phones other than Apple, therefore it’s a monopoly?

Nope. My point to is neither “there are no high priced smart phones” (there are) not “it’s a monopoly” (I have no position on that question). You should be able to tell that because I say nothing similar to either of those anywhere in the post you are responding to.

My point is that the existence of other players in a particular popular framing of what the market is has very little to do with the anti-trust definition of “monopoly”, which has to do with empirical competition (what do consumers move to or from in response to, particularly, price changes), not how markets are popularly described.

> If your point instead is that any brand that builds up a “pocket market” via good products and marketing is an abusive monopoly that just be stopped

Again, no. Building a non-competitive market via branding or other means makes you a monopoly, not necessarily an abusive one. You still have to abuse that monopoly to be an abusive monopoly.

That's a lot of words that take literally no stance on the issue. You've slowly shifted back to an incredibly vague definition of a market and an incredibly vague definition of what an abusive monopoly is. At this point it's literally impossible to discuss the issue with you, because the definitions have become uselessly broad.

> You've slowly shifted back to an incredibly vague definition of a market and an incredibly vague definition of what an abusive monopoly is

I've had two posts in this thread, both of which take exactly the same position despite your ridiculous misinterpretation in between them; I haven't shifted, slowly or otherwise.

And the definition of market/monopoly is exactly the one used in antitrust law. That may be inconvenient for your desire to argue, but that doesn't change the facts. And I haven't presented any definition of an abusive monopoly, just corrected your claim that a particular definition of monopoly also meant any company that meant it was also an abusive monopoly, noting that “abusive” actually does have meaning.

So your argument is that only Apple can make an app store?

I feel like Google might disagree.

Only Apple can make an iPhone app store, which is the market they are alleged to have monopolized

It’s their market! Only Apple can make an Apple App Store for the Apple iPhone.

Every company deserves a “monopoly” on their own product, unless they are a common carrier, a designation which is based on the government enabling them to exist through preferential treatment in the first place.

Apple created the Apple iPhone and they should be allowed to determine its destiny, as long as other companies are also allowed a fair shot to create their own competing devices.

> "Every company deserves a “monopoly” on their own product,"


Because anything else is silly. Am I allowed to release a Ford car because allowing Ford to have a monopoly on Ford cars is unjust?

That's not remotely comparable. Comparable would be asking if you're allowed to release third party components (hardware or software) for Ford cars without asking for Ford's permission.

Obviously hardware third party components without Ford's permission is a well established industry. Nobody questions it, except maybe a few lunatic capitalists like Elon Musk. Even then though, I think people like him keep their head on straight for the most part. They don't do much to facilitate it, but if you started selling compatible hardware components for his cars, I doubt they'd actually consider trying to stop you. (If I'm wrong about that, he's worse than I thought.)

Your supposition is that software components should be a special case, unlike hardware components, where companies have a right to deny anybody else the right to produce components without their permission. I say that's bullshit. They should be legally forbidden from creating products for which third party components can only exist if given permission.

If they are big enough then yes it can be a monopoly too.

Apple has about 45% of the US smart phone market with their next competitor being Samsung at about 29%, with both of those companies on track to lose share to Huawei. Globally Apple is about 11% of the market, with Samsung selling well over double of what Apple sells.

Those numbers, combined with the fact that a new competitor is arising to take market share away from Apple strongly implies that Apple is not currently behaving in an anticompetitive fashion.

The iPhone app-approval market leaps to mind.

If you mean Apple owns and runs an App Store, then yes, but that’s not a monopoly. It’s just a store that they control. Would you say that their physical stores are a monopoly that they abuse by not selling samsung phones?

Then why are we seeing eg https://techcrunch.com/2019/05/13/supreme-court-rules-agains... ?

Regardless, their behaviour with their own Dev relations is very much like what they're accusing Qualcomm of.

The Supreme Court didn't rule on the merits of the case. The actual antitrust case is extremely weak against Apple. You'd need a nonsensical market definition to say Apple has monopoly power, and it wouldn't hold up in court.

The supreme court didn't say Apple was a monopoly it said that people could sue them to see if they were.

If I want to sell an app to users on iOS, and those users want my app, there's no way for them to get it without Apple. That's a vertical market monopoly. And even if it doesn't qualify as a monopoly in your mind, it can still run afoul of anti-trust law.

If you want to sell a product in Apple’s physical store, there’s no way to do it without Apple.

That’s not a monopoly. That’s them exercising control over their own property.

What’s the rationale to treat Apple’s digital store different than Apple’s physical store?

You could say the same about console makers. They have been doing the same thing since the 80s....

The monopoly is that no other software distributor can be used on iOS other than Apple.

That is a monopoly in their own product, not in a market.

In this case, I think it's both. Contrast Android, where other people have set up app stores: https://en.wikipedia.org/wiki/List_of_Android_app_stores

Doesn't matter, Apple is the only supplier of iOS apps and forbids any competition here.

LG is the only supplier of apps for their TVs and forbids any competition there.


So does LG take a 30% rake of all app sales? It's not illegal to have a monopoly. It is illegal to take advantage of that fact with non-competitive pricing.


This poor understanding of what a monopoly is on your part is a testament to how uncommon true monopolies have become thanks to modern regulation.

