Naked Power

Apple and Google’s mobile duopoly is held together with chewing gum, twine, and moral vacuity.

Twitter’s descent into a gutter of the lowest order has been gut-wrenching:

Elon Musk’s Grok and the Mass Undressing Scandal

As I draft this, a week later, it appears pressure from civil society, investigations by regulators, and outright bans on multiple continents have forced Musk to back down to an uncertain degree.

As this scandal roiled, Twitter’s apps have been continuously available in Google’s Play and Apple’s App Store, marking new lows in moral cowardice and non-enforcement of the duopolist’s own policies.

Now we sit on tenterhooks, wondering if the worst has actually passed. What outrage the will valley’s billionaire man-children unleash next? Meanwhile, we brace for this episode to embolden censorious authoritarians keen to extract more governmental power over the media and legitimate speech.

This is the backdrop to Elizabeth Lopatto’s must-read denunciation in The Verge:

It is genuinely unbelievable to me that I wasted hours of my actual life on a court case where Apple explained it needed total control of its App Store to protect its users. Total control of the App Store was Apple’s main argument against antitrust enforcement: The company insisted that its monopolistic control of what users could install on their phones was essential to create a walled garden where it could protect children from unsafe content.

Ha! Ha ha ha!!

— Elizabeth Lopatto,"Tim Cook and Sundar Pichai are cowards, The Verge"

Failure to react to the “everything app” going all-in on abuse of women and girls for weeks reveals the illegitimacy of their mobile monopolies1; anyone pretending otherwise is a fool or a dupe.

We don’t need to guess why they did it. Acting against Musk’s abusive apps might put Apple and Google out of favour with an erratic, power-tripping administration which in turn could impact short-term business prospects. Choosing their own stated principles is incompatible with maximizing shareholder value under competitive authoritarianism.

Recall that both firms lent their monopolies on software distribution to ICE, citing the implausible claim that federal agents are a “vulnerable group.”

This new scandal is just the same choice in relief. Publishing the unthinkable at the behest of administration allies pays homage to power through obscenity. If they offend Musk…who knows what could happen? So maybe let it play out; let others take the heat. Surely somebody will do something. The internal monologue of the quisling scarcely needs exposition.

And so abuse at scale was amplified through their channels, against their own policies, for weeks.

Vertigo

The duopolist’s justification for the necessity of monarchical app stores have always been bullshit, top to bottom, stem to stern.

App stores are not sui generis; they’re just programs that install other programs, and "apps" are whatever the OS says they are.

As Cory Doctorow observed:

Apps interact with law in precisely the way that web-pages don’t. “An app is just a web-page wrapped in enough IP to make it a crime to defend yourself against corporate predation”

It sure looks like Apple and Google failed to protect women and girls in order to preserve the rents they extract from the ecosystems these IP wrappers give them control over.

Gatekeepers like to point out that the wrapper comes with treats, but this is misdirection. Web apps could provide safe, privacy-enhancing versions of every capability they currently reserve to “native” apps, and the gatekeepers know it.

That’s why Apple has worked furiously to keep APIs away from browsers through legal wrangling and subversion of standards. Cryptography and lawyers have also been enlisted to keep other programs-that-install-programs out and a safe, powerful open web at bay. Without those shields, we’d see the deeper failures clearly.

Consider the justifications Apple and its merry band of astroturfers trot out like clockwork to delay browser choice. Cupertino argues it must exclusively control browsers and software distribution to:

• Ensure device security
• Prevent frauds and scams
• Provide a bulwark for privacy
• Simplify software acquisition and distribution
• Keep a lid on the most objectionable content

We can see now that real protection on the last point comes not from the stores, but from governments. This realisation provides a template: each justification is an admission, a misdirection to cover for their own failures.

Let’s take it from the top.

The Security Argument

Stores don’t ensure security, runtimes do.

Operating systems and browsers — the platforms that sandbox code and mediate permissions — protect users to the extent they’re designed to; app stores are just overwrought “beware of dog“ signs meant to scare off easily intimidated ne’er-do-wells. So it’s no surprise that whenever app stores are trusted with the role, a trail of embarrassing failures follows.

