It is unlikely that by the end of 2017 at least one device will appear on the market that will manage to bypass the iPhone 8, iPhone 8 Plus and iPhone X in terms of speed.

We decided to figure out how Apple is able to do such productive processors in this way, although it simply does not have its own capacities for this.

A11 - the most productive chip in the mobile market

The A11 processor is built on two large productive cores and four small energy-efficient ones. Its main difference from the A10 is the presence of a special chip for working with neural networks.

This processor is installed in the iPhone 8, iPhone 8 Plus and iPhone X, the three fastest smartphones to date. Its performance is almost 2 times higher than its flagship competitors.

The fastest in the multi-core Geekbench 4 test:

• Apple iPhone 8 Plus - 10,472
• Apple iPhone 8 - 10,170
• Apple iPhone X - 10,051
• Samsung Galaxy Note 8 - 6,564
• One Plus 5 - 6.542
• Samsung Galaxy S8 - 6,295

Each A11 processor is manufactured using a 10-nanometer process technology that guarantees not only maximum speed, but also maximum energy efficiency.

The Taiwanese company TSMC, which works under the contract, is responsible for the production of A11 chips. And Apple itself does not have the necessary capacity for mass production processors.

iPhone X is even more powerful than some MacBook Pros

the iPhone 8, iPhone 8 Plus, and iPhone X are the first mobile devices to break the 10,000 mark in the multi-core Geekbench 4 test.

And to make these numbers even more grandiose, we note that the 13-inch 2017 MacBook Pro with a dual-core Intel Core i5 Kaby Lake processor is gaining about 9 thousand points.

In theory A11 is faster than Intel Core i5 Kaby Lake on 10%

Of course, in practice, the mobile processor installed in smartphones is unlikely to be able to outperform a full-fledged one, because they initially have completely different tasks.

A11 is sharpened not only for speed, but also for energy efficiency. And Intel’s solution, on the contrary, is primarily for performance. But the test results are still indicative.

Why none of the rivals can catch up with Apple

Since Taiwanese TSMC is a contract manufacturer, it does not directly develop processors for Apple. She is only responsible for mass production.

It turns out that Apple actually brings its partner something like ready-made “clichés”, on the basis of which he includes his machines in the work. There is simply no time to merge technology.

A similar situation was with Samsung, which released processors for Apple smartphones a year ago. It turned out for a competitor, she made chips better than for herself.

Of course, in addition to being able to quickly copy technology for yourself or for sale, the work of engineers who receive money at Apple for good reason is very important.

But the work of engineers would not have made any sense without software, so I want to shake hands with developers who work for Apple.

On the net today, there are thoughts that Apple has turned software into hardware. Its devices work quickly thanks to not only hardware, but also software.

Yes, iOS 11 doesn’t work so well, and even with Jobs it wasn’t exactly like that. But otherwise, the joint production of software and hardware clearly makes themselves felt.

It’s also very important that Apple develops processors itself, and does not buy them from market giants like Qualcomm. Competitors are trying to do something different, but they cannot count on Snapdragon.

Why aren't most competitors releasing processors on their own or following Apple's contractual path? There is nothing strange here.

And in this case, we are returning to the notorious fragmentation of Android devices. There are too many of them, and if almost every flagship did not have a Snapdragon 835, it would be very tight.

iPad Pro 2018 will turn out to be performance monsters

The first data on processors that Apple will use in the iPad Pro in 2018 has already begun to appear on the network. And this is a plague, comrades. It seems that the company suffered a core.

Most likely, the new generation iPad Pro will have the A11x Bionic processor installed. He will get as many as eight cores instead of six. And another 7 nm process technology and "water cooling".

The main question 2018 - how to calm down the power of the new iPad Pro

Sources close to TSMC, which will remain responsible for the mass production of processors under a contractual scheme, specify that three cores will be fast and five economical.

Most likely, along with ultra-fast chips, the iPad Pro 2018 will also receive the TrueDepth front camera module to be more trendy and generally recognizable.

What are the prospects for new Apple mobile devices

Of course, the higher the performance of the processor installed in a modern mobile device, the better it seems - especially considering the benchmark for energy efficiency.

Nevertheless, today it seems to me that new capacities do not force developers to make better software - on the contrary.

New nonsense - the faster the processors, the less attention to optimization

And the most striking example in this case is augmented reality on a special Apple engine, which appeared with the release of iOS 11. It works on the iPhone 6s and above. But tupit on iPhone 7.

It seems that the developers are already just broke to work on optimization. The chips of the latest iPhone pull, well, the rules. I still hope for positive developments in iOS software in the future. All.

Why is iPhone 7 faster than Samsung Galaxy S7 and iPhone 8 faster than Galaxy S8? This is due to the different ideology of operating systems, and in addition, one of the main advantages of Apple has been and remains unique systems on a chip. The A10 and A11 processors noticeably outperform similar offers from Qualcomm in the face of Snapdragon 820/821 and Snapdragon 835, respectively. Why is this happening? What is the "magic of Apple"? Having left behind the arguments in the style of “Android is better!”, Let's try to figure out the reasons that led to the superiority of Apple's mobile processors over Qualcomm.

