processor

125
http://en.wikipedia.org/wiki/ARM_architecture "ARM" redirects here. For other uses, see ARM (disambiguation) . Logo ARM Designer ARM Holdings Bits 32/64 Introduce d 1983 Version ARMv8 [1] Design RISC Type Register-Register Encoding Fixed Branching Condition code Endiannes s Bi (Little as default) Extension s NEON, Thumb, Jazelle , VFP, A64 Registers 16/31 [2] ARM is a 32-bit reduced instruction set computer (RISC) instruction set architecture (ISA) developed by ARM Holdings . It was named the Advanced RISC Machine and, before that, the Acorn RISC Machine. The ARM architecture is the most widely used 32-bit instruction set architecture in numbers produced. [3] [4] Originally conceived by Acorn Computers for use in its personal computers ,

Upload: pritam-desai

Post on 24-Aug-2014

337 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Processor

http://en.wikipedia.org/wiki/ARM_architecture

"ARM" redirects here. For other uses, see ARM (disambiguation).

LogoARM

Designer ARM HoldingsBits 32/64Introduced 1983Version ARMv8[1]

Design RISCType Register-RegisterEncoding FixedBranching Condition codeEndianness Bi (Little as default)Extensions NEON, Thumb, Jazelle, VFP, A64

Registers16/31[2]

ARM is a 32-bit reduced instruction set computer (RISC) instruction set architecture (ISA) developed by ARM Holdings. It was named the Advanced RISC Machine and, before that, the Acorn RISC Machine. The ARM architecture is the most widely used 32-bit instruction set architecture in numbers produced.[3][4] Originally conceived by Acorn Computers for use in its personal computers, the first ARM-based products were the Acorn Archimedes range introduced in 1987.

Contents [hide] 

1 Features and applications 2 Licencees 3 History

o 3.1 Acorn RISC Machine: ARM2

Page 2: Processor

o 3.2 Apple, DEC, Intel, Marvell: ARM6, StrongARM, XScale o 3.3 Licensing

4 ARM cores 5 Example applications of ARM cores 6 Architecture

o 6.1 Instruction set 6.1.1 Conditional execution 6.1.2 Other features 6.1.3 Pipelines and other implementation issues 6.1.4 Coprocessors

o 6.2 Debugging o 6.3 DSP enhancement instructions o 6.4 Jazelle o 6.5 Thumb o 6.6 Thumb-2 o 6.7 Thumb Execution Environment (ThumbEE) o 6.8 VFP o 6.9 Advanced SIMD (NEON) o 6.10 Security Extensions (TrustZone) o 6.11 No-execute page protection

7 ARM licensees o 7.1 Approximate licensing costs

8 Operating systems o 8.1 Acorn systems o 8.2 Embedded operating systems o 8.3 Unix-like

8.3.1 Linux 8.3.2 BSD 8.3.3 Solaris

o 8.4 Windows 9 See also 10 References 11 Further reading 12 External links

[edit] Features and applicationsThe relative simplicity of ARM processors makes them suitable for low power applications. As a result, they have become dominant in the mobile and embedded electronics market, as relatively low-cost, small microprocessors and microcontrollers. In 2005, about 98% of the more than one billion mobile phones sold each year used at least one ARM processor.[5] As of 2009, ARM processors account for approximately 90% of all embedded 32-bit RISC processors[6] and are used extensively in consumer electronics, including personal digital assistants (PDAs), tablets,

Page 3: Processor

mobile phones, digital media and music players, hand-held game consoles, calculators and computer peripherals such as hard drives and routers.

[edit] LicenceesThe ARM architecture is licensable. Companies that are current or former ARM licensees include Alcatel-Lucent, Apple Inc., AppliedMicro, Atmel, Broadcom, Cirrus Logic, Digital Equipment Corporation, Ember, Energy Micro, Freescale, Intel (through DEC), LG, Marvell Technology Group, Microsemi, Microsoft, NEC, Nintendo, Nuvoton, Nvidia, Sony, NXP (formerly Philips), Oki, ON Semiconductor, Psion, Qualcomm, Samsung, Sharp, Silicon Labs, STMicroelectronics, Symbios Logic, Texas Instruments, VLSI Technology, Yamaha, Fuzhou Rockchip, and ZiiLABS.

ARM processors are developed by ARM and by ARM licensees. Prominent ARM processor families developed by ARM Holdings include the ARM7, ARM9, ARM11 and Cortex. Notable ARM processors developed by licensees include AppliedMicro X-Gene, DEC StrongARM, Freescale i.MX, Marvell (formerly Intel) XScale, Nvidia Tegra, ST-Ericsson Nova and NovaThor, Qualcomm Snapdragon, the Silicon Labs Precision32 MCU product line, the Texas Instruments OMAP product line, the Samsung Hummingbird and the Apple A4 and A5.

[edit] HistoryAfter achieving success with the BBC Micro computer, Acorn Computers Ltd considered how to move on from the relatively simple MOS Technology 6502 processor to address business markets like the one that would soon be dominated by the IBM PC, launched in 1981. The Acorn Business Computer (ABC) plan required a number of second processors to be made to work with the BBC Micro platform, but processors such as the Motorola 68000 and National Semiconductor 32016 were unsuitable, and the 6502 was not powerful enough for a graphics based user interface.[citation needed]

Acorn would need a new architecture, having tested all of the available processors and found them wanting. Acorn then seriously considered designing its own processor, and their engineers came across papers on the Berkeley RISC project. They felt it showed that if a class of graduate students could create a competitive 32-bit processor, then Acorn would have no problem. A trip to the Western Design Center in Phoenix, where the 6502 was being updated by what was effectively a single-person company, showed Acorn engineers Steve Furber [7] and Sophie Wilson that they did not need massive resources and state-of-the-art R&D facilities.

Wilson set about developing the instruction set, writing a simulation of the processor in BBC Basic that ran on a BBC Micro with a second 6502 processor. It convinced the Acorn engineers that they were on the right track. Before they could go any further, however, they would need more resources. It was time for Wilson to approach Acorn's CEO, Hermann Hauser, and explain what was afoot. Once the go-ahead had been given, a small team was put together to implement Wilson's model in hardware.

Page 4: Processor

A Conexant ARM processor used mainly in routers

[edit] Acorn RISC Machine: ARM2

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (March 2011)

The official Acorn RISC Machine project started in October 1983. VLSI Technology, Inc was chosen as silicon partner, since it already supplied Acorn with ROMs and some custom chips. The design was led by Wilson and Furber, and was consciously designed with a similar efficiency ethos as the 6502.[8] It had a key design goal of achieving low-latency input/output (interrupt) handling like the 6502. The 6502's memory access architecture had allowed developers to produce fast machines without the use of costly direct memory access hardware. VLSI produced the first ARM silicon on 26 April 1985 – it worked the first time and came to be termed ARM1 by April 1985.[9] The first "real" production systems named ARM2 were available the following year.

The ARM1 second processor for the BBC Micro

Its first practical application was as a second processor to the BBC Micro, where it was used to develop the simulation software to finish work on the support chips (VIDC, IOC, MEMC) and to speed up the operation of the CAD software used in developing ARM2. Wilson subsequently coded BBC Basic in ARM assembly language, and the in-depth knowledge obtained from designing the instruction set allowed the code to be very dense, making ARM BBC Basic an extremely good test for any ARM emulator. The original aim of a principally ARM-based computer was achieved in 1987 with the release of the Acorn Archimedes.

Page 5: Processor

Such was the secrecy surrounding the ARM CPU project that when Olivetti were negotiating to take a controlling share of Acorn in 1985, they were not told about the development team until after the negotiations had been finalised.

In 1992 Acorn once more won the Queen's Award for Technology for the ARM.

The ARM2 featured a 32-bit data bus, a 26-bit address space and twenty-seven 32-bit registers. Program code had to lie within the first 64 Mbyte of the memory, as the program counter was limited to 24 bits because the top 6 and bottom 2 bits of the 32-bit register served as status flags. The ARM2 was possibly the simplest useful 32-bit microprocessor in the world, with only 30,000 transistors (compare the transistor count with Motorola's six-year older 68000 model which was aptly named, since it contained 68,000 transistors). Much of this simplicity comes from not having microcode (which represents about one-quarter to one-third of the 68000) and, like most CPUs of the day, not including any cache. This simplicity led to its low power usage, while performing better than the Intel 80286.[10] A successor, ARM3, was produced with a 4 KB cache, which further improved performance.

[edit] Apple, DEC, Intel, Marvell: ARM6, StrongARM, XScale

In the late 1980s Apple Computer and VLSI Technology started working with Acorn on newer versions of the ARM core. The work was so important that Acorn spun off the design team in 1990 into a new company called Advanced RISC Machines Ltd. Advanced RISC Machines became ARM Ltd when its parent company, ARM Holdings plc, floated on the London Stock Exchange and NASDAQ in 1998.[11]

The new Apple-ARM work would eventually turn into the ARM6, first released in early 1992. Apple used the ARM6-based ARM 610 as the basis for their Apple Newton PDA. In 1994, Acorn used the ARM 610 as the main central processing unit (CPU) in their Risc PC computers. DEC licensed the ARM6 architecture and produced the StrongARM. At 233 MHz this CPU drew only one watt (more recent versions draw far less). This work was later passed to Intel as a part of a lawsuit settlement, and Intel took the opportunity to supplement their aging i960 line with the StrongARM. Intel later developed its own high performance implementation named XScale which it has since sold to Marvell.

[edit] Licensing

The ARM core has remained largely the same size throughout these changes. ARM2 had 30,000 transistors, while the ARM6 grew to only 35,000. ARM's business has always been to sell IP cores, which licensees use to create microcontrollers and CPUs based on this core. The original design manufacturer combines the ARM core with a number of optional parts to produce a complete CPU, one that can be built on old semiconductor fabs and still deliver substantial performance at a low cost. The most successful implementation has been the ARM7TDMI with hundreds of millions sold. Atmel has been a precursor design center in the ARM7TDMI-based embedded system.

Page 6: Processor

ARM licensed about 1.6 billion cores in 2005. In 2005, about 1 billion ARM cores went into mobile phones.[12] By January 2008, over 10 billion ARM cores had been built, and in 2008 iSuppli predicted that by 2011, 5 billion ARM cores will be shipping per year.[13] As of January 2011, ARM states that over 15 billion ARM processors have shipped.[14]

The ARM architectures used in smartphones, personal digital assistants and other mobile devices range from ARMv5, in obsolete/low-end devices, to the ARM M-series, in current high-end devices. XScale and ARM926 processors are ARMv5TE, and are now more numerous in high-end devices than the StrongARM, ARM9TDMI and ARM7TDMI based ARMv4 processors, but lower-end devices may use older cores with lower licensing costs. ARMv6 processors represented a step up in performance from standard ARMv5 cores, and are used in some cases, but Cortex processors (ARMv7) now provide faster and more power-efficient options than all those prior generations. Cortex-A targets applications processors, as needed by smartphones that formerly used ARM9 or ARM11. Cortex-R targets real-time applications, and Cortex-M targets microcontrollers.

In 2009, some manufacturers introduced netbooks based on ARM architecture CPUs, in direct competition with netbooks based on Intel Atom.[15] According to analyst firm IHS iSuppli, by 2015, ARM ICs are estimated to be in 23% of all laptops.[16]

In 2011, HiSilicon Technologies Co. Ltd. licensed a variety of ARM technology to be used in communications chip designs. These included 3G/4G basestations, networking infrastructure and mobile computing applications.[17]

[edit] ARM coresMain article: List of ARM microprocessor coresArchitecture FamilyARMv1 ARM1ARMv2 ARM2, ARM3ARMv3 ARM6, ARM7ARMv4 StrongARM, ARM7TDMI, ARM9TDMIARMv5 ARM7EJ, ARM9E, ARM10E, XScaleARMv6 ARM11, ARM Cortex-MARMv7 ARM Cortex-A, ARM Cortex-M, ARM Cortex-RARMv8 No cores available yet. Will support 64-bit data and addressing [18][19]

A summary of the numerous vendors who implement ARM cores in their design is provided by ARM.[20]

[edit] Example applications of ARM coresMain article: List of applications of ARM cores

Page 7: Processor

ARM cores are used in a number of products, particularly various smartphones. Some computing examples are the Acorn Archimedes, Apple iPad and ASUS Eee Pad Transformer. Some other uses are the Apple iPod portable media player, Canon PowerShot A470 digital camera, Nintendo DS handheld games console and TomTom automotive navigation system.

Since 2005, ARM was also involved in Manchester University's computer, SpiNNaker, which used ARM cores to simulate the human brain.[21]

[edit] Architecture

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (March 2011)

From 1995, the ARM Architecture Reference Manual has been the primary source of documentation on the ARM processor architecture and instruction set, distinguishing interfaces that all ARM processors are required to support (such as instruction semantics) from implementation details that may vary. The architecture has evolved over time, and starting with the Cortex series of cores, three "profiles" are defined:

"Application" profile: Cortex-A series "Real-time" profile: Cortex-R series "Microcontroller" profile: Cortex-M series

Profiles are allowed to subset the architecture. For example, the ARMv6-M profile (used by the Cortex-M0) is a subset of the ARMv7-M profile (it supports fewer instructions).

[edit] Instruction set

To keep the design clean, simple and fast, the original ARM implementation was hardwired without microcode, like the much simpler 8-bit 6502 processor used in prior Acorn microcomputers.

The ARM architecture includes the following RISC features:

Load/store architecture. No support for misaligned memory accesses (although now supported in ARMv6 cores,

with some exceptions related to load/store multiple word instructions). Uniform 16 × 32-bit register file. Fixed instruction width of 32 bits to ease decoding and pipelining, at the cost of

decreased code density. Later, the Thumb instruction set increased code density. Mostly single-cycle execution.

To compensate for the simpler design, compared with contemporary processors like the Intel 80286 and Motorola 68020, some additional design features were used:

Page 8: Processor

Conditional execution of most instructions, reducing branch overhead and compensating for the lack of a branch predictor.

Arithmetic instructions alter condition codes only when desired. 32-bit barrel shifter which can be used without performance penalty with most arithmetic

instructions and address calculations. Powerful indexed addressing modes. A link register for fast leaf function calls. Simple, but fast, 2-priority-level interrupt subsystem with switched register banks.

[edit] Conditional execution

The conditional execution feature (called predication) is implemented with a 4-bit condition code selector (the predicate) on every instruction; one of the four-bit codes is reserved as an "escape code" to specify certain unconditional instructions, but nearly all common instructions are conditional. Most CPU architectures only have condition codes on branch instructions.

This cuts down significantly on the encoding bits available for displacements in memory access instructions, but on the other hand it avoids branch instructions when generating code for small if statements . The standard example of this is the subtraction-based Euclidean algorithm: ARM address mode

In the C programming language, the loop is:

while(i != j) { if (i > j) i -= j; else j -= i; }

In ARM assembly, the loop is:

loop CMP Ri, Rj ; set condition "NE" if (i != j), ; "GT" if (i > j), ; or "LT" if (i < j) SUBGT Ri, Ri, Rj ; if "GT" (greater than), i = i-j; SUBLT Rj, Rj, Ri ; if "LT" (less than), j = j-i; BNE loop ; if "NE" (not equal), then loop

which avoids the branches around the then and else clauses. Note that if Ri and Rj are equal then neither of the SUB instructions will be executed, optimising out the need for a conditional branch to implement the while check at the top of the loop, for example had SUBLE (less than or equal) been used.

One of the ways that Thumb code provides a more dense encoding is to remove that four bit selector from non-branch instructions.

[edit] Other features

Page 9: Processor

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (March 2011)

Another feature of the instruction set is the ability to fold shifts and rotates into the "data processing" (arithmetic, logical, and register-register move) instructions, so that, for example, the C statement

a += (j << 2);

could be rendered as a single-word, single-cycle instruction on the ARM.

ADD Ra, Ra, Rj, LSL #2

This results in the typical ARM program being denser than expected with fewer memory accesses; thus the pipeline is used more efficiently. Even though the ARM runs at what many would consider to be low speeds, it nevertheless competes quite well with much more complex CPU designs.[citation needed]

The ARM processor also has some features rarely seen in other RISC architectures, such as PC-relative addressing (indeed, on the 32-bit[1] ARM the PC is one of its 16 registers) and pre- and post-increment addressing modes.

Another item of note is that the ARM has been around for a while, with the instruction set increasing somewhat over time. Some early ARM processors (before ARM7TDMI), for example, have no instruction to store a two-byte quantity, thus, strictly speaking, for them it's not possible to generate efficient code that would behave the way one would expect for C objects of type "int16_t".

[edit] Pipelines and other implementation issues

The ARM7 and earlier implementations have a three stage pipeline; the stages being fetch, decode, and execute. Higher performance designs, such as the ARM9, have deeper pipelines: Cortex-A8 has thirteen stages. Additional implementation changes for higher performance include a faster adder, and more extensive branch prediction logic. The difference between the ARM7DI and ARM7DMI cores, for example, was an improved multiplier (hence the added "M").

[edit] Coprocessors

For those familiar with the Intel x86 family of CPUs, the ARM family of processors does not support or have any instructions similar to CPUID. There are, however, mechanisms for addressing coprocessors in the ARM architecture.

The ARM architecture provides a non-intrusive way of extending the instruction set using "coprocessors" which can be addressed using MCR, MRC, MRRC, MCRR, and similar instructions. The coprocessor space is divided logically into 16 coprocessors with numbers from

Page 10: Processor

0 to 15, coprocessor 15 (cp15) being reserved for some typical control functions like managing the caches and MMU operation (on processors that have one).

In ARM-based machines, peripheral devices are usually attached to the processor by mapping their physical registers into ARM memory space or into the coprocessor space or connecting to another device (a bus) which in turn attaches to the processor. Coprocessor accesses have lower latency so some peripherals (for example an XScale interrupt controller) are designed to be accessible in both ways (through memory and through coprocessors).

In other cases, chip designers only integrate hardware using the coprocessor mechanism. For example, an image processing engine might be a small ARM7TDMI core combined with a coprocessor that has specialized operations to support a specific set of HDTV transcoding primitives.

[edit] Debugging

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (March 2011)

All modern ARM processors include hardware debugging facilities; without them, software debuggers could not perform basic operations like halting, stepping, and breakpointing of code starting from reset. These facilities are built using JTAG support, though some newer cores optionally support ARM's own two-wire "SWD" protocol. In ARM7TDMI cores, the "D" represented JTAG debug support, and the "I" represented presence of an "EmbeddedICE" debug module. For ARM7 and ARM9 core generations, EmbeddedICE over JTAG was a de-facto debug standard, although it was not architecturally guaranteed.

The ARMv7 architecture defines basic debug facilities at an architectural level. These include breakpoints, watchpoints, and instruction execution in a "Debug Mode"; similar facilities were also available with EmbeddedICE. Both "halt mode" and "monitor" mode debugging are supported. The actual transport mechanism used to access the debug facilities is not architecturally specified, but implementations generally include JTAG support.

There is a separate ARM "CoreSight" debug architecture, which is not architecturally required by ARMv7 processors.

[edit] DSP enhancement instructions

To improve the ARM architecture for digital signal processing and multimedia applications, a few new instructions were added to the set.[22] These are signified by an "E" in the name of the ARMv5TE and ARMv5TEJ architectures. E-variants also imply T,D,M and I.

The new instructions are common in digital signal processor architectures. They are variations on signed multiply–accumulate, saturated add and subtract, and count leading zeros.