This unearths the lie behind the obfuscation: iOS and Android didn’t create app stores to deliver unheard of security — iOS 1.0 did that by removing unsafe native code and replacing it with the web; ChromeOS did the same for desktops shortly thereafter — no, the gatekeepers built the app stores because their OSes are insecure places to run native code.

There was no App Store on the original iPhone because it only ran third-party code on the world’s most secure platform: the browser. When apps ran in the tightly sandboxed web, the presumption of safety reigned. It was only because iOS and Android were (and remain) fundamentally insecure for native apps that it became a necessity to introduce a store with iOS 2.0. It’s no accident that this was when Apple retreated from making the safe, open, and interoperable web more capable.

Apple’s App Store was a marketing answer to a brand-promise problem: what to do about a hole below the water line that Product and Engineering aren’t just failing to patch, but are enthusiastically expanding instead?

The whole facade of the duopolist’s power hinges on the false claim that stores create security. Without the need to paper over the disaster of carelessly dispensed power tools, none of the rest of the services the stores provide could be justified; certainly not at the ruinous prices they demand.

More recent, chest-thumping pronouncements need to be evaluated in the same light; these aren’t heroic explorers of new frontiers, they’re embarrassed students bluffing book reports for tomes they didn’t read.

Instead of protecting us, app stores reward platform vendors for failure and foster centralising, anti-Open Source ecosystems. Open societies cannot abide closed platforms that assert ownership of this much of our lives, particularly not when claims of security are based on misrepresentations.

Frauds and Scams

In the narrow conception, the app stores are feckless. Taking a wider view, they’re complicit, if not enabling.

In a strict legalistic sense of “fraud," the track record of app stores is abysmal. Take just one recent example: while loudly proclaiming to protect users from scams, Apple simultaneously facilitated wide-scale app impersonation at the launch of Sora.

For its part, Google routinely facilitates shocking amounts of ad fraud via Play. Stores failed to catch clearly fraudulent fronts for sanctioned Russian banks, and even bald-faced imposters have been a long-running problem. This is just the tip of the proverbial iceberg.

If we widen the aperture to let in adjacent classes of user abuse, it gets immeasurably worse.

Apple’s policies purport to disallow use of the ultra-low-friction IAP systems for gambling:

5.3.3 Apps may not use in-app purchase to purchase credit or currency for use in conjunction with real money gaming of any kind.

This text is lawyered to sound like a curb on gambling addiction’s worst effects. In reality, it’s designed to facilitate the predatory “gambling lite” systems Apple and Google gleefully fostered.

For most of the mobile duopoly’s existence, the primary revenue driver has been the problematic, gambling-adjacent behaviour of “digital whales” in so-called “casual games.”

And don’t imagine the wilfully predatory behaviour is limited to adults. By allowing “bait apps” — even after previous FTC settlements that should have forbidden them — the app stores have shown us the duopolist’s true colours. Serial disregard for the financial health of users is literally baked into their model.

This is the rotten core of app stores. Understood in POSIWID terms, mobile app stores exist to tax the problematic gambling of vulnerable users.

Privacy

App stores safeguard privacy the way packs of wolves safeguard flocks of sheep.

The only appropriate response to the two-faced, duplicitous claims by Apple and Google towards privacy in recent years is incandescent rage.

I’ve covered before how Apple’s posturing against Facebook is nothing but kayfabe and how Cupertino’s privacy arguments regarding alternative browsers are steaming piles of illogical nonsense.

In reality, our privacy problems have been multipled because of Apple and Google.

It was the duopolists that created APIs for persistent background access to your contacts, calendar, location, radios, battery levels, and much else besides. And they did this knowing full well it was going to lead to abuse. Remember, they exposed this information to all comers after having built browsers and web-based alternatives that could have been extended more thoughtfully.

It was the duopolists who handed those APIs to native apps from shady publishers like Facebook with less-than-thoughtful controls. And it was these very companies that failed to police even their mildest policies.

These same trillion-dollar market-cap firms simultaneously declined to do the one thing that had a chance of actually improving privacy: using their incredible lobbying capacity to call forcefully for privacy regulations worth a damn, preferring a market structure where they can posture against each other.