The first factor: it happened

Recall the year 2013. Qualcomm's arsenal includes very successful Snapdragon 800 chips based on 32-bit Krait 400 cores of its own design. On this chip (and its follower, Snapdragon 801) dozens, if not hundreds, of a wide variety of models have been released. At the time of the announcement, the top Qualcomm chipset simply had no alternatives: based on the ARM Cortex A15 cores, the solutions were voracious to the extreme and could not compete with the four custom Krait cores. Everything seems to be fine, Qualcomm is the king of the mountain, it’s enough to continue to develop successful architecture. It would seem that what could go wrong?

But - in order. In 2011, ARM Holdings announced the architecture of ARMv8, the use of which opened up numerous opportunities for accelerating some of the special types of computing - for example, stream encryption, which (running ahead) is used today in almost all smartphones. The first mobile cores of this architecture were Cortex A53 and A57, announced by ARM in 2012. At the same time, ARM predicted the release of finished processors on new cores only for 2014. That's just Apple, the owners of the architectural license ARM, managed to be the first - almost a year ahead of the competition.

So, in November 2013, Apple releases the iPhone 5s. In addition to the fingerprint sensor and Secure Enclave integrated security system, the new iPhone is for the first time on the market equipped with a 64-bit Apple A7 ARMv8 processor. The new processor shows performance wonders at Geekbench: the result of a dual-core processor in single-threaded computing is one and a half times higher than the results of Krait 400 cores, in multi-threaded parity is observed.

An extended set of ARMv8 commands came in handy: it was in the iPhone 5s that Apple built in the Secure Enclave hardware security system, which is also responsible for data encryption. From the point of view of Apple, the choice of 64-bit architecture was quite logical: only in kernels with support for ARMv8 did instructions appear to accelerate stream encryption, which Apple had been using for quite some time at that time. Later, the use of new cores allowed Apple to achieve unprecedented speeds of access to encrypted data - the Nexus 6, released a year later, based on 32-bit Qualcomm Snapdragon 805 (ARMv7), showed terrifying crypto streaming performance: access to encrypted data was 3-5 times slower than unencrypted.

At first, 64-bit architecture in smartphones was perceived by ordinary people - and by many experts - as pure marketing. This is what users thought, and Qualcomm executives said so - at least in their official speeches.

In 2014, the iPhone 6 comes out, equipped with an A8 processor, also working with the ARMv8 command system. What does Qualcomm say? A small update: smartphones running on the Snapdragon 801 (32 bits, ARMv7) dominate the market. The Snapdragon 805 also comes out, using the same Krait 400 cores, but with a more powerful GPU. Apple processors are faster than Qualcomm counterparts in both single-threaded and multi-threaded computing, and in specific applications - for example, in the implementation of stream encryption - bypass the decisions of competitors simply at times. Qualcomm strenuously pretends that nothing unusual is happening, but manufacturers, stepping on the throat, require a competitive SoC. Qualcomm has no choice but to enter the race.

In 2015, Apple launches the iPhone 6s and A8, Qualcomm - the Snapdragon 810 chip and its stripped-down version of the Snapdragon 808. These processors were Qualcomm's response to the requirements of partners. However, the lack of experience in developing 64-bit chips played a cruel joke with the company: both processors were extremely unsuccessful. From the very first days, processors showed a tendency to excessive power consumption, overheating and throttling, as a result of which their established performance after several minutes of operation did not differ much from the performance of Snapdragon 801.

Which of all this can be concluded? There is only one conclusion: Apple took the industry by surprise, using cores with a new architecture then and where, it would seem, there is no need for this. As a result, Qualcomm turned out to be catching up, and Apple got a head start in a year and a half. Why did it happen?

Here you need to consider the features of the development cycle of mobile processors.

Factor Two: Differences in Development Cycles

So, we found out that Apple managed to get ahead, a year and a half ahead of the competition. How could this happen? The reason is the difference in development cycles between Apple and the manufacturers of smartphones running Android.

As you know, Apple has full control over the development and production of the iPhone, starting from the lowest level - processor design. And until recently Apple licensed graphic cores from Imagination Technologies, then the company preferred to develop processor cores on its own.

What does Apple's development cycle look like? Based on the ARM architectural license, a processor is designed that is compatible with a given instruction system (ARMv8). At the same time, a smartphone is being developed in which this processor will be used. In parallel, all the necessary drivers, OS are created for it, optimization is performed. Everything happens within one company; OS developers have no problem getting access to the driver source codes, and driver developers, in turn, have the opportunity to communicate with people who designed the processor.

The production cycle of devices on Android looks completely different.

First of all, ARM, the developer of the systems of the same name teams and processor architectures, comes into play. It is ARM that designs reference processor cores. So, back in 2012, the ARM Cortex A53 cores were announced, on which the vast majority of smartphones released in 2015, 2016 and 2017 are based.