Page 11: Processor

[edit] Jazelle

Main article: Jazelle

Jazelle is a technique that allows Java Bytecode to be executed directly in the ARM architecture as a third execution state (and instruction set) alongside the existing ARM and Thumb-mode. Support for this state is signified by the "J" in the ARMv5TEJ architecture, and in ARM9EJ-S and ARM7EJ-S core names. Support for this state is required starting in ARMv6 (except for the ARMv7-M profile), although newer cores only include a trivial implementation that provides no hardware acceleration.

[edit] Thumb

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (March 2011)

To improve compiled code-density, processors since the ARM7TDMI have featured Thumb instruction set, which have their own state. (The "T" in "TDMI" indicates the Thumb feature.) When in this state, the processor executes the Thumb instruction set, a compact 16-bit encoding for a subset of the ARM instruction set.[23] Most of the Thumb instructions are directly mapped to normal ARM instructions. The space-saving comes from making some of the instruction operands implicit and limiting the number of possibilities compared to the ARM instructions executed in the ARM instruction set state.

In Thumb, the 16-bit opcodes have less functionality. For example, only branches can be conditional, and many opcodes are restricted to accessing only half of all of the CPU's general purpose registers. The shorter opcodes give improved code density overall, even though some operations require extra instructions. In situations where the memory port or bus width is constrained to less than 32 bits, the shorter Thumb opcodes allow increased performance compared with 32-bit ARM code, as less program code may need to be loaded into the processor over the constrained memory bandwidth.

Embedded hardware, such as the Game Boy Advance, typically have a small amount of RAM accessible with a full 32-bit datapath; the majority is accessed via a 16 bit or narrower secondary datapath. In this situation, it usually makes sense to compile Thumb code and hand-optimise a few of the most CPU-intensive sections using full 32-bit ARM instructions, placing these wider instructions into the 32-bit bus accessible memory.

The first processor with a Thumb instruction decoder was the ARM7TDMI. All ARM9 and later families, including XScale, have included a Thumb instruction decoder.

[edit] Thumb-2

Thumb-2 technology made its debut in the ARM1156 core, announced in 2003. Thumb-2 extends the limited 16-bit instruction set of Thumb with additional 32-bit instructions to give the

Page 12: Processor

instruction set more breadth, thus producing a variable-length instruction set. A stated aim for Thumb-2 is to achieve code density similar to Thumb with performance similar to the ARM instruction set on 32-bit memory. In ARMv7 this goal can be said to have been met.[citation needed]

Thumb-2 extends both the ARM and Thumb instruction set with yet more instructions, including bit-field manipulation, table branches, and conditional execution. A new "Unified Assembly Language" (UAL) supports generation of either Thumb-2 or ARM instructions from the same source code; versions of Thumb seen on ARMv7 processors are essentially as capable as ARM code (including the ability to write interrupt handlers). This requires a bit of care, and use of a new "IT" (if-then) instruction, which permits up to four successive instructions to execute based on a tested condition. When compiling into ARM code this is ignored, but when compiling into Thumb-2 it generates an actual instruction. For example:

; if (r0 == r1)CMP r0, r1ITE EQ ; ARM: no code ... Thumb: IT instruction; then r0 = r2;MOVEQ r0, r2 ; ARM: conditional; Thumb: condition via ITE 'T' (then); else r0 = r3;MOVNE r0, r3 ; ARM: conditional; Thumb: condition via ITE 'E' (else); recall that the Thumb MOV instruction has no bits to encode "EQ" or "NE"

All ARMv7 chips support the Thumb-2 instruction set. Other chips in the Cortex and ARM11 series support both "ARM instruction set state" and "Thumb-2 instruction set state".[24][25][26]

[edit] Thumb Execution Environment (ThumbEE)

ThumbEE, also termed Thumb-2EE, and marketed as Jazelle RCT (Runtime Compilation Target), was announced in 2005, first appearing in the Cortex-A8 processor. ThumbEE is a fourth processor mode, making small changes to the Thumb-2 extended Thumb instruction set. These changes make the instruction set particularly suited to code generated at runtime (e.g. by JIT compilation) in managed Execution Environments. ThumbEE is a target for languages such as Limbo, Java, C#, Perl and Python, and allows JIT compilers to output smaller compiled code without impacting performance.

New features provided by ThumbEE include automatic null pointer checks on every load and store instruction, an instruction to perform an array bounds check, access to registers r8-r15 (where the Jazelle/DBX Java VM state is held), and special instructions that call a handler.[27] Handlers are small sections of frequently called code, commonly used to implement a feature of a high level language, such as allocating memory for a new object. These changes come from repurposing a handful of opcodes, and knowing the core is in the new ThumbEE mode.

[edit] VFP

VFP (Vector Floating Point) technology is an FPU coprocessor extension to the ARM architecture. It provides low-cost single-precision and double-precision floating-point computation fully compliant with the ANSI/IEEE Std 754-1985 Standard for Binary Floating-Point Arithmetic. VFP provides floating-point computation suitable for a wide spectrum of

Page 13: Processor

applications such as PDAs, smartphones, voice compression and decompression, three-dimensional graphics and digital audio, printers, set-top boxes, and automotive applications. The VFP architecture was intended to support execution of short "vector mode" instructions but these operated on each vector element sequentially and thus did not offer the performance of true single instruction, multiple data (SIMD) vector parallelism. This vector mode was therefore removed shortly after its introduction,[28] to be replaced with the much more powerful NEON Advanced SIMD unit.

Some devices such as the ARM Cortex-A8 have a cut-down VFPLite module instead of a full VFP module, and require roughly ten times more clock cycles per float operation.[29] Other floating-point and/or SIMD coprocessors found in ARM-based processors include FPA, FPE, iwMMXt. They provide some of the same functionality as VFP but are not opcode-compatible with it.

[edit] Advanced SIMD (NEON)

The Advanced SIMD extension (aka NEON or "MPE" Media Processing Engine) is a combined 64- and 128-bit single instruction multiple data (SIMD) instruction set that provides standardized acceleration for media and signal processing applications. NEON is included in all Cortex-A8 devices but is optional in Cortex-A9 devices.[30] NEON can execute MP3 audio decoding on CPUs running at 10 MHz and can run the GSM adaptive multi-rate (AMR) speech codec at no more than 13 MHz. It features a comprehensive instruction set, separate register files and independent execution hardware.[31] NEON supports 8-, 16-, 32- and 64-bit integer and single-precision (32-bit) floating-point data and operates in SIMD operations for handling audio and video processing as well as graphics and gaming processing. In NEON, the SIMD supports up to 16 operations at the same time. The NEON hardware shares the same floating-point registers as used in VFP. Devices such as the ARM Cortex-A8 and Cortex-A9 support 128-bit vectors but will execute with just 64 bits at a time,[29] whereas newer Cortex-A15 devices can execute 128 bits at once.

[edit] Security Extensions (TrustZone)

The Security Extensions, marketed as TrustZone Technology, is found in ARMv6KZ and later application profile architectures. It provides a low cost alternative to adding an additional dedicated security core to an SoC, by providing two virtual processors backed by hardware based access control. This enables the application core to switch between two states, referred to as worlds (to reduce confusion with other names for capability domains), in order to prevent information from leaking from the more trusted world to the less trusted world. This world switch is generally orthogonal to all other capabilities of the processor, thus each world can operate independently of the other while using the same core. Memory and peripherals are then made aware of the operating world of the core and may use this to provide access control to secrets and code on the device.

Typical applications of TrustZone Technology are to run a rich operating system in the less trusted world, and smaller security-specialized code in the more trusted world (named TrustZone Software, a TrustZone optimized version of the Trusted Foundations Software developed by

Page 14: Processor

Trusted Logic), allowing much tighter digital rights management for controlling the use of media on ARM-based devices,[32] and preventing any unapproved use of the device.

In practice, since the specific implementation details of TrustZone are proprietary and have not been publicly disclosed for review, it is unclear what level of assurance is provided for a given threat model.

[edit] No-execute page protection

As of ARMv6, the ARM architecture supports no-execute page protection, which is referred to as XN, for eXecute Never.[33]

[edit] ARM licensees

This section needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed. (March 2011)

ARM Ltd does not manufacture and sell CPU devices based on its own designs, but rather, licenses the processor architecture to interested parties. ARM offers a variety of licensing terms, varying in cost and deliverables. To all licensees, ARM provides an integratable hardware description of the ARM core, as well as complete software development toolset (compiler, debugger, SDK), and the right to sell manufactured silicon containing the ARM CPU.

Fabless licensees, who wish to integrate an ARM core into their own chip design, are usually only interested in acquiring a ready-to-manufacture verified IP core. For these customers, ARM delivers a gate netlist description of the chosen ARM core, along with an abstracted simulation model and test programs to aid design integration and verification. More ambitious customers, including integrated device manufacturers (IDM) and foundry operators, choose to acquire the processor IP in synthesizable RTL (Verilog) form. With the synthesizable RTL, the customer has the ability to perform architectural level optimizations and extensions. This allows the designer to achieve exotic design goals not otherwise possible with an unmodified netlist (high clock speed, very low power consumption, instruction set extensions, etc.). While ARM does not grant the licensee the right to resell the ARM architecture itself, licensees may freely sell manufactured product (chip devices, evaluation boards, complete systems, etc.). Merchant foundries can be a special case; not only are they allowed to sell finished silicon containing ARM cores, they generally hold the right to re-manufacture ARM cores for other customers.

Like most IP vendors, ARM prices its IP based on perceived value. In architectural terms, lower performing ARM cores command lower license costs than higher performing cores. In implementation terms, a synthesizable core costs more than a hard macro (blackbox) core. Complicating price matters, a merchant foundry which holds an ARM license (such as Samsung and Fujitsu) can offer reduced licensing costs to its fab customers. In exchange for acquiring the ARM core through the foundry's in-house design services, the customer can reduce or eliminate payment of ARM's upfront license fee. Compared to dedicated semiconductor foundries (such as TSMC and UMC) without in-house design services, Fujitsu/Samsung charge 2 to 3 times more

Page 15: Processor

per manufactured wafer. For low to mid volume applications, a design service foundry offers lower overall pricing (through subsidization of the license fee). For high volume mass produced parts, the long term cost reduction achievable through lower wafer pricing reduces the impact of ARM's NRE (Non-Recurring Engineering) costs, making the dedicated foundry a better choice.

Many semiconductor or IC design firms hold ARM licenses: Analog Devices, AppliedMicro, Atmel, Broadcom, Cirrus Logic, Energy Micro, Faraday Technology, Freescale, Fujitsu, Intel (through its settlement with Digital Equipment Corporation), IBM, Infineon Technologies, Marvell Technology Group, Nintendo, NXP Semiconductors, OKI, Qualcomm, Samsung, Sharp, STMicroelectronics, and Texas Instruments are some of the many companies who have licensed the ARM in one form or another.

[edit] Approximate licensing costs

ARM's 2006 annual report and accounts state that royalties totalling £88.7 million ($164.1 million) were the result of licensees shipping 2.45 billion units.[34] This is equivalent to £0.036 ($0.067) per unit shipped. However, this is averaged across all cores, including expensive new cores and inexpensive older cores.

In the same year ARM's licensing revenues for processor cores were £65.2 million (US$119.5 million),[35] in a year when 65 processor licenses were signed,[36] an average of £1 million ($1.84 million) per license. Again, this is averaged across both new and old cores.

Given that ARM's 2006 income from processor cores was approximately 60% from royalties and 40% from licenses, ARM makes the equivalent of £0.06 ($0.11) per unit shipped including both royalties and licenses. However, as one-off licenses are typically bought for new technologies, unit sales (and hence royalties) are dominated by more established products. Hence, the figures above do not reflect the true costs of any single ARM product.

[edit] Operating systems

Page 16: Processor

Android, a popular operating system running on the ARM architectureThis section relies on references to primary sources or sources affiliated with the subject, rather than references from independent authors and third-party publications. Please add citations from reliable sources. (September 2011)

[edit] Acorn systems

The very first ARM-based Acorn Archimedes personal computers ran an interim operating system called Arthur, which evolved into RISC OS, used on later ARM-based systems from Acorn and other vendors.

[edit] Embedded operating systems

The ARM architecture is supported by a large number of embedded and real-time operating systems, including Windows CE, .NET Micro Framework, Symbian, ChibiOS/RT, FreeRTOS, eCos, Integrity, Nucleus PLUS, MicroC/OS-II, QNX, RTEMS, BRTOS, RTXC Quadros, ThreadX, Unison Operating System, uTasker, VxWorks, MQX and OSE.[37]

[edit] Unix-like

The ARM architecture is supported by Unix and Unix-like operating systems such as:

Android Bada

Page 17: Processor

BSD iOS Linux Plan 9 from Bell Labs

o Inferno Solaris webOS

[edit] Linux

The following Linux distributions support ARM processors:

APEXAR PDK [38] Android [39] Arch Linux Arm [40] Ångström [41] CRUX ARM [42] BackTrack Chrome OS [43] DSLinux [citation needed]

Debian [44] ELinOS [45] Fedora [46] Gentoo [47] GoboLinux [48] iPodLinux [citation needed]

Maemo [citation needed]

MeeGo [citation needed]

Mer [49] MontaVista [50] Slackware [51] T2 SDE [52] TimeSys [53] Ubuntu [54] [55] webOS [citation needed]

Wind River Linux [56]

[edit] BSD

The following BSD derivatives support ARM processors:

RISC iX (Acorn ARM2/ARM3-based systems only)

FreeBSD [57] NetBSD [58] OpenBSD [59] iOS

[edit] Solaris

OpenSolaris [60]

[edit] Windows

Microsoft announced on 5 January 2011 that the next major version of the Windows NT family (Windows 8)[61] will include support for ARM processors. Microsoft demonstrated a preliminary version of Windows (version 6.2.7867) running on an ARM-based computer at the 2011 Consumer Electronics Show.[62] The ARM architecture is also supported by Microsoft's mobile operating systems, Windows Phone and Windows Mobile. ARM is also supported on Microsoft's

Page 18: Processor

Embedded OS, Windows Embedded CE which is now called Windows Embedded Compact. This latest version supports ARMv 5,6 and 7. Windows CE 5 is the underlying OS for Windows Mobile and Windows Embedded Compact 7 is the underlying OS for Windows Phone 7. The smaller Microsoft OS .NET Microframework uses ARM exclusively.

[edit] See also

Electronics portal

AMULET , family of asynchronous ARMs ARMulator, ARM Instruction Set Simulator ARMware , a virtual machine that emulates an ARM-based PDA. QEMU , a virtual machine which supports some ARM processors cores SkyEye simulator – an open source ARM Instruction Set Simulator Amber (processor core) – open source ARM core developed in Verilog HDL. Smartbook Symbian Windows CE and Windows Phone iOS Inferno Android Plug computer architectures Linaro Raspberry Pi , ARM11-based computer

http://en.wikipedia.org/wiki/System-on-a-chip

System on a chipA system on a chip or system on chip (SoC or SOC) is an integrated circuit (IC) that integrates all components of a computer or other electronic system into a single chip. It may contain digital, analog, mixed-signal, and often radio-frequency functions—all on a single chip substrate. A typical application is in the area of embedded systems.

The contrast with a microcontroller is one of degree. Microcontrollers typically have under 100 kB of RAM (often just a few kilobytes) and often really are single-chip-systems, whereas the term SoC is typically used with more powerful processors, capable of running software such as the desktop versions of Windows and Linux, which need external memory chips (flash, RAM) to

Page 19: Processor

be useful, and which are used with various external peripherals. In short, for larger systems system on a chip is hyperbole, indicating technical direction more than reality: increasing chip integration to reduce manufacturing costs and to enable smaller systems. Many interesting systems are too complex to fit on just one chip built with a process optimized for just one of the system's tasks.

When it is not feasible to construct an SoC for a particular application, an alternative is a system in package (SiP) comprising a number of chips in a single package. In large volumes, SoC is believed to be more cost-effective than SiP since it increases the yield of the fabrication and because its packaging is simpler.[1]

Another option, as seen for example in higher end cell phones and on the Beagle Board, is package on package stacking during board assembly. The SoC chip includes processors and numerous digital peripherals, and comes in a ball grid package with lower and upper connections. The lower balls connect to the board and various peripherals, with the upper balls in a ring holding the memory buses used to access NAND flash and DDR2 RAM. Memory packages could come from multiple vendors.

Contents [hide] 

1 Structure 2 Design flow 3 Fabrication 4 Books 5 See also 6 Notes 7 External links

[edit] StructureA typical SoC consists of:

A microcontroller, microprocessor or DSP core(s). Some SoCs—called multiprocessor system on chip (MPSoC)—include more than one processor core.

Memory blocks including a selection of ROM, RAM, EEPROM and flash memory. Timing sources including oscillators and phase-locked loops. Peripherals including counter-timers, real-time timers and power-on reset generators. External interfaces including industry standards such as USB, FireWire, Ethernet,

USART, SPI. Analog interfaces including ADCs and DACs. Voltage regulators and power management circuits.

Page 20: Processor

These blocks are connected by either a proprietary or industry-standard bus such as the AMBA bus from ARM Holdings. DMA controllers route data directly between external interfaces and memory, bypassing the processor core and thereby increasing the data throughput of the SoC.

Microcontroller-based system on a chip

[edit] Design flowAn SoC consists of both the hardware described above, and the software that controls the microcontroller, microprocessor or DSP cores, peripherals and interfaces. The design flow for an SoC aims to develop this hardware and software in parallel.

Most SoCs are developed from pre-qualified hardware blocks for the hardware elements described above, together with the software drivers that control their operation. Of particular importance are the protocol stacks that drive industry-standard interfaces like USB. The hardware blocks are put together using CAD tools; the software modules are integrated using a software-development environment.

Chips are verified for logical correctness before being sent to foundry. This process is called functional verification and it accounts for a significant portion of the time and energy expended in the chip design life cycle (although the often quoted figure of 70% is probably an exaggeration).[2] With the growing complexity of chips, hardware verification languages like SystemVerilog, SystemC, e, and OpenVera are being used. Bugs found in the verification stage are reported to the designer.

Page 21: Processor

System-on-a-chip design flow

Often, one step in the verification flow is emulation: The hardware is mapped onto an emulation platform based on a field-programmable gate array (FPGA) that mimics the behavior of the SoC, and the software modules are loaded into the memory of the emulation platform. Once programmed, the emulation platform enables the hardware and software of the SoC to be tested and debugged at close to its full operational speed. Emulation is generally preceded by extensive software simulation. In fact, sometimes the FPGAs are used primarily to speed up some parts of the simulation work.

After emulation the hardware of the SoC follows the place-and-route phase of the design of an integrated circuit before it is fabricated.

[edit] FabricationSoCs can be fabricated by several technologies, including:

Full custom Standard cell FPGA

SoC designs usually consume less power and have a lower cost and higher reliability than the multi-chip systems that they replace. And with fewer packages in the system, assembly costs are reduced as well.

Page 22: Processor

However, like most VLSI designs, the total cost is higher for one large chip than for the same functionality distributed over several smaller chips, because of lower yields and higher NRE costs.

[edit] Books (2003) Wael Badawy, Graham Jullien (2003). System-on-chip for real-time applications.

Kluwer. ISBN 1402072546, 9781402072543. 465 pages Furber, Stephen B. (2000). ARM system-on-chip architecture. Boston: Addison-Wesley.

ISBN 0-201-67519-6.