Solving the root issue might deprive them of a marketing tool, after all. And they have got away with it. Their press and product shops are keenly aware reporters don’t understand privacy deeply enough to call their bluff, and that so-called privacy experts will happily clap for symbolic gestures.

Humiliatingly for the fourth estate, Cupertino and Mountain View’s self-issued privacy participation prizes were never questioned. Indeed, credulous journalists continue to shower them with praise for steps away from the very worst excesses best measured in angstroms.

Apple have been allowed to take credit for foisting responsibility onto users while Google has faced no sustained questioning for just giving up, having never launched anything at all to structurally curb Android abuses.

Cynics might be inclined to think this was very much the point.

Indeed, the POSIWID description of these monopolies-on-apps-that-install-apps is that they exist to squash competition. Apple’s not trying to keep alternative browsers off of iOS because they’ll hurt privacy, they’re keeping them at bay because they could provide an alternative. One that might challenge the (low) quality of Apple’s offer while eliding Cupertino’s ability to extract usurious taxes along the way.

Software acquisition and distribution

You know what the easiest way to get an app is? Clicking a link.

Apple literally pioneered this model with iOS 1.0, only to walk away from it when it chose to expose new, carelessly overpowered, unsafe-by-default APIs to developers with the introduction of native apps. Throwing away privacy and security made software harder to build and distribute, too, but deposited power over developers with OS gatekeepers. Over time, that power became addictive.

A more secure and privacy preserving model is still possible but the duopolists continue to suppress it. I can’t speak out of school about all the ways Android and Play mirrored Apple’s underhanded tactics to suppress PWAs, but suffice to say it was a lot.

Industrial-scale suppression of safe, privacy-respecting platforms has been packaged up in florid terms as an advantage for developers. Except developers hate mobile app stores. But you don’t have to take my word for it.

Given the choice, developers would do exactly what the gatekeepers do when constructing billing, distribution, and marketing systems: shop around in an open market, based on standards-oriented technologies, and select the best fit for their needs.

This is exactly the model that gave rise to the web and to web search. Discovery isn’t impossible; it isn’t even that hard. If we can build search engines for web pages, we can also highlight sites in those engines that want to be installable. None of this is magic, and none of it requires a 30% take from the developer’s budget.

Content Moderation

For the sake of completeness, we should stipulate here that an end to app stores, including a potential flowering of web apps and alternative browsers, would not meaningfully change the content moderation landscape.

We now have a powerful example of this counterfactual thanks to the Twitter/Grok episode. There is no safety to be lost when we replace the gatekeeper’s app stores with a powerful, open, interoperable web. The app stores stand for nothing and will stand up to no-one. Good riddance.

The Rot of “Who Should Rule?”

Before the 2024 US elections, tech titans were well-enough advised to know which way the winds were blowing. But that did not stir them to defend truth, the rule-of-law, or even the employees that enabled their success. Instead, they hurried to capitulate. Today they sponsor coup-excusers pay vigs, grovel to people they surely loathe, and fund the literal destruction of America’s institutions.

This month’s failure to stand up for basic decency is just another link in that chain.

Having narrowed the running to two choices, mobile’s masters always ask us to consider governing our phones through the authoritarian frame of "who should rule?"

But these aren’t our only choices. As Popper retorts, the better question is "How can we so organize political institutions that bad or incompetent rulers can be prevented from doing too much damage?"

This isn’t purely a political question, but applies to all of society’s power structures. The callous indifference of the app store’s billionaire managers (1, 2) when faced with an even moderately difficult call tells us that they cannot be trusted; this was the test, and the mobile overlords failed by their own terms.2

What’s left for the rest of us to take on is how we dismantle the mechanisms our misplaced trust helped them build. This will not be easy, and an insightful commenter at The Verge restates the core problem:

This is true and fantastic reporting and why we need to pay for The Verge.

But, it begs the question, what do we do?

Do we opt out of the tech of the modern world to protest? Commitment to values isn’t what we talk about, it’s what we are willing to give up. A key problem is that we don’t have any real competition vs Apple or Google as platforms if we want to exist in the modern world or even have this conversation.