Wait a minute! 2012? Exactly so: the 64-bit A53 cores were announced in October 2012. But the core architecture is one thing, and the real processors are another: ARM Holdings simply does not release them, offering reference designs to partners, but not putting SoC on the market. Before a smartphone based on this or that architecture appears on the market, someone must develop and release a finished system on a chip, SoC.

Despite the public statements of their representatives, in 2013 Qualcomm worked hard on the release of a 64-bit processor. There was no time left to develop its own core; I had to take what they give. They gave - the big.LITTLE architecture, which at that time included the "small" Cortex A53 cores (successful) and the "large" A57 cores (quite controversial in terms of energy efficiency and throttling).

The first Qualcomm processors based on these cores were announced in 2014. But the processor is not all! At a minimum, you also need a body, a screen ... All this is produced by OEMs, which, in fact, are engaged in the development and production of smartphones. And this is also time, and considerable time.

Finally, the operating system. In order to “get” Android on the device, you need a set of drivers for the new chipset. Drivers are developed by the chipset developer (for example, Qualcomm), providing them to smartphone manufacturers for integration. It takes a certain amount of time for the manufacturer to figure out and integrate the drivers.

But this is not the end! An already finished smartphone with a working version of Android must also be certified in one of the laboratories of Google for compatibility and compliance with Android Compatibility Definition. This is also time, which is already catastrophically small.

In other words, the fact that we saw smartphones on the Snapdragon 808/810 only in 2015, there is absolutely nothing surprising. Qualcomm's first flagship chips based on 64-bit architecture lagged behind Apple's SoC by a year and a half. This is a historical fact, and it is a real advantage for Apple.

In 2015, the long development cycle and the requirements of partners played a trick on Qualcomm: the first pancake turned out to be lumpy. However, the company managed to rehabilitate with the release of the Snapdragon 820. But was it too late?

Third factor: size issue

Consider a table comparing the two latest generations of Apple and Qualcomm processors.

What do we see from this table? It is easy to notice that the performance per one core in Apple processors is more than two times higher than Qualcomm solutions, and the multi-threaded performance of current generations of processors differs almost one and a half times. Why is this so? You can try to find the answer in the following plate.

If we discard a pair of A10 Fusion / Snapdragon 820 processors that use different technological processes, we can compare the area of \u200b\u200bthe A11 Bionic and Snapdragon 835 chips. Apple's chip surface area is 1.2 times that of Qualcomm. What does this mean in practice? Ability to use more transistors, more advanced core architecture. In particular, the researchers found that in A11 Bionic, “weak” processor cores are several times larger than the small A53 cores (sorry - Kryo 280) used in Snapdragon 835. This means that even the “small” A11 Bionic cores support extraordinary execution of instructions, which allows you to get more performance per cycle compared to straight-line A53 cores.

The area of \u200b\u200bthe processor directly affects its price. The larger the area (when using one process technology), the higher the cost. Which brings us to another factor: processor costs for the manufacturer.

Fourth factor: price issue

According to the Android Authority report, the area of \u200b\u200bthe Apple A10 Fusion processor cores is twice the size of the cores of its closest competitor, Snapdragon 820.

“Apple’s advantage is that the company can afford to spend money on increasing the area of \u200b\u200bthe processor built using the latest 16-nanometer FinFET technology ... A few extra dollars will not play a big role in the final cost of the device - and after all, Apple will be able to sell significantly more than 600- dollar devices thanks to such great performance, ”writes Linley Gwennap, director of The Linley Group.

Indeed, the extra five or six dollars will not play a big role in the final cost of the iPhone - these are shares, in the worst case, a few percent of its cost to the consumer. But if these five or six dollars are able to double the performance of the device compared to competitors on Android - this is a great argument in favor of Apple.

Why doesn't Qualcomm do this? There are too many stakeholders in the processor development chain for Android devices. This is ARM, which develops and licenses processor cores, and Qualcomm, which designs ready-made processors under license, and manufacturers of Android smartphones. OEMs, forced to compete with each other in prices, have every dollar in their accounts. Manufacturers want the cheapest SoC possible (therefore, by the way, solutions based on the archaic weak A53 cores are still so popular), and Qualcomm has to reckon with this. But both Qualcomm and ARM want to bite off a piece of the pie, having received their share of the profit - so that the cost of a solution similar to Apple processors would come out even higher than that of Apple. As a result, OEMs would not be able to afford bulk purchases of such processors, which would further increase their cost. (By the way, this is exactly what happened with the MTK Helio X30 processor - it was not in demand, and on its basis only two smartphones were released.)

Of course, here you can argue that Samsung and Huawei have their own line of processors - Exynos and Kirin, respectively. But Huawei does not have its own developments, the company takes ready-made ARM Cortex cores and ready-made ARM Mali graphics accelerators, assembling “proprietary” processors based on them. It is clear that the computing cores of these processors cannot be more powerful than those offered by ARM. Samsung, on the other hand, is trying to follow the path of Apple, releasing its own customized cores - whose performance, however, is not far from the usual "stock" ARM cores.

Fifth factor: control issue

Last year, Apple did an interesting thing: a strong-willed decision removed support for 32-bit applications from iOS 11. It just so happened that it was on this version of the OS that the new iPhone line was released: 8, 8 Plus and X. What does this mean in terms of performance ?

