What is an arm processor. Comparative analysis of microcontrollers with ARM core

Due to the ever-increasing number of applications that place increased demands on data processing performance, there is a trend towards increasing demand for 32-bit. microcontrollers. This conclusion was made by the marketing company Semico, which predicts a 2-fold predominance of the 32-bit market capacity. microcontrollers over 8 and 16-bit. in 2007 . In this regard, the purpose of this article is to present the general development trends of one of the most common 32-bit. ARM cores and give a comparative assessment of microcontrollers based on them from the most affordable manufacturers in the CIS markets.

Overview of the ARM architecture

The ARM microcontroller core was developed by the English company of the same name, organized in 1990. The name ARM comes from "Advanced RISC Machines". It should be noted that the company specializes exclusively in the development of microprocessor cores and peripheral units, while it does not have production facilities for the production of microcontrollers. ARM delivers its designs in electronic form, on the basis of which customers design their own microcontrollers. The company's clients are over 60 semiconductor manufacturing companies, among which are such popular manufacturers in the CIS semiconductor market as Altera, Analog Devices, Atmel, Cirrus Logic, Fujitsu, MagnaChip (Hynix), Intel, Motorola, National Semiconductor, Philips , ST Microelectronics and Texas Instruments.

Currently, the ARM architecture is in the lead and covers 75% of the 32-bit market. embedded RISC microprocessors. The prevalence of this core is explained by its standard nature, which allows the developer to more flexibly use both his own and third-party software developments, both when switching to a new ARM processor core, and when migrating between different types of ARM microcontrollers.

Currently, six major families have been developed (see Figure 1): ARM7™, ARM9™, ARM9E™, ARM10™, ARM11™ and SecurCore™. The XScale™ and StrongARM® families have also been developed with Intel.

As an addition to the ARM architecture, several extensions can be integrated:

Thumb® - 16-bit an instruction set that improves program memory efficiency;
DSP - a set of arithmetic instructions for digital signal processing;
Jazelle™ - extension for hardware direct execution of Java instructions;
Media - an extension for 2-4 times the speed of processing audio and video signals.

Figure 1. ARM processor cores

The record levels that the ARM architecture has crossed are the speed of over 1 GHz and the specific consumption of 1 μW / MHz. Depending on the purpose, ARM processors are divided into three groups (see Figure 2):

Processors for open platform operating systems in wireless communications, imaging, and consumer electronics applications.
Processors for embedded real-time operating systems for mass storage, industrial, automotive, and network applications.
Data protection system for smart cards and SIM cards.

0.18µm (0.13µm)
Core	Cache	Area, mm 2	Specific consumption mW/MHz	Frequency, MHz
ARM7TDMI	-	0,53 (0,26)	0.24 (0,06)	100 (133)
ARM7TDMI-S	-	0,62 (0,32)	0,39 (0,11)	80-100 (100-133)
SC100	-	0,50	0.21	80
SC200	-	0,70	0.30	110
ARM7EJ-S	-	1,25 (0,65)	0,45 (0,16)	80-100 (100-133)
ARM946E-S	8k + 8k	5,8 (3,25)	0,9 (0,45)	150-170 (180-210)
ARM966E-S	16k+16k TCM	4,0 (2,25)	0,65 (0,4)	180-200 (220-250)
ARM1026EJ-S	8k + 8k	7,5 (4,2)	1,15 (0,5)	190-210 (266-325)
ARM1136J(F)-S	16k/16k+ 16/16k TCM	- (8,2; 9,6)	1,30 (0,4)	250-270 (333-400)

Figure 2. Technical data for processor cores
ISE - in-circuit emulator, RT - real time, DSP - digital signal processor, SIMD - multiple data in one instruction, TCM - tightly coupled memory (cache), ETM - built-in trace macrocells, VIC - vectorized interrupt controller, ASB, AHB - types of internal tires

The promise of the ARM core becomes apparent after Atmel's revolutionary announcement at the ARM microcontroller developer conference held in Santa Clara (USA) in October 2004. The essence of the announcement was the intention of Atmel to release 32-bit. AT91SAM7S microcontrollers at the price of 8-bit, targeting 8-bit. applications to expand the functionality of information processing, while maintaining their competitive cost at the same level.

Thumb instruction set

32-bit ARM processors support previous 16-bit. development by supporting the Thumb instruction set. Using 16-bit instructions can save up to 35% memory compared to the equivalent 32-bit. code, while retaining all the benefits of 32-bit. system, for example, accessing memory with 32-bit. address space.

SIMD technology

SIMD (multiple data in one instruction) technology is used in media expansion and is aimed at increasing the speed of data processing in applications where low power consumption is required. SIMD extensions are optimized for a wide range of software, incl. audio / video codecs, where they allow you to increase the processing speed by 4 times.

DSP instruction set (DSP)

Many applications place high demands on the speed of real-time signal processing. Traditionally, in such situations, developers resort to the use of a digital signal processor (DSP), which increases the power consumption and cost of both the development itself and the end device. To eliminate these shortcomings, a number of ARM processors have integrated DSP instructions that execute 16-bit. and 32-bit. arithmetic operations.

Jazelle® Technology

The ARM Jazelle technology is targeted at applications that support the Java programming language. It offers a unique combination of high performance, low system cost, and low power requirements that cannot be achieved simultaneously using a coprocessor or a dedicated Java processor.

ARM Jazelle technology is an extension to 32-bit. A RISC architecture that allows an ARM processor to execute Java code in hardware. At the same time, unsurpassed performance of Java code execution using the ARM architecture is achieved. Thus, developers have the opportunity to freely implement Java applications, incl. operating systems and application code, on the same processor.

Jazelle technology is currently integrated into the following ARM processors: ARM1176JZ(F)-S, ARM1136J(F)-S, ARM1026EJ-S, ARM926EJ-S, and ARM7EJ-S.

Traditional ARM processors support 2 instruction sets: in ARM mode, 32-bit instructions, and in Thumb mode, the most popular instructions are compressed to 16-bit. format. The Jazelle technology expands on this concept by adding a third Java instruction set that is invoked in the new Java mode.

Intelligent Energy Management Technology

One of the main challenges faced by developers of portable devices (such as smart phones, personal digital assistants, and audio/video players) is to optimize power consumption, which can improve performance characteristics finished device by extending the battery life or reducing the size of the device.

The traditional method of reducing power consumption is the use of economical modes of operation, such as idle (idle) or sleep (sleep), which differ in the depth of deactivation of internal elements. As a rule, the active mode of operation of such a system is designed for the worst operating conditions and is characterized by maximum load, thereby unnecessarily reducing battery life. Thus, in order to further optimize battery power consumption, the developers pay special attention to power management in active mode.

To facilitate this process, Intelligent Energy Manager (IEM) technology has been developed for ARM processors. This technology is a combination of hardware and software components that work together to perform dynamic power scaling.

The essence of the method of dynamic control of the supply voltage is based on the expression of the power consumption of CMOS processors:

where P is the total power consumption, C is the switched capacitance, fc is the processor frequency, is the supply voltage, is the leakage current in static mode. It follows from the expression that the frequency and supply voltage can be varied to adjust the power consumption.

Frequency reduction to reduce power consumption is widely used in microcontrollers and systems on chips (PSoC), but not the disadvantage of this method is the reduction in performance. The method of dynamic control of the supply voltage is based on varying the supply voltage, however, if the possibilities of adjustment are exhausted, then the method of adjusting the processor frequency is used as an additional method.

Microcontrollers based on ARM architecture

Table 1 presents the general Comparative characteristics ARM microcontrollers from the most well-known and affordable manufacturers: Analog Device, Atmel, Philips Semiconductors and Texas Instruments, and Table 2 presents their technical data in more detail.

Table 1. Comparison of ARM microcontrollers from different manufacturers by key features

TMS 470 (Texas Instruments)	AT91 (Atmel)	Micro Converter (AD)	LPC2000 (Philips)
Systemic:
ARM7TDMI core An external clock divider (ECP) allows you to clock an external device at the desired frequency	ARM7TDMI/ARM920T core Optional built-in calibrated RC oscillator DMA for data exchange between peripheral devices and memory significantly offloads the processor Separate clock enable/disable for each peripheral (250 uA if all off) Built-in 1.8V stabilizer	ARM7TDMI core Built-in calibrated RC generator (± 3%)	ARM7TDMI-S core (slightly worse specific consumption than ARM7TDMI) Versions for temperature range -40…+105°C
Memory:
Flash memory up to 1 MB Memory Protection Module (MSM)	Economical static RAM (for example, AT91M40800 at 40 MHz with external RAM (12ns) draws 120 mA, and AT91R40807 with internal RAM draws 50 mA under the same conditions).	Flash program/data memory up to 62 KB with data retention up to 100 years and endurance of 100 thousand write/erase cycles	Flash memory with 128-bit accelerator interface for operation at a frequency of 60 MHz
Analog Peripherals:
Multibuffer ADC: - 10 bits, 16 channels, 1.75 µs (sample-hold, conversion); - the ability to organize groups of channels; - programming models: TMS470R1X ADC compatibility and conversion result buffering (FIFO); - modes of operation: conversion, calibration (search for the midpoint of the ADC bias error); - self-test (fault check on analog inputs); shutdown. - three PDP channels; - start of conversion, incl. group, by external event or timer; - outputs for setting both transformation boundaries (outputs ADREFHI and ADREFLO).		Built-in temperature sensor (± 3°С) 12-bit ADC - 1 million converters in sec - single-ended or full differential modes 12-bit DAC - output signal: voltage - optional output buffer amplifier; - full power swing (rail-to-rail) Comparator (K) - 2 inputs and output connected to the pins of the microcircuit	10-bit multiplexed ADC: - conversion time 2.44 µs (400 thousand conversions per second) - conversion range 0 ... 3V - triggered by an external signal or timer
Digital Peripherals:
High quality timer (HET): - a set of 21 specialized instructions for controlling the timer; - a specialized predefined timer micromachine associated with input-output lines. Hardware angle generator (HWAG) for solving motor control tasks: - collaboration with NO	Real time clock with separate backup power input Periodic interval timer (PIT) for generating time-synchronized interrupts	Programmable logic matrix - two blocks connecting 16 inputs and 14 outputs - 2 input logic element with any given conversion function Three-phase 16-bit PWM generator for inverter/motor control - anti-phase outputs on each phase with adjustable non-overlapping pause - adjustable PWM frequency	32-bit timers (edge capture channels and comparison channels), PWM block (6 outputs), real time clock
Interfaces:
Serial interfaces of the 2nd class (C2SIa and C2SIb) - receiving and transmitting data in a multi-master network; - connection of TMS470R1Vx with external analog interface microcircuits; - buffering, detection of errors and breaks, calibration, etc. CAN controllers - standard (SCC): mailbox for 16 messages; - high quality (HECC): mailbox for 32 messages. DSP interface - connects TMS470R1x and TMS320C54x DSP mega module Bus Expansion Module (EBM): - support 8 or 16-bit. external memory; - general purpose I/O function when external bus is not used	USB 2.0 interface External memory interface with configurable external chip select outputs High Speed Programming Controller: - Serial and Parallel Flash Memory Programming Modes Flash Card Interfaces (RM9200)	Standard serial interfaces (UART, SPI, I2C)	16C550 compatible UART - support for modem signals on one of the UARTs Configurable external memory interface with 4 banks and 8/16/32 data width