[edit] See also List of system-on-a-chip suppliers PSoC ASIC Microcontroller Electronic design automation Post silicon validation System in package Single-board computer Network On Chip Radio-on-a-chip

[edit] Notes1. ̂ "The Great Debate: SOC vs. SIP". EE Times. http://www.eetimes.com/electronics-

news/4052047/The-Great-Debate-SOC-vs-SIP. Retrieved 2009-08-12.2. ̂ "Is verification really 70 percent?". Eetimes.com.

http://www.eetimes.com/showArticle.jhtml?articleID=21700028. Retrieved 2009-08-12.

[edit] External links SOCC Annual IEEE International SOC Conference

[hide]

v t e

CPU technologiesArchitecture ISA  : CISC

EDGE

Page 23: Processor

EPIC MISC OISC RISC VLIW NISC ZISC Harvard architecture von Neumann architecture 1-bit 4-bit 8-bit 12-bit 16-bit 18-bit 24-bit 31-bit 32-bit 33-bit 36-bit 48-bit 60-bit 64-bit 128-bit Comparison of CPU architectures

Parallelism

Pipeline

Instruction pipelining In-order & out-of-order execution Register renaming Speculative execution Hazards

Level

Bit Instruction Superscalar Data Task

Threads

Multithreading Simultaneous multithreading Hyperthreading Superthreading

Flynn's taxonomy

SISD SIMD

Page 24: Processor

MISD MIMD

Types

Digital signal processor Microcontroller System-on-a-chip Vector processor

Components

Arithmetic logic unit (ALU) Barrel shifter Floating-point unit (FPU) Back-side bus Multiplexer Demultiplexer Registers Memory management unit (MMU) Translation lookaside buffer (TLB) Cache Register file Microcode Control unit Clock rate

Power managemen

t

APM ACPI Dynamic frequency scaling Dynamic voltage scaling Clock gating

http://en.wikipedia.org/wiki/Nvidia_Tegra

Tegra, developed by Nvidia, is a system on a chip series for mobile devices such as smartphones, personal digital assistants, and mobile Internet devices. The Tegra integrates the ARM architecture processor central processing unit (CPU), graphics processing unit (GPU), northbridge, southbridge, and memory controller onto one package. The series emphasizes low power consumption and high performance for playing audio and video.

Contents [hide] 

1 History

Page 25: Processor

2 Specifications o 2.1 Tegra APX series

2.1.1 Tegra APX 2500 2.1.2 Tegra APX 2600

o 2.2 Tegra 6xx series 2.2.1 Tegra 600 2.2.2 Tegra 650

o 2.3 Tegra 2 serieso 2.4 Tegra 3 (Kal-El) serieso 2.5 Tegra (Wayne) serieso 2.6 Tegra (Grey) serieso 2.7 Tegra (Logan) serieso 2.8 Tegra (Stark) series

3 Similar platforms 4 See also 5 References 6 External links

[edit] HistoryThe Tegra APX 2500 was announced on February 12, 2008; the Tegra 6xx product line was revealed on June 2, 2008[1]and the APX 2600 was announced in February 2009. The APX chips were designed for smartphones, while the Tegra 600 and 650 chips were intended for smartbooks and mobile Internet devices (MID).[2]

The first product to use the Tegra was Microsoft's Zune HD media player in September 2009, followed by the Samsung M1.[3] Microsoft's KIN was the first cellular phone to use the Tegra,[4] however the phone does not have an app store, so the Tegra's power does not provide much advantage. In September 2008, Nvidia and Opera Software announced that they will produce a version of the Opera 9.5 browser optimised for the Tegra on Windows Mobile and Windows CE.[5][6] At Mobile World Congress 2009, Nvidia introduced its port of Google's Android to the Tegra.

On January 7, 2010, Nvidia officially announced and demonstrated its next generation Tegra system-on-a-chip, the Nvidia Tegra 250, at Consumer Electronics Show 2010.[7] Nvidia primarily supports Android on Tegra 2, but booting other ARM-supporting operating systems is possible on devices where the bootloader is accessible. Tegra 2 support for the Ubuntu GNU/Linux distribution was also announced on the Nvidia developer forum.[8]

On February 15, 2011, Nvidia announced the first quad-core SoC that will be used in many of the tablets to be released in the second half of 2011. The announcement was made at the 2011 Mobile World Congress event in Barcelona. Though the chip has currently been codenamed Kal-El, it will likely be branded as Tegra 3. Early benchmark results show impressive gains over Tegra 2. Nvidia initially claimed that Tegra 3 could outperform a Core 2 Duo processor from Intel, and released benchmarks with an underclocked Tegra 3 to that effect; later investigations

Page 26: Processor

proved that the Intel chip had also been handicapped by compiling settings (although the handicap to the Intel chip was noted in the details initially released). Code running on the underclocked Kal-El (running at 2/3 speed) had been compiled with a modern version of GNU Compiler Collection (GCC) and aggressive optimizations while that running on the Intel chip was produced by an obsolete version of GCC and only minimal optimizations. When the Intel code was compiled using the same flags as the code running on Kal-El, the Core 2 Duo was appreciably faster than at least an underclocked upcoming Tegra 3.[9][10]

Codenames of all upcoming releases in the Tegra series are references to comic book superheroes. Specifically, Superman (Kal-El), Batman (Wayne), Jean Grey (Grey), Wolverine (Logan), and Iron Man (Stark).[11]

In January 2012, Nvidia announced that Audi had selected the Tegra 3 processor for its in-vehicle infotainment systems and digital instruments display.[12] The processor will be integrated into Audi's entire line of vehicles worldwide, beginning in 2013.

[edit] Specifications

[edit] Tegra APX series

[edit] Tegra APX 2500

Processor: ARM11 600 MHz MPCore (originally GeForce ULV) o suffix: APX (formerly CSX)

Memory: NOR or NAND flash, Mobile DDR Graphics: Image processor (FWVGA 854×480 pixels)

o Up to 12 megapixels camera supporto LCD controller supports resolutions up to 1280×1024

Storage: IDE for SSD Video codecs: up to 720p H.264 and VC-1 decoding Includes GeForce ULV support for OpenGL ES 2.0, Direct3D Mobile, and

programmable shaders Output: HDMI, VGA, composite video, S-Video, stereo jack, USB USB On-The-Go

[edit] Tegra APX 2600

Enhanced NAND flash Video codecs:[13]

o 720p H.264 Baseline Profile Encode or Decodeo 720p VC-1/WMV9 Advanced Profile Decodeo D1 MPEG-4 Simple Profile Encode or Decode

[edit] Tegra 6xx series

Page 27: Processor

[edit] Tegra 600

Targeted for GPS segment and automotive Processor: ARM11 700 MHz MPCore Memory: low-power DDR (DDR-333, 166 MHz) SXGA, HDMI, USB, stereo jack HD camera 720p

[edit] Tegra 650

Targeted for GTX of handheld and notebook Processor: ARM11 800 MHz MPCore Low power DDR (DDR-400, 200 MHz) Less than 1 watt envelope HD image processing for advanced digital still camera and HD camcorder functions Display supports 1080p at 24 frame/s, HDMI v1.3, WSXGA+ LCD and CRT, and

NTSC/PAL TV output Direct support for Wi-Fi, disk drives, keyboard, mouse, and other peripherals A complete board support package (BSP) to enable fast time to market for Windows

Mobile-based designs

[edit] Tegra 2 series

The second generation Tegra SoC has a dual-core ARM Cortex-A9 CPU (lacking ARMs advanced SIMD extension, marketed as NEON), an ultra low power (ULP) GeForce GPU with 4 pixel shaders + 4 vertex shaders,[14] a single-channel memory controller with either LPDDR2 at 600 MHz or DDR2 at 667 MHz, a 32KB/32KB L1 cache per core and a shared 1MB L2 cache.[15] There is also a version of the SoC supporting 3D displays; this SoC uses a higher clocked CPU and GPU.

Model number

Semiconductor technology

CPU instruction

setCPU GPU Memory

technology Availability Utilizing Devices

Tegra 250 AP20H

40 nm ARMv7 1 GHz dual-core ARM Cortex-A9

ULP GeForce 300 MHz

Single-channel LPDDR2 600 MHz or DDR2 667 MHz

Q1 2010 LG Optimus 2X, Motorola Atrix 4G, Motorola Droid X2, Motorola Photon, Samsung Galaxy R, Samsung Captivate Glide, Tesla Model S, ZTE

Page 28: Processor

Mimosa X, Micromax Superfone A85

Tegra 250 T20

40 nm ARMv7 1 GHz dual-core ARM Cortex-A9

ULP GeForce 333 MHz

Single-channel LPDDR2 600 MHz or DDR2 667 MHz

Q1 2010 Acer Iconia Tab A100, A200 and A500, Asus Slider, LG Optimus Pad, Avionic Design Tamonten Processor Board,[16] Exper EasyPad, Notion Ink Adam tablet, Olivetti OliPad 100, Point of View Mobii 10.1, ViewSonic G Tablet, Motorola Xoom,[17] Toshiba AC100, Toshiba Folio 100, ASUS Eee Pad Transformer, Advent Vega, Hannspree Hannspad, Aigo n700, CompuLab Trim-Slice nettop, Dell Streak 7, E-Noa Interpad, Malata Tablet Zpad, MSI 10-inch (250 mm)

Page 29: Processor

tablet, Toradex Colibri Tegra 2, Toshiba Thrive tablet, Samsung Galaxy Tab 10.1, T-Mobile G-Slate, Lenovo IdeaPad Tablet K1, Lenovo ThinkPad Tablet, Velocity Micro Cruz Tablet L510, Dell Streak Pro,[18] Zyrex Onepad SP1110, Zyrex Onepad SP1113G, Sony Tablet S

Tegra 250 3D AP25

40 nm ARMv7

1.2 GHz dual-core ARM Cortex-A9

ULP GeForce 400 MHz

Single-channel LPDDR2 600 MHz or DDR2 667 MHz

Q1 2011Fusion Garage Grid 10[citation

needed]

Tegra 250 3D T25

40 nm ARMv7

1.2 GHz dual-core ARM Cortex-A9

ULP GeForce 400 MHz

Single-channel LPDDR2 600 MHz or DDR2 667 MHz

Q1 2011

Samsung GT-P7320, Motorola MOTO XT882

[edit] Tegra 3 (Kal-El) series

The Tegra 3 is functionally a SoC with a quad-core CPU, but includes a fifth "companion" core. All cores are Cortex-A9s, but the companion core is manufactured with a special low power silicon process that doesn't allow for high clock frequencies; hence it is limited to 500 MHz. There is also special logic to allow running state to be quickly transferred between the companion core and one of the normal cores. The goal is for a mobile phone or tablet to be able to power down all the normal cores and run on only the companion core, using comparatively

Page 30: Processor

little power, during standby mode or when otherwise underutilizing the CPU. According to Nvidia, this includes playing music or even video content.[19] Compared to Tegra 2, the ARM Cortex-A9s in Tegra 3 now supports ARMs SIMD extension, marketed as NEON. It can also output video up to 2560x1600 resolution and supports 1080p MPEG-4 AVC/h.264 40 Mbps High-Profile, VC1-AP and DivX 5/6 video decode.[20]

The Tegra 3 was officially released on November 9, 2011.[21]

Model numbe

r

Semiconductor technology

CPU instructio

n setCPU CPU

Cache GPUMemory technolog

y

Availability

Utilizing Devices

Tegra 3 40 nm LPG by TSMC ARMv7

1.3 GHz* Quad-core ARM Cortex-A9

L1: 32KB Instruction + 32KB Data, L2: 1MB

ULP GeForce (improved over Tegra 2)

32bit Single-channel LPDDR2-1066 or DDR3-L up to 1600 MHz

Q4 2011

Asus Eee Pad Transformer Prime [22] , IdeaPad K2 / LePad K2[23], Acer Iconia Tab A700[24], LG Optimus 4X HD, HTC One X, ZTE Era, ASUS Transformer Pad Infinity 700 Series (WiFi-version), ASUS Transformer Pad 300 Series, ZTE PF 100, ZTE T98, Toshiba AT270

Up to 1.4 GHz in single-core mode

[edit] Tegra (Wayne) series

Page 31: Processor

Processor: Quad ARM Cortex-A15 MPCore + low power companion core Improved 24 (for the quad-core) and 32 to 64 (for the octa-core) GPU cores with support

for Directx 11+, OpenGL 4.X, and PhysX 28 nm HPL[25]

About 10 times faster than Tegra 2 To be released in 2012 Jun

[edit] Tegra (Grey) series

Processor: ARM Cortex MPCore 28 nm HPM Integrated Icera 3G/4G baseband To be released in 2012

[edit] Tegra (Logan) series

Processor: ARM v7 Improved GPU core 28 nm[25]

About 50 times faster than Tegra 2 To be released in 2013

[edit] Tegra (Stark) series

Processor: ARM v8 Improved GPU core About 75 times faster than Tegra 2[26]

To be released in 2014

[edit] Similar platforms Snapdragon by Qualcomm OMAP by Texas Instruments Exynos by Samsung Ax by Apple NovaThor by ST-Ericsson Atom by Intel

[edit] See also ZiiLABS ZMS series i.MX by Freescale Imageon Loongson Mobile Internet device

Page 32: Processor

Nomadik by ST-Ericsson PXA by Marvell Renesas Electronics Corporation

[edit] References1. ̂ "Nvidia Rolls out "Tegra" Processors".

http://www.techtree.com/India/News/Nvidia_Rolls_out_Tegra_Processors/551-89833-581.html. Retrieved 2008-06-02.

2. ̂ "Nvidia Tegra FAQ" (PDF). http://www.nvidia.com/docs/IO/55043/NVIDIA_Tegra_FAQ_External.pdf. Retrieved 2008-06-04.

3. ̂ New Nvidia Tegra 3 at 1.5GHz4. ̂ Microsoft KIN first to use tegra5. ̂ "Nvidia and Opera team to accelerate the full Web on mobile devices" (Press release).

Opera Software. 2008-09-09. http://www.opera.com/press/releases/2008/09/09/. Retrieved 2009-01-09.

6. ̂ "Nvidia And Opera Team To Accelerate The Full Web On Mobile Devices" (Press release). Nvidia. 2008-09-09. http://www.nvidia.com/object/io_1220962341614.html. Retrieved 2009-04-17.

7. ̂ "New Nvidia Tegra Processor Powers The Tablet Revolution". Nvidia. 7Jan2010. http://www.nvidia.com/object/io_1262837617533.html. Retrieved 2010-03-19.

8. ̂ "What operating systems does Tegra support?" (Press release). Nvidia. 2011-08-17. http://developer.nvidia.com/beta-forum#/discussion/46/what-operating-systems-does-tegra-support. Retrieved 2011-09-14.

9. ̂ "Nvidia's Kal-El Demonstration Marred By Benchmark Confusion". http://hothardware.com/News/Nvidias-KalEl-Demonstration--Marred-By-Benchmark-Shenanigans/. Retrieved 2011-02-22.

10. ̂ "Why nVidia’s Tegra 3 is faster than a Core 2 Duo T7200". Brightsideofnews.com. 2011-02-21. http://www.brightsideofnews.com/news/2011/2/21/why-nvidiae28099s-tegra-3-is-faster-than-a-core-2-duo-t7200.aspx. Retrieved 2011-08-12.

11. ̂ Trenholm, Rich. "Nvidia Kal-El quad-core phone chip is faster than a speeding bullet in video". CBS Interactive Limited. http://crave.cnet.co.uk/mobiles/nvidia-kal-el-quad-core-phone-chip-is-faster-than-a-speeding-bullet-in-video-50003933/. Retrieved 2 June 2011.

12. ̂ Peter Clarke, EE Times. "Audi selects Tegra processor for infotainment, dashboard." January 17, 2012. Retrieved January 17, 2012.

13. ̂ "Nvidia Tegra APX Specifications". http://www.nvidia.com/object/product_tegra_apx_us.html. Retrieved 2011-02-17.

14. ̂ "LG Optimus 2X & Nvidia Tegra 2 Review: The First Dual-Core Smartphone". AnandTech. http://www.anandtech.com/show/4144/lg-optimus-2x-nvidia-tegra-2-review-the-first-dual-core-smartphone/5. Retrieved 2011-08-12.

15. ̂ "NVidia Tegra 2 Product Information". NVidia. http://www.nvidia.com/object/tegra-2.html. Retrieved 2011-09-05.

Page 33: Processor

16. ̂ "Avionic Design Tegra 2 (T290) Tamonten Processor Module - Product Brief". Avionic Design. http://www.avionic-design.de/uploads/pdf/Tamonten-Product-Brief-1373-101-000-02.pdf. Retrieved 2011-02-23.

17. ̂ "Motorola Xoom Specifications Table". Motorola Mobility, Inc. February 16, 2011. http://developer.motorola.com/products/xoom-mz601/. Retrieved 2011-02-16.

18. ̂ Dell Streak Pro Honeycomb tablet pictured, likely to be with us in June, Engadget, 19 May 2011

19. ̂ "Variable SMP – A Multi-Core CPU Architecture for Low Power and High Performance". 2011-09-19. http://www.nvidia.com/content/PDF/tegra_white_papers/tegra-whitepaper-0911b.pdf.

20. ̂ "ASUS Transformer Prime introduced and examined". HEXUS.net. http://hexus.net/mobile/news/tablets/32531-asus-transformer-prime-introduced-examined/. Retrieved 2011-11-11.

21. ̂ "NVIDIA Quad-Core Tegra 3 Chip Sets New Standards of Mobile Computing Performance, Energy Efficiency". 2011-11-09. http://pressroom.nvidia.com/easyir/customrel.do?easyirid=A0D622CE9F579F09&version=live&releasejsp=release_157&xhtml=true&prid=819304.

22. ̂ "Asus Eee Pad Transformer Prime (Nvidia Tegra 3 Processor; 10.1-inch display) Review". 2011-12-30. http://asia.cnet.com/product/asus-eee-pad-transformer-prime-nvidia-tegra-3-processor-10-1-inch-display-45735891.htm.

23. ̂ LePad K2 performance test by GLBenchmark24. ̂ [1]25. ^ a b "Nvidia's Tegra Kal-El Will Be 40 nm, Not 28 nm". Tomshardware.com.

http://www.tomshardware.com/news/Tegra-Kal-El-Tegra-2-Snapdragon-ARM-CPU,12229.html. Retrieved 2011-08-12.

26. ̂ "Tegra Roadmap Revealed: Next Chip to be World’s First Quad-Core Mobile Processor". Blogs.nvidia.com. http://blogs.nvidia.com/2011/02/tegra-roadmap-revealed-next-chip-worlds-first-quadcore-mobile-processor/. Retrieved 2011-08-12.

[edit] External links Nvidia's Tegra APX website Nvidia's Tegra FAQ Tegra 2 Whitepaper New Chips Could Boost iPhone Rivals Nvidia Unveils Mobile Graphics Powerhouse in Barcelona Nvidia Surprises With First Mobile CPU Nvidia dialing into mobile phones Nvidia APX 2500 chip enables 3D and hours of high-def playback on Windows Mobile

handsets Nvidia rolls out Tegra chips aimed at tiny PCs Nvidia Announces Tegra "Grey" SoC with Built-in 3G, 4G Communication Capabilities

Page 34: Processor

http://en.wikipedia.org/wiki/Texas_Instruments_OMAP

OMAP developed by Texas Instruments is a category of proprietary system on chips (SoCs) for portable and mobile multimedia applications. OMAP devices generally include a general-purpose ARM architecture processor core plus one or more specialized co-processors. Earlier OMAP variants commonly featured a variant of the Texas Instruments TMS320 series digital signal processor.