You can’t (easily) read this or participate from a Kobo or Lightphone. Anyone have any suggestions?

I dropped off Twitter and Meta, but I’m running out of options.

— Anonymous commenter,"Tim Cook and Sundar Pichai are cowards, The Verge, Comments"

We aren’t going to get anywhere by throwing our iPhones and Androids into the sea.

Credible, incremental steps that remove power from the gatekeepers are now demanded, and as I have previewed throughout this piece, the open web is that next step. It has all the properties we need to attenuate misgiven power: no single vendor control, based in standards, multiple OSS implementations, and most of all, portability.

The web is an abstraction that holds the power to liberate our computing in-situ, removing superfluous gatekeepers from the loop an increasing fraction of the time. As use of the web grows, so do the prospects for alternatives OSes and hardware ecosystems. They know this, and that’s why they’re trying to keep the web from winning.

Moving our computing to browsers and web apps won’t protect us from Musk, but neither will Apple or Google.3 Now that we know that, we can at least start to claw back at the corrosive power of monopolists in our pockets by building for a future that doesn’t depend exclusively on them.

FOOTNOTES

Some folks like to continue to pretend that the mobile duopoly still includes any serious competition for either player. I assume those people are paid to review phones for a living.

As I outlined in this year’s instalment of the Performance Inequality Gap series, the mobile market is actually two distinct markets: iOS for the rich, and Android for the rest. The average price for iPhones is hovering nearly $1K, while the average Android costs $300 new, unlocked. There is no functional competition between these ecosystems, and though they’ll never admit it, that’s a situation the duopolists are more than comfortable in, even if they don’t particularly love it. ⇐ 1.

If it always falls to regulators to protect women and girls from Elon Musk and his Trumpian alliance, what is the point of Tim and Sundar? Of Play and the App Store?

And if their policies are just fig leaves to justify rent extraction, why should any regulator listen to anything they say?

These questions should be hair-on-fire in the capitols of functioning democracies. ⇐ 1.

It is not the most offensive thing about this episode by a country mile, but I am driven to distraction by how unbelievably stupid Apple and Google have become.

Did Tim and Sundar really think that, having sniffed weakness once, Trumpist shake downs would pass them over the next time a pro quo could be extracted for the quid?

Did they not understand that by participating in oligarchy they signed on to authoritarianism?

Did they really fail to calculate that capitulation didn’t lower their risks, only centralised them?

This was all predictable. You don’t have to look as far as Russia to understand that autocrats grant temporary loans of state power towards undemocratic ends to create leverage for themselves, not the borrower. And whatever the price, autocrats never stay bought.

Everyone but them knew that domination is a ladder, and now we’re all paying the price. ⇐

The Performance Inequality Gap, 2026

Have we finally rounded the corner? A look at the device and network landscape.

The Budget, 2026 Edition

Let’s cut to the chase, shall we? Updated network test parameters for 2026 are:

• 9 Mbps downlink
• 3 mbps uplink
• 100 millisecond RTT

Regarding devices, my updated recommendations are the Samsung Galaxy A24 4G (or equivalent) and the HP 14. The goal of these recommendations is to emulate a 75th percentile user experience, meaning a full quarter of devices and networks will perform worse than this baseline.

Plugging these parameters into the updated budget calculator, we can derive critical-path resource thresholds for three and five second page load targets. Per usual, we consider pages built in two styles: JS-light, where only 15% of critical-path bytes are JavaScript, and JS-heavy, comprised of 50% JavaScript:

Time	JS-light (MiB)	JS-heavy	Total	JS	Other	Total	JS	Other
3 sec	2.0	0.3	1.7	1.2	0.62	0.62
5 sec	3.7	0.57	3.2	2.3	1.15	1.15

Note: Budgets account for two TLS connections.

Many sites initiate more early connections, reducing time available to download resources. Using four connections cuts the three-second budget by 350 KiB, to 1.5 MiB / 935 KiB. The five-second budget loses nearly half a megabyte, dropping to 3.2 / 1.9 MiB.