Table 2. Technical data for ARM microcontrollers from Atmel, Analog Device, Texas Instruments, Philips Semiconductors

Name	Core	Frame	Memory		Peripherals						I/O	Max. h-ta, MHz
Name	Core	Frame	Flash, KB	RAM, KB	Timer	ADC, ch / res	SPI/U(S)APP/ I2C	USB Dev/Host	CAN	Other	I/O	Max. h-ta, MHz
Microcontrollers of the TMS470 family from Texas Instruments
TMS470R1A64	ARM7TDMI	80 LQFPs	64	4	13	8/10	2/2/-	-	2	C2SI	40	48
	ARM7TDMI	100 LQFPs	128	8	16	16/10	2/2/-	-	1	C2SI	50	48
	ARM7TDMI	100 LQFPs	256	12	16	16/10	2/2/-	-	1	C2SI	50	48
	ARM7TDMI	100/144LQFP	288	16	12	12/10	2/2/1	-	2	C2SI, RAP, EBM, MSM	93	48
	ARM7TDMI	144 LQFP	512	32	32	16/10	2/2/-	-	2	RAP	87	60
	ARM7TDMI	144 LQFP	768	48	32	16/10	5/2/-	-	3	RAP	87	60
TMS470R1A1024	ARM7TDMI	144 LQFP	1024	64	12	12/10	5/2/1	-	2	DMA, EBM, MSM	93	60
Atmel's AT91 ARM Thumb family
	ARM7TDMI	QFP100	-	8	3		-/2/-			EBI	32	40
	ARM7TDMI	QFP100	-	256	3		-/2/-			EBI	32	70
	ARM7TDMI	BGA121	512	256	3		-/2/-			EBI	32	70
	ARM7TDMI	BGA121	2048	256	3		-/2/-			EBI	32	70
	ARM7TDMI	QFP144 BGA144	-	8	6		2/2/-			EBI, PIT, RTT	54	33
	ARM7TDMI	QFP176 BGA176	-	8	6	8/10	1/3/-			EBI, RTC, 2x10 rubles DAC	58	33
	ARM7TDMI	QFP100	256	96	6	1/4/1	1/-			SSC, PIT, RTC, RTT	63	66
	ARM7TDMI	BGA256	1	16	3		1/2/-			EBI, int. SDRAM, 2xEthernet	48	36
	ARM7TDMI	QFP144	-	4	9	8/10	1/3/-			EBI, 4 PWM, CAN	49	40
	ARM7TDMI	QFP176	-	16	10	16/10	1/2/-		4	EBI	57	30
	ARM7TDMI	QFP100	256	32	9	16/10	2/4/1	1/-	1	8 PWM, RTT, PIT, RC Gen., SSC, MCI	62	60
	ARM7TDMI	QFP48	32	8	3	8/10	1/1/1				21	55
	ARM7TDMI	QFP64	64	16	3	8/10	1/2/1	1/-		4 PWM, RTT, PIT, RC Gen., SSC	32	55
	ARM7TDMI	QFP64	128	32	3	8/10	1/2/1	1/-		4 PWM, RTT, PIT, RC Gen., SSC	32	55
	ARM7TDMI	QFP64	256	64	3	8/10	1/2/1	1/-		4 PWM, RTT, PIT, RC Gen., SSC	32	55
	ARM7TDMI	QFP100	128	32	3	8/10	1/2/1	1/-	1	4 PWM, RTT, PIT, RC Gen., SSC, Ethernet	60	55
	ARM920T	QFP208 BGA256	128	16	6		1/4/1	1/2		EBI, RTC, RTT, PIT, SDRAM, 3xSSC, MCI, Ethernet	94	180
AT91SAM9261	ARM7TDMI	BGA217	32	160	3		3/3/1	1/2		EBI, RTT, PIT, int.SDRAM, 3xSSC, MCI	96	200
Microcontrollers of the MicroConverter family from Analog Device
	ARM7TDMI	CP-40	62	8		5/12	1/1/2			4 x 12r. DAC, K, PLM	14	45
	ARM7TDMI	CP-40	62	8		8/12	1/1/2			2 x 12r. DAC, K, PLM	13	45
	ARM7TDMI	CP-40	62	8		10/12	1/1/2			K, PLM	13	45
	ARM7TDMI	CP-64	62	8		10/12	1/1/2			2 x 12p.DAC, 3ph. PWM, K, PLM	30	45
	ARM7TDMI	CP-64	62	8		12/12	1/1/2			3f. PWM, K, PLM	30	45
	ARM7TDMI	ST-80	62	8		12/12	1/1/2			4 x 12p.DAC, 3-phase PWM, K, PLM	40	45
	ARM7TDMI	ST-80	62	8		16/12	1/1/2			3f. PWM, K, PLM	40	45
Microcontrollers of the LPC2000 family from Philips Semiconductors
	ARM7TDMI-S	LQFP48	128	16	4		1/2/1			6 ch. PWM	32	60
	ARM7TDMI-S	LQFP48	128	32	4		1/2/1			6 ch. PWM	32	60
	ARM7TDMI-S	LQFP48	128	64	4		1/2/1			6 ch. PWM	32	60
	ARM7TDMI-S	LQFP64	128	16	4	4/10	2/2/1			6 ch. PWM	46	60
	ARM7TDMI-S	LQFP64	128	16	4	4/10	2/2/1			6 ch. PWM	46	60
	ARM7TDMI-S	LQFP64	256	16	4	4/10	2/2/1			6 ch. PWM	46	60
	ARM7TDMI-S	LQFP64	256	16	4	4/10	2/2/1			6 ch. PWM	46	60
2/2/1			6 ch. PWM	112	60
	ARM7TDMI-S	LQFP144	256	16	4	8/10	2/2/1		2	6 ch. PWM	112	60
	ARM7TDMI-S	LQFP144	256	16	4	8/10	2/2/1		4	6 ch. PWM	112	60

Despite the use of the common ARM7TDMI core in most microcontrollers, microcontrollers from different manufacturers have a pretty clear portrait. Analog Device is the undisputed leader in analog peripherals with 12-bit. ADC and DAC class 1MHz. Atmel noticeably lags behind in this direction, which in its development of individual ADCs has already taken the 2GHz barrier, but to integrate a decent ADC in 32-bit. microcontroller, and could not. However, this disadvantage of Atmel microcontrollers overcomes their "friendliness" (when using the built-in RC generator and stabilizer, only one supply voltage is required to start the microcontroller), efficiency, and most importantly, low cost. Among the microcontrollers in question, the Atmel microcontrollers are the only ones that contain a USB interface. TI microcontrollers are characterized by excessive representativeness at a moderate cost. Working with TMS470 microcontrollers, you can be sure that peripheral resources are sufficient. Microcontrollers LPC2000 (Philips) can be called the golden mean according to the considered criteria. They are distinguished by the presence of a UART made in the tradition of Philips and which is compatible with the standard 16C550 UART, and also has a modem interface and a hardware communication control mode with FIFO buffering. Among Philips ARM microcontrollers, you can find representatives for an extended temperature range of -40…+105°C.

32-bit microcontrollers with alternative cores

When it comes to 32-bit. microcontrollers, it would be unfair not to mention other 32-bit. alternatives to the ARM core. In this regard, the FR core from Fujitsu and M68000/M68300 from Motorola should be singled out.

The FR core is used in a vast number of microcontrollers (over 40) that form several families, and has a 16-bit instruction set mode to optimize program memory usage with minimal performance degradation, which is identical to the ARM core. The size of ROM and RAM reaches up to 512 kB, depending on the type, a variety of standard peripherals are supported, incl. 10-bit ADC, 12-bit PWM, CAN interface, UART, etc. Just like in the case of ARM microcontrollers, microcontrollers based on the FR core are distinguished by common traditions that the developer lays down and which are recognizable throughout the line of microcontrollers. In the case of Fujitsu, this is hardware support for endianism, a hardware bit search function, many channels of the same type of peripheral devices, and a non-maskable interrupt input. A fairly decent 10-bit is integrated into many microcontrollers. ADC (conversion time 1.7 µs) and DAC (0.9 µs). In the FRLite family, a record for specific power consumption of 1mA / MHz has been set. The FR 65E family has the maximum speed, in which the clock frequency reaches 66 MHz.

32-bit Motorola microcontrollers are characterized by implementation from a set of standard functional modules. The microcontrollers of the 68300 family include: 32-bit processor (CPU32), internal memory modules, system integration interface module (SIM), serial interface module (QSM), timer processor (TPU) or timer module (GPT), analog digital converter (ADC) and a number of others. The modules are connected to each other by means of an intermodule bus. The CPU32 processor used in microcontrollers of the 68300 family is similar in its main functions to the 32-bit microprocessor MC68020 of the 68000 family. For use in communication systems, microcontrollers are produced that contain a communication RISC processor module that has a set of special tools for data exchange. Such communication controllers (68360, 68302, 68356) are also part of the 68300 family. of the 68000 family is the division of their resources and capabilities depending on the class of tasks being solved. This implies the implementation of two classes of tasks: control of the operation of the microprocessor system itself with the help of system software (operating system - supervisor) and solution of applied user tasks. This gives rise to modes of operation: supervisor mode or user mode. Depending on the mode, when programs are executed, access to all or part of the microcontroller resources is allowed. Supervisor mode allows execution of any instructions implemented by the processor and access to all registers. In user mode, execution of some commands and access to some registers is prohibited to limit the possibility of such changes in the state of the system that may interfere with the execution of other programs or violate the mode of operation of the processor set by the supervisor. A strong argument in favor of choosing Motorola microcontrollers is the high popularity of the M68000 family in its time and the software compatibility of the M68000 and more modern M68300 microcontrollers, which allows using existing software developments in new developments, thereby reducing design time.