The ability to take and refuse support for 32-bit commands gives a lot, a lot. Decoding and execution blocks are simplified, the required number of transistors is reduced. Where does this saving go? It can be spent on reducing the area of \u200b\u200bthe processor (which directly translates into reduced cost and reduced power consumption), or you can add transistors to other units with a constant area and power consumption, thereby increasing performance. Most likely, events developed according to the second scenario and the A11 Bionic processor received an additional 10-15% of the performance precisely due to the rejection of support for 32-bit code.

Is this possible in the Android world? Yes, perhaps, but not completely and very soon. Only in August 2019, requirements for developers come into force, which will have to include 64-bit versions of binary libraries when adding or updating applications in the Google Play Store. (We note here that far from all - and not even the majority! - Android applications generally use some kind of binary library, often content with dynamically translated bytecode.) Recall that Apple introduced a similar requirement in February 2015 - again an advantage in time, this time in four and a half years.

Sixth factor: optimization and use of available resources

Optimization is an essential part of performance. Traditionally, Apple had everything with optimization either perfect or exemplary (users who complain about the declining performance of older devices upgraded to the latest version of iOS just don’t understand what the hell would be on such weak hardware if they were running Android). But Android with optimization is all ... colorful. Variously. We can say - enchanting.

Most often, clean Android assemblies work quite quickly on fresh hardware - such as those used in Google Nexus and Pixel smartphones, Motorola and Nokia devices. But even here, everything is not good: for example, the Google smartphone (Motorola) Nexus 6 had absolutely stunning imagination problems with the speed of access to the drive that arose due to the illiterate implementation of encryption (Google developers could not cope with the hardware accelerator of cryptographic operations of the Snapdragon 805 processor, after which they stated that “software implementation is better”). Here in this article we have analyzed in detail the speed of reading and writing encrypted data with the Nexus 6 smartphone, comparing it with the speed of similar operations in the iPhone 5s. Here are the numbers:

• Nexus 6, sequential read, unencrypted data: 131.65 MB / s;
• Nexus 6, sequential read, encrypted data: 25.17 MB / s (39 MB / s in the upgrade to Android 7);
• iPhone 5s sequential read, encrypted data: 183 MB / s.

Impressive? With similar hardware characteristics, the Google developers (Google, not Krivorukov OEM!) Managed in a reference device, which was supposed to promote secure encryption to the masses, to make such a mistake. Will you be surprised to learn that other manufacturers with optimization may have problems? And they arise. So, equipped to the maximum HTC U Ultra (Snapdragon 821) manages to slow down and overheat during the most routine operations; it seems that the processor performs at least twice as many calculations as it should. Well, about Samsung smartphones that manage to slow down the little things even on the most powerful hardware available, it’s not even worth talking in detail.

Seventh: Screen Resolution

There is one more point worth mentioning. This is the resolution of the display. As you know, standard iPhone models are equipped with HD screens, Plus models - Full HD. Manufacturers of Android smartphones using Qualcomm's flagship chipsets are trying to install screens with a QHD resolution of 2560 × 1440. Well, at the very least - Full HD, but this is, unfortunately, not common in flagship smartphones.

Why alas? Because the resolutions are higher than Full HD on screens with IPS-matrix diagonal up to 5.7 ″ inclusive, more than enough. For AMOLED screens, which, firstly, have the PenTile subpixel structure, and secondly, can have Google VR support for virtual reality glasses (by the way, what percentage of users did it really come in handy for?), The QHD resolution can still be justified then argue.

A bit aloof is the iPhone X with a resolution of 2436 × 1125 - however, this, in fact, differs little from Full HD. For comparison: the screen resolution of the Samsung Galaxy S8 is 2960 × 1440, that is, one and a half times more pixels than in the iPhone X.

Now imagine that we are comparing the performance of the iPhone 8 with its HD resolution and some Nokia 8 with QHD. Presented? Nokia has to process almost four times as many pixels as the iPhone, which cannot but affect the power consumption and performance (at least in those tests that use screen output). I now do not in any way justify the old screens that Apple continues to install with manic persistence in devices worth a thousand dollars, but just focus on the fact that the performance and energy efficiency of devices with low-resolution screens, even other things being equal, will be higher than that of smartphones with QHD screens.

Something is suspected and manufacturers. So, Sony Xperia Z5 Premium, whose screen (by the way, IPS, which is useless for VR purposes) has a physical resolution of 4K (in fact, no, even here marketers cheated), but the logical one is “only” Full HD, which allowed the manufacturer and the consumer cheat, and not kill the productivity too much. Samsung did likewise, allowing it to use lower logical resolution on screens with a high pixel density. Obviously, the interests of marketers are contrary to the interests of both users of these devices and their own developers.

Instead of a conclusion: do our phone need 64 bits?

Are 64-bit processors needed on mobile devices? Indeed, 32-bit computing cores have their own advantages. Such processors can run faster than 64-bit due to the shorter instruction length due to the shorter address length, and as a result, they are less demanding on the amount of RAM; they can implement a shorter queue of commands, which can also give a performance gain in certain scenarios.