Contents [hide] 

1 OMAP family o 1.1 High-performance applications processors

1.1.1 OMAP 1 1.1.2 OMAP 2 1.1.3 OMAP 3 1.1.4 OMAP 4 1.1.5 OMAP 5

o 1.2 Basic multimedia applications processorso 1.3 Integrated modem and applications processorso 1.4 OMAP L-1x

2 Products using OMAP processors 3 Similar platforms 4 See also 5 References 6 External links

[edit] OMAP family

Page 35: Processor

The Galaxy Nexus, example of a smartphone with an OMAP 4460 SoC

The OMAP family consists of three product groups classified by performance and intended application:

High-performance applications processors Basic multimedia applications processors Integrated modem and applications processors

Further, two main distribution channels exist, and not all parts are available in both channels. The genesis of the OMAP product line is from partnership with cell phone vendors, and the main distribution channel involves sales directly to such wireless handset vendors. Parts developed to suit evolving cell phone requirements are flexible and powerful enough to support sales through less specialized catalog channels; some OMAP 1 parts, and many OMAP 3 parts, have catalog versions with different sales and support models. Parts that are obsolete from the perspective of handset vendors may still be needed to support products developed using catalog parts and distributor-based inventory management.

Recently, the catalog channels have received more focus, with OMAP35x and OMAP-L13x parts being marketed for use with various applications where capable and power-efficient processors are useful.

[edit] High-performance applications processors

These are parts originally intended for use as application processors in smartphones, with processors powerful enough to run significant operating systems (such as Linux, Android or Symbian), support connectivity to personal computers, and support various audio and video applications.

[edit] OMAP 1

The OMAP 1 family started with a TI-enhanced ARM core, and then changed to a standard ARM926 core. It included many variants, most easily distinguished according to manufacturing technology (130 nm except for the OMAP171x series), CPU, peripheral set, and distribution channel (direct to large handset vendors, or through catalog-based distributors). In March 2009, the OMAP1710 family chips are still available to handset vendors.

Products using OMAP 1 processors include hundreds of cell phone models, and the Nokia 770 Internet tablets.

OMAP171x - 220 MHz ARM926EJ-S + C55x DSP, low-voltage 90 nm technology OMAP162x - 204 MHz ARM926EJ-S + C55x DSP + 2 MB internal SRAM, 130 nm

technology OMAP5912 - catalog availability version of OMAP1621 (or OMAP1611b in older

versions) OMAP161x - 204 MHz ARM926EJ-S + C55x DSP, 130 nm technology

Page 36: Processor

OMAP1510 - 168 MHz ARM925T (TI-enhanced) + C55x DSP OMAP5910 - catalog availability version of OMAP 1510

[edit] OMAP 2

These parts were only marketed to handset vendors. Products using these include both Internet tablets and mobile phones:

OMAP2431 - 330 MHz ARM1136 + 220 MHz C64x DSP OMAP2430 - 330 MHz ARM1136 + 220 MHz C64x DSP + PowerVR MBX lite GPU OMAP2420 - 330 MHz ARM1136 + 220 MHz C55x DSP + PowerVR MBX GPU

[edit] OMAP 3

The 3rd generation OMAP, The OMAP 3[1] is broken into 3 distinct groups: the OMAP34x, the OMAP35x, and the OMAP36x. OMAP34x and OMAP36x are distributed directly to large handset (such as cell phone) manufacturers. OMAP35x is a variant of OMAP34x intended for catalog distribution channels. The OMAP36x is a 45 nm version of the 65 nm OMAP34x with higher clock speed.[2]

The video technology in the higher end OMAP 3 parts is derived in part from the DaVinci product line, which first packaged higher end C64x+ DSPs and image processing controllers with ARM9 processors last seen in the older OMAP 1 generation or ARM Cortex-A8[3].

Not highlighted in the list below is that each OMAP 3 SoC has an "Image, Video, Audio" (IVA2) accelerator. These units do not all have the same capabilities. Most devices support 12 megapixel camera images, though some support 5 or 3 megapixels. Some support HD imaging.

Model number

Semiconductor technology

CPU instruction

setCPU GPU Utilizing devices

OMAP3410 65 nm ARMv7 600 MHz ARM Cortex-A8

PowerVR SGX530

Motorola Charm, Motorola Flipside,[citation

needed] Motorola Flipout

OMAP3420 65 nm ARMv7 600 MHz ARM Cortex-A8

PowerVR SGX530

OMAP3430 65 nm ARMv7 600 MHz ARM Cortex-A8

PowerVR SGX530

Motorola Droid/Milestone, Nokia N900, Palm Pre, Samsung i8910, Sony Ericsson Satio

OMAP3440 65 nm ARMv7 800 MHz ARM Cortex-A8

PowerVR SGX530

Archos 5 (Gen 7), Motorola Milestone XT720, Motorola Titanium XT800,[citation

needed] Samsung Galaxy

Page 37: Processor

A (SHW-M100S)[citation

needed]

OMAP3503 65 nm ARMv7 600 MHz ARM Cortex-A8 N/A Gumstix Overo Earth

OMAP3515 65 nm ARMv7 600 MHz ARM Cortex-A8

PowerVR SGX530

OMAP3525 65 nm ARMv7 600 MHz ARM Cortex-A8 N/A

OMAP3530 65 nm ARMv7 720 MHz ARM Cortex-A8

PowerVR SGX530

Alico's Kinetic 3500,[4] Always Innovating Touch Book, BeagleBoard, Embest DevKit8000,[5] Gumstix Overo Water, IGEPv2, OpenSourceMID K7 MID,[6] Oswald,[citation

needed] Overo Water,[citation needed] Pandora, phyCARD-L OMAP-3530 SOM,[7] TianyeIT CIP312[8]

OMAP3611 45 nm ARMv7 800 MHz ARM Cortex-A8

PowerVR SGX530

Cybook Odyssey[citation

needed]

OMAP3621; OMAP3622 45 nm ARMv7

3621: 800 MHz, 3622: 1 GHz; ARM Cortex-A8

PowerVR SGX530

Barnes & Noble Nook Color, Barnes & Noble Nook Simple Touch, Lenovo IdeaPad A1

OMAP3630 45 nm ARMv7 600 MHz~1.2 GHz ARM Cortex-A8

PowerVR SGX530

3630-800: Motorola Bravo,[citation needed] Motorola Defy [9]

3630-1000: Archos 28, Archos 32, Archos 43, Archos 70, Archos 101, LG Optimus Black, LG Optimus Bright, LG Optimus Mach,[citation needed] Motorola Cliq 2, Motorola Droid 2/Milestone 2, Motorola Droid X, Motorola Defy+, Nokia N9, Nokia N950,[citation needed] Palm

Page 38: Processor

Pre 2, Panasonic P-07C, Panasonic Sweety 003P,[citation needed] Samsung Galaxy S LCD (GT-I9003),[citation

needed] Sony Ericsson Vivaz[citation needed]

3630-1200: Motorola Droid 2 Global

[edit] OMAP 4

The 4th generation OMAPs, OMAP 4430, 4460 (formerly named 4440),[10] and 4470 all use dual-core ARM Cortex-A9s. The 4470 additionally contains two Cortex-M3s running at 266 MHz to offload the A9s in less computionally intensive tasks to increase power efficiency.[11]

[12][13] 4430 and 4460 use a PowerVR SGX540 integrated 3D graphics accelerator which runs at a clock frequency of 304 and 384 MHz respectively compared to prior versions of SGX540 typically at 200 MHz making them theoretically much faster.[14] 4470 has a PowerVR SGX544 GPU that supports DirectX 9 which enables it for use in Windows 8 as well as a dedicated 2D graphics core for increased power efficiency. All OMAP 4 comes with an IVA3 multimedia hardware accelerator with a programmable DSP that enables 1080p Full HD and multi-standard video encode/decode.[15][16][17][18][19] OMAP 4 uses ARM-Cortex A9s with ARMs SIMD engine (Media Processing Engine, aka NEON) which may have a significant performance advantage in some cases over Nvidia Tegra 2s Cortex-A9s with non-vector floating point units.[20] It also uses a dual-channel LPDDR2 memory controller compared to Nvidia Tegra 2s single-channel memory controller.

Model number

Semiconductor

technology

CPU instructio

n setCPU GPU

Memory technolog

y

Availability Utilizing devices

OMAP4430

45 nm ARMv7 1-1.2 GHz dual-core ARM Cortex-A9

PowerVR SGX540 @ 304 MHz

Dual-channel LPDDR2 memory controller

Q1 2011 Amazon Kindle Fire, Archos 80 (Gen 9), Archos 101 (Gen 9), Barnes and Noble Nook Tablet, BlackBerry PlayBook,[21] Fujitsu Arrows Tab LTE F-01D, Fujitsu Arrows X LTE F-05D, Fujitsu Arrows Z ISW11F, Panasonic Lumix Phone 101P, Panasonic Lumix

Page 39: Processor

Phone P-02D, Fujitsu Regza Phone T-01D, LG Prada 3.0, LG Optimus 3D P920, Motorola Atrix 2, Motorola Droid 3/Milestone 3, Motorola Droid Bionic, Motorola Droid RAZR,[22] PandaBoard, phyCORE-OMAP4460/OMAP4430 SOM,[23] Samsung Galaxy S II (GT-I9100G), Sharp Aquos Phone SH-01D, Sharp Aquos Phone 102SH, TianyeIT CIP411,[24] Toshiba AT200 Excite[citation

needed],LGP925 Thrill AT&T

OMAP4460 45 nm ARMv7

1.2-1.5 GHz dual-core ARM Cortex-A9

PowerVR SGX540 @ 384 MHz

Dual-channel LPDDR2 memory controller

Q4 2011

Archos 80 Turbo (Gen 9), Archos 101 Turbo (Gen 9), Galaxy Nexus, Huawei Ascend D1,[citation needed] Huawei Ascend P1/P1S,[citation

needed] Pandaboard ES,[25] Sharp Aquos Phone 104SH, Variscite VAR-SOM-OM44[26]

OMAP4470 45 nm ARMv7

1.5-1.8 GHz dual-core ARM Cortex-A9

PowerVR SGX544 @ 384 MHz + dedicated 2D graphics core

Dual-channel LPDDR2 memory controller, 466 MHz

Q2 2012

Page 40: Processor

[edit] OMAP 5

The 5th generation OMAP, OMAP 5 SoC uses a dual-core ARM Cortex-A15 CPU with two additional Cortex-M4 cores to offload the A15s in less computionally intensive tasks to increase power efficiency, two PowerVR SGX544MP graphics cores and a dedicated TI 2D BitBlt graphics accelerator, a multi-pipe display sub-system and a signal processor.[27] They respectively support 24 and 20 megapixel cameras for front and rear 3D HD video recording. The chip also supports up to 8 GB of dual channel DDR3 memory, output to four HD 3D displays and 3D HDMI 1.4 video output. OMAP 5 also includes 3 USB 2.0 ports and a SATA 2.0 controller.

Model number

Semiconductor technology

CPU instruction

setCPU GPU Memory

technology Availability

OMAP5430 28 nm ARMv7

2 GHz dual-core ARM Cortex-A15

Dual-core PowerVR SGX544MP + dedicated 2D graphics accelerator

Dual-channel package on package LPDDR2

Q3 2012

OMAP5432 28 nm ARMv7

2 GHz dual-core ARM Cortex-A15

Dual-core PowerVR SGX544MP + dedicated 2D graphics accelerator

Dual-channel DDR3 controller

Q3 2012

[edit] Basic multimedia applications processors

These are marketed only to handset manufacturers. They are intended to be highly integrated, low cost chips for consumer products. The OMAP-DM series are intended to be used as digital media coprocessors for mobile devices with high megapixel digital still and video cameras.

OMAP331 - ARM9 OMAP310 - ARM9 OMAP-DM270 - ARM7 + C54x DSP OMAP-DM299 - ARM7 + ISP + stacked mDDR SDRAM OMAP-DM500 - ARM7 + ISP + stacked mDDR SDRAM OMAP-DM510 - ARM926 + ISP + 128 MB stacked mDDR SDRAM OMAP-DM515 - ARM926 + ISP + 256 MB stacked mDDR SDRAM OMAP-DM525 - ARM926 + ISP + 256 MB stacked mDDR SDRAM

[edit] Integrated modem and applications processors

Page 41: Processor

An OMAP 850 in a HTC Wizard

These are marketed only to handset manufacturers. Many of the newer versions are highly integrated for use in very low cost cell phones.

OMAPV1035 - single-chip EDGE (was discontinued in 2009 as TI announced baseband chipset market withdrawal).

OMAPV1030 - EDGE digital baseband OMAP850 - 200 MHz ARM926EJ-S + GSM/GPRS digital baseband + stacked EDGE

co-processor OMAP750 - 200 MHz ARM926EJ-S + GSM/GPRS digital baseband + DDR Memory

support OMAP733 - 200 MHz ARM926EJ-S + GSM/GPRS digital baseband + stacked SDRAM OMAP730 - 200 MHz ARM926EJ-S + GSM/GPRS digital baseband + SDRAM

Memory support OMAP710 - 133 MHz ARM925 + GSM/GPRS digital baseband

[edit] OMAP L-1x

The OMAP L-1x parts are marketed only through catalog channels, and have a different technological heritage than the other OMAP parts. Rather than deriving directly from cell phone product lines, they grew from the video-oriented DaVinci product line by removing the video-specific features while using upgraded DaVinci peripherals. A notable feature is use of a floating point DSP, instead of the more customary fixed point one.

The Hawkboard uses the OMAP-L138

OMAP-L137 - 300 MHz ARM926EJ-S + C674x floating point DSP OMAP-L138 - 300 MHz ARM926EJ-S + C674x floating point DSP

[edit] Products using OMAP processorsMany mobile phones use OMAP SoCs, including the Nokia N90, N91, N92, N95, N82, E61, E62, E63 and E90 mobile phones, as well as the N800, N810 and N900 Internet tablets, Motorola Droid, Droid X, and Droid 2. The Palm Pre, Pandora, Touch Book also use an OMAP SoC (the OMAP3430). Others to use an OMAP SoC include Sony Ericsson's Satio and Vivaz,

Page 42: Processor

the Samsung Omnia HD, Sony Ericsson Idou, the Nook Color, and some Archos tablets (such as Archos 80 gen 9 and Archos 101 gen 9).

OMAP SoCs are also used as the basis for a number of hobbyist and prototyping boards, such as the Beagle Board, Panda Board and Gumstix.

[edit] Similar platforms Snapdragon by Qualcomm Tegra by Nvidia Exynos by Samsung Ax by Apple NovaThor by ST-Ericsson Atom by Intel

[edit] See also OpenMAX IL (Open Media Acceleration Integration Layer) - a royalty-free cross-

platform media abstraction API from the Khronos Group Distributed Codec Engine (libcde) is a Texas Instruments API for the video codec engine

in OMAP based embedded systems

[edit] References1. ̂ OMAP34xx series in TI Web site2. ̂ OMAP36x3. ̂ DaVinci Digital Video Processor - TMS320DM37x SOC - DM3730 - TI.com4. ̂ http://www.alicosystems.com/Alico%20FSDK%203500%200311A.pdf5. ̂ "Embest DevKit8000 OMAP3530 Evaluation Kit". Embest.

http://www.armkits.com/Product/devkit8000.asp. Retrieved 17 Feb 2012.6. ̂ "OpenSourceMID K7 MID". OpenSourceMID.

http://www.opensourcemid.org/k7mid.htm. Retrieved 17 Feb 2012.7. ̂ "TI OMAP3530 ARM Cortex A8 System on Module". Phytec America, LLC.

http://www.phytec.com/products/som/Cortex-A8/phyCARD-L-CortexA8-OMAP3525.html. Retrieved 17 Feb 2012.

8. ̂ "CIP312 TI DM3730/OMAP3530 Computer in Package". TianyeIT LTD. http://www.tianyeit.com/?p=34&a=view&r=31. Retrieved 17 Feb 2012.

9. ̂ http://focus.ti.com/pdfs/wtbu/OMAP36xx_ES1.x_PUBLIC_TRM_vN.zip10. ̂ http://www.linuxfordevices.com/c/a/News/Variscite-VARSOMOM44/ Computer

module taps 1.5GHz, dual-core OMAP4460 SoC11. ̂ http://focus.ti.com/pdfs/wtbu/OMAP4430_ES2.x_Public_TRM_vK.zip12. ̂ "OMAP4460 Public TRM vE (pdf)"13. ̂ Texas Instruments announces multi-core, 1.8GHz OMAP4470 ARM processor for

Windows 8 - Engadget

Page 43: Processor

14. ̂ AnandTech - TI Announces OMAP4470 and Specs: PowerVR SGX544, 1.8 GHz Dual Core Cortex-A9

15. ̂ OMAP44xx series in TI Web site16. ̂ http://www.linuxfordevices.com/c/a/News/TI-OMAP4430-and-OMAP4440/ TI speeds

up its OMAP 4 for 3D video17. ̂ http://www.engadget.com/2010/02/02/tis-omap4-prototype-drives-three-independent-

displays-without-b/ TI's OMAP 4 prototype drives three independent displays without breaking a sweat

18. ̂ http://www.engadget.com/2009/02/17/tis-omap-4-bringing-1080p-support-to-smartphones-and-mids/ TI's OMAP 4 bringing 1080p support to smartphones and MIDs

19. ̂ http://www.engadget.com/2010/02/15/texas-instruments-introduces-arm-based-omap-4-soc-blaze-develop/ Texas Instruments introduces ARM-based OMAP 4 SOC, Blaze development platform

20. ̂ AnandTech - NVIDIA's Tegra 2 Take Two: More Architectural Details and Design Wins

21. ̂ Hunter Skipworth (23 March 2011). "Blackberry confirms PlayBook specs and launch date". The Telegraph. Telegraph Media Group. http://www.telegraph.co.uk/technology/blackberry/8400612/Blackberry-confirms-PlayBook-specs-and-launch-date.html. Retrieved 17 Feb 2012.

22. ̂ "Droid Razr by Motorola, XT912". MotoDev. Motorola Mobility. http://developer.motorola.com/products/droid-razr-xt912/. Retrieved 17 Feb 2012.

23. ̂ "OMAP4460/OMAP4430: OMAP 4 Cortex A9 System on Module". Phytec America, LLC. http://www.phytec.com/products/som/Cortex-A9/phyCORE-OMAP4460-OMAP4430.html. Retrieved 17 Feb 2012.

24. ̂ "CIP Ti OMAP4430/4460 Computer In Package". TianyeIT. http://www.tianyeit.com/?p=34&a=view&r=30. Retrieved 17 Feb 2012.

25. ̂ "PandaBoard ES Technical Specs". PandaBoard. http://pandaboard.org/node/300/#PandaES. Retrieved 17 Feb 2012.

26. ̂ "VAR-SOM-OM44 CPU: TI OMAP4460". Variscite. http://www.variscite.com/products/item/76-var-som-om44-ti-omap4460. Retrieved 17 Feb 2012.

27. ̂ "Not Just a Faster Horse: TI’s OMAP 5 Platform Transforms the Concept of ‘Mobile’". Texas Instruments. 11-02-07. http://newscenter.ti.com/Blogs/newsroom/archive/2011/02/07/not-just-a-faster-horse-ti-s-omap-5-platform-transforms-the-concept-of-mobile-615064.aspx. Retrieved 2011-02-09. "The OMAP 5 processor leverages two ARM Cortex-A15 MPCores [...] [It] also includes two ARM Cortex-M4 processors [...]"