The undoubted advantage of the ARM core is its standard nature, which allows you to use software from other compatible microcontrollers, have wider access to design tools, or more easily migrate between microcontrollers.
Despite the use of the same ARM core in microcontrollers from different manufacturers, nevertheless, each of them has its own face, which is achieved by the original "recipe" of peripheral devices and occupying leadership positions in some types of peripheral devices, for example, for Analog Device this is a digital -analog converters.
ARM cores have a representative nomenclature and development dynamics, however, it follows from the comparison that microcontrollers based on the ARM7TDMI core are mainly available for the general public. This can be explained, for example, by the fact that the main area of consumption of ARM microcontrollers is household, office, user electronic devices and equipment, which, unfortunately, are mainly produced by foreign OEMs.
The market of 32-bit microcontrollers has a high capacity, which will dynamically grow in the coming years, therefore, we just have to follow the struggle of microcontroller manufacturers for the share of this market, follow the announcements and have time to master new technologies.

Literature

J. Wilbrink. Facilitation the Migration from 8-bit to 32-bit Microcontrollers/Atmel Corporation -2004.
"Atmel Introduces World"s First Sub $3 ARM7 Flash Microcontroller", Atmel news on 10/19/04, www.atmel.com.
Processor Cores Flyer//Ref: ARM DOI 0111-4/05.03, Issued: May 2003.
Site materials www.arm.com

The name ARM has certainly been heard by everyone interested in mobile technology. Many understand this abbreviation as a type of processor for smartphones and tablets, while others specify that this is not a processor at all, but its architecture. And certainly few people delved into the history of the emergence of ARM. In this article, we will try to understand all these nuances and tell you why modern gadgets need ARM processors.

A brief excursion into history

When asked for "ARM", Wikipedia gives two meanings for this abbreviation: Acorn RISC Machine and Advanced RISC Machines. Let's start in order. In the 1980s, Acorn Computers was founded in the UK, which began its activities by creating personal computers. At that time, Acorn was also called the "British Apple". A decisive period for the company came in the late 1980s, when its chief engineer took advantage of the decision of two local university graduates to come up with a new kind of reduced instruction set (RISC) processor architecture. This is how the first computer based on the Acorn Risc Machine processor appeared. Success was not long in coming. In 1990, the British entered into an agreement with Apple and soon began work on a new version of the chipset. As a result, the development team formed a company called Advanced RISC Machines, similar to the processor. Chips with the new architecture also became known as the Advanced Risc Machine, or ARM for short.

Since 1998, Advanced Risc Machine has become known as ARM Limited. At the moment, the company is not engaged in the production and sale of its own processors. The main and only activity of ARM Limited is the development of technologies and the sale of licenses to various companies to use the ARM architecture. Some manufacturers buy a license for off-the-shelf cores, others a so-called "architectural license" to produce processors with their own cores. These companies include Apple, Samsung, Qualcomm, nVidia, HiSilicon and others. According to some reports, ARM Limited earns $0.067 on each such processor. This figure is average and also outdated. Every year there are more and more cores in chipsets, and new multi-core processors outperform obsolete samples at cost.

Technical features of ARM chips

There are two types of modern processor architectures: CISC(Complex Instruction Set Computing) and RISC(Reduced Instruction Set Computing). The CISC architecture refers to the x86 processor family (Intel and AMD), while the RISC architecture refers to the ARM family. The main formal difference between RISC and CISC and, accordingly, x86 and ARM is the reduced instruction set used in RISC processors. So, for example, each instruction in the CISC architecture is transformed into several RISC instructions. In addition, RISC processors use fewer transistors and thus consume less power.

The main priority of ARM processors is the ratio of performance to power consumption. ARM has a higher performance-per-watt ratio than x86. You can get the power you need from 24 x86 cores or from hundreds of small, low power ARM cores. Of course, even the most powerful processor on the ARM architecture will never be comparable in power to the Intel Core i7. But the same Intel Core i7 needs an active cooling system and will never fit in a phone case. Here ARM is out of competition. On the one hand, it looks like an attractive option for building a supercomputer using a million ARM processors instead of a thousand x86 processors. On the other hand, the two architectures cannot be unambiguously compared. In some ways, the advantage will be for ARM, and in some ways - for x86.

However, calling ARM architecture chips processors is not entirely correct. In addition to several processor cores, they also include other components. The most appropriate term would be "single-chip system" or "system on a chip" (SoC). Modern single-chip systems for mobile devices include a RAM controller, a graphics accelerator, a video decoder, an audio codec, and wireless communication modules. As mentioned earlier, individual chipset components can be developed by third-party manufacturers. The most striking example of this is the graphics cores, which are being developed in addition to ARM Limited (Mali graphics), by Qualcomm (Adreno), NVIDIA (GeForce ULP) and Imagination Technologies (PowerVR).

In practice, it looks like this. Most budget Android mobile devices come with chipsets manufactured by the company. MediaTek, which almost invariably follows the instructions of ARM Limited and completes them with Cortex-A cores and Mali graphics (less often PowerVR).

A-brands for their flagship devices often use chipsets manufactured by Qualcomm. By the way, the latest Qualcomm Snapdragon chips (,) are equipped with fully custom Kryo cores for the central processor and Adreno for the graphics accelerator.

Concerning Apple, then for the iPhone and iPad, the company uses its own A-series chips with PowerVR graphics accelerator, which are produced by third-party companies. So, a 64-bit quad-core A10 Fusion processor and a PowerVR GT7600 graphics processor are installed.

The architecture of processors of the family is considered relevant at the time of writing the article. ARMv8. It was the first to use a 64-bit instruction set and support more than 4 GB of RAM. The ARMv8 architecture is backward compatible with 32-bit applications. The most efficient and most powerful processor core developed by ARM Limited so far is Cortex-A73, and most SoC manufacturers use it unchanged.

Cortex-A73 delivers 30% faster performance than Cortex-A72 and supports the full set of ARMv8 architectures. The maximum frequency of the processor core is 2.8 GHz.

Scope of use of ARM

The greatest glory of ARM brought the development of mobile devices. In anticipation of the mass production of smartphones and other portable equipment, energy-efficient processors came in handy. The culmination of the development of ARM Limited was 2007, when the British company renewed its partnership with Apple, and some time later, the Cupertino company introduced its first iPhone with an ARM architecture processor. Subsequently, the single-chip system based on the ARM architecture has become an invariable component of almost all smartphones on the market.

ARM Limited's portfolio is not limited to the Cortex-A family of cores. In fact, under the Cortex brand, there are three series of processor cores, which are denoted by the letters A, R, M. Core family Cortex-A, as we already know, is the most powerful. They are mainly used in smartphones, tablets, set-top boxes, satellite receivers, automotive systems, robotics. Processor cores Cortex-R are optimized to perform high-performance tasks in real time, so such chips are found in medical equipment, autonomous security systems, and storage media. The main task of the family Cortex-M is simplicity and low cost. Technically, these are the weakest processor cores with the lowest power consumption. Processors based on such cores are used almost everywhere where a device requires minimal power and low cost: sensors, controllers, alarms, displays, smart watches and other equipment.

In general, most of today's devices, from small to large, requiring a CPU use ARM chips. A huge plus is the fact that the ARM architecture is supported by many operating systems based on Linux (including Android and Chrome OS), iOS, and Windows (Windows Phone).

Competition in the market and prospects for the future

Admittedly, at the moment ARM has no serious competitors. And by and large, this is due to the fact that ARM Limited made the right choice at a certain time. But at the very beginning of its journey, the company produced processors for PCs and even tried to compete with Intel. After ARM Limited changed the direction of its activities, it was also not easy for her. Then the software monopolist represented by Microsoft, having entered into a partnership agreement with Intel, left no chance for other manufacturers, including ARM Limited - Windows simply did not work on systems with ARM processors. No matter how paradoxical it may sound, but now the situation may change dramatically, and Windows is already ready to support processors based on this architecture.

In the wake of the success of ARM chips, Intel made an attempt to create a competitive processor and entered the market with a chip Intel Atom. To do this, it took her much more time than ARM Limited. The chipset entered production in 2011, but, as they say, the train has already left. The Intel Atom is an x86 CISC processor. The company's engineers have achieved lower power consumption than ARM, but currently a variety of mobile software has poor adaptation to the x86 architecture.

Last year, Intel abandoned several key decisions in the further development of mobile systems. Actually a company for mobile devices as they have become unprofitable. The only major manufacturer that bundled their smartphones with Intel Atom chipsets was ASUS. However, Intel Atom still received massive use in netbooks, nettops and other portable devices.

ARM Limited's position in the market is unique. At the moment, almost all manufacturers use its developments. At the same time, the company does not have its own factories. This does not prevent her from standing on a par with Intel and AMD. The history of ARM includes another curious fact. It is possible that now ARM technology could belong to Apple, which was at the heart of the formation of ARM Limited. Ironically, in 1998, the Cupertinos, going through times of crisis, sold their stake. Now Apple is forced, along with other companies, to buy a license for the ARM processors used in the iPhone and iPad.

Now ARM processors are capable of performing serious tasks. In the short term, they will be used in servers, in particular, Facebook and PayPal data centers already have such solutions. In the era of the Internet of Things (IoT) and smart home devices, ARM chips have become even more in demand. So the most interesting thing for ARM is yet to come.

The vast majority of modern gadgets use processors based on the ARM architecture, which is being developed by the ARM Limited company of the same name. Interestingly, the company itself does not produce processors, but only licenses its technologies to third-party chip manufacturers. In addition, the company also develops Cortex processor cores and Mali graphics accelerators, which we will definitely touch on in this material.