It pays to adopt H/2 or H/3 and consolidate connections.

These budgets are extremely generous. Even the target of three seconds is lavish; most sites should be able to put up interactive content much sooner for nearly all users.

Meanwhile, sites are ballooning. The median mobile page is now 2.6 MiB, blowing past the size of DOOM (2.48 MiB) in April. The 75th percentile site is now larger than two copies of DOOM. P90+ sites are more than 4.5x larger, and sizes at each point have doubled over the past decade. Put another way, the median mobile page is now 70 times larger than the total storage of the computer that landed men on the moon.

Median page weights are more than 2.5x larger for mobile sites than a decade ago, and sites at the 75th percentile are now 4x their 2015 weight.

An outsized contributor to this bloat comes from growth in JavaScript. Mobile JavaScript payloads have more than doubled since 2015, reaching 680 KiB and 1.3 MiB at P50 and P75 (respectively). This compositional shift exacerbates latent inequality and hurts businesses trying to grow.

When JavaScript grows as a proportion of critical-path resources, the impact of higher CPU cost per byte reduces budgets. This coffin corner effect explains why image and CSS-heavy experiences perform better byte-for-byte than sites built with the failed tools of frontend’s lost decade.

Indeed, the latest CrUX data shows not even half of origins have passing Core Web Vitals scores for mobile users. More than 40% of sites still perform poorly for desktop users, and progress in both cohorts is plateauing:

This is a technical and business challenge, but also an ethical crisis. Anyone who cares to look can see the tragic consequences for those who most need the help technology can offer. Meanwhile, the lies, half-truths, and excuses made by frontend’s influencer class are in defence of these approaches are, if anything, getting worse.

Through no action of their own, frontend developers have been blessed with more compute and bandwidth every year. Instead of converting that bounty into delightful experiences and positive business results, the dominant culture of frontend has leant into self-aggrandising narratives that venerate failure as success. The result is a web that increasingly punishes the poor for their bad luck while paying developers huge salaries to deliver business-undermining results.

Nobody comes to work wanting to do a bad job, but low-quality results are now the norm. This is a classic case of under-priced externalities created by induced demand from developers and PMs living in a privilege bubble.

The interactive budget calculator has been updated and revised for 2026, allowing you to see the impact of networks, devices, connections, and JavaScript on site performance.

Embedded in this year’s estimates is hopeful news about the trajectory of devices and networks. Compared with early 2024’s estimates, we’re seeing budget growth of 600+KiB for three seconds, and a full megabyte of extra headroom at five seconds.1

While this is not enough to overcome continued growth in payloads, budgets are now an order of magnitude more generous than those first sketched in 2017. It has never been easier to deliver pages quickly, but we are not collectively hitting the mark.

To get back to a healthy, competitive web, developers will need to apply considerably more restraint. If past is prologue, moderation is unlikely to arise organically. It’s also unhelpful to conceive of ecosystem-level failures as personal failings. Yes, today’s frontend culture is broken, but we should not expect better while incentives remain misaligned.

Browsers, search engines, and developer tools will need to provide stronger nudges, steering users away from bloated sites where possible, and communicating the problem to decision-makers. This will be unpopular, but it is necessary for the web to thrive.

Contents

• The Budget, 2026 Edition

• Recommended Test Devices and Settings

• The Big Story Is Still Low-End Android

• The Landscape

• Mobile

• Desktop

• Networks

• Looking Forward

• Content Trends

• Are SPAs Working?

• Analysis and Conclusions

Recommended Test Devices and Settings

This series has continually stressed that today’s P75 device is yesterday’s mid-market Android, and that trend continues.

The explosive smartphone market growth of the mid 2010s is squarely in the rearview mirror, and so historical Average Selling Prices (ASPs) and replacement dynamics now dominate any discussion of fleet performance.

Hardware upgrade timelines are elongating. Previous estimates of 18 months for replacement on average is now too rosy, with the median smartphone now living longer than two years. P75 devices may be nearly 3 years old, and TechInsights estimates a 23.7% annual replacement rate.