Some of these advantages will remain theoretical, but in a number of modern usage scenarios, support for ARMv8 commands is no longer possible. This streaming encryption, and gluing HDR in real time, and many other inconspicuous things. Be that as it may, processor manufacturers switched to 64-bit kernels with support for ARMv8, and this is a fait accompli.

But smartphone manufacturers are in no hurry to switch to 64-bit builds of operating systems.

So, in nature, there is not a single smartphone running Windows 10 Mobile, in which the operating system would work in 64-bit mode. Both the Lumia 950 (Snapdragon 808), and the Lumia 950 XL (Snapdragon 810), and even the relatively fresh Alcatel Idol 4 Pro (Snapdragon 820) are running a 32-bit build of Windows 10 Mobile.

Android phone manufacturers are not far behind. For example, Lenovo, which manufactures smartphones under the Motorola brand, has only two devices with the “correct” 64-bit Android: these are the flagships of the Moto Z line (the regular version and version of Force) and the Moto Z2 Force. All other devices - both the budget Moto G5 on the Snapdragon 430, and the fresh sub-flagship Moto Z2 Play on the Snapdragon 626 - work in 32-bit mode.

A number of devices from other manufacturers (for example, the BQ Aquaris X5 Plus) use the powerful Snapdragon 652 in 32-bit mode. Needless to say, such devices do not squeeze the maximum out of the available hardware capabilities?

On the other hand, not everything is perfect with Apple. Even 64-bit applications compiled into native code, due to backward compatibility requirements, are forced to limit themselves to the instruction set available in the company's earliest processors - the 2013 Apple A7. But the ART byte-code compiler, which has been used in Android since version 5, has no such problems: application byte-code is compiled into optimized native code, using all the instructions available on the current hardware.

However, we will live with what we have. For maximum processor core performance and guaranteed optimization - to Apple. The same thing, only one and a half to two times worse and as much as cheaper - to a host of handset manufacturers on Android.

1. All declared battery specifications depend on network settings and other factors; Actual operating time may not be as specified. The battery allows a limited number of charge cycles. After some time, the battery may need to be replaced at an Apple Authorized Service Center. Battery life and number of charge cycles depend on the settings and usage of the device. More on pages and.
2. the iPhone 8, iPhone 8 Plus, iPhone XR, iPhone 11 Pro, iPhone 11 Pro Max and iPhone 11 are resistant to splashes, water and dust and are tested in specially supported laboratory conditions. iPhone 11 Pro and iPhone 11 Pro Max have an IP68 rating of IEC 60529 (immersion in water to a depth of 4 meters for up to 30 minutes is allowed); iPhone 11 has an IP68 rating of IEC 60529 (immersion in water to a depth of 2 meters for up to 30 minutes is allowed). iPhone 8, iPhone 8 Plus and iPhone XR are rated IP67 according to IEC 60529 (immersion in water to a depth of 1 meter for up to 30 minutes is allowed). Resistance to splashing, water, and dust may decrease with normal wear. Do not attempt to charge a wet iPhone: Wipe and dry it according to the instructions in the user guide. Liquid damage is not covered by the warranty.
3. The display is a rounded rectangle. Diagonal of this rectangle without taking into account rounding is 5.85 inches (for iPhone 11 Pro), 6.46 inches (for iPhone 11 Pro Max) or 6.06 inches (for iPhone 11, iPhone XR). Actual viewing area is smaller.
4. The subscription price is 199 rubles per month at the end of the trial period. Single subscription to a Family Access group. The offer is valid for 3 months after activation of the corresponding device. Subscription renews automatically until canceled. There are certain restrictions and other conditions.
5. Subscription fee is 199 rubles per month at the end of the trial period. Subscription renews automatically until canceled. You can cancel your subscription at any time.
6. The cost of an individual subscription is 169 rubles. per month at the end of the trial period. You can unsubscribe at any time. At the end of the trial period, the subscription renews automatically until canceled.

• Symbols of the NHL and NHL Teams are the property of NHL and their respective teams. All rights reserved.
• Officially licensed United States Women’s National Team Players Association © 2019.
• Officially licensed NFL Players Inc. products © 2019

After a series of lawsuits with Samsung, Apple finally took the long-awaited step and decided to reduce its dependence on the Korean manufacturer in the printing of chips. After the iFixit team disassembled the new Apple smartphones, the same work and the Chipworks team, trying to identify the manufacturers of various components of the iPhone 6.

The publication itself is quite interesting and replete with photographs of the mass of components, but the most interesting conclusions about the A8 processor. Experts believe that the manufacturer is no longer Samsung. Employees of Chipworks believe that the Taiwanese TSMC most likely acts as the new contract manufacturer, but they do not undertake to claim this 100%. However, the shutter sizes resemble those of the Qualcomm MDM9235 chip, which is also manufactured in compliance with the 20-nm TSMC standards.