ARM Limited

The ARM company, in fact, is a monopolist in its field, and the vast majority of modern smartphones and tablets on various mobile operating systems use processors based on the ARM architecture. Chipmakers license individual cores, instruction sets and related technologies from ARM, and the cost of licenses varies significantly depending on the type of processor cores (from low-power budget solutions to cutting-edge quad-core and even eight-core chips) and additional components. ARM Limited's 2006 annual income statement showed revenues of $161 million for licensing about 2.5 billion processors (up from $7.9 billion in 2011), which translates into approximately $0.067 per chip. However, for the reason stated above, this is a very average figure due to the difference in prices for various licenses, and since then the company's profit should have grown many times over.

Currently, ARM processors are very widespread. Chips on this architecture are used everywhere, right down to servers, but most often ARM can be found in embedded and mobile systems, from hard drive controllers to modern smartphones, tablets and other gadgets.

Cortex cores

ARM develops several families of cores that are used for various tasks. For example, processors based on Cortex-Mx and Cortex-Rx (where "x" is a digit or a number indicating the exact core number) are used in embedded systems and even consumer devices such as routers or printers.

We will not dwell on them in detail, because we are primarily interested in the Cortex-Ax family - chips with such cores are used in the most productive devices, including smartphones, tablets and game consoles. ARM is constantly working on new cores from the Cortex-Ax line, but at the time of this writing, smartphones use the following ones:

The larger the number, the higher the processor performance and, accordingly, the more expensive the class of devices in which it is used. However, it is worth noting that this rule is not always observed: for example, chips based on Cortex-A7 cores have higher performance than those based on Cortex-A8. Nevertheless, if Cortex-A5 processors are already considered almost obsolete and are almost never used in modern devices, then Cortex-A15 processors can be found in flagship communicators and tablets. Not so long ago, ARM officially announced the development of new, more powerful and, at the same time, energy-efficient Cortex-A53 and Cortex-A57 cores, which will be combined on a single chip using ARM big.LITTLE technology and support the ARMv8 instruction set (“architecture version”) , but they are not currently used in mass consumer devices. Most chips with Cortex cores can be multi-core, and quad-core processors are ubiquitous in modern high-end smartphones.

Large manufacturers of smartphones and tablets usually use processors from well-known chipmakers like Qualcomm or their own solutions that have already become quite popular (for example, Samsung and its family of Exynos chipsets), but among the technical characteristics of gadgets of most small companies, you can often find descriptions like “processor based on Cortex-A7 @ 1 GHz" or "Dual Core Cortex-A7 @ 1 GHz", which won't tell the average user anything. In order to understand what the differences between such nuclei are, let's focus on the main ones.

The Cortex-A5 core is used in inexpensive processors for the most budget devices. Such devices are designed only to perform a limited range of tasks and run simple applications, but they are not at all designed for resource-intensive programs and, especially, games. An example of a gadget with a Cortex-A5 processor is the Highscreen Blast, which received a Qualcomm Snapdragon S4 Play MSM8225 chip containing two Cortex-A5 cores clocked at 1.2 GHz.

Cortex-A7 processors are more powerful than Cortex-A5 chips and are more common. Such chips are made on a 28-nanometer process technology and have a large second-level cache up to 4 megabytes. Cortex-A7 cores are found mainly in budget smartphones and low-cost mid-range devices like the iconBIT Mercury Quad, and, as an exception, in the Samsung Galaxy S IV GT-i9500 with an Exynos 5 Octa processor - this chipset uses an energy-saving quad-core processor on Cortex-A7.

The Cortex-A8 core is not as common as its “neighbors”, Cortex-A7 and Cortex-A9, but is still used in various entry-level gadgets. The operating clock frequency of Cortex-A8 chips can range from 600 MHz to 1 GHz, but sometimes manufacturers overclock processors to higher frequencies. A feature of the Cortex-A8 core is the lack of support for multi-core configurations (that is, processors on these cores can only be single-core), and they are executed on a 65-nanometer process technology, which is already considered obsolete.

Cortex-A9

A couple of years ago, Cortex-A9 cores were considered the top solution and were used in both traditional single-core and more powerful dual-core chips, such as Nvidia Tegra 2 and Texas Instruments OMAP4. Currently, processors based on Cortex-A9, made according to the 40-nanometer process technology, do not lose popularity and are used in many mid-range smartphones. The operating frequency of such processors can be from 1 to 2 or more gigahertz, but usually it is limited to 1.2-1.5 GHz.

In June 2013, ARM officially introduced the Cortex-A12 core, which is based on a new 28nm process technology and is designed to replace Cortex-A9 cores in mid-range smartphones. The developer promises a 40% increase in performance compared to Cortex-A9, and in addition, Cortex-A12 cores will be able to participate in the ARM big.LITTLE architecture as productive ones along with energy-saving Cortex-A7, which will allow manufacturers to create inexpensive eight-core chips. True, at the time of this writing, all this is only in the plans, and mass production of Cortex-A12 chips has not yet been established, although RockChip has already announced its intention to release a quad-core Cortex-A12 processor with a frequency of 1.8 GHz.

For 2013, the Cortex-A15 core and its derivatives are the top solution and are used in flagship communicator chips from various manufacturers. Among the new processors made according to the 28-nm process technology and based on Cortex-A15 are Samsung Exynos 5 Octa and Nvidia Tegra 4, and this core often acts as a platform for modifications from other manufacturers. For example, Apple's latest A6X processor uses Swift cores, which are a modification of the Cortex-A15. Chips based on Cortex-A15 are capable of operating at a frequency of 1.5-2.5 GHz, and support for many third-party standards and the ability to address up to 1 TB of physical memory makes it possible to use such processors in computers (how can one not recall a mini-computer the size of a bank Raspberry Pi card).

Cortex-A50 series

In the first half of 2013, ARM introduced a new line of chips called the Cortex-A50 series. The cores of this line will be made according to the new version of the architecture, ARMv8, and will support new instruction sets, and will also become 64-bit. The transition to a new bit depth will require optimization of mobile operating systems and applications, but, of course, support for tens of thousands of 32-bit applications will remain. Apple was the first to switch to 64-bit architecture. The company's latest devices, such as the iPhone 5S, run on just such an Apple A7 ARM processor. It is notable that it does not use Cortex cores - they are replaced with the manufacturer's own cores called Swift. One of the obvious reasons for the need to switch to 64-bit processors is the support for more than 4 GB of RAM, and, in addition, the ability to operate with much larger numbers when calculating. Of course, while this is relevant, first of all, for servers and PCs, but we will not be surprised if smartphones and tablets with this amount of RAM appear on the market in a few years. To date, nothing is known about plans to release chips on a new architecture and smartphones using them, but it is likely that such processors will receive flagships in 2014, as Samsung has already announced.

The Cortex-A53 core opens the series, which will be the direct “successor” of the Cortex-A9. Processors based on Cortex-A53 are noticeably superior to chips based on Cortex-A9 in performance, but at the same time, low power consumption is maintained. Such processors can be used both individually and in the ARM big.LITTLE configuration, being combined on the same chipset with a Cortex-A57 processor

Performance Cortex-A53, Cortex-A57

Processors on Cortex-A57, which will be made on a 20-nanometer process technology, should become the most powerful ARM processors in the near future. The new core significantly outperforms its predecessor, the Cortex-A15, in various performance metrics (you can see the comparison above), and according to ARM, which is seriously targeting the PC market, will be a profitable solution for mainstream computers (including laptops), not just mobile ones. devices.

ARM big.LITTLE

As a high-tech solution to the problem of power consumption of modern processors, ARM offers the big.LITTLE technology, the essence of which is to combine different types of cores on one chip, usually the same number of energy-saving and high-performance ones.

There are three schemes for the operation of different types of cores on a single chip: big.LITTLE (migration between clusters), big.LITTLE IKS (migration between cores), and big.LITTLE MP (heterogeneous multiprocessing).

big.LITTLE (migration between clusters)

The first chipset based on the ARM big.LITTLE architecture was the Samsung Exynos 5 Octa processor. It uses the original big.LITTLE “4+4” scheme, which means combining into two clusters (hence the name of the scheme) on one chip four high-performance Cortex-A15 cores for resource-intensive applications and games and four energy-saving Cortex-A7 cores for everyday work with most programs, and at one time only one type of kernel can work. Switching between groups of cores occurs almost instantly and imperceptibly for the user in a fully automatic mode.

big.LITTLE IKS (migration between cores)

A more complex implementation of the big.LITTLE architecture is the combination of several real cores (usually two) into one virtual one, controlled by the operating system kernel, which decides which cores to use - energy efficient or productive. Of course, there are also several virtual cores - the illustration shows an example of an IKS scheme, where each of the four virtual cores contains one Cortex-A7 and Cortex-A15 core.

big.LITTLE MP (heterogeneous multiprocessing)

The big.LITTLE MP scheme is the most "advanced" one - in it each core is independent and can be turned on by the OS core as needed. This means that if four Cortex-A7 cores and the same number of Cortex-A15 cores are used, in a chipset built on the ARM big.LITTLE MP architecture, all 8 cores will be able to work simultaneously, even though they are of different types. One of the first processors of this type was Mediatek's eight-core chip - MT6592, which can operate at a clock frequency of 2 GHz, as well as record and play videos in UltraHD resolution.

Future

According to currently available information, in the near future, ARM, together with other companies, plans to launch the release of next-generation big.LITTLE chips that will use the new Cortex-A53 and Cortex-A57 cores. In addition, Chinese manufacturer MediaTek is going to release budget processors on ARM big.LITTLE, which will work according to the “2 + 2” scheme, that is, use two groups of two cores.

Mali graphics accelerators

In addition to processors, ARM also develops graphics accelerators of the Mali family. Like processors, graphics accelerators are characterized by many parameters, such as the level of anti-aliasing, bus interface, cache (ultra-fast memory used to increase speed) and the number of “graphics cores” (although, as we wrote in a previous article, this figure, despite the similarity with the term used to describe the CPU has little to no effect on performance when comparing two GPUs).

The first ARM graphics accelerator was the now unused Mali 55, which was used in the LG Renoir touch phone (yes, the most ordinary cell phone). The GPU was not used in games - only for drawing the interface, and had primitive characteristics by today's standards, but it was he who became the "ancestor" of the Mali series.