Time will tell.

> If you mean Apple owns and runs an App Store, then yes, but that’s not a monopoly.

They also refuse to permit their customers to run apps not approved by Apple on the phones they have bought. That's the monopoly.

Huh? That’s still not a monopoly. Argue that they’re violating anti-consumers laws to have access to device you purchased, sure, but what you’re describing is in no way a monopoly.

I think it comes down to how narrow you want to define the market. And even if it is a monopoly, that doesn't mean anything by itself. Monopolies aren't illegal.

That’s factually false. You can side load apps, plenty of enterprises do it.

last i looked that requires a couple hundred dollar per year special developer account from apple, with special checks to make sure you're actually a business doing it for business reasons.

and they still sign your apps, they just do it automatically, i think.

So? You can still side load. Developers do it all the time. The approval process is only for distributing your private app to hundreds of people using Apple’s servers and hardware.

look, buddy, i'm with you on the monopoly issue, but this whole side loading thing is bullshit. i don't think these mechanisms work the way you think they do, and they're certainly not reasonable for even expert users to keep an app up to date on their device. it just isn't legally relevant.

Having side loaded apps personally, I'm not really sure why you're so pissed off.

That's like saying McDonald's has a monopoly on what food appears on their menu. It's non-sensical as a complaint.

The question is how you want to characterize an iPhone. If the phone is Apple's property that they're letting you use, then Apple's app store restrictions are analogous to McDonalds deciding what going on their menu. If the iPhone is a product that you bought from Apple, then Apple's app store policies are equivalent to GE selling me an oven that will only cook food purchased from GE's grocery store.

Luckily, oven technology doesn't really make that business model feasible so GE doesn't try to do it. But I think there are reasons for antitrust authorities to take a careful look at the behavior of companies that do have this kind of power over their customers.

But none of those things are a monopoly. A GE oven that only cooks GE food isn’t a monopoly, it’s just a shitty oven.

Plus, monopolies aren’t illegal, using them anti-competitively is. So the analogy would be GE buying or putting out of business all the other oven makers, then saying you can only cook GE food in a GE oven.

It's maybe not a monopoly, but it's definitely tying, which is still covered under anti-trust law.

In United States v. Microsoft Corp.[0] it was determined that Microsoft abused monopoly it's power by simply preventing OEMs from removing internet explorer.

Apple demands that all web browsing is done via Safari on IOS as I understand it.


Microsoft had a 90%+ market share in computer operating systems, aka a monopoly. So there's that.

But Microsoft did have a monopoly in the browser market at that time. Apple doesn’t have a monopoly in the browser market or the phone market or the app market. You can’t abuse a monopoly you don’t have.

QCOM doesn't have a monopoly in the baseband market either, yet Koh's court found QCOM's monopoly there. The FTC simply redefined a market where QCOM had a dominant marketshare -- the "premium" modem market (LTE) -- for Koh to come to that ruling.

So I guess if someone can prove AAPL's dominance in "premium" PC, education, or some niche subgroup in PC market, it too can be in trouble.

> So I guess if someone can prove AAPL's dominance in "premium" PC, education, or some niche subgroup in PC market, it too can be in trouble.

If Apple used its dominance in some niche subgroup PC market to force buyers to pay a percentage of the total cost of all USB-C devices they buy, including non-Apple devices, to Apple to pay for Apple's patents related to USB-C. And if those buyers were companies like Best Buy that needed to be able to buy Apple devices or else lose access to over half their market. And if Apple refused to sell buyers any devices unless they agreed to those terms. And if Apple promised during standardization of USB-C to offer licenses to said patents on FRAND (fair, reasonable, and non-discriminatory) terms. Then that analogy might hold a little water.

The monopoly was in the OS market. Microsoft got in trouble for trying to use that monopoly to create a new one in the nascent browser market.

And Apple has no monopoly. Having a monopoly is a pre-requisite for being found guilty of using a monopoly to leverage your way into an adjacent market.

End of story.

You can use Chrome on iOS.

Chrome on iOS still uses Safari's renderer and JS engine.

My understanding is that Google is allowed to use their own renderer and their own Javascript engine - as long as it doesn't do JIT compilation. The ability to mark pages or files as executable is a platform security model thing.

Chrome uses WebKit because Google would rather use a WebKit/JSC engine than a slow Blink/V8 engine - especially since Blink is a WebKit fork.

If the alternative is something so slow that nobody would actually use it, does the why really matter that much in practice?

What? That's not the same thing at all. McDonald's doesn't let third party companies sell food on their menu. McDonalds is a place, not a platform. Your analogy falls short in so many ways I have to assume you're being disingenuous.

No, the case of Apple is more like McDonald's being able to prevent its customers from ever eating food not signed-off on by McDonald's.

They can do that, if you are sitting in a McDonald’s dining room.

They can’t stop you from walking over to Burger King.

It's analogous to McDonald's selling you a dining room and then regulating what food you can eat in it.

Except you're perfectly aware of that fact when you buy the dining room.

How so? Apple can’t stop you from using a different phone.

I think people are confusing mild inconvenience with monopolistic oppression.

Registration is open for Startup School 2019. Classes start July 22nd.

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