The A8 chip itself includes 2 billion transistors (twice as much as A7), but at the same time its area is 89.25 mm 2 (8.5 × 10.5 mm), that is, it is 13% smaller than A7 (102 mm 2) . Apple claims that the CPU is 25% more powerful and the graphics 50% more powerful. At the same time, the chip in CPU tasks is 50 times more productive than the single-chip system used in the original iPhone, and 84 times - in graphic tasks. Finally, according to Apple, the chip is 50% more energy efficient compared to the A7, which should have a beneficial effect on battery life.

The Chipworks team also confirmed that the A8 single-chip system is supplemented with only 1 GB of RAM, while many modern flagship Android devices use 3 GB of RAM. Judging by the code on a particular chip that fell into Chipworks hands, it was packed only six weeks ago (probably in Taiwan), after which it managed to go through Foxconn production lines in China and appear on the shelf of an American store in Ottawa.

After Apple used a self-made A4 processor in the first iPad, rumors spread that in the future, the company could abandon Intel processors in Macs and switch to ARM architecture. This has its advantages, however, such a migration is fraught with many consequences that Apple will need to overcome. Is the game worth the candle?

Why modern Macs work on Intel processors

Starting in 2006, all new Apple computers run on x86 architecture processors in conjunction with Nvidia or AMD GPUs (or Intel's integrated graphics core in entry-level models). Thanks to Open GL, Mac software can interact with GPUs of various architectures, which allows Apple to easily change the graphics provider.

After switching to Intel processors, Apple released two more major OS X releases, which implemented support for the old (PowerPC) and new (x86) architectures, but Snow Leopard released in 2009 only worked with Intel processors.

How does switching to ARM differ from switching to Intel with PowerPC

In the period from 1994 to 2005, all software for Mac OS was “tailored” exclusively to work on PowerPC processors, the architecture of which was radically different from x86. Even earlier, over the previous ten years, Macs were running Motorola processors called 68k (68000, 68020, 60030 and 68040).

The first change of architecture was caused by the desire to switch to more modern and productive processors with support for 64-bit computing. With PowerPC performance far superior to 68k, it could easily emulate existing code.

Apple’s second move from PowerPC to Intel no longer looked like such a big step forward. PowerPC chip makers (IBM and Motorola / Freescale) have virtually abandoned the PC market, "playing occasional roles" in the niches of the automotive and gaming consoles. Apple was their last customer, but the company sold less than 4 million of its computers a year.

But on the Windows-PC market, life was in full swing, all computers used Intel x86 architecture or compatible analogs from AMD. Switching from PowerPC to Intel, Apple left a sinking ship and chose a developing ecosystem where, thanks to large volumes of production, innovations and technologies developed very quickly.

However, the affordable x86 architecture was essentially a step backward. After all, at that time all Intel processors were 32-bit, while the PowerPC, which Apple has used in its PowerMac G5 since 2003, supported 64-bit computing. Only in 2006, when Intel introduced the Core 2 line, Apple returned to 64-bit processors in their computers.

There were other shortcomings in the transition to Intel architecture, but they were covered by the high pace of development due to the large market. At that time, Intel processors were a little more powerful than PowerPC, but their performance was enough to emulate most of the code written for PowerPC. This was made possible thanks to Rosetta technology, which Apple bought and refined to smooth the difficulties of moving to a new platform.

In addition, the change of architecture to x86 meant the possibility of starting Windows (Linux and other x86 operating systems). This greatly expanded the potential audience, attracting users who needed to run specific Windows applications to purchase Macs. Boot Camp allowed Windows to install the second system on disk, and third-party applications made it possible to run Windows programs directly in OS X. Both methods were much faster than simply emulating Windows code on PowerPC, which was the only option available to Mac users before switching to Intel processors. .

What Apple might be interested in avoiding Intel processors

Cost savings

The main reason Apple could consider creating Macs without using Intel processors is the high price of the latter. Intel chips are too high-tech and complex enough to copy, so they are unrivaled and allow Intel to charge them such a high price.

It is difficult to determine the exact price Apple pays for Intel processors. Analysts from IHS iSuppli believe that the Intel Core i5 used in Microsoft Surface Pro is 4-5 times more expensive than ARM chips in Surface RT. A6 processors for the iPad, in their opinion, cost Apple $ 25 apiece, while the Intel chips used in Macs cost $ 180-300. The idea that Apple could replace Intel's 200-dollar chips with one or two 25-dollar chips and gave rise to rumors about the possibility of switching Apple computers to ARM architecture.

However, such a comparison is not entirely correct, because modern ARM processors are significantly inferior in performance to Intel Core i5 chips, even entry-level ones. There is a huge gap between the processing power of Intel processors and the fastest ARM processors, as Microsoft's experiment in porting Windows to the Surface RT ARM chip proved.

Apple is able to create more productive ARM-processors

Apple has been actively increasing the processing power of its Ax series processors, thanks to funding from economies of scale. Each year, the company sells about 70 million iPads and nearly 170 million iPhones.