Since then, progress has come a long way, and now supported APIs and game standards are of no small importance. For example, support for OpenGL ES 3.0 is now announced only in the most powerful processors like the Qualcomm Snapdragon 600 and 800, and, if we talk about ARM products, the standard is supported by such accelerators as the Mali-T604 (it was he who became the first ARM GPU made on new microarchitecture Midgard), Mali-T624, Mali-T628, Mali-T678 and some other chips similar in characteristics. One or another GPU, as a rule, is closely related to the core, but, nevertheless, is indicated separately, which means that if the quality of graphics in games is important to you, then it makes sense to look at the name of the accelerator in the specifications of a smartphone or tablet.

ARM also has graphics accelerators for mid-range smartphones, the most common of which are Mali-400 MP and Mali-450 MP, which differ from their older brothers in relatively low performance and a limited set of APIs and supported standards. Despite this, these GPUs continue to be used in new smartphones, for example, the Zopo ZP998, which received the Mali-450 MP4 graphics accelerator (an improved modification of the Mali-450 MP) in addition to the eight-core MTK6592 processor.

Presumably, at the end of 2014, smartphones with the latest ARM graphics accelerators should appear: Mali-T720, Mali-T760 and Mali-T760 MP, which were introduced in October 2013. Mali-T720 should be the new GPU for low-end smartphones and the first GPU in this segment to support Open GL ES 3.0. Mali-T760, in turn, will become one of the most powerful mobile graphics accelerators: according to the declared characteristics, the GPU has 16 processing cores and has a truly huge processing power, 326 Gflops, but at the same time, four times less power consumption than Mali-T604 mentioned above.

The role of CPU and GPU from ARM in the market

Despite the fact that ARM is the author and developer of the architecture of the same name, which, we repeat, is now used in the vast majority of mobile processors, its solutions in the form of cores and graphics accelerators are not popular with major smartphone manufacturers. For example, it is rightly believed that flagship communicators on Android OS should have a Snapdragon processor with Krait cores and an Adreno graphics accelerator from Qualcomm, chipsets from the same company are used in Windows Phone smartphones, and some gadget manufacturers, for example, Apple, develop their own cores. . Why is this the current situation?

Perhaps some of the reasons may lie deeper, but one of them is the lack of a clear positioning of the CPU and GPU from ARM among the products of other companies, as a result of which the company's developments are perceived as basic components for use in B-brand devices, low-cost smartphones and creating based on them more mature decisions. For example, Qualcomm repeats at almost every presentation that one of its main goals when creating new processors is to reduce power consumption, and its Krait cores, being modified by Cortex cores, consistently show higher performance results. A similar statement is true for Nvidia chipsets, which are focused on games, but as for the Exynos processors from Samsung and the A-series from Apple, they have their own market due to the installation in smartphones of the same companies.

The above does not mean at all that ARM developments are significantly worse than third-party processors and cores, but competition in the market ultimately only benefits smartphone buyers. We can say that ARM offers some blanks, by purchasing a license for which, manufacturers can already modify them on their own.

Conclusion

ARM-based microprocessors have successfully conquered the mobile device market due to their low power consumption and relatively large processing power. Previously, other RISC architectures, such as MIPS, competed with ARM, but now it has only one serious competitor left - Intel with the x86 architecture, which, by the way, although actively fighting for its market share, is not yet perceived by either consumers or most manufacturers seriously, especially when there are actually no flagships on it (Lenovo K900 can no longer compete with the latest top-end smartphones on ARM processors).

What do you think, will anyone be able to push ARM, and how will the fate of this company and its architecture develop further?

How is the processor. Why is ARM the future? The modern consumer of electronics is very difficult to surprise. We are already accustomed to the fact that our pocket is legally occupied by a smartphone, a laptop is in a bag, a “smart” watch obediently counts steps on the hand, and headphones with an active noise reduction system caress our ears.

It's a funny thing, but we are used to carrying not one, but two, three or more computers at once. After all, that's what you can call a device that has a processor. And it doesn’t matter what a particular device looks like. A miniature chip is responsible for its work, having overcome a turbulent and rapid path of development.

Why did we bring up the topic of processors? Everything is simple. Over the past ten years, there has been a real revolution in the world of mobile devices.

There are only 10 years difference between these devices. But Nokia N95 then seemed to us a space device, and today we look at ARKit with a certain mistrust

But everything could have turned out differently, and the battered Pentium IV would have remained the ultimate dream of an ordinary buyer.

We tried to do without complicated technical terms and tell how the processor works and find out which architecture is the future.

1. How it all started

The first processors were completely different from what you can see when you open the lid of your PC system unit.

Instead of microcircuits in the 40s of the XX century, electromechanical relays were used, supplemented by vacuum tubes. The lamps acted as a diode, the state of which could be regulated by lowering or increasing the voltage in the circuit. The structures looked like this:

For the operation of one gigantic computer, hundreds, sometimes thousands of processors were needed. But, at the same time, you would not be able to run even a simple editor like NotePad or TestEdit from the standard set of Windows and macOS on such a computer. The computer would simply not have enough power.

2. The advent of transistors

The first field-effect transistors appeared in 1928. But the world changed only after the appearance of the so-called bipolar transistors, discovered in 1947.

In the late 1940s, experimental physicist Walter Brattain and theorist John Bardeen developed the first point transistor. In 1950, it was replaced by the first junction transistor, and in 1954, the well-known manufacturer Texas Instruments announced a silicon transistor.

But the real revolution came in 1959, when the scientist Jean Henri developed the first silicon planar (flat) transistor, which became the basis for monolithic integrated circuits.

Yes, it's a bit tricky, so let's dig a little deeper and deal with the theoretical part.

3. How a transistor works

So, the task of such an electrical component as a transistor is to control the current. Simply put, this little tricky switch controls the flow of electricity.

The main advantage of a transistor over a conventional switch is that it does not require the presence of a person. Those. such an element is capable of independently controlling the current. In addition, it works much faster than you would turn on or off the electrical circuit yourself.

The task of the computer is to represent the electric current in the form of numbers.

And if earlier the task of switching states was performed by clumsy, bulky and inefficient electrical relays, now the transistor has taken over this routine work.

From the beginning of the 60s, transistors began to be made from silicon, which made it possible not only to make processors more compact, but also to significantly increase their reliability.

But first, let's deal with the diode

Silicon (also known as Si - “silicium” in the periodic table) belongs to the category of semiconductors, which means that, on the one hand, it transmits current better than a dielectric, on the other hand, it does it worse than metal.

Whether we like it or not, but to understand the work and the further history of the development of processors, we will have to plunge into the structure of one silicon atom. Don't be afraid, let's make it short and very clear.

The task of the transistor is to amplify a weak signal due to an additional power source.

The silicon atom has four electrons, thanks to which it forms bonds (or, to be more precise, covalent bonds) with the same nearby three atoms, forming a crystal lattice. While most of the electrons are in bond, a small part of them is able to move through the crystal lattice. It is because of this partial transfer of electrons that silicon was classified as a semiconductor.

But such a weak movement of electrons would not allow the use of a transistor in practice, so the scientists decided to increase the performance of transistors by doping, or, more simply, adding atoms to the silicon crystal lattice with a characteristic arrangement of electrons.

So they began to use a 5-valent impurity of phosphorus, due to which n-type transistors were obtained. The presence of an additional electron made it possible to accelerate their movement, increasing the current flow.

When doping p-type transistors, boron, which includes three electrons, became such a catalyst. Due to the absence of one electron, holes appear in the crystal lattice (they play the role of a positive charge), but due to the fact that electrons are able to fill these holes, the conductivity of silicon increases significantly.

Suppose we took a silicon wafer and doped one part of it with a p-type impurity, and the other with an n-type impurity. So we got a diode - the basic element of a transistor.

Now the electrons located in the n-part will tend to go to the holes located in the p-part. In this case, the n-side will have a slight negative charge, and the p-side will have a positive charge. The electric field formed as a result of this "gravity" - the barrier - will prevent the further movement of electrons.

If you connect a power source to the diode in such a way that "-" touches the p-side of the plate, and "+" touches the n-side, current flow will not be possible due to the fact that the holes will be attracted to the negative contact of the power source, and the electrons to positive, and the bond between the p and n electrons will be lost due to the expansion of the combined layer.

But if you connect the power supply with sufficient voltage the other way around, i.e. "+" from the source to the p-side, and "-" to the n-side, electrons placed on the n-side will be repelled by the negative pole and pushed to the p-side, occupying holes in the p-region.

But now the electrons are attracted to the positive pole of the power source and they continue to move through the p-holes. This phenomenon is called forward bias of the diode.

diode + diode = transistor

By itself, the transistor can be thought of as two diodes docked to each other. In this case, the p-region (the one where the holes are located) becomes common for them and is called the “base”.

The N-P-N transistor has two n-regions with additional electrons - they are also the “emitter” and “collector” and one, weak region with holes - the p-region, called the “base”.

If you connect a power supply (let's call it V1) to n-regions of the transistor (regardless of the pole), one diode will be reverse-biased and the transistor will be in the off state.

But, as soon as we connect another power source (let's call it V2), setting the "+" contact to the "central" p-region (base), and the "-" contact to the n-region (emitter), some of the electrons will flow through again formed chain (V2), and the part will be attracted by the positive n-region. As a result, electrons will flow into the collector region, and a weak electric current will be amplified.

Exhale!

4. So how does a computer actually work?

And now the most important thing.

Depending on the applied voltage, the transistor can be either open or closed. If the voltage is insufficient to overcome the potential barrier (the same one at the junction of p and n plates) - the transistor will be in the closed state - in the “off” state or, in the language of the binary system, “0”.

With enough voltage, the transistor turns on, and we get the value "on" or "1" in binary.

This state, 0 or 1, is called a "bit" in the computer industry.

Those. we get the main property of the very switch that opened the way to computers for mankind!

In the first electronic digital computer ENIAC, or, more simply, the first computer, about 18 thousand triode lamps were used. The size of the computer was comparable to a tennis court, and its weight was 30 tons.

To understand how the processor works, there are two more key points to understand.

Moment 1. So, we have decided what a bit is. But with its help, we can only get two characteristics of something: either "yes" or "no". In order for the computer to learn to understand us better, they came up with a combination of 8 bits (0 or 1), which they called a byte.