With all of this in mind, we update our target test device to the Samsung Galaxy A24 4G, a mid-2023 release featuring an SoC fabbed on a 6 nm process; a notable improvement over previous benchmark devices.

Readers of this blog are unlikely to have used a phone as slow as the A24 in at least a decade.

The A24 sold for less than the global Average Selling Price for smartphones at launch ($250 vs. $353). Because that specific model may be hard to acquire for testing, anything based on the MediaTek Helio G99 or Samsung Exynos 1330 will do just fine; e.g.:

Samsung Galaxy A16 4G

Samsung Galaxy A07 4G

Xiaomi Redmi Note 13 Pro 4G

Teams serious about performance should track the low-end cohort instead, sticking with previously acquired Samsung Galaxy A51’s, or any late-model device from the Moto E range.2

For link-accurate network throttling, I recommend spending $3 for Throttly for Android. It supports custom network profiles, allowing you to straightforwardly emulate a 9/3/100 network. DevTools throttling will always be inaccurate, and this is the best low-effort way to correctly condition links on your primary test device.

Desktops are not the limiting factor, but it’s still helpful to have physical test devices. You should not spend more than $250 (new) on a low-end test laptop. It should have a Celeron processor, eMMC storage, and run Windows. The last point is not an effort to sell more licences, but rather to represent the nasty effects of defender, NTFS, and parasitic background services on system performance. Something like the HP 14 dq3500nr.

Behold, the HP 14! A Celeron N4500 laptop, sporting a 4-core chip first released in 2021. This CPU packs less than a quarter as much cache as a late-model iPhone.

Desktop network throttling remains fraught, and the best solutions are still those from Pat Meenan’s 2016 article announcing winShaper.

The Big Story Is Still Low-End Android

What we see in our recommended test setups is an echo of the greatest influence of the past decade on smartphone performance: the spread of slow, ever-cheaper Androids with ageing chipsets, riding the cost curve downward, year-on-year.

The explosive growth of this segment drove nearly all market growth between 2011 and 2017. Now that smartphones have reached global saturation, flat sales volumes mirror the long-term trends in desktop device ownership:

At no point in the past dozen years has iOS accounted for more than 20% of new-device shipments. Quarter-by-quarter fluctuations have pushed that number as high as 25% when new models are released, but the effect never lasts.

Most phones — indeed, most computers — are 24+ month old Androids. This is the result of a price segmented market: a preponderance of smartphones sold for more than $600USD (new, unlocked) are iPhones, and the overwhelming majority of devices sold for less than that are slow Androids.

The “i” in “iPhone” stands for “inequality.”

Global ASPs show the low-end isn’t just alive-and-well, it’s many multiples of high-end device volume. To maintain a global ASP near $370, an outsized number of cheaper Androids must be sold for every $1K (average) iOS device.

The Landscape

To understand how the payload budget estimate is derived, we need to peer deeper into the device and network situation. Despite huge, unpredictable shocks in the market (positive: Reliance Jio; negative: a pandemic), the market trends this series tracks have allowed us to forecast accurately.

75th+ percentile users are almost always on older devices, meaning we don’t need to divine what will happen, just remember the recent past.

Mobile

The properties of today’s mobile devices define how our sites run in practice. From the continued prevalence of 4G radios, to the shocking gaps in CPU performance, the reality of the modern web is best experienced through real devices. The next best way to understand it is through data.

Per usual, I’ve built single and multicore CPU performance charts for each price point; we track four groups:

• Fastest iOS device
• Fastest Android
• Mid-range Android ($300-350)
• Low-end Android ($100-150)

The last two cohorts account for more than 2/3 of new device sales:

The performance of JavaScript-based web experiences is heavily correlated with single-core speed, meaning that the P75 device places a hard cap on the amount of JavaScript that is reasonable for any website to rely on.

Depressingly, budget device CPUs have not meaningfully improved since 2022. But the nearly-identical SoCs in each year’s device are getting cheaper. Reduced bill-of-materials costs mean declining retail prices for low-spec phones.