This year, Apple could have created even more powerful A8 chips, if not for the restrictions imposed by the thickness of the case, the limited battery size and the problem of heat dissipation in iOS devices. The company made it clear that when designing the A8, the main priority was energy efficiency, which is so important for the iPad Air 2 (which has a smaller battery than its predecessor), to keep the device’s autonomy at the same level.

Mac mini and even MacBook Air are much less constrained by energy consumption and heat sink restrictions, which would allow Apple to increase the operating frequencies of the processors, the number of cores or add other hardware to them, supplying large amounts of memory and cache.

Given all these circumstances, Apple may even be interested in creating some kind of new specific Mac running on an ARM processor, which in performance will not be so far from budget desktop processors. Indeed, now ARM is already bypassing Intel x86 mobile chips.

Before switching to Intel processors, Apple produced about 4 million Macs per year. At the moment, Mac's annual production volume is almost 20 million, about the same amount of iPad the company sold in the first four quarters. Initially, Apple considered using Intel Atom chips in the iPad, but abandoned this idea in favor of ARM.

Creating proprietary silicon IC technology

Based on the fact that Apple uses the optimization of chips used in iOS-devices, we can assume that the company is also interested in optimizing processors for the Mac. It can remove unused logic sets and implement additional ones to implement encryption, audio processing or video decoding at the hardware level.

Using a single architecture in Macs and iOS devices can greatly simplify the use of hardware and software, as well as transferring APIs and other software between systems.

Moreover, when developing proprietary technologies used only in Ah processors, all Apple investments will remain inside the company and will bring profit exclusively to it. Now, by purchasing processors from Intel, Apple indirectly contributes to the development of the entire PC industry. Intel is creating new generations of processors that are accessible to everyone, and their development costs are reduced due to production volumes provided by Apple.

Considering Intel’s not very impressive successes in motivating PC manufacturers to create ultrabooks, Mac mini clones and Android Atom tablets, the loss of a client like Apple will have disastrous consequences not only for Intel, but also for everyone who uses x86-based processors.

What keeps Apple from switching to ARM

Apple took a step towards Intel for reasonable reasons. In 2006, she did not have a serious team for the development of chips, nor sufficient capital to develop her own technology for their creation. Intel has already done this work and buying a turnkey solution not only made sense, but was the best of the few options available to Apple at that time.

Despite the fact that now Apple is one of the leading manufacturers of mobile processors and has 150 billion dollars of capital, allowing to implement the most ambitious projects, the use of Intel chips still makes sense for several reasons.

Existing Intel Technologies and Capabilities

Today, Intel owns the world's leading processor technology and has amazing manufacturing capabilities to meet Apple’s needs. As a customer of Intel, Apple receives not only them, but also future developments of the chip manufacturer, in which he invests in order to remain the most advanced processor manufacturer in the world.

Large orders give Apple chip selection priorities, as well as discounts due to large volumes. The company's profit that it receives from each Mac sold is simply unattainable for PC manufacturers, even taking into account the considerable cost of Intel processors.

For Apple, there are no half measures that other manufacturers can go for; it selects only the most advanced technologies. The company buys the best LCD panels, uses the licensed Helvetica font. While Microsoft and Google use low-quality displays, copies of Helvetica, and also do not use fingerprint scanners in their products, because of their high cost.

AMD loss as a supplier

Leaving Intel, Apple may lose a potential supplier of x86 compatible AMD video chips.

Now the company buys GPUs from both AMD and Nvidia, choosing the best available solutions depending on new technologies and prices. Thanks to OpenGL, changing the video chip vendor is straightforward.

Apple did not play into AMD’s hands in their confrontation with Intel, but theoretically it could - if Intel made a mistake and AMD managed to create a more affordable and superior competitor processor capable of executing x86 code on Macs. The departure of Apple from Intel to ARM-processors will exclude even this theoretical possibility to change Intel chips to cheaper AMD.

Doubtful savings with a partial switch to ARM

Apple now will not be able to replace Intel-based processors with ARM in the entire Mac line, especially in the top-end families and modifications of the MacBook Pro and Mac Pro, and this is precisely the segment from which the company receives most of its profit and maintains community loyalty due to minimal competition.

If Apple releases only one new Mac model running on the ARM architecture, this will reduce its dependence on Intel, but also increase the cost of buying processors for x86-Macs by reducing the volume. Thus, a partial switch to ARM will not give Apple anything in terms of savings.

The fact of creating an ARM-Mac does not guarantee its popularity. Microsoft has already made an attempt to port Windows to ARM, but this did not bring a new audience. Two years were wasted, not counting the deterioration in relations with Intel. The processor giant responded by announcing support for Android and Meego / Tizen, spending billions of dollars subsidized by tablet makers to introduce Atom, which was aimed at the same goal as Microsoft with its Surface RT - a significant expansion of the market.

Of course, Microsoft was not going to save money and the main reason for using ARM was the desire to increase energy efficiency compared to Intel's desktop and mobile alternatives. But these wonderful beginnings were cut short by the harsh reality - existing Windows applications could not work on the ARM architecture.

Apple has extensive experience in porting software to new architectures. The company proved that it can simultaneously support various hardware platforms, but despite this, it always tried to complete such transitions quickly in order to bring everything to a single standard and avoid the problem of hardware fragmentation.