[edit] External links OMAP Application Processors OMAPWorld OMAPpedia Linux OMAP Mailing List Archive OMAP3 Boards OMAP4 Boards

Page 44: Processor

http://en.wikipedia.org/wiki/Samsung_Hummingbird#Exynos_3110

Exynos refers to the name of a series of ARM based System-on-Chip's (SoCs) from Samsung Electronics. Exynos 'SoCs' are purpose-built for mobile devices.

Samsung has a long history of designing and producing SoCs and has been manufacturing SoCs for its own devices as well as for others, for example Apple. The first Samsung SoC, the S3C44B0, was built around an ARM7 CPU which operated at 66 MHz clock frequency. Later, several SoCs (S3C2xxx) containing an ARM9 CPU were produced.[1]

In 2010 Samsung launched the S5PC110 also known as 'Hummingbird' (now Exynos 3110) in its Samsung Galaxy S mobile phone which featured a licensed ARM Cortex-A8 CPU.

In early 2011, Samsung first launched the Exynos 4210 SoC within its Samsung Galaxy S II mobile smartphone. The driver code for the Exynos 4210 was made available in the Linux kernel.[2] and kernel 3.2 support for the Exynos 4210[3][4]

In later 2011 Samsung introduced Exynos 4212, as a successor to the 4210, that features a higher clock frequency and “50 percent higher 3D graphics performance over the previous processor generation". Built with 32 nm High-K Metal Gate (HKMG) low-power process it promises “30 percent lower power-level over the previous process generation.”

Contents [hide] 

1 List of historical Samsung SoC and current Exynos SoC 2 Similar platforms 3 References 4 External links

[edit] List of historical Samsung SoC and current Exynos SoC

Model Number

Semiconductor

Technology

CPU Instruction Set

CPU GPUMemory Technol

ogy

Availability Utilizing Devices

S3C44B0 0.25 um ARMv4 66 MHz LCD 2001 Juice Box

Page 45: Processor

CMOS

Single-Core ARM7 (ARM7TDMI)

Controller

S3C2410 0.18 um CMOS ARMv5

200/266 MHz Single-Core ARM9 (ARM920T)

LCD Controller

2003

HP iPAQ H1930/H1937/H1940/rz1717, Everex E500, E-TEN InfoTouch P300, Acer n30/n35/d155, Palm Z22, LG LN600, Typhoon MyGuide 3610 GO

S3C2440 0.13 um CMOS ARMv5

300/400 MHz Single-Core ARM9 (ARM920T)

LCD controller

2004

HP iPAQ rx3115/3415/3417/3715, Everex E900, Airis T470i/920/930, Acer n300/311,Typhoon MyPhone M500,Mio p550/P350/C710 Digi-Walker

S3C2443 ARMv5

400/533 MHz Single-Core ARM9 (ARM920T)

LCD Controller

4KB iSRAM 2007

Asus R300/R600/R700, Mio Digi-Walker (C620T/...), Mio Moov 2xx/3xx, LG LN8xx, JoinTech JPro Mini Laptop JL7220, Navigon 8300/8310

S3C2450 65 nm ARMv5

400/533 MHz Single-Core ARM9 (ARM926EJ)

2D graphic acceleration

64KB SRAM 2008

Mio Moov 500/510/560/S568/580, Getac PS535F, MENQ EasyPC E720/E790, Archos Arnova 10, Hivision PWS0890AW,SMiT MTV-PND530 8GB

S3C2416 65 nm ARMv5

400 MHz Single-Core ARM9 (ARM926EJ)

2D graphic acceleration

64KB SRAM 2009 iconX G310

S5L8900 90 nm ARMv6 412 MHz Single-Core ARM11

PowerVR MBX Lite

eDRAM 2007 Apple iPhone, Apple iPhone 3G, Apple iPod touch 1G, Apple

Page 46: Processor

iPod touch 2G

S3C6410 65 nm ARMv6533/667/800 MHz Single-core ARM11

FIMG-3DSE DDR 2009

S5P6442 45 nm ARMv6533/667 MHz Single-core ARM11

FIMG-3DSE 2010

S5PC100 65 nm ARMv7

667/833 MHz Single-core ARM Cortex-A8

PowerVR SGX535

LPDDR2, DDR2 2009 Apple iPhone 3GS,

Apple iPod touch 3G

Exynos 3110 [5] (previously S5PC110 / Hummingbird)

45 nm ARMv7

1 GHz Single-core ARM Cortex-A8

200 Mhz PowerVR SGX540

LPDDR1, LPDDR2, or DDR2

2010

Samsung Galaxy S line, Samsung GT-S8500 Wave,Samsung Wave II S8530 , Nexus S, Meizu M9, Samsung Galaxy Tab, Samsung Droid Charge, Exhibit 4G, Samsung Infuse

Exynos 4210 [6] 45 nm ARMv7

1.2 / 1.4 GHz Dual-core ARM Cortex-A9

ARM Mali-400 MP4

LPDDR2, DDR2 or DDR3

2011

Samsung Galaxy S II, Samsung Galaxy Note, Samsung Galaxy Tab 7.7, Hardkernel ODROID-A, Meizu MX, Cotton Candy by FXI Tech

Exynos 4212 [7]

32 nm HKMG ARMv7

1.2 / 1.5 GHz Dual-core ARM Cortex-A9

ARM Mali-400 MP4

LPDDR, LPDDR2, DDR2 or DDR3

2011

Exynos 4412

32 nm HKMG ARMv7

1.5-1.8 GHz Quad-core ARM Cortex-A9

ARM Mali-T604 MP4

2012

Exynos 5250[8]

32 nm HKMG ARMv7

2.0 GHz Dual-core ARM Cortex-A15 MPCore

Q2 2012

Exynos 5450

32 nm HKMG

ARMv7 2.0 GHz Quad-core ARM

ARM Mali-T658

2012

Page 47: Processor

Cortex-A15 MPCore MP4

[edit] Similar platforms OMAP by Texas Instruments Snapdragon by Qualcomm Tegra by Nvidia Ax by Apple NovaThor by ST-Ericsson Atom by Intel

[edit] References1. ̂ "SAMSUNG Semiconductor - Products - Application Processor - Products".

Samsung.com. http://www.samsung.com/global/business/semiconductor/productList.do?fmly_id=844&xFmly_id=229. Retrieved 13 January 2012.

2. ̂ "[RFC[PATCH v3] DRM: add DRM Driver for Samsung SoC EXYNOS4210"]. freedesktop.org. 26 August 2011. http://lists.freedesktop.org/archives/dri-devel/2011-August/013846.html. Retrieved 13 January 2012.

3. ̂ Linux kernel 3.2 support Exynos 421 "Linux 3.2 DriverArch Linux kernel 3.2 support Exynos 4210 - Linux Kernel Newbies". kernelnewbies.org. http://kernelnewbies.org/Linux_3.2_DriverArch Linux kernel 3.2 support Exynos 421. Retrieved 13 January 2012.

4. ̂ Larabel, Michael (06 November 2011). "Samsung Keeps Working On Its Linux DRM". Phoronix. http://www.phoronix.com/scan.php?page=news_item&px=MTAxMjE. Retrieved 13 January 2012.

5. ̂ Exynos 3310 Product Webpage; Samsung.6. ̂ Exynos 4210 Product Webpage; Samsung.7. ̂ Exynos 4212 Product Webpage; Samsung.8. ̂ http://www.samsung.com/global/business/semiconductor/minisite/Exynos/

news_11.html#mainClick

[edit] External links Samsung Exynos official website

http://en.wikipedia.org/wiki/Apple_Ax

Page 48: Processor

Ax is a series of ARM-based System on Chips (SoC) designed by Apple Inc. for use in their consumer electronic devices, such as the iPod, iPad, iPhone and Apple TV.

Contents [hide] 

1 Apple A4 2 Apple A5, A5X 3 List of Apple chipsets 4 Similar platforms 5 See also 6 References

[edit] Apple A4Main article: Apple A4

The Apple A4 SoC

The Apple A4 is a package on package (PoP) system-on-a-chip (SoC) designed by Apple and manufactured by Samsung.[1] It combines an ARM Cortex-A8 CPU with a PowerVR GPU, and emphasizes power efficiency.[2] The chip commercially debuted with the release of Apple's iPad tablet;[3] followed shortly by the iPhone 4 smartphone,[4] the 4th generation iPod Touch and the 2nd generation Apple TV. It was superseded in the second-generation iPad, released the following year, by the Apple A5 processor.

Apple A4 is based on the ARM processor architecture.[5] The first version released runs at 1 GHz for the iPad and contains an ARM Cortex-A8 CPU core paired with a PowerVR SGX 535 graphics processor (GPU)[3][6][7][8] built on Samsung's 45-nanometer (nm) silicon chip fabrication process.[9] Clock speed for the units used in the iPhone 4, iPod Touch, and Apple TVs have not been revealed.

The Cortex-A8 core used in the A4 is thought to use performance enhancements developed by chip designer Intrinsity (which was subsequently acquired by Apple)[10] in collaboration with Samsung.[11] The resulting core, dubbed "Hummingbird", is able to run at far higher clock rates than other implementations while remaining fully compatible with the Cortex-A8 design

Page 49: Processor

provided by ARM.[12] Other performance improvements include additional L2 cache. The same Cortex-A8 CPU core used in the A4 is also used in Samsung's S5PC110A01 SoC.[13][14]

The A4 processor package does not contain RAM, but supports PoP installation. The top package of the A4 used in the iPad, in the iPod Touch [15] 4th gen and in the Apple TV [16] 2nd gen contains two low-power 128 MB DDR SDRAM chips for a total of 256MB RAM. For the iPhone 4 there are two chips of 256 MB for a total of 512 MB.[17][18][19] RAM is connected to the processor using ARM's 64-bit-wide AMBA 3 AXI bus. This is twice the width of the RAM data bus used in previous ARM 11 and ARM 9 based Apple devices, to support the greater need for graphics bandwidth in the iPad.[20]

[edit] Apple A5, A5XMain article: Apple A5

The Apple A5 SoC

The Apple A5X SoC

The Apple A5 is a package on package (PoP) system-on-a-chip (SoC) designed by Apple and manufactured by Samsung [21] that replaced the A4. The chip commercially debuted with the release of Apple's iPad 2 tablet in March 2011[22] , followed by its release in the iPhone 4S smartphone later that year. (This is consistent with how Apple debuted the A4 chip: first in the original iPad, then in the iPhone 4, and finally in the 4th generation iPod touch.[23])

The A5 contains a dual-core ARM Cortex-A9 CPU[24] with ARM's advanced SIMD extension, marketed as NEON, and a dual core PowerVR SGX543MP2 GPU. This GPU can push between 70 and 80 million polygons/second and has a pixel fill rate of 2 billion pixels/second. The SGX535 in the A4 could only push 35 million polygons/second and 500 million pixels/second, although in real world tests it struggled to pull off even 7 million flat shaded polygons. [25] Apple

Page 50: Processor

lists the A5 to be clocked at 1 GHz on the iPad 2's technical specifications page,[26] though it can dynamically adjust its frequency to save battery life.[27][24]

Apple claims that the CPU is twice as powerful and the GPU up to nine times as powerful as its predecessor, the Apple A4. The A5 package contains 512 MB of low-power DDR2 RAM clocked at 533 MHz.[28][29][30][31] The A5 is estimated to cost 75% more than its predecessor; the price difference is supposed to diminish as production increases.[32]

Apple announced on March 7, 2012 the new A5X SoC for the iPad (3rd Generation). It is a dual-core chip with quad-core graphics.

[edit] List of Apple chipsets

Name

Model no.

Fabrication

process

Instruction set CPU CPU

cache GPUMemory technolo

gy

Introduced Devices

A4 S5L8930X

45 nm,

by Samsung

ARMv7

800–1000 MHz single-core ARM Cortex-A8

L1: 32 kB Instruction + 32 kB Data, L2: 512 kB

PowerVR SGX535

LPDDR 200 MHz (Effective 400 MHz)

March 2010

iPad iPhone

4 iPod

Touch (4th gen.)

Apple TV (2nd gen.)

A5

S5L8940X

800–1000 MHz dual-core ARM Cortex-A9

PowerVR SGX543MP2 (dual-core)

LPDDR2 266 MHz (Effective 533 MHz)

March 2011

iPad 2 iPhone

4S

single-core ARM Cortex-A9

March 2012

Apple TV (3rd gen.)

A5X S5L8945X

dual-core ARM Corte

PowerVR SGX543MP4 (quad-core)

iPad (3rd gen.)

Page 51: Processor

x-A9

[edit] Similar platforms Snapdragon by Qualcomm OMAP by Texas Instruments Tegra by Nvidia Exynos by Samsung NovaThor by ST-Ericsson Atom by Intel

[edit] See also PWRficient , a processor designed by P.A. Semi, a company Apple acquired to form an

in-house custom chip design department PowerVR SGX GPUs were also used in the iPhone 3GS and the third-generation iPod

touch ARM Cortex-A9 MPCore

[edit] References1. ̂ Clark, Don (2010-04-05). "Apple iPad Taps Familiar Component Suppliers -

WSJ.com". Online.wsj.com. http://online.wsj.com/article/SB10001424052702303912104575164112770784290.html?mod=rss_Today's_Most_Popular. Retrieved 2010-04-15.

2. ̂ "iPad - It's thin, light, powerful, and revolutionary". Apple. http://www.apple.com/ipad/design/#performance. Retrieved 2010-07-07.

3. ^ a b "Apple Launches iPad" (Press release). Apple. 2010-01-27. http://www.apple.com/pr/library/2010/01/27ipad.html. Retrieved 2010-01-28.

4. ̂ "iPhone 4 design". Apple. 2010-07-06. http://www.apple.com/iphone/design/index.html.

5. ̂ Vance, Ashlee (2010-02-21). "For Chip Makers, the Next Battle Is in Smartphones". New York Times. http://www.nytimes.com/2010/02/22/technology/22chip.html. Retrieved 2010-02-25.

6. ̂ Wiens, Kyle (2010-04-05). "conclusion from both hard and software analysis it uses an ARM Cortex-A8 core". Ifixit.com. http://www.ifixit.com/Teardown/Apple-A4-Teardown/2204/3. Retrieved 2010-04-15.

7. ̂ "iPad — Technical specifications and accessories for iPad". Apple. http://www.apple.com/ipad/specs/. Retrieved 2010-01-28.

8. ̂ Melanson, Donald (2010-02-23). "iPad confirmed to use PowerVR SGX graphics". Engadget. http://www.engadget.com/2010/02/23/ipad-confirmed-to-use-powervr-sgx-graphics/.

9. ̂ "Chipworks Confirms Apple A4 iPad chip is fabbed by Samsung in their 45-nm process". Chipworks. http://www.chipworks.com/A4_is_Samsung_45nm.aspx.

Page 52: Processor

10. ̂ Stokes, Jon (2010-04-28). "Apple purchase of Intrinsity confirmed". Ars Technica. http://arstechnica.com/apple/news/2010/04/apple-purchase-of-intrinsity-confirmed.ars. Retrieved 2010-04-28.

11. ̂ Merritt, Rick. "Samsung, Intrinsity pump ARM to GHz rate". EETimes.com. http://www.eetimes.com/news/latest/showArticle.jhtml?articleID=218600577. Retrieved 2010-04-23.

12. ̂ Keizer, Gregg (2010-04-06). "Apple iPad smokes past the iPhone 3GS in speed". PC World. http://www.pcworld.com/article/193597/Apple_iPad_Smokes_Past_the_iPhone_3GS_in_Speed_Test.html. Retrieved 2010-04-11.

13. ̂ Boldt, Paul; Scansen, Don; Whibley, Tim (16 June 2010). "Apple's A4 dissected, discussed...and tantalizing". EETimes.com. http://www.eetimes.com/showArticle.jhtml?articleID=225700447. Retrieved 2010-07-07.

14. ̂ "Microsoft PowerPoint - Apple A4 vs SEC S5PC110A01" (PDF). http://www.ubmtechinsights.com/uploadedFiles/Apple%20A4%20vs%20SEC%20S5PC110A01.pdf. Retrieved 2010-07-07.

15. ̂ "Teardown of Apple's 4th-gen iPod touch finds 256MB of RAM". Appleinsider.com. 2010-09-08. http://www.appleinsider.com/articles/10/09/08/teardown_of_apples_4th_gen_ipod_touch_finds_256mb_of_ram.html. Retrieved 2010-09-10.

16. ̂ "Apple TV 2nd Generation Teardown". iFixit. 2010-09-30. http://www.ifixit.com/Teardown/Apple-TV-2nd-Generation-Teardown/3625/2. Retrieved 2011-01-04.

17. ̂ "Apple reveals iPhone 4 has 512MB RAM, doubling iPad - report". Appleinsider.com. 2010-06-17. http://www.appleinsider.com/articles/10/06/17/apple_reveals_iphone_4_has_512mb_ram_doubling_ipad_report.html. Retrieved 2010-07-07.

18. ̂ "A Peek Inside Apple’s A4 Processor". iFixit. 2010-04-05. http://www.ifixit.com/blog/2010/04/a-peek-inside-apples-a4-processor/.

19. ̂ Greenberg, Marc (2010-04-09). "Apple iPad: no LPDDR2?". Denali. http://www.denali.com/wordpress/index.php/dmr/2010/04/09/apple-ipad-no-lpddr2.

20. ̂ Merritt, Rick (2010-04-09). "iPad equipped to deliver richer graphics". EE Times Asia. http://www.eetasia.com/ART_8800603321_499495_NP_1e1373d9.HTM. Retrieved 2010-04-14.

21. ̂ "Updated: Samsung fabs Apple A5 processor". EETimes.com. March 12, 2011. http://www.eetimes.com/electronics-news/4213981/Samsung-fabs-Apple-A5-processor. Retrieved 2011-03-15.

22. ̂ "Apple Announces iPad 2 with New Design, Faster A5 Processor". http://www.appleinsider.com/articles/11/03/02/apple_announces_ipad_2_with_new_design_faster_a5_processor.html.

23. ̂ "iPhone 5 expected to have same A5 chip as iPad 2". Macworld.co.uk. http://www.macworld.co.uk/ipod-itunes/news/index.cfm?newsid=3264638. Retrieved 2012-01-25.

24. ^ a b "Apple iPad 2 Preview - AnandTech   :: Your Source for Hardware Analysis and News". AnandTech. http://www.anandtech.com/show/4215/apple-ipad-2-benchmarked-dualcore-cortex-a9-powervr-sgx-543mp2/2. Retrieved 2011-03-15.

Page 53: Processor

25. ̂ "Apple iPad 2 GPU Performance Explored: PowerVR SGX543MP2 Benchmarked - AnandTech   :: Your Source for Hardware Analysis and News" . AnandTech. http://www.anandtech.com/show/4216/apple-ipad-2-gpu-performance-explored-powervr-sgx543mp2-benchmarked. Retrieved 2011-03-15.

26. ̂ "iPad - View the technical specifications for iPad". Apple. http://www.apple.com/ipad/specs/. Retrieved 2011-03-15.

27. ̂ "Inside Apple's iPad 2 A5: fast LPDDR2 RAM, costs 66% more than Tegra 2". AppleInsider. http://www.appleinsider.com/articles/11/03/13/inside_apples_ipad_2_a5_fast_lpddr2_ram_costs_66_more_than_tegra_2.html. Retrieved 2011-03-15.