Using a byte, you can encode a number from zero to 255. Using these 255 numbers - combinations of zeros and ones, you can encode anything.

Moment 2. The presence of numbers and letters without any logic would give us nothing. That is why the concept of logical operators appeared.

By connecting just two transistors in a certain way, you can achieve several logical actions at once: “and”, “or”. The combination of the amount of voltage on each transistor and the type of their connection allows you to get different combinations of zeros and ones.

Through the efforts of programmers, the values \u200b\u200bof zeros and ones, the binary system, began to be translated into decimal so that we could understand what exactly the computer “says”. And to enter commands, our usual actions, such as entering letters from the keyboard, are represented as a binary chain of commands.

Simply put, imagine that there is a correspondence table, say, ASCII, in which each letter corresponds to a combination of 0 and 1. You pressed a button on the keyboard, and at that moment on the processor, thanks to the program, the transistors switched so that the following appeared on the screen the most written letter on the key.

5. And the transistor race began

After the British radio engineer Geoffrey Dahmer proposed in 1952 to place the simplest electronic components in a monolithic semiconductor crystal, the computer industry took a leap forward.

From the integrated circuits proposed by Dahmer, engineers quickly switched to microchips, which were based on transistors. In turn, several of these chips already formed the processor itself.

Of course, the dimensions of such processors are not much similar to modern ones. In addition, until 1964, all processors had one problem. They required an individual approach - their own programming language for each processor.

1964 IBM System/360. Universal Programming Code compatible computer. An instruction set for one processor model could be used for another.

70s. The appearance of the first microprocessors. Single chip processor from Intel. Intel 4004 - 10 µm TPU, 2300 transistors, 740 kHz.

1973 Intel 4040 and Intel 8008. 3,000 transistors, 740 kHz for the Intel 4040 and 3,500 transistors at 500 kHz for the Intel 8008.

1974 Intel 8080. 6 µm TPU and 6000 transistors. The clock frequency is about 5,000 kHz. It was this processor that was used in the Altair-8800 computer. The domestic copy of the Intel 8080 is the KR580VM80A processor, developed by the Kiev Research Institute of Microdevices. 8 bits

1976 Intel 8080. 3 µm TPU and 6500 transistors. Clock frequency 6 MHz. 8 bits

1976 Zilog Z80. 3 micron TPU and 8500 transistors. Clock frequency up to 8 MHz. 8 bits

1978 Intel 8086. 3 µm TPU and 29,000 transistors. The clock frequency is about 25 MHz. The x86 instruction set that is still in use today. 16 bits

1980 Intel 80186. 3 µm TPU and 134,000 transistors. Clock frequency - up to 25 MHz. 16 bits

1982 Intel 80286. 1.5 µm TPU and 134,000 transistors. Frequency - up to 12.5 MHz. 16 bits

1982 Motorola 68000. 3 µm and 84,000 transistors. This processor was used in the Apple Lisa computer.

1985 Intel 80386. 1.5 µm Tp and 275,000 transistors. Frequency - up to 33 MHz in the 386SX version.

It would seem that the list could be continued indefinitely, but then Intel engineers faced a serious problem.

Out in the late 80s. Back in the early 60s, one of the founders of Intel, Gordon Moore, formulated the so-called "Moore's Law". It sounds like this:

Every 24 months, the number of transistors on an integrated circuit chip doubles.

It is difficult to call this law a law. It would be more accurate to call it empirical observation. Comparing the pace of technology development, Moore concluded that a similar trend could form.

But already during the development of the fourth generation of Intel i486 processors, engineers were faced with the fact that they had already reached the performance ceiling and could no longer accommodate more processors in the same area. At that time, technology did not allow this.

As a solution, a variant was found using a number of additional elements:

cache memory;

conveyor;

built-in coprocessor;

multiplier.

Part of the computational load fell on the shoulders of these four nodes. As a result, the appearance of cache memory, on the one hand, complicated the design of the processor, on the other hand, it became much more powerful.

The Intel i486 processor already consisted of 1.2 million transistors, and the maximum frequency of its operation reached 50 MHz.

In 1995, AMD joined the development and released the fastest i486-compatible Am5x86 processor at that time on a 32-bit architecture. It was already manufactured according to the 350 nanometer process technology, and the number of installed processors reached 1.6 million pieces. The clock frequency has increased to 133 MHz.

But the chipmakers did not dare to pursue further increasing the number of processors installed on a chip and developing the already utopian CISC (Complex Instruction Set Computing) architecture. Instead, American engineer David Patterson suggested optimizing the operation of processors, leaving only the most necessary computational instructions.

So processor manufacturers switched to the RISC (Reduced Instruction Set Computing) platform. But even this was not enough.

In 1991, the 64-bit R4000 processor was released, operating at a frequency of 100 MHz. Three years later, the R8000 processor appears, and two years later, the R10000 with clock speeds up to 195 MHz. In parallel, the market for SPARC processors developed, the architecture feature of which was the absence of multiplication and division instructions.

Instead of fighting over the number of transistors, chip manufacturers began to rethink the architecture of their work. The rejection of "unnecessary" commands, the execution of instructions in one cycle, the presence of registers of general value and pipelining made it possible to quickly increase the clock frequency and power of processors without distorting the number of transistors.

Here are just a few of the architectures that appeared between 1980 and 1995:

They were based on the RISC platform, and in some cases, a partial, combined use of the CISC platform. But the development of technology once again pushed chipmakers to continue building up processors.

In August 1999, the AMD K7 Athlon entered the market, manufactured using a 250 nm process technology and including 22 million transistors. Later, the bar was raised to 38 million processors. Then, up to 250 million, the technological processor increased, the clock frequency increased. But, as physics says, there is a limit to everything.

7. The end of the transistor competition is near

In 2007, Gordon Moore made a very blunt statement:

Moore's Law will soon cease to apply. It is impossible to install an unlimited number of processors indefinitely. The reason for this is the atomic nature of matter.

It is noticeable to the naked eye that the two leading chip manufacturers AMD and Intel have clearly slowed down the pace of processor development over the past few years. The accuracy of the technological process has increased to only a few nanometers, but it is impossible to place even more processors.

And while semiconductor manufacturers threaten to launch multilayer transistors, drawing a parallel with 3DNand memory, a serious competitor appeared at the walled x86 architecture 30 years ago.

8. What awaits "regular" processors

Moore's Law has been invalidated since 2016. This was officially announced by the largest processor manufacturer Intel. Doubling computing power by 100% every two years is no longer possible for chipmakers.

And now processor manufacturers have several unpromising options.

The first option is quantum computers. There have already been attempts to build a computer that uses particles to represent information. There are several similar quantum devices in the world, but they can only cope with algorithms of low complexity.

In addition, the serial launch of such devices in the coming decades is out of the question. Expensive, inefficient and… slow!

Yes, quantum computers consume much less power than their modern counterparts, but they will also be slower until developers and component manufacturers switch to new technology.

The second option is processors with layers of transistors. Both Intel and AMD have seriously thought about this technology. Instead of one layer of transistors, they plan to use several. It seems that in the coming years, processors may well appear in which not only the number of cores and clock frequency will be important, but also the number of transistor layers.

The solution has the right to life, and thus the monopolists will be able to milk the consumer for another couple of decades, but, in the end, the technology will again hit the ceiling.

Today, realizing the rapid development of the ARM architecture, Intel made a quiet announcement of the Ice Lake family of chips. The processors will be manufactured on a 10-nanometer process and will become the basis for smartphones, tablets and mobile devices. But it will happen in 2019.

9. ARM is the future So, the x86 architecture appeared in 1978 and belongs to the CISC platform type. Those. by itself, it implies the existence of instructions for all occasions. Versatility is the main strong point of x86.

But, at the same time, versatility played a cruel joke with these processors. x86 has several key disadvantages:

the complexity of commands and their frank confusion;

high energy consumption and heat release.

For high performance, I had to say goodbye to energy efficiency. Moreover, two companies are currently working on the x86 architecture, which can be safely attributed to monopolists. These are Intel and AMD. Only they can produce x86 processors, which means that only they rule the development of technologies.

At the same time, several companies are involved in the development of ARM (Arcon Risk Machine). Back in 1985, developers chose the RISC platform as the basis for further development of the architecture.

Unlike CISC, RISC involves designing a processor with the minimum required number of instructions, but maximum optimization. RISC processors are much smaller than CISC, more power efficient and simpler.

Moreover, ARM was originally created solely as a competitor to x86. The developers set the task to build an architecture that is more efficient than x86.

Ever since the 40s, engineers have understood that one of the priority tasks is to work on reducing the size of computers, and, first of all, the processors themselves. But almost 80 years ago, hardly anyone could have imagined that a full-fledged computer would be smaller than a matchbox.

For skeptical users who trawl through the top lines of Geekbench, I just want to remind you: in mobile technology, size is what matters first of all.

Place a candy bar with a powerful 18-core processor that “rips the ARM architecture to shreds” on the table, and then put your iPhone next to it. Feel the difference?

11. Instead of output

It is impossible to cover the 80-year history of the development of computers in one material. But after reading this article, you will be able to understand how the main element of any computer is arranged - the processor, and what to expect from the market in the coming years.

Of course, Intel and AMD will work on further increasing the number of transistors on a single chip and promoting the idea of multilayer elements.

But do you, as a customer, need such power?

I don't think you're dissatisfied with the performance of an iPad Pro or a flagship iPhone X. I don't think you're dissatisfied with the performance of your multicooker in your kitchen or the picture quality of a 65-inch 4K TV. But all these devices use processors on the ARM architecture.

Windows has already officially announced that it is looking towards ARM with interest. The company included support for this architecture back in Windows 8.1, and is now actively working on a tandem with the leading ARM chipmaker Qualcomm.

Google also managed to look at ARM - the Chrome OS operating system supports this architecture. Several Linux distributions have appeared at once, which are also compatible with this architecture. And this is just the beginning.

And just try for a moment to imagine how pleasant it will be to combine an energy-efficient ARM processor with a graphene battery. It is this architecture that will make it possible to obtain mobile ergonomic gadgets that can dictate the future.