Meanwhile, the high end continues to pull away. As previewed in prior instalments of this series, top-end Androids are beginning to close the massive performance lead that Apple’s A-series chips have opened up over the past decade. This is largely thanks to Qualcomm and MediaTek finally starting to address the cache-gap I have harped on since 2016:

Some cache sizings are estimates, particularly in the Android ecosystem, where vendor documentation is lacking. Android SoC vendors have a habit of implementing the smallest values ARM allows for a licensed core design.

The latest iPhone chip (A19 Pro) features truly astonishing amounts of cache. For a sense of scale, the roughly 50 MiB of L1, L2, and L3 cache in an iPhone 17 Pro provides 8.3 MiB of cache per core. This is more than double the per core cache of Intel’s latest high-end desktop part, the 285K, which provides a comparatively skimpy 3.3 MiB combined cache per core.

The gobsmacking caches of A-series parts allow Apple to keep the beast fed, leading to fewer stalls and more efficient use of CPU time. This advantage redounds to better battery life because well-fed CPUs can retire work faster, returning to lower-power states sooner.

That it has taken this long for even the top end of the Android ecosystem to begin to respond is a scandal.

Single-Core performance per $

Source: GSMArena, Geekbench, and vendor documentation. Geekbench 6 points per dollar at each price point over time. Prices are MSRP at date of launch.

If there’s good news for buyers of low-end devices, it’s that performance per dollar continues to improve across the board.

Frustratingly, though, low-end single core performance is still 9x slower than contemporary iPhones, and mid-tier devices remain more than 3.5x slower:

Multicore performance tells a similar tale. As a reminder, this metric is less correlated with web application performance than single-core throughput:

Geekbench 6 Multi-Core scores

Source: GSMArena, Geekbench. When Geekbench reports multiple scores for an SoC, recent non-outlier values are selected.

Performance per dollar looks compelling for the lower tier parts, but recall that they are 1/3 to 1/5 the performance:

Market Trends

What makes iPhones so bloody fast, while Androids have languished? Several related factors in Apple’s chip design and development strategy provided a commanding lead over the past decade:

In-house tuning of all ARM-designed cores, thanks to a long-ago negotiated Architecture licence, leading to aggressive cache sizings.

Concentration in fewer SoC SKUs enabled focused yield optimisation.

Large, early orders with TSMC secured exclusive access to the latest fabrication nodes.

Android vendors, meanwhile, have spread their SoC development budgets in penny-wise, pound-foolish fashion. Even Google and Samsung’s in-house efforts have failed to replicate the virtuous effects of Apple’s disciplined CPU designs.

Feature sizes are a fudge below 10 nanometres, but marketing names usually reflect real increases in transistor density and frequencies, along with reductions in power use. High-end Android and iOS parts have generally been produced on comparable nodes, with Apple’s lead lasting less than a year. But that’s less than half the story. Android SoC vendors have avoided adding competitively sized caches, dedicating the same mm^2 on die to higher core counts and on-die radios. From a performance perspective, this has been catastrophic:

Core counts are a headline fixture of device marketing, but even the cheapest phones have featured eight (slow, memory-starved) cores since 2019. Speed comes from other properties; namely appropriate cache sizing, memory latency, frequency scaling, and chip architecture. Apple’s core-count restraint and focus on other aspects should have been a lesson to the entire industry long before 2024.

Smaller transistors also allow for higher peak frequencies, giving Apple a perennial advantage thanks to early access to TSMC’s latest, smallest, power-sipping processes:

Maximum CPU Frequency (gHZ)

Source: GSMArena, Wikipedia, and vendor documentation. Maximum advertised frequency of the fastest on-package core.

These trade-offs have allowed Apple to charge a premium for devices which no other vendor can justify:

Global ASPs are conservative estimates; some research groups estimate values 10-15% higher, but the trends are consistent. The premium end continues to pull away in both price and performance, dragging ASPs slightly upward. Meanwhile, high-end devices continue to be outsold more than 3:1 by low-end phones that aren’t getting faster. Those slow devices are, however, getting increasingly inexpensive, dropping below $100 in the last two years. That’s 1/10th the price of the cheapest iPhone with Apple’s fastest chip.

Rise of the Refurbs

Thanks to market trends, the recent-spec iPhones many web developers carry don’t even represent the experience of most iOS users.