28. ̂ "Apple iPad 2 feature page". Apple.com. http://www.apple.com/ipad/features/. Retrieved 2011-03-15.

29. ̂ "Apple's A5 CPU in iPad 2 has 512MB of RAM, same as iPhone 4". Appleinsider.com. 2011-03-03. http://www.appleinsider.com/articles/11/03/03/apples_a5_cpu_in_ipad_2_has_512mb_of_ram_same_as_iphone_4_report.html. Retrieved 2011-03-15.

30. ̂ "TiPb Answers: Apple A5 chip — what we know and what we guess". Tipb.com. 2011-03-03. http://www.tipb.com/2011/03/03/tipb-answers-apple-a5-chip-guess/. Retrieved 2011-03-15.

31. ̂ "The iPad 2 – Daring Fireball". Daringfireball.net. 2011-03-09. http://daringfireball.net/2011/03/the_ipad_2. Retrieved 2011-03-15.

32. ̂ Pascal-Emmanuel Gobry (14 March 2011). "It Costs $326.60 To Make An iPad 2 — Why That Matters". Business Insider, Inc.. http://www.businessinsider.com/how-much-does-it-cost-to-build-the-ipad-2-2011-3. Retrieved March 14, 2011.

Retrieved from "http://en.wikipedia.org/w/index.php?title=Apple_Ax&oldid=480881207"

http://en.wikipedia.org/wiki/NovaThor

NovaThor is a platform consisting of integrated System on Chips (SoC) and modems for smartphones and tablets developed by ST-Ericsson, a 50/50 joint venture of Ericsson and STMicroelectronics established on February 3, 2009. ST-Ericsson also sells the SoCs (Nova) and the modems (Thor) separately.[1]

On November 2, 2011, Nokia announced that NovaThor SoCs will power Nokia's future Windows Phone based smartphones, making a deviation from the standard use of Qualcomm Snapdragon SoCs in Windows Phones.[2]

Page 54: Processor

On February 28, 2012, ST-Ericsson announced that they will switch to fully depleted silicon on insulator (FD-SOI) transistors in their upcoming NovaThor U8540 SoC which compared to L8540 will use 35% less power.[3]

Contents [hide] 

1 List of Nova and NovaThor SoCs 2 Similar platforms 3 References 4 External links

[edit] List of Nova and NovaThor SoCs

Model Number

Semiconductor

Technology

CPU Instructi

on SetCPU GPU

Memory Technolo

gy

Wireless Radio

Technologies

Availability

Utilizing Devices

NovaThor U5500 [4]

45nm ARMv7

Dual-core ARM Cortex-A9

ARM Mali 400 MP

LP-DDR2 EDGE, HSPA+ 2011

NovaThor U8500 [5]

45nm ARMv7

Dual-core ARM Cortex-A9

ARM Mali 400 MP

LP-DDR2 HSPA 2011

Sony Xperia P[6], Sony Xperia U, Samsung Galaxy Beam, Samsung Galaxy S Advance[7], Ontim WP8500[8]

NovaThor U9500 [9]

45nm ARMv7

Dual-core ARM Cortex-A9

ARM Mali 400 MP

LP-DDR2

GSM/EDGE, WCDMA/HSPA+ (HSDPA 21 Mbps, HSUPA 5.76 Mbps)

2011

Nova 45nm ARMv7 1.2 ARM LP-DDR2 None 2011

Page 55: Processor

A9500[10]

GHz Dual-core ARM Cortex-A9

Mali 400 MP

NovaThor L8540[11]

28nm ARMv7

1.85 GHz Dual-core ARM Cortex-A9

PowerVR SGX 544 @ 500 MHz[12]

Dual-channel LP-DDR2

LTE FDD/TDD, HSPA+, TD-SCDMA, EDGE (integrated modem)

2013

NovaThor L9540[13]

28nm ARMv7

1.85 GHz Dual-core ARM Cortex-A9

PowerVR SGX 544

Dual-channel LP-DDR2

LTE FDD/TDD, HSPA+, TD-SCDMA, EDGE (external Thor M7400 modem)

2013

Nova A9540[14]

28nm ARMv7

1.85 GHz Dual-core ARM Cortex-A9

PowerVR SGX 544

Dual-channel LP-DDR2

None 2013

Nova A9600[10]

28nm ARMv7

2.5 GHz Dual-core ARM Cortex-A15

PowerVR Series6 (Rogue)

None 2013

[edit] Similar platforms Snapdragon by Qualcomm OMAP by Texas Instruments Tegra by Nvidia Exynos by Samsung Ax by Apple Atom by Intel

Page 56: Processor

[edit] References1. ̂ NOVA TM - Highest performance application processors2. ̂ ST-Ericsson's NovaThor to power Nokia's Windows Phone devices, loosens

Qualcomm's grip - Engadget3. ̂ AnandTech - ST-E Will Have FD-SOI Based U8540: 35% Lower Power, Much Higher

Frequencies4. ̂ NovaThor U5500 Product Webpage; ST Ericsson.5. ̂ NovaThor U8500 Product Webpage; ST Ericsson.6. ̂ Specifications - Sony Smartphones (Global UK English)7. ̂ ST-Ericsson NovaThor U8500 powers new Samsung GALAXY S Advance8. ̂ ST-Ericsson NovaThor U8500 powers new tablet from Ontim9. ̂ NovaThor U9500 Product Webpage; ST Ericsson.10. ^ a b Changing the game: ST-Ericsson Unveils NovaThor™ Family of Smartphone

Platforms Combining its Most Advanced Application Processors with the Latest Generation of Modems

11. ̂ NOVATHOR™ L854012. ̂ ST-Ericsson announces new highly integrated LTE NovaThor platform13. ̂ NOVATHOR™ L954014. ̂ Nova™ A9540

[edit] External links Page on ST-Ericsson website about NovaThor.

http://en.wikipedia.org/wiki/Atom_%28system_on_chip%29

The new Atom system on chip (SoC) platform released in 2012[1] is an attempt by Intel to compete with existing SoCs developed for the smartphone and tablet market from companies like Texas instruments, Nvidia and Samsung. Unlike these companies, which uses a CPU inside their SoCs built on licence from ARM that has been designed from the beginning to consume very low amount of power to house hold with the limited power of a cell-phone battery, Intel has tried to adapt the x86 based Atom line CPU developed for low power usage in netbooks, to even lower power usage. Even though the architecture by its own most likely still is a bit less power efficient than ARMs designs, the advantage Intel has is their own manufacturing capabilities which makes it possible to stay one step ahead in semiconductor technology, currently using a 32 nm High-K/metal gate process whereas most competitors SoCs as of Q1 2012 are built with a 40/45 nm process by for example TSMC.

[edit] List of Intel Atom SoCs

Model Number

Semiconductor

CPU Instructio

CPU CPU Cache

GPU Memory Technolog

Availability

Utilizing

Page 57: Processor

Technology n Set y Devices

Atom Z2000

32 nm High-K/metal gate x86

1 GHz single-core Saltwell

L1: 32KB Instruction + 24KB Data, L2: 512KB

PowerVR SGX 540 @ 320 MHz[2]

Dual-channel 32bit LPDDR2

1H 2013

Atom Z2460[3] (codename Medfield)

32 nm High-K/metal gate x86

1.6–2 GHz single-core Saltwell with HT

L1: 32KB Instruction + 24KB Data, L2: 512KB

PowerVR SGX 540 @ 400 MHz[4]

Dual-channel 32bit LPDDR2 400 MHz

2012

Lenovo K800, Orange "Santa Clara", XOLO X900

Atom Z2580[5]

32 nm High-K/metal gate x86

1.8 GHz dual-core Saltwell with HT

L1: 32KB Instruction + 24KB Data, L2: 512KB

PowerVR SGX 544MP2 @ 533 MHz

Dual-channel 32bit LPDDR2

1H 2013

[edit] Similar platforms OMAP by Texas Instruments Snapdragon by Qualcomm Tegra by Nvidia Exynos by Samsung Ax by Apple NovaThor by ST-Ericsson

[edit] References1. ̂ http://newsroom.intel.com/community/intel_newsroom/blog/2012/01/10/intel-raises-

bar-on-smartphones-tablets-and-ultrabook-devices2. ̂ http://www.anandtech.com/show/5592/intel-atom-z2580-z20003. ̂ http://download.intel.com/newsroom/kits/ces/2012/pdfs/AtomprocessorZ2460.pdf4. ̂ http://download.intel.com/newsroom/kits/ces/2012/pdfs/AtomprocessorZ2460.pdf5. ̂ http://www.engadget.com/2012/02/27/intel-details-medfield-plans-announces-a-trio-of-

phone-friendly/

Page 58: Processor

http://en.wikipedia.org/wiki/Qualcomm

Qualcomm (NASDAQ: QCOM) is an American global telecommunication corporation that designs, manufactures and markets digital wireless telecommunications products and services based on its code division multiple access (CDMA) technology and other technologies. Headquartered in San Diego, CA, USA. The company operates through four segments: Qualcomm CDMA Technologies (QCT); Qualcomm Technology Licensing (QTL); Qualcomm Wireless & Internet (QWI), and Qualcomm Strategic Initiatives (QSI).

Contents [hide] 

1 Corporate history 2 Acquisitions 3 Mobile phone standards 4 Satellite phone network 5 Legal issues

o 5.1 Qualcomm's role in 3G 6 Products

o 6.1 Software 7 Locations 8 See also 9 References 10 Further reading 11 External links

[edit] Corporate historyQualcomm was founded in 1985 by MIT Alumnus and UC San Diego Professor Irwin Jacobs, MIT and USC Alumnus Andrew Viterbi, Harvey White, Adelia Coffman, Andrew Cohen, Klein Gilhousen, and Franklin Antonio. Jacobs and Viterbi had previously founded Linkabit. Qualcomm's first products and services included the OmniTRACS satellite locating and messaging service, used by long-haul trucking companies, developed from a product called Omninet owned by Parviz Nazarian and Neil Kadisha, and specialized integrated circuits for digital radio communications such as a Viterbi decoder.

Page 59: Processor

In 1990, Qualcomm began the design of the first CDMA-based cellular base station, based upon calculations derived from the CDMA-based OmniTRACS satellite system. This work began as a study contract from AirTouch which was facing a shortage of cellular capacity in Los Angeles. Two years later Qualcomm began to manufacture CDMA cell phones, base stations, and chips. The initial base stations were not reliable and the technology was licensed wholly to Nortel in return for their work in improving the base station switching. The first CDMA technology was standardized as IS-95. Qualcomm has since helped to establish the CDMA-2000 cellular standard.

In 1997, Qualcomm paid $18 million for the naming rights to the Jack Murphy Stadium in San Diego, renaming it to Qualcomm Stadium. The naming rights will belong to Qualcomm until 2017.[3]

In 1999, Qualcomm sold its base station business to Ericsson, and later, sold its cell phone manufacturing business to Kyocera. The company was now focused on developing and licensing wireless technologies and selling ASICs that implement them.

In 2011, Qualcomm announced that Steve Mollenkopf has been promoted to president and chief operating officer of the company, effective November 12.[4]

[edit] AcquisitionsIn 2000, Qualcomm acquired SnapTrack, the inventor of the assisted-GPS system for cellphones, branded as gpsOne. The Snaptrack patents describe how a cellphone can acquire a GPS signal rapidly using timing information sent from the base station. This reduces the searching time for geolocation from minutes down to roughly one second.

In October 2004, Qualcomm acquired Trigenix Ltd, a mobile user interface (UI) software development company, based in Cambridge, UK. After integrating the company, Qualcomm re-branded their interface markup language and its accompanying integrated development environment (IDE) as uiOne. In March 2009, Qualcomm informed their Cambridge engineering staff, mostly from the division working on uiOne, that they were going to be eliminated, and, in April that year, (after a legally required 30 day consultancy period) around 45 staff were let go. The rationale was stated as being a greater focus on deploying Flash Lite as a UI solution for Qualcomm-chipset-powered mobile phones. During 2004 Qualcomm also acquired Iridigm Corporation to form Qualcomm MEMS Technologies to develop low power reflective displays for mobile applications.

In 2006, Qualcomm purchased Flarion Technologies. Flarion is the creator of the Flash-OFDM wireless base station, and the inventor of the "flash" beaconing method and several other innovations in OFDM communications.

In 2007, Qualcomm purchased Open Interface North America, a Bluetooth stack software provider.

Page 60: Processor

In 2009, Qualcomm purchased AMD's handset division.[5] This acquisition formed the basis for the later Adreno chips.[6]

In 2010, Qualcomm announced acquisition of San Francisco based iSkoot Technologies Inc. Qualcomm did not disclose financial details of the acquisition.[7]

In 2011, Qualcomm announced acquisition of Atheros Communications Inc. for about $3.2 billion in cash, broadening its lineup of Wi-Fi networking technology.[8] In early February 2011, Qualcomm acquired the Canadian company of Sylectus.[9]. The name is now Qualcomm Atheros.

In July 2011, Qualcomm acquired some assets of GestureTek. The company plans to use the gesture recognition technology in its Snapdragon processors.[10]

In November 2011, Qualcomm acquired a substantial portfolio of assets and technology from HaloIPT. The company provides wireless charging technology for electric road vehicles.[11]

[edit] Mobile phone standardsQualcomm is the inventor of CDMAone (IS-95), CDMA 2000, and CDMA 1xEV-DO, which are wireless cellular standards used for communications. The company also owns significant number of key patents on the widely adopted 3G technology, W-CDMA.[12] The license streams from the patents on these inventions, and related products are a major component of Qualcomm's business.

In June 2011, Qualcomm announced that it will be releasing a set of applications programming interfaces.[13]

[edit] Satellite phone networkMain article: Globalstar

Qualcomm participated in the development of the Globalstar satellite system along with Loral Space & Communications. It uses a low-earth orbit (LEO) satellite constellation consisting of 44 active satellites. The system is used for voice telephony via hand-held satellite phones, asset tracking and data transfer using mobile satellite modems. The system was designed as a normal IS-95 system, and used the satellite as a "bent pipe" or "repeater" to transfer cellular signals from the handset to the terrestrial base station. Unlike the Iridium system, which routes phone calls between satellites, the Globalstar satellite must always be able to see both the handset and the base station to establish a connection, therefore, there is no coverage over the Earth's poles where there are no satellite orbits. Some of the Globalstar hardware is manufactured by Qualcomm. Like other satellite phone networks Globalstar went bankrupt in 1999, only to be bought up by a group of investors who are currently running the system. Those investors plan to launch a constellation supporting EV-DO in 2009.

[edit] Legal issues

Page 61: Processor

In April 2006, a dispute between Reliance Communications and Qualcomm over royalty fees cost Qualcomm approximately $11.7b in market capitalization.[14] In July 2007, Reliance and Qualcomm decided to make peace and agreed to expand the use of CDMA technology in India.[15]

In June 2007, the U.S. International Trade Commission blocked the import of new cell phone models based on particular Qualcomm microchips. They found that these Qualcomm microchips infringe patents owned by Broadcom. Broadcom has also initiated patent litigation in U.S. courts over this issue.

At issue is software designed to extend battery life in chips while users make out-of-network calls. In October, an ITC administrative judge made an initial ruling that Qualcomm violated the Broadcom patent covering that feature and the commission later affirmed the decision.

Sprint Nextel Corp. is using a software patch from Qualcomm to get around a U.S. government agency ban on new phones with Qualcomm chips.

In August 2007, Judge Rudi Brewster held that Qualcomm had engaged in litigation misconduct by withholding relevant documents during the lawsuit it brought against Broadcom and that Qualcomm employees had lied about their involvement.[16][17]

[edit] Qualcomm's role in 3G

The current UMTS air interfaces are for the most part based on Qualcomm patents, and royalties from these patents represent a significant part of Qualcomm's revenue.

This followed a series of patent-related lawsuits and antitrust complaints, spearheaded by Broadcom, in the US. In 2006, Broadcom started a series of patent-related lawsuits and antitrust complaints against Qualcomm to get what Broadcom regarded fair terms for access to the W-CDMA technologies. Broadcom was soon joined by Nokia and others, and complaints were also filed in the European Commission.[18]

The Chinese TDSCDMA 3G technology was developed primarily to avoid Qualcomm licensing fees, although Qualcomm claims that the Chinese technology still infringes on many Qualcomm patents.

October 2008, Nokia announced it will make a one time payment of $2.29 billion (US) to Qualcomm as part of its patent agreement with the company.

[edit] Products

Page 62: Processor

Qualcomm dual-band mobile phone

Tracking devices - OmniTRACS is a two-way satellite communications and geolocation trailer tracking technology designed for the over-the-road transport market. As of summer 2005, over 567,000 units have been shipped to transport companies on 4 continents.

Semiconductors - Qualcomm designs various ARM architecture CDMA and UMTS modem chipsets designated Mobile Station Modem (MSM), baseband radio processors, and power processor chips. These chipsets are sold to mobile phone manufacturers such as Kyocera, HTC Corporation, Motorola, Sharp, Sanyo, LG and Samsung for integration into CDMA and UMTS cell phones. Although a "fabless" semiconductor company, meaning Qualcomm does not engage in the actual manufacturing process, the chips the firm has designed are powering a significant number of handsets and devices world wide, both in CDMA and UMTS markets. As of summer of 2007, Qualcomm is among the top-ten semiconductor firms, after Intel, Texas Instruments, Samsung, and a few others.

Satellite phones - Qualcomm manufactures some of the handsets used on the Globalstar network.

MediaFLO - Qualcomm is the inventor of the MediaFLO system, based upon OFDM, which transmits 12-15 television channels within 6 MHz of spectrum. Qualcomm has standardized the lower layers of this design in TIA, and manufactures chips and software to add this television capability to cellphones.

Page 63: Processor

QChat - QChat is a cellular/data 2-way push-to-talk voice communications program. Nextel's original push-to-talk technology operates on the iDen network, but Qualcomm's Qchat push-to-talk operates on the EV-DO Revision A mobile broadband network. Sprint-Nextel's first Qchat phones were released in June 2008. Both iDen and Qchat handsets are sold under the Nextel brand. On November 29, 2009 Sprint issued a statement to PhoneNews.com that there are no new QChat handsets on the product development roadmap, but it will continue supporting its existing QChat subscribers.

Qualcomm Gobi - Qualcomm Gobi is a mobile broadband chipset used mainly for cellular data networking and it is also now used in a few enterprise smart phones (e.g. Motorola ES4000). It currently is a 3G technology capable up to HSPA on GSM and EV-DO rev.A on CDMA carriers. The Gobi chipset is a microprocessor that can load a specific carrier image so that the device appears to be specifically designed for that carrier's network. Since, GSM and CDMA are quite different and since Gobi devices and switch between them both using the same silicon their solution is considered to be innovative. Gobi Technology is best suited for large enterprise customers where a single mobile operator cannot serve all of their wireless modem needs since there is not one carrier that was provide the same level of service in all the places they need that service. The Gobi solution allows the IT department to roll out a single module on their laptop builds which can be configured to behave exactly like a device that is locked to the carrier that they want to use in that area. In the United States the exact same hardware can be used on the CDMA network or the GSM network of their choice. For GSM users that travel out of the United States the Gobi solution can be used to avoid international roaming charges by switching the SIM and the device's carrier image to a local provider instead of incurring the roaming charges. In both scenario's the customer must have different wireless accounts with each provider they wish to use natively. It typically takes 20 seconds for the device to load the carrier image into NVRAM and reset and come back online.