The computer world is changing rapidly. Desktop PCs have lost the top spot in the sales rankings to laptops, and they are about to give up the market to tablets and other mobile devices. 10 years ago we valued pure megahertz, real power and performance. Now, in order to conquer the market, the processor must be not only fast, but also economical. Many consider the ARM to be the architecture of the 21st century. Is it so?

New - well forgotten old

Journalists, following ARM PR people, often present this architecture as something completely new, which should bury the gray-haired x86.

In fact, ARM and x86, on the basis of which they are built Intel processors, AMD and VIA, installed in laptops and desktop PCs, are practically the same age. The first x86 chip was released in 1978. The ARM project officially started in 1983, but it was based on developments that were carried out almost simultaneously with the creation of x86.

The early ARMs impressed specialists with their finesse, but with their relatively low performance, they could not conquer a market that demanded high speeds and did not pay attention to performance. There had to be certain conditions for ARM's popularity to skyrocket.

At the turn of the eighties and nineties, with their relatively inexpensive oil, huge SUVs with powerful 6-liter engines were in demand. Few people were interested in electric cars. But nowadays, when a barrel of oil costs more than $100, big cars with voracious engines are only for the rich, the rest are in a hurry to switch to fuel-efficient cars. A similar thing happened with ARM. When the question of mobility and efficiency arose, the architecture turned out to be in great demand.

"Risk" processor

ARM is a RISC architecture. It uses a reduced set of commands - RISC (reduced instruction set computer). This type of architecture appeared in the late seventies, around the same time that Intel introduced its x86.

While experimenting with various compilers and microcoded processors, the engineers noticed that in some cases, sequences of simple instructions were faster than a single complex operation. It was decided to create an architecture that would involve working with a limited set of simple instructions, decoding and execution of which would take a minimum of time.

One of the first projects for RISC processors was implemented by a group of students and teachers from the University of Berkeley in 1981. Just at this time, the British company Acorn faced the challenge of the times. It produced BBC Micro educational computers, which were very popular in Foggy Albion, based on the 6502 processor. But soon these home PCs began to lose to more advanced machines. Acorn risked losing the market. The company's engineers, having become acquainted with student work on RISC processors, decided that it would be quite simple to cope with creating their own chip. In 1983, the Acorn RISC Machine project started, which later turned into ARM. Three years later, the first processor was released.

First ARMs

He was extremely simple. The first ARM chips were even devoid of multiplication and division instructions, which seemed to be a set of more simple instructions. Another feature of the chips was the principles of working with memory: all operations with data could be carried out only in registers. At the same time, the processor worked with the so-called register window, that is, it could access only a part of all available registers, which were basically universal, and their work depended on the mode in which the processor was. This allowed the very first versions of ARM to abandon the cache.

In addition, by simplifying the instruction sets, the architects were able to do without a number of other blocks. For example, in the first ARM, there was no microcode at all, as well as a floating point unit, the FPU. The total number of transistors in the first ARM was 30,000. In similar x86, there were several times, or even an order of magnitude more. Additional energy savings are achieved by conditionally executing commands. That is, this or that operation will be performed if there is a corresponding fact in the register. This helps the processor to avoid "excessive gestures". All instructions are executed sequentially. As a result, ARM lost in performance, but not significantly, while gaining significantly in power consumption.

The basic principles of building the architecture remain the same as in the first ARMs: work with data only in registers, a reduced set of commands, a minimum of additional modules. All this provides the architecture with low power consumption at relatively high performance.

In order to increase it, ARM has introduced several additional instruction sets over the past years. Along with the classic ARM, there are Thumb, Thumb 2, Jazelle. The latter is designed to speed up the execution of Java code.

Cortex - the most advanced ARM

Cortex - modern architectures for mobile devices, embedded systems and microcontrollers. Accordingly, CPUs are designated as Cortex-A, embedded - Cortex-R and microcontrollers - Cortex-M. All of them are based on the ARMv7 architecture.

The most advanced and powerful architecture in the ARM line is Cortex-A15. It is assumed that on its basis, mainly two or four-core models will be produced. Cortex-A15 of all previous ARMs is the closest to x86 in terms of the number and quality of blocks.

The Cortex-A15 is based on processor cores equipped with an FPU and a set of NEON SIMD instructions designed to speed up the processing of multimedia data. The cores have a 13-stage pipeline, they support the execution of instructions in free order, ARM-based virtualization.

Cortex-A15 supports extended memory addressing system. ARM remains a 32-bit architecture, but the company's engineers have learned how to convert 64-bit or other extended addressing into a 32-bit understandable processor. The technology is called Long Physical Address Extensions. Thanks to her, Cortex-A15 can theoretically address up to 1 TB of memory.

Each core is equipped with a first level cache. In addition, there is up to 4 MB of low-latency distributed L2 cache. The processor is equipped with a 128-bit coherent bus that can be used to communicate with other blocks and peripherals.

The cores that underlie the Cortex-A15 are an evolution of the Cortex-A9. They have a similar structure.

Cortex-A9, unlike Cortex-A15, can be produced in both multi- and single-core versions. The maximum frequency is 2.0 GHz, Cortex-A15 suggests the possibility of creating chips operating at a frequency of 2.5 GHz. Chips based on it will be manufactured using 40 nm and thinner manufacturing processes. Cortex-A9 is available in 65 and 40 nm process technologies.

Cortex-A9, like Cortex-A15, is designed for use in high-performance smartphones and tablets, but it is too tough for more serious applications, for example, in servers. Only Cortex-A15 has hardware virtualization, extended memory addressing. In addition, the NEON Advanced SIMD instruction set and the FPU in the Cortex-A9 are optional elements, while they are mandatory in the Cortex-A15.

Cortex-A8 will gradually disappear from the scene in the future, but for now this single-core variant will find use in budget smartphones. A low cost solution with frequencies from 600 MHz to 1 GHz is a balanced architecture. It has an FPU, supports the first version of SIMD NEON. Cortex-A8 assumes a single manufacturing process - 65 nm.

ARM previous generations

ARM11 processors are quite common in the mobile market. They are based on the ARMv6 architecture and its modifications. It is characterized by 8-9-stage pipelines, Jazelle support, which accelerates the processing of Java code, streaming SIMD instructions, Thumb-2.

XScale, ARM10E, ARM9E processors are based on the ARMv5 architecture and its modifications. Maximum pipeline length is 6 stages, Thumb, Jazelle DBX, Enhanced DSP. XScale chips have a second level cache. Processors were used in smartphones from the mid-2000s, and today they can be found in some inexpensive mobile phones.

ARM9TDMI, ARM8, StrongARM are representatives of ARMv4, which has a 3-5 stage pipeline, supports Thumb. ARMv4, for example, was found in early classic iPods.

ARM6 and ARM7 are ARMv3. In this architecture, the FPU block appeared for the first time, 32-bit memory addressing was implemented, and not 26-bit, as in the first samples of the architecture. Formally, ARMv2 and ARMv1 were 32-bit chips, but in reality they actively worked only with a 26-bit address space. Cache first appeared in ARMv2.

Their name is legion

Acorn was not originally going to become a player in the processor market. The task of the ARM project was to be the creation of a chip of its own production for the production of computers - it was the creation of a PC at Acorn that was considered its main business.

From a group of developers, ARM has become a company thanks to Apple. In 1990, Apple partnered with VLSI and Acorn to develop an economical processor for the first Newton handheld computer. For these purposes, a separate company was created, which received the name of the internal Acorn project - ARM.

With the participation of Apple, the ARM6 processor was created, which is the closest to the modern chips of the English developer. At the same time, DEC was able to patent the ARM6 architecture and began producing chips under the StrongARM brand. A couple of years later, the technology was transferred to Intel as part of another patent dispute. The microprocessor giant has created its own analogue based on ARM - the XScale processor. But in the middle of the previous decade, Intel got rid of this "non-core asset", focusing exclusively on x86. XScale has been taken over by Marvell, which has already licensed ARM.

Newly appeared to the world ARM at first was not able to engage in the production of processors. Her management chose a different way to make money. The ARM architecture was characterized by simplicity and flexibility. At first, the core was even deprived of a cache, therefore, subsequently, additional modules, including FPU, controllers were not closely integrated into the processor, but, as it were, hung on the base.

Accordingly, ARM got its hands on an intelligent designer that allowed technologically advanced companies to create processors or microcontrollers for their needs. This is done with the help of so-called coprocessors, which can extend the standard functionality. In total, the architecture supports up to 16 coprocessors (numbered from 0 to 15), but number 15 is reserved for a coprocessor that performs cache and memory management functions.

Peripherals connect to the ARM chip by mapping their registers to the memory space of the processor or coprocessor. For example, an image processing chip may consist of a relatively simple ARM7TDMI-based core and a coprocessor that provides HDTV decoding.

ARM has begun licensing its architecture. Other companies have already been involved in its implementation in silicon, including Texas Instruments, Marvell, Qualcomm, Freescale, but also completely non-core ones like Samsung, Nokia, Nintendo or Canon.

The lack of own factories, as well as impressive royalties, allowed ARM to be more flexible in developing new versions of the architecture. The company baked them like hot cakes, entering new niches. In addition to smartphones and tablets, the architecture is used in specialized processors, such as GPS navigators, digital cameras and camcorders. On its basis, industrial controllers and other chips for embedded systems are created.

The ARM licensing system is a real hypermarket of microelectronics. The company licenses not only new, but also obsolete architectures. The latter can be used to create microcontrollers or chips for low-cost devices. Naturally, the level of royalties depends on the degree of novelty and complexity of the architecture variant of interest to the manufacturer. Traditionally, the technical processes for which ARM develops processors are 1-2 steps behind those that are considered relevant for x86. The high energy efficiency of the architecture makes it less dependent on the transition to new technical standards. Intel and AMD are looking to make thinner chips to increase clock speeds and core counts while maintaining physical size and power consumption. ARM has lower power requirements natively and also delivers more performance per watt.

Features of NVIDIA, TI, Qualcomm, Marvell processors

By licensing ARM right and left, the developers strengthened the position of their architecture at the expense of the competencies of partners. NVIDIA Tegra can be considered a classic example in this case. This line of systems-on-a-chip is based on the ARM architecture, but NVIDIA already had its own very serious developments in the field of three-dimensional graphics and system logic.