It may seem incongruous that the ASP of iOS devices is bumping up against the $1K mark while most developed countries experience bouts of slow growth post-pandemic, along with well-documented cost-of-living crises among middle-class buyers.

One possible solution to this riddle is that Android sales remain strong, as Apple is cordoned into the segment of the market it can justify to shareholders with a 30% net margin. Another is growth in the resale market, particularly for iOS devices. Thanks to premium components and better-than-Android software support lifetimes, the longevity of iPhones has created a vibrant and growing market below the $400 price point.

This also helps to explain the flat-to-slightly-declining market for new smartphones, as refurbished devices accelerate past $40BN/yr in sales.

What does this mean? We should expect, and see, higher-than-first-sale volumes of iOS use in various aggregate statistics. Wealth effects have historically explained much of this, but the scale is growing. Refurbishment and resale are now likely to be driving growing discontinuity in the data:

Wikimedia reports more than 40% of mobile site visits come from iOS devices over the past decade, despite global sales ratios never breeching 20% annually.

Not only is an iPhone 17 Pro not real life in the overall market, it isn’t even real life in the iOS market any more.

Desktop

The situation on desktop is one of overwhelming stasis, modulo the Windows 11 upgrade cycle resulting from Windows 10’s EOL at the end of 2025. That has driven an unusually strong one-off cycle of upgrades that will reverberate through the data in coming years.

IDC’s recent sales analysis shows that only 20% of ‘desktop’ devices are fully wired, with performance of the vast majority subject to power and thermal limits arising from battery power.

This impetus to upgrade is cross-pressured by pricing headwinds. Economic uncertainty, tariffs, scrambled component pricing from AI demand for silicon of all sorts, and ever-longer device replacement cycles all mean that new PCs may not provide more than incremental performance gains. As a result, an increase in recent worldwide PC Average Selling Prices from ~$650 in our previous estimate to ~$750 in 2025 (per IDC) may not indicate premiumisation. In a globally connected economy, inflation comes for us all.

Overall, the composition and trajectory of the desktop market remains stable. Despite 2025’s device replacement boomlet, IDC predicts stasis in “personal computing device” volumes, with growth bumping along at ±1-2% a year for the next 5 years, and landing just about where things are today. The now-stable baseline of ~410MM devices per year is predicted to be entirely flat into 2030.

Top-line things to remember about desktops are:

80% of “desktop” devices are laptops, tablets, and other battery-powered form-factors. The performance impact of thermal and power envelopes for these devices is drastic; many cores are spun down to low power states most of the time; symmetric multiprocessing is now a datacentre curio.

Flaky Wi-Fi, rather than wired Ethernet, is now the last mile to most desktops.

Desktops (including laptops) are only 15-18% of web-capable device sales; a pattern that has been stable for a decade.

Put another way: if you spend a majority of your time in front of a computer looking at a screen that’s larger than 10", you live in a privilege bubble.

Evidence From the Edge User Base

Per previous instalments, we can use Edge’s population-level data to understand the evolution of the ecosystem. As of late 2025, the rough breakdown looks like:

Device Tier	Fleet %	Definition
Low-end	30%	Either: <= 4 cores, or <= 4GB RAM
Medium	60%	HDD (not SSD), or 4-16 GB RAM, or 4-8 cores
High	10%	SSD + > 8 cores + > 16GB RAM

Compared with the data from early 2024, we see some important shifts:

Low-end devices have fallen from ~45% to ~30% of the fleet.

Most growth is in the middle tier, growing from 48% to 60%.

The high-end is growing slowly.

Because the user base is also growing, it’s worth mentioning that the apparent drop in the low-end is a relative change. In absolute terms, the ecosystem is seeing a slower absolute removal of low-spec machines. This matches what we should intuitively expect from incremental growth of the Edge user base, which is heavily skewed to Windows 11.

Older and slower devices likely constitute an even larger fraction of the total market, but may be invisible to us. Indeed, computers with spinning rust HDDs, <= 4GB of RAM, and <= 2 cores still represent millions of active devices in our data. Alarmingly, they have dropped by les