Gobi 3000 is the next hardware revision of the Gobi platform and it natively supports HSPA+. The model for Gobi 3000 is different. It is a reference design the OEMs can licence and produce their own Gobi 3000 compliant modules with their own extensions. Qualcomm does not sell any Gobi 3000 silicon. The reference design allows the same boiler plate hardware and software components for the basis of OEM chips which allow the OEMs to focus on innovations on the mobile broadband platform rather than getting bogged down with low-level RF implementations. Future Gobi platforms will support LTE natively. Currently, some Gobi 3000 modules support LTE through their own extensions.

mirasol Displays - mirasol Displays are the world's first and only reflective, bistable display based on IMOD technology. Qualcomm's mirasol displays use ambient light as their source of illumination and consume almost no power when the image is unchanged. This results in a very low power display solution that is visible even in direct sunlight.

[edit] Software

Page 64: Processor

Operating system - BREW (Binary Runtime Environment for Wireless) is a proprietary cell phone application platform. BREW is designed so that the platform rejects unsigned applications. In order to have an application signed, a developer must pay a testing fee to National Software Testing Labs (NSTL), which then can approve or deny the request. This allows carriers to maintain control over the applications that run on their customers' phones. BitPim is a popular open source program which can access the embedded filesystem on phones using Qualcomm MSMs via a cable or Bluetooth. It should be pointed out that signing systems are also used in Apple IOS, Java ME, and signing is often required by carriers and OEMs.

Speech codec - Qualcomm has developed an audio codec for speech called PureVoice,[19]

which besides use on mobile phones was also licensed for use in the very popular Chinese instant messaging software Tencent QQ.[20]

FEC codec - After its acquisition of Fremont-based Digital Fountain in 2009, Qualcomm developed the latest generation of Raptor codes called RaptorQ.[21]

Eudora client - Qualcomm formerly developed and distributed Eudora, which it acquired in 1991 from its author Steve Dorner. Qualcomm ceased sales of Eudora on May 1, 2007.[22] Qualcomm has committed to co-operating with Mozilla developers to develop a Eudora-like version of Thunderbird, called Project Penelope.[23]

Eudora servers - Qualcomm formerly developed and sold email servers for multiple platforms, including WorldMail for Windows and EIMS (Eudora Internet Mail Server) for Macintosh. Qualcomm no longer sells these products. Qualcomm continues to maintain and distribute the popular open-source Qpopper for Unix and Linux.

[edit] LocationsQualcomm offices are present in Australia, Austria, Belgium, Brazil, Canada, China, Finland, France, Germany, Hong Kong, India, Indonesia, Ireland, Israel, Italy, Japan, Mexico, Netherlands, Philippines, Russia, Saudi Arabia, Singapore, South Africa, South Korea, Spain, Sweden, Taiwan, Thailand, Turkey, UAE, United Kingdom, United States, and Vietnam.

[edit] See alsoSan Diego portal

Companies portal

Business incubator Smartbook Snapdragon DTRACS

http://en.wikipedia.org/wiki/Snapdragon_%28system_on_chip%29

Page 65: Processor

Snapdragon is a family of mobile system on chips by Qualcomm. Qualcomm considers Snapdragon a "platform" for use in smartphones, tablets, and smartbook devices.

The original Snapdragon CPU, dubbed Scorpion,[1] is Qualcomm's own design. It has many features similar to those of the ARM Cortex-A8 core and it is based on the ARM v7 instruction set, but theoretically has much higher performance for multimedia-related SIMD operations.[2] The successor to Scorpion, found in S4 Snapdragon SoCs is named Krait and has many similarities with the ARM Cortex-A15 CPU and is also based on the ARMv7 instruction set.

All Snapdragon processors contain the circuitry to decode high-definition video (HD) resolution at 720p or 1080p depending on the Snapdragon chipset.[3] Adreno, the company's proprietary GPU technology, integrated into Snapdragon chipsets (and certain other Qualcomm chipsets) is Qualcomm's own design, using assets the company acquired from AMD.[4]

Snapdragon chipsets, as of 2011, include a CPU "Krait" (with speeds up to 2.5 GHz), GPU "Adreno 225", 2G/3G/4G modem, and several Hexagon DSP coprocessors.[1]

Contents [hide] 

1 History 2 Current and future specifications 3 Similar platforms 4 See also 5 References 6 External links

[edit] History Q4 2008

The first chipsets in the Snapdragon family, the QSD8650 and the QSD8250, were made available.

June 2009

Qualcomm presented the Compal smartbook and ASUS Eee PC at Computex using the Snapdragon SoC and running Google's Android operating system.[5][6] At the same event, ASUS also showed a Snapdragon-based device, then withdrew it abruptly.[7][8]

7 December 2009

Page 66: Processor

The LG eXpo was the first US phone to utilize the Snapdragon SoC.[9]

5 January 2010

The Nexus One was released, manufactured by HTC, and featured Android OS 2.1 powered by a Snapdragon running at 1 GHz (Qualcomm QSD8250).[10]

29 April 2010

The HTC Droid Incredible was released, using the Snapdragon QSD8650 1 GHz SoC, and was the first Snapdragon device available on the Verizon Wireless network.

1 June 2010

Qualcomm announced sampling of the MSM8x60 series of Snapdragon SoC's.[11]

4 June 2010

The HTC Evo 4G was released, using the Snapdragon QSD8650 1 GHz SoC, and was available on the Sprint network. The HTC EVO 4G was the United States' first WiMAX phone.[12][13]

22 October 2010

The HTC Desire HD is released, featuring the MSM8255 SoC with Adreno 205 GPU.[citation needed]

17 November 2010

Qualcomm announces the roadmap for Next-Gen Snapdragon SoC development, including the MSM8960, citing future improvements in CPU and GPU performance and lower power consumption.[14]

5 January 2011

A version of Microsoft Windows compiled for ARM is shown running on the Snapdragon SoC at CES 2011.[15]

13 February 2011

The HTC Inspire 4G is released, featuring the MSM8255 SoC.[16]

21 March 2011

The HTC Evo 3D features the MSM8660 Dual-Core SoC with Adreno 220 GPU.

Page 67: Processor

The HTC Thunderbolt features the MSM8655 SoC.

3 August 2011

Qualcomm announces plan to use simple names (S1, S2, S3 and S4) for Snapdragon processors so that the public can better understand the products. The bigger the number is, the more advanced functions that the processor has, which means S4 can perform better than S3.[17]

[edit] Current and future specifications[3][11][18]

Family / generati

on

Model Number

Process

CPU Instruction Set

CPU

CPU

Cache

GPUMemory Technol

ogy

Wireless Radio

Technologies

Sampling

Availability

Utilizing Devices

Snapdragon S1

QSD8250 65 nm

ARMv7

1 GHz Scorpion

Adreno 200

GSM (GPRS, EDGE), UMTS/WCDMA (HSDPA, HSUPA), MBMS

Q4 2008

Acer Stream/Liquid, Acer neoTouch S200, Dell Venue Pro (Lightning), Dell Streak, Fujitsu Toshiba Mobile REGZA Phone T-01C, HP Compaq AirLife 100, HTC Desire, HTC HD2, HTC 7 Mozart, HTC 7 Surround, HTC 7 Trophy, HTC HD7, HTC 7 Pro, Nexus One, Huawei

Page 68: Processor

SmaKit S7, Lenovo LePhone, LG Optimus Q, LG Optimus Z, LG Quantum, LG Panther, Pantech IM-A600S, Pantech IM-A650S, Sharp LYNX SH-10B, Sharp LYNX 3D SH-03C, Samsung Focus, Samsung Omnia 7, Sony Ericsson Xperia X10, Toshiba dynapocket T-01B/KG01, Toshiba TG01/TG02/TG03.

QSD8650 65 nm

ARMv7

1 GHz Scorpion

Adreno 200

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO

Q4 2008

Fujitsu Toshiba Mobile REGZA Phone IS04(TSI04), Fujitsu Toshiba Mobile T006 (TS006)/iida X-Ray (TSX06), HTC

Page 69: Processor

MC Rev.A) Arrive, HTC Droid Incredible, HTC Evo 4G, LG Apollo GW990, LG Fathom VS750, LG GW820 eXpo, LG GW825 IQ, LG Optimus 7, Sharp IS01 (SHI01)/IS03 (SHI03), Sony Ericsson S004 (SO004)/S005 (SO005)/S006 (SO006)/iida G11 (SOX02)/S007 (SO007), Toshiba dynapocket IS02 (TSI01)/K01, Toshiba T004 (TS004), Pantech Sirius α IS06 (PTI06), Kyocera Echo, Fujitsu Toshiba Mobile T007

Page 70: Processor

(TS007), Fujitsu Toshiba Mobile T008 (TS008), NEC Casio CA007, Kyocera K009 (KY009), Fujitsu F001 (FJ001), Sony Ericsson Arbano Affare (SOY05)

QSD8250A 45 nm

ARMv7

1.3 GHz Scorpion

Adreno 205

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA), MBMS

Q4 2009  

QSD8650A 45 nm

ARMv7

1.3 GHz Scorpion

Adreno 205

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A)

Q4 2009

Lenovo LePad

MSM7225A

45 nm

ARMv7

600 MHz CortexA5

Adreno 200 (enhanced)

GSM (GPRS, EDGE), W-CDMA/UM

Q4 2011

HTC Explorer, Motorola Defy Mini

Page 71: Processor

TS (HSDPA, HSUPA), MBMS

MSM7625A

45 nm

ARMv7

800 MHz CortexA5

Adreno 200 (enhanced)

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A)

Q4 2011

MSM7227A

45 nm

ARMv7

800-1000 MHz CortexA5

Adreno 200 (enhanced)

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA), MBMS

Q4 2011

Motorola XT615, Motorola Motoluxe, Samsung Galaxy Ace Plus, Samsung Galaxy Mini 2

MSM7627A

45 nm

ARMv7

800 MHz CortexA5

Adreno 200 (enhanced)

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A)

Q4 2011

Snapdragon S2

MSM7230 45 nm

ARMv7

800 MHz

Adreno 205

GSM (GPRS,

Q2 2010

Acer Liquid Metal, HP

Page 72: Processor

Scorpion

EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS

Veer, HTC Desire Z, Huawei U8800, NEC Casio Medias N-04C

MSM7630 45 nm

ARMv7

800 MHz Scorpion

Adreno 205

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A, SV-DO)

Q2 2010

Casio G'zOne Commando, HTC Evo Shift 4G, HTC Merge

MSM8255 45 nm

ARMv7

1 GHz Scorpion

Adreno 205

Dual-channel 333 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS

Q2 2010

Acer Iconia Smart, Acer Allegro, Fujitsu F-12C, HTC Desire HD, HTC Desire S, HTC Incredible S, HTC Inspire 4G, HTC Radar, Huawei U9000 Ideos X6, Huawei Vision, LG Eclypse,[19] Samsung Exhibit II 4G, Sharp Galapagos

Page 73: Processor

003SH/005SH, Sharp Aquos Phone f (SH-13C), Sharp Aquos Phone the Hybrid (007SH/007SH J), Sharp Aquos Phone the Premium (009SH), Sony Ericsson Live with Walkman, Sony Ericsson Xperia Active, Sony Ericsson Xperia Arc, Sony Ericsson Xperia Acro (SO-02C), Sony Ericsson Xperia Neo, Sony Ericsson Xperia Play, Sony Ericsson Xperia Pro, Sony Ericsson Xperia Mini, Sony Ericsson

Page 74: Processor

Xperia Mini Pro, Sony Ericsson Xperia Ray, T-Mobile myTouch 4G, ZTE Tania, ZTE 008Z

MSM8655 45 nm

ARMv7

1.2 GHz Scorpion

Adreno 205

Dual-channel 333 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B)

Q2 2010

BlackBerry Bold 9900/9930, BlackBerry Torch 9810, BlackBerry Torch 9860, Fujitsu Toshiba IS12T, HTC Droid Incredible 2,[20] HTC Evo Design 4G,[21] HTC Thunderbolt, LG Revolution, Motorola Triumph, Pantech Mirach IS11PT, Samsung Conquer 4G,[22] Sharp IS05 (SHI05), Sony Ericsson Xperia Acro (IS11S)

MSM8255T

45 nm

ARMv7

1.4-1.5 GHz Scorpi

Adreno 205

Dual-channel 333 MHz

GSM (GPRS, EDGE), W-CDMA/UM

2011 HP Pre 3, HTC Flyer, HTC Titan, HTC Titan

Page 75: Processor

on LPDDR2

TS (HSDPA, HSUPA, HSPA+), MBMS

II, Nokia Lumia 710, Nokia Lumia 800, Samsung Focus S, Samsung Galaxy S Plus,[23] Samsung Galaxy W,[24] Samsung Omnia W, Sharp Aquos Phone SH-12C,[25] Sharp Aquos Phone 006SH, Sony Xperia arc S[26]

MSM8655T

45 nm

ARMv7

1.4-1.5 GHz Scorpion

Adreno 205

Dual-channel 333 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B)

2011 HP Pre 3, Kyosera Digno ISW11K, NEC Casio Medias BR IS11N, Sharp Aquos Phone IS11SH,[citation needed] Sharp Aquos Phone IS12SH,[citation needed] Sharp Aquos Phone IS13SH, Toshiba

Page 76: Processor

Regza Phone IS11T[citation

needed]

Snapdragon S3

APQ8060 45 nm

ARMv7

1.2-1.5 GHz Dual-core Scorpion

L2: 512 kB

Adreno 220

Single-channel 333 MHz ISM/266 MHz LPDDR2

Connectivity features not included

2011

HP TouchPad, HTC Jetstream, HTC Vivid, LG Nitro HD, Samsung Galaxy Note (GT-N7003), Samsung Galaxy S II (SGH-T989), Samsung Galaxy S II LTE

MSM8260 45 nm

ARMv7

1.2-1.5 GHz Dual-core Scorpion

Adreno 220

Single-channel 333 MHz ISM/266 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS

Q3 2010

Asus Eee Pad Memo,[citation needed] HTC Amaze 4G, HTC Evo 3D (GSM), HTC Sensation, HTC Sensation XE, Huawei Mediapad,[citation needed] Samsung Galaxy S II Skyrocket, Sony Xperia Ion,[citation needed] Sony Xperia S, T-Mobile myTouch

Page 77: Processor

4G Slide, Xiaomi MI-One, ZTE V71A[citation

needed]

MSM8660 45 nm

ARMv7

1.2-1.5 GHz Dual-core Scorpion

Adreno 220

Single-channel 333 MHz ISM/266 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A)

Q3 2010

HTC Evo 3D (CDMA), HTC Rezound, LG Optimus LTE LU6200,[citation needed] Pantech Vega Racer

Snapdragon S4

MSM8230 28 nm

ARMv7

1.0-1.2 GHz Dual-core Krait

L2: 1 MB

Adreno 305

Single-channel 533 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+ cat.14), MBMS

Q3 2012

MSM8260A

28 nm

ARMv7

1.5-1.7 GHz Dual-core Krait

L2: 1 MB

Adreno 225

Dual-channel 500 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, DC-HSPA+ cat.28, MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B,

1H 2012

Asus Padfone,[ citation needed] HTC One S,[27] Panasonic Eluga Power,[citation

needed] ZTE V9S[citation

needed]

Page 78: Processor

1xEV-DO MC Rev.A),TD-SCDMA

MSM8960[28][29][30]

28 nm

ARMv7

1.5-1.7 GHz Dual-core Krait

L2: 1 MB

Adreno 225

Dual-channel 500 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+, DC-HSPA+ cat.28), MBMS,LTE cat.3,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A),TD-SCDMA

1H 2012

Asus Transformer Pad Infinity (3G/4G version),[31] HTC One XL,[27] ZTE V96[citation

needed]

MSM8960 Pro[32]

28 nm

ARMv7

1.5-1.7 GHz Dual-core Krait

L2: 1 MB

Adreno 320

Dual-channel 500 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+, DC-HSPA+ cat.28), MBMS,LTE cat.3,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A),TD-SCDMA

2H 2012

Page 79: Processor

MSM8270[29]

28 nm

ARMv7

1.5-1.7 GHz Dual-core Krait

L2: 1 MB

Adreno 225

Dual-channel 500 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, DC-HSPA cat.21), MBMS

1H 2012  

APQ8064[28][29][30]

28 nm

ARMv7

2.5 GHz Quad-core Krait

Adreno 320

Connectivity features not included

2012

Snapdragon ??

QSD8672 45 nm

ARMv7

1.5 GHz Dual-core Scorpion

Adreno 220

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+), MBMS,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A)

originally Q1 2010; cancelled

 

MSM8930[28][29][30]

28 nm

ARMv7

1.0-1.2 GHz Single-core Krait

L2: 1 MB

Adreno 305

Single-channel 533 MHz LPDDR2

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+, DC-HSPA+ cat.28), MBMS,LTE cat.2,CDMA2000 (1xRTT, 1xEV-DO

Q3 2012

 

Page 80: Processor

Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A),TD-SCDMA

MSM8974[30]

28 nm

ARMv7

2.0-2.5 GHz Quad-core Krait

L2: 2 MB

Adreno 320

Dual-channel 667/800 MHz LPDDR3

GSM (GPRS, EDGE), W-CDMA/UMTS (HSDPA, HSUPA, HSPA+, DC-HSPA+ cat.28), MBMS,LTE cat.4,CDMA2000 (1xRTT, 1xEV-DO Rel.0/Rev.A/Rev.B, 1xEV-DO MC Rev.A, 1xAdv Rev.A/Rev.B),TD-SCDMA

Q1 2013  

Some Snapdragon designs like QSD8672 were announced but have never made it into production and were superseded by a newer generation of chips.

[edit] Similar platforms OMAP by Texas Instruments Tegra by Nvidia Exynos by Samsung Ax by Apple NovaThor by ST-Ericsson Atom by Intel

[edit] See also

Page 81: Processor

Qualcomm MSM7000 Smartbook , a new netbook-like class of devices, first models of which are powered by

Snapdragon PXA by Marvell i.MX by Freescale SH-Mobile by Renesas Nomadik (discontinued) by ST-Ericsson ZiiLabs ZMS series Snapdragon Stadium , temporary name used for Qualcomm Stadium in December 2011

[edit] References1. ^ a b "Snapdragon S4 Processors: System on a Chip Solution for a New Mobile Age;

White Paper". Qualcomm. 2011. http://www.qualcomm.com/documents/files/snapdragon-s4-processors-system-on-chip-solutions-for-a-new-mobile-age-white-paper.pdf. Retrieved 21 Jan 2012.

2. ̂ "Analysis: Qualcomm's 1 GHz ARM "Snapdragon"". EE Times. UBM Electronics. 5 December 2007. http://eetimes.com/design/microcontroller-mcu/4017566. Retrieved 29 Jan 2012.

3. ^ a b "Snapdragon - Technical Features". Qualcomm. http://www.qualcomm.com/products_services/chipsets/snapdragon.html. Retrieved 29 Dec 2009.

4. ̂ "Qualcomm Acquires Handheld Graphics and Multimedia Assets from AMD". Qualcomm. 20 January 2009. http://www.qualcomm.com/news/releases/2009/01/20/qualcomm-acquires-handheld-graphics-and-multimedia-assets-amd. Retrieved 14 Sept 2010.

5. ̂ jkkmobile (3 June 2009). "Compal Snapdragon smartbook at computex". YouTube. http://www.youtube.com/watch?v=fSAh4sjnX40. Retrieved 29 Jan 2012.

6. ̂ Sumner Lemon (1 June 2009). "Qualcomm shows Eee PC running Android OS". PC World AU. IDG. http://www.goodgearguide.com.au/article/305316/qualcomm_shows_eee_pc_running_android_os. Retrieved 29 Jan 2012.