ARM gives its licensors wide authority to redesign the architecture. Accordingly, NVIDIA engineers were able to combine the strengths of ARM (CPU computing) and their own products in Tegra - work with three-dimensional graphics, etc. As a result, Tegra has the highest 3D performance in its class. They are 25-30% faster than the PowerVR used by Samsung and Texas Instruments, and are almost twice as fast as Qualcomm's Adreno.

Other manufacturers of processors based on the ARM architecture are strengthening certain additional blocks, improving chips in order to achieve higher frequencies and performance.

For example, Qualcomm does not use the ARM reference design. The company's engineers seriously redesigned it and called it Scorpio - it is he who underlies the Snapdragon chips. In part, the design has been redesigned in order to master the more subtle technical processes than provided by the standard IP ARM. As a result, the first Snapdragons were produced at 45 nm standards, which provided them with higher frequencies. And the new generation of these processors with the declared 2.5 GHz may even become the fastest among analogues based on ARM Cortex-A9. Qualcomm also uses its own Adreno graphics core, based on designs acquired from AMD. So in a way, Snapdragon and Tegra are enemies at the genetic level.

Samsung, when creating Hummingbird, also took the path of optimizing the architecture. The Koreans, together with Intrinsity, changed the logic, which reduced the number of instructions needed to perform some operations. Thus it was possible to win 5-10% of productivity. In addition, a dynamic second level cache and an ARM NEON multimedia extension were added. The Koreans used the PowerVR SGX540 as a graphics module.

Texas Instruments in the new OMAP series based on the ARM Cortex-A architecture has added a special IVA module responsible for accelerating image processing. It allows you to quickly process the data coming from the sensor built-in camera. In addition, it is connected to the ISP and contributes to video acceleration. OMAP also uses PowerVR graphics.

The Apple A4 has a large 512KB cache, PowerVR graphics, and the ARM core itself is based on a variant of the architecture redesigned by Samsung.

The dual-core Apple A5, which debuted in the iPad 2 in early 2011, is based on the ARM Cortex-A9 architecture, just as it was optimized by Samsung last time. Compared to the A4, the new chip has twice the amount of L2 cache - it has been increased to 1 MB. The processor contains a dual-channel RAM controller and has an improved video block. As a result, its performance in some tasks is twice that of the Apple A4.

Marvell offers chips based on its own Sheeva architecture, which upon closer examination turns out to be a hybrid of XScale, once bought from Intel, and ARM. These chips have a large amount of cache memory compared to their counterparts and are equipped with a special multimedia module.

Currently, ARM licensees only produce chips based on the ARM Cortex-A9 architecture. At the same time, although it allows you to create quad-core variants, NVIDIA, Apple, Texas Instruments and others are still limited to models with one or two cores. In addition, the chips operate at frequencies up to 1.5 GHz. Cortex-A9 allows you to make two-GHz processors, but again, manufacturers do not strive to quickly increase frequencies - for now, the market will have enough dual-core processors at 1.5 GHz.

Processors based on Cortex-A15 should become truly multi-core, but if they are announced, then on paper. Their appearance in silicon should be expected next year.

Current Cortex-A9 based ARM licensee processors:

x86 - the main rival

x86 is a representative of CISC architectures. They use the full set of commands. One instruction in this case performs several low-level operations. Program code, unlike ARM, is more compact, but it does not run as fast and requires more resources. In addition, from the very beginning, x86s were equipped with all the necessary blocks, which suggested both their versatility and gluttony. Additional energy was spent on unconditional, parallel execution of commands. This allows you to achieve a speed advantage, but some operations are idle because they do not satisfy the previous conditions.

These were the classic x86, but starting with 80486, Intel de facto created an internal RISC core that executed CISC instructions, previously decomposed into simpler instructions. Modern Intel and AMD processors have the same design.

Windows 8 and ARM

ARM and x86 today differ less than 30 years ago, but are still based on different principles, which separates them into different niches of the processor market. The architectures might never have crossed if the computer itself hadn't changed.

Mobility and efficiency came to the fore, more attention was paid to smartphones and tablets. Apple makes a lot of money on mobile gadgets and the infrastructure tied to them. Microsoft does not want to be left behind and has been trying to gain a foothold in the tablet market for the second year. Google is doing quite well.

The desktop PC becomes first of all a working tool, the niche of a household computer is occupied by tablets and specialized devices. Under these conditions, Microsoft is going to take an unprecedented step. . It is not entirely clear what this will lead to. We will get two versions of the operating system, or one that will work with both architectures. Will Microsoft's x86 support bury ARM, or not?

There is little information yet. Microsoft demonstrated Windows 8 on an ARM-based device during CES 2011. Steve Ballmer showed that on the ARM platform, you can use Windows to watch videos, work with images, use the Internet - Internet Explorer even worked with hardware acceleration - connect a USB devices to print documents. Most important in this demo was having Microsoft Office running on ARM without a virtual machine. The presentation showed three gadgets based on Qualcomm, Texas Instruments and NVIDIA processors. Windows had a standard "seven" shell, but Microsoft representatives announced a new, redesigned system kernel.

However, Windows is not only an OS made by Microsoft engineers, it is also millions of programs. Some software is critical for people in many professions. For example, the Adobe CS package. Will the company support the ARM-Windows version of the software, or will the new kernel allow Photoshop and other popular applications to run on computers with NVIDIA Tegra or similar chips without additional code modifications?

In addition, there is a question with video cards. Now video cards for laptops are made by optimizing the power consumption of desktop graphics chips - they are architecturally the same. At the same time, now a video card is something like a "computer within a computer" - it has its own ultra-fast RAM and its own computing chip, which significantly outperforms conventional processors in specific tasks. It goes without saying that corresponding optimization of applications working with 3D graphics has been made for them. Yes, and various video editing programs and graphic editors (in particular, Photoshop from version CS4), and more recently, browsers also use GPU hardware acceleration.

Of course, in Android, MeeGo, BlackBerry OS, iOS and other mobile systems, the necessary optimization has been made for various mobile (more precisely, ultra-mobile) accelerators on the market. However, they are not supported in Windows. Drivers, of course, will be written (and already written - Intel Atom Z500 series processors are supplied with a chipset, where the "smartphone" graphics core PowerVR SGX 535 is integrated), but the optimization of applications for them may be late, if at all happens.

Obviously, "ARM on the desktop" won't really take off. Unless in low-power systems on which they will access the Internet and watch movies. On nettops in general. So ARM is only trying to take a swing at the niche that Intel Atom has occupied and where AMD is now actively pushing with its Brazos platform. And she seems to be part of it. Unless both processor companies "shoot" with something very competitive.

In places, Intel Atom and ARM are already competing. They are used to create network storages and low-power servers that can serve a small office or apartment. There are also several commercial cluster projects based on economical Intel chips. The characteristics of the new processors based on ARM Cortex-A9 allow them to be used to support the infrastructure. Thus, in a couple of years we can get ARM servers or ARM-NAS for small local networks, and the emergence of low-powered web servers cannot be ruled out.

First sparring

The main rival of ARM from the x86 side is the Intel Atom, and now you can add the . Comparison of x86 and ARM was conducted by Van Smith, who created test packages OpenSourceMark, miniBench and one of the co-authors of SiSoftware Sandra. Atom N450, Freescale i.MX515 (Cortex-A8), VIA Nano L3050 took part in the race. The frequencies of x86 chips were reduced, but they still had an advantage due to more advanced memory.

The results were very interesting. The ARM chip proved to be as fast as its competitors in integer operations, while consuming less power. There is nothing surprising here. Initially, the architecture was both quite fast and economical. In floating point operations, ARM lost x86. The traditionally powerful FPU block available for Intel and AMD chips has affected here. Recall that it appeared in ARM relatively recently. The tasks that fall on the FPU occupy a significant place in the life of a modern user - these are games, video and audio encoding, and other streaming operations. Of course, the tests conducted by Van Smith are no longer so relevant today. ARM has significantly increased the weaknesses of its architecture in the Cortex-A9 and especially Cortex-A15 versions, which, for example, can already execute instructions unconditionally, parallelizing the execution of tasks.

ARM outlook

So what architecture should you end up using, ARM or x86? It would be best to bet on both. Today we live in conditions of reformatting of the computer market. In 2008, netbooks predicted a bright future. Cheap compact laptops should have become the main computer for most users, especially against the background of the global crisis. But then the economic recovery began and the iPad came along. Tablets are now the kings of the market. However, the tablet is good as an entertainment console, but not very convenient to use primarily because of the touch input - writing this article on the iPad would be very difficult, and long. Will tablets stand the test of time? Perhaps in a couple of years we will come up with a new toy.

But still, in the mobile segment, where high performance is not required, and user activity is mainly limited to entertainment and not related to work, ARM looks preferable to x86. They provide an acceptable level of performance, as well as great battery life. Intel's attempts to bring Atom to mind have so far been unsuccessful. ARM sets a new bar for performance per watt. Most likely, ARM will be successful in compact mobile gadgets. In the netbook market, they can also become leaders, but here everything depends not so much on processor developers, but on Microsoft and Google. If the first implements normal ARM support in Windows 8, and the second will bring Chrome OS to mind. So far, smartbooks offered by Qualcomm have not made a market. Netbooks based on x86 survived.

A breakthrough in this direction, as planned by ARM, should be made by the Cortex-A15 architecture. The company recommends dual- and quad-core processors based on it with a frequency of 1.0-2.0 GHz for home entertainment systems that will combine a media player, 3D TV and Internet terminal. Quad-core chips with a frequency of 1.5-2.5 GHz can become the basis of home and web servers. Finally, the most ambitious use case for the Cortex-A15 is wireless infrastructure. It can use chips with four or more cores, with a frequency of 1.5-2.5 GHz.

But for now, these are just plans. Cortex-A15 was introduced to ARM in September last year. Cortex-A9 was shown by the company in October 2007, two years later the company presented the A9 version with the ability to increase the frequency of the chips up to 2.0 GHz. For comparison, NVIDIA Tegra 2 - one of the most popular solutions based on Cortex-A9 - was released only in January last year. Well, the first gadgets based on it, users were able to feel after another six months.

The segment of working PCs and high-performance solutions will remain for x86. This will not mean the death of the architecture, but in terms of money, Intel and AMD should prepare for the loss of part of the income that will go to the manufacturers of ARM processors.