Big risks

To top it off, the development of ARM chips for the Mac can cause additional problems, such as complications and a slowdown in the development of mobile processors used in the iPhone, iPad and other new products.

Apple mobile device sales account for most of its revenue. Over the past year, the company has sold 244 million iOS devices and only 18.9 million Macs. The transition to the ARM architecture will inevitably cause a change in the development priorities of the mobile segment and theoretically can allow competitors to become leaders. It is unlikely that Apple has hundreds of free engineers sitting idle to spray the efforts of the ARM chip development team into two different directions.

Moving away from a key supplier, Apple can also embarrass existing customers and runs the risk of casting a shadow over its name. When Microsoft introduced the Surface RT, it lost the trust of customers because the “uncompromising Windows computer” was virtually unable to run Windows applications and had performance limitations for ARM processors. Potential buyers of ARM Macs will have even greater inquiries and expectations from the new Apple product.

Incompatibility with x86 architecture

Apple has extensive experience porting its own operating systems, frameworks, applications, and development tools to new architectures. The company transferred Mac OS from 68k to PowerPC, ported NeXT software from Intel to PowerPC, and iOS, in fact, is adapted for OS X mobile realities.

Apple definitely knows how to create an ARM version of OS X and, if necessary, can provide developers with tools that would help them rebuild their Mac applications on the ARM architecture, but this will require a lot of work and considerable effort from the developers themselves. The costs and expenses associated with creating application ports may not live up to expectations, especially if Apple sells less than 20 million Macs per year.

Apple TV Experience

Like Surface RT, Apple TV can be seen as a recent example of a change of architecture. The original version of Apple TV, sold from 2007 to 2009, was actually a truncated Mac with an Intel x86 processor and Nvidia graphics running on a modified version of OS X.

In 2010, Apple introduced the second generation of a set-top box, running iOS running its own A4 processor, which had integrated graphics. This transition, which entailed a complete redesign of the hardware architecture, allowed to reduce the price of the product from $ 299 to $ 99.

But Apple TV is a very specific example - the set-top box is released in relatively small volumes and does not bring much profit to the company, in addition, it does not have third-party applications, and therefore there are no problems with their adaptation. Her transition to iOS and ARM was a fairly simple task. With a price of $ 300, Apple TV simply had no chance in the market, but when it dropped to $ 99, the set-top box sold very well, bringing Apple about a billion dollars a year (including media content that boosts its sales). In 2010, Apple came up with a source of rejected A4 chips (and then A5), which were not suitable for the iPad, so Apple TV was the perfect candidate for the transition of the ARM architecture.

Do not wait for the ARM-MacBook in the near future

The issue of switching traditional Macs to ARM architecture is not whether Apple can replace Intel, but rather whether it will be commercially viable.

If Apple really decides to introduce an ultra-budget MacBook Air “netbook format”, it will be easier for it to abandon the expensive Core i5 chips and create an inexpensive product that runs on iOS or a stripped-down version of OS X. Such a MacBook would take a place next to Surface RT and “Chromebooks” from HP and Samsung, powered by Samsung's ARM chips.

However, at the moment there are too few convincing arguments that would prove that Apple is interested in selling laptops with low performance. Now a record number of Macs are on sale in the price range of $ 900-3000, in addition there is an iPad covering a more budgetary range from 200 to 800 dollars.

Despite the fact that last year iPad sales fell by 4%, you can’t say the form factor of the tablet is losing popularity and needs to be replaced. In fact, everything looks as if Apple made potential Mac buyers from iPad users, and this is already much greater success (and profit) than the motivation of Mac users to buy an iPad.

Nevertheless, the technology industry is constantly on the move and often familiar patterns are crumbling with new products that cost and do less than existing ones. The proof of this can be the iPhone, which was much less able than the existing smartphones at that time, as well as the iPad and Apple TV, lacking the functions that were in the tablets of the set-top boxes that preceded them. Apple simply cut off the "necessary" features and thereby created new, affordable and attractive product categories.

By creating a Mac based on an ARM processor, Apple can greatly discredit its own business of premium computers. Theoretically, the company could create an inexpensive MacBook, say, for education, but this is too small a market that is now saturated with Google Chromebooks.

In a year or two circumstances may change. It is possible that Apple will reach a point of development when the premium Mac business is difficult to expand further. During this time, the company can develop a technology that would allow it to create an ARM processor that is close in performance to Intel, but having a lower price. Apple could create hardware support for emulating x86 applications, thereby minimizing costs and accelerating the transition to ARM.

While Intel does not expect big breakthroughs in the development of x86-processors, so it may be more reasonable for Apple to invest in the development and development of its own modern ARM-chips (or even a completely new architecture) for desktop computers and laptops.

In general, it seems that the market for traditional computers and laptops has stopped in development. Apple is expanding its share among premium-class computers and it has every chance to develop this trend without making radical changes to Macs. The company can use its huge, but still limited resources for a more profitable investment than replacing Intel as a processor supplier for several million Macs. At least over the next few years.