7. ̂ Steven J. Vaughan-Nichols (2 June 2009). "Microsoft strikes back at Linux netbook push". Computerworld. IDG. http://blogs.computerworld.com/microsoft_strikes_back_at_linux_netbook_push. Retrieved 20 June 2009.

8. ̂ Charlie Demerjian (12 June 2009). "MS steps on a Snapdragon". SemiAccurate. Stone Arch Networking Services, Inc.. http://www.semiaccurate.com/2009/06/12/ms-steps-snapdragon/. Retrieved 20 June 2009.

9. ̂ "AT&T and LG Mobile Phones Announce the First 1   GHz Smartphone in the United States, the LG Expo". AT&T. http://www.att.com/gen/press-room?pid=4800&cdvn=news&newsarticleid=27621. Retrieved 29 Jan 2012.

10. ̂ "Nexus One - Google Phone Gallery". Google. https://www.google.com/phone/detail/nexus-one. Retrieved 29 Jan 2012.

Page 82: Processor

11. ^ a b "Qualcomm Ships First Dual-CPU Snapdragon Chipset". Qualcomm. 1 June 2010. http://www.qualcomm.com/news/releases/2010/06/01/qualcomm-ships-first-dual-cpu-snapdragon-chipset. Retrieved 29 Jan 2012.

12. ̂ Paul Miller (23 March 2010). "HTC EVO 4G is Sprint's Android-powered knight in superphone armor, we go hands-on". Engadget. AOL. http://www.engadget.com/2010/03/23/htc-evo-4g-is-sprints-android-powered-knight-in-superphone-armo/. Retrieved 29 Jan 2012.

13. ̂ Matt Buchanan (5 November 2010). "The Dirty Secret of Today's 4G: It's not 4G". Gizmodo. Gawker Media. http://gizmodo.com/5680755/. Retrieved 29 Jan 2012.

14. ̂ Anand Lal Shimpi (17 November 2010). "Qualcomm Reveals Next-Gen Snapdragon MSM8960: 28nm, dual-core, 5x Performance Improvement". AnandTech. http://www.anandtech.com/show/4024/. Retrieved 29 Jan 2012.

15. ̂ "Windows runs on Arm's mobile phone chips". BBC. 6 January 2011. http://www.bbc.co.uk/news/technology-12125541. Retrieved 29 Jan 2012.

16. ̂ "HTC Inspire 4G at AT&T". HTC. http://www.htc.com/us/products/inspire-att. Retrieved 29 Jan 2012.

17. ̂ Anand Lal Shimpi (3 August 2011). "Qualcomm's Updated Brand: Introducing Snapdragon S1, S2, S3 & S4 Processors". AnandTech. http://www.anandtech.com/show/4565/. Retrieved 28 Jan 2012.

18. ̂ "PDAdb.net - The World's Largest Smartphone, Tablet, Netbook, PDA, PNA & Mobile Device Database". PDAdb. http://www.pdadb.net. Retrieved 29 Jan 2012.

19. ̂ "LG Eclypse C800G". LG. http://www.lg.com/ca_en/mobile-phones/all-phones/LG-C800G.jsp. Retrieved 29 Jan 2012.

20. ̂ "Droid Incredible 2 by HTC". HTC. http://www.htc.com/us/products/incredible2-verizon/#tech-specs. Retrieved 29 Jan 2012.

21. ̂ "HTC Evo Design 4G (Sprint)". HTC. http://www.htc.com/us/products/evodesign4g-sprint#tech-specs. Retrieved 1 Feb 2012.

22. ̂ Natalie Papaj (5 August 2011). "Samsung Conquer 4G fact sheet". Sprint Nextel. http://newsroom.sprint.com/news/samsung-conquer-fact-sheet.htm. Retrieved 29 Jan 2012.

23. ̂ Vlad Savov (31 March 2011). "Samsung Galaxy S getting a 1.4GHz '2011 edition' next month (update: confirmed)". Engadget. AOL. http://www.engadget.com/2011/03/31/samsung-galaxy-s-rumored-to-be-getting-a-1-4ghz-2011-edition-n/. Retrieved 28 Jan 2012.

24. ̂ "Samsung Galaxy W I8150 - Full phone specifications". GSMArena. http://www.gsmarena.com/samsung_galaxy_w_i8150-4114.php. Retrieved 29 Jan 2012.

25. ̂ Myriam Joire (20 June 2011). "Sharp Aquos SH-12C 3D smartphone hands-on (video)". Engadget. AOL. http://www.engadget.com/2011/06/20/sharp-aquos-sh-12c-3d-smartphone-hands-on-video/. Retrieved 29 Jan 2012.

26. ̂ "Sony Ericsson unveils its fastest entertainment experiences to date with Xperia arc S". Sony Ericsson. 31 August 2011. http://www.sonyericsson.com/cws/corporate/press/pressreleases/pressreleasedetails/xperiaarcspressreleasefinal-20110831. Retrieved 29 Jan 2012.

27. ^ a b Anand Lal Shimpi (26 February 2012). "HTC's New Strategy - The HTC One". AnandTech. http://www.anandtech.com/show/5584. Retrieved 27 Feb 2012.

Page 83: Processor

28. ^ a b c "Qualcomm Announces Next-generation Snapdragon Mobile Chipset Family". Qualcomm. 14 February 2011. http://www.qualcomm.com/media/releases/2011/02/14/qualcomm-announces-next-generation-snapdragon-mobile-chipset-family. Retrieved 29 Jan 2012.

29. ^ a b c d Makram E. Daou (25 April 2011). "Qualcomm roadmap detailed: Quad Core CPU and GPU chipsets coming later this year". MobileTechWorld. http://www.mobiletechworld.com/2011/04/25/qualcomm-roadmap-detailed-quad-core-cpu-and-gpu-chipsets-coming-later-this-year/. Retrieved 29 Jan 2012.

30. ^ a b c d Makram E. Daou (5 July 2011). "New Qualcomm 2011 / 2012 roadmap and SoC specifications". MobileTechWorld. http://www.mobiletechworld.com/2011/07/05/new-qualcomm-2011-2012-roadmap-and-soc-specifications/. Retrieved 29 Jan 2012.

31. ̂ Anand Lal Shimpi (27 February 2012). "The ASUS Transformer Pad Infinity: 1920 x 1200 Display, Krait Optional". AnandTech. http://www.anandtech.com/show/5586. Retrieved 27 Feb 2012.

32. ̂ Sharif Sakr (27 February 2012). "Qualcomm unleashes Snapdragon S4 'Pro' chip with superior graphics". Engadget. AOL. http://www.engadget.com/2012/02/27/qualcomm-unleashes-snapdragon-s4-pro/. Retrieved 27 Feb 2012.

[edit] External links Snapdragon Chipset Product Page

http://en.wikipedia.org/wiki/XScale#PXA

The XScale, a microprocessor core, is Intel's and Marvell's implementation of the ARMv5 architecture, and consists of several distinct families: IXP, IXC, IOP, PXA and CE (see more below). Intel sold the PXA family to Marvell Technology Group in June 2006.[1]

The XScale architecture is based on the ARMv5TE ISA without the floating point instructions. XScale uses a seven-stage integer and an eight-stage memory superpipelined microarchitecture. It is the successor to the Intel StrongARM line of microprocessors and microcontrollers, which Intel acquired from DEC's Digital Semiconductor division as the side effect of a lawsuit between the two companies. Intel used the StrongARM to replace its ailing line of outdated RISC processors, the i860 and i960.

All the generations of XScale are 32-bit ARMv5TE processors manufactured with a 0.18 µm or 0.13 µm (as in IXP43x parts) process and have a 32 kB data cache and a 32 kB instruction cache. First and second generation XScale cores also have a 2 kB mini-data cache. Products based on the 3rd generation XScale have up to 512 kB unified L2 cache.[2]

Contents [hide] 

1 Processor families

Page 84: Processor

o 1.1 PXA 1.1.1 PXA210/PXA25x 1.1.2 PXA26x 1.1.3 PXA27x 1.1.4 PXA3xx

o 1.2 PXA16xo 1.3 PXA90xo 1.4 PXA930/935o 1.5 PXA940o 1.6 IXC

1.6.1 IXC1100o 1.7 IOPo 1.8 IXP network processoro 1.9 CE

2 Applications 3 Sale of PXA processor line 4 See also 5 References 6 External links

[edit] Processor familiesThe XScale core is used in a number of microcontroller families manufactured by Intel and Marvell, notably:

Application Processors (with the prefix PXA). There are four generations of XScale Application Processors, described below: PXA210/PXA25x, PXA26x, PXA27x, and PXA3xx.

I/O Processors (with the prefix IOP) Network Processors (with the prefix IXP) Control Plane Processors (with the prefix IXC). Consumer Electronics Processors (with the prefix CE).

There are also standalone processors: the 80200 and 80219 (targeted primarily at PCI applications).

[edit] PXA

[edit] PXA210/PXA25x

The PXA210 was Intel's entry-level XScale targeted at mobile phone applications. It was released with the PXA250 in February 2002 and comes clocked at 133 MHz and 200 MHz.

Page 85: Processor

The PXA25x family consists of the PXA250 and PXA255. The PXA250 was Intel's first generation of XScale processors. There was a choice of three clock speeds: 200 MHz, 300 MHz and 400 MHz. It came out in February 2002. In March 2003, the revision C0 of the PXA250 was renamed to PXA255. The main differences were a doubled internal bus speed (100 MHz to 200 MHz) for faster data transfer, lower core voltage (only 1.3 V at 400 MHz) for lower power consumption and writeback functionality for the data cache, the lack of which had severely impaired performance on the PXA250.

[edit] PXA26x

The PXA26x family consists of the PXA260 and PXA261-PXA263. The PXA260 is a stand-alone processor clocked at the same frequency as the PXA25x, but features a TPBGA package which is about 53% smaller than the PXA25x's PBGA package. The PXA261-PXA263 are the same as the PXA260 but have Intel StrataFlash memory stacked on top of the processor in the same package; 16 MB of 16-bit memory in the PXA261, 32 MB of 16-bit memory in the PXA262 and 32 MB of 32-bit memory in the PXA263. The PXA26x family was released in March 2003.

[edit] PXA27x

e-con Systems eSOM270 Marvell XScale PXA270 Computer on module

The PXA27x family (code-named Bulverde) consists of the PXA270 and PXA271-PXA272 processors. This revision is a huge update to the XScale family of processors. The PXA270 is clocked in four different speeds: 312 MHz, 416 MHz, 520 MHz and 624 MHz and is a stand-alone processor with no packaged memory. The PXA271 can be clocked to 13, 104, 208 MHz or 416 MHz and has 32 MB of 16-bit stacked StrataFlash memory and 32 MB of 16-bit SDRAM in the same package. The PXA272 can be clocked to 312 MHz, 416 MHz or 520 MHz and has 64 MB of 32-bit stacked StrataFlash memory.

Intel also added many new technologies to the PXA27x family such as:

Wireless SpeedStep: the operating system can clock the processor down based on load to save power.

Wireless MMX: 43 new SIMD instructions containing the full MMX instruction set and the integer instructions from Intel's SSE instruction set along with some instructions unique to the XScale. Wireless MMX provides 16 extra 64-bit registers that can be treated as an array of two 32-bit words, four 16-bit halfwords or eight 8-bit bytes. The

Page 86: Processor

XScale core can then perform up to eight adds or four MACs in parallel in a single cycle. This capability is used to boost speed in decoding and encoding of multimedia and in playing games.

Additional peripherals, such as a USB-Host interface and a camera interface. Internal 256 kB SRAM to reduce power consumption and latency.

The PXA27x family was released in April 2004. Along with the PXA27x family Intel released the 2700G embedded graphics co-processor.

[edit] PXA3xx

Toradex Colibri XScale Monahans PXA290 SODIMM-module (Prototype Of Marvell PXA320 SODIMM-module)

In August 2005 Intel announced the successor to Bulverde, codenamed Monahans.

They demonstrated it showing its capability to play back high definition encoded video on a PDA screen.

The new processor was shown clocked at 1.25 GHz but Intel said it only offered a 25% increase in performance (800 MIPS for the 624 MHz PXA270 processor vs. 1000 MIPS for 1.25 GHz Monahans). An announced successor to the 2700G graphics processor, code named Stanwood, has since been canceled. Some of the features of Stanwood are integrated into Monahans. For extra graphics capabilities, Intel recommends third-party chips like the NVIDIA GoForce chip family.

In November 2006, Marvell Semiconductor officially introduced the Monahans family as Marvell PXA320, PXA300, and PXA310.[3] PXA320 is currently shipping in high volume, and is scalable up to 806 MHz. PXA300 and PXA310 deliver performance "scalable to 624 MHz", and are software-compatible with PXA320.

[edit] PXA16x

Page 87: Processor

PXA168 System On Module by tianyeit.com

The pxa16x delivers strong performance at a mass market price point for cost sensitive consumer and embedded markets such as digital picture frames, E Readers, multifunction printer user interface (UI) displays, interactive VoIP phones, IP surveillance cameras, and home control gadgets..[4]

[edit] PXA90x

The PXA90x was released by Marvell and combines an XScale Core with a GSM/CDMA communication module.[5] The PXA90x is build using a 130 nm process[6]

[edit] PXA930/935

The PXA930 and PXA935 processor series were built using an architecture developed by Marvell,[7] instead of using an Xscale or ARM design. This design, called the Sheeva core,[8] is ARM-compliant. The Sheeva core is a so-called Tri-core architecture[8] codenamed Tavor; Tri-core means it supports the ARMv5TE, ARMv6 and ARMv7 instruction sets.[8][9] This new architecture was a significant leap from the old Xscale architecture. The PXA930 uses 65 nm technology[10] while the PXA935 is build using the 45 nm process.[9]

The PXA930 is used in the Blackberry Bold 9700.

[edit] PXA940

Little is known about the PXA940, although it is known to be ARM Cortex-A8 compliant.[11] It is utilized in the Blackberry Torch 9800[12][13] and is built using 45 nm technology.

[edit] IXC

[edit] IXC1100

Page 88: Processor

The IXC1100 processor features clock speeds at 266, 400, and 533 MHz, a 133 MHz bus, 32 kB of instruction cache, 32 kB of data cache, and 2 kB of mini-data cache. It is also designed for low power consumption, using 2.4 W at 533 MHz. The chip comes in the 35 mm PBGA package.

[edit] IOP

The IOP line of processors is designed to allow computers and storage devices to transfer data and increase performance by offloading I/O functionality from the main CPU of the device. The IOP3XX processors are based on the XScale architecture and designed to replace the older 80219 processor and i960 family of chips. There are ten different IOP processors currently available: IOP303, IOP310, IOP315, IOP321, IOP331, IOP332, IOP333, IOP341, IOP342 and IOP348. Clock speeds range from 100 MHz to 1.2 GHz. The processors also differ in PCI bus type, PCI bus speed, memory type, maximum memory allowable, and the number of processor cores.

[edit] IXP network processor

The XScale core is utilized in the second generation of Intel's IXP network processor line, while the first generation used StrongARM cores. The IXP network processor family ranges from solutions aimed at small/medium office network applications , IXP4XX, to high performance network processors such as the IXP2850, capable of sustaining up to OC-192 line rates. In IXP4XX devices the XScale core is used as both a control and data plane processor, providing both system control and data processing. The task of the XScale in the IXP2XXX devices is typically to provide control plane functionality only, with data processing performed by the microengines, examples of such control plane tasks include routing table updates, microengine control, memory management.

[edit] CE

In April 2007, Intel announced an XScale based processor targeting consumer electronics markets, the Intel CE 2110.[14]

[edit] ApplicationsXScale microprocessors can be found in products such as the popular RIM BlackBerry handheld, the Dell Axim family of Pocket PCs, most of the Zire, Treo and Tungsten Handheld lines by Palm, later versions of the Sharp Zaurus, the Motorola A780, the Acer n50, the Compaq iPaq 3900 series and many other PDAs. It is used as the main CPU in the Iyonix PC desktop computer running RISC OS, and the NSLU2 (Slug) running a form of Linux. The XScale is also used in devices such as PVPs (Portable Video Players), PMCs (Portable Media Centres), including the Creative Zen Portable Media Player and Amazon Kindle E-Book reader, and industrial embedded systems. At the other end of the market, the XScale IOP33x Storage I/O processors are used in some Intel Xeon-based server platforms.

[edit] Sale of PXA processor line

Page 89: Processor

On June 27, 2006, the sale of Intel's XScale PXA mobile processor assets was announced. Intel agreed to sell the XScale PXA business to Marvell Technology Group for an estimated $600 million in cash and the assumption of unspecified liabilities. The move was intended to permit Intel to focus its resources on its core x86 and server businesses. Marvell holds a full Architecture License for ARM, allowing it to design chips to implement the ARM instruction set, not just license a processor core.[15]

The acquisition was completed on November 9, 2006. Intel was expected to continue manufacturing XScale processors until Marvell secures other manufacturing facilities, and would continue manufacturing and selling the IXP and IOP processors, as they were not part of the deal.[16]

The XScale effort at Intel was initiated by the purchase of the StrongARM division from Digital Equipment Corporation in 1998.[17] Intel still holds an ARM license even after the sale of XScale.[17]

[edit] See also RedBoot open-source bootloader, the standard boot firmware shipped with XScale boards

[edit] References1. ̂ Marvell buys Intel's handheld processor unit for $600 million2. ̂ 3rd Generation Intel XScale(R) Microarchitecture Developer's Manual -

http://www.intel.com/design/intelxscale/316283.htm3. ̂ Marvell Introduces Next Generation Application Processors4. ̂ Marvell ARMADA 100 Processors product page5. ̂ Marvell Communications Processors product page6. ̂ http://pdadb.net/index.php?m=cpu&id=a900&c=intel_xscale_pxa9007. ̂ http://translate.google.nl/translate?js=y&prev=_t&hl=nl&ie=UTF-

8&layout=1&eotf=1&u=http%3A%2F%2Ftweakers.net%2Freviews%2F1766%2F4%2Fblackberry-pearl-3g-klein-en-kwiek-hardware-en-bouwkwaliteit.html&sl=nl&tl=en

8. ^ a b c http://www.marvell.com/company/news/press_detail.html?releaseID=13299. ^ a b http://pdadb.net/index.php?m=cpu&id=a935&c=marvell_pxa93510. ̂ http://pdadb.net/index.php?m=cpu&id=a930&c=marvell_pxa93011. ̂ http://extranet.marvell.com/technologies/cpu_history/cpu_history.jsp12. ̂ http://www.ubmtechinsights.com/reports-and-subscriptions/investigative-analysis/

blackberry-torch-9800/teardown/13. ̂ http://www.ubmtechinsights.com/uploadedImages/Public_Website/Content_-

_Primary/Investigative_Analysis/2010/Blackberry_Torch_9800/torch-front-web.jpg14. ̂ Intel System-On-A-Chip Media Processor Powers New Generation Of Consumer

Electronics Devices15. ̂ http://www.marvell.com/products/cellular/xscale_micro.jsp16. ̂ Intel ditches mobile phone processors

Page 90: Processor

17. ^ a b http://www.reghardware.co.uk/2006/06/27/intel_sells_xscale/

[edit] External links Intel XScale Technology Overview IXP4XX Toolkits Intel StrataFlash Memory Marvell PXA168 high-performance processor product brief Optimized Linux Code for Intel XScale Microarchitecture