Turion 64 X2

Turion 64 X2 is AMD's 64-bit dual-core mobile CPU, intended to compete with Intel's Core and Core 2 CPUs. The Turion 64 X2 was launched on May 17, 2006[1], after several delays. These processors use Socket S1, and feature DDR2 memory. They also include AMD Virtualization Technology and more power-saving features.

AMD first produced the Turion 64 X2 on IBM's 90 nm Silicon on insulator (SOI) process (cores with the Taylor codename). As of May 2007, they have switched to a 65 nm Silicon-Germanium stressed process[citation needed], which was recently achieved through the combined effort of IBM and AMD, with 40% improvement over comparable 65 nm processes[citation needed]. The earlier 90 nm devices were codenamed Taylor and Trinidad, while the newer 65 nm cores have codename Tyler.

Cores Taylor & Trinidad (90 nm SOI)

* Dual AMD64 core
* L1 cache: 64 + 64 KiB (data + instructions) per core
* L2 cache: 256 KiB (Taylor) or 512 KiB (Trinidad) per core, fullspeed
* Memory controller: dual channel DDR2-667 MHz
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, PowerNow!, NX bit, AMD-V
* Socket S1, HyperTransport (800 MHz, 1600 MT/s, 10.7 GB/s CPU-RAM + 6.4 GB/s CPU-I/O transfer rate)[1]
* Power consumption (TDP): 31, 33, 35 watt max
* First release: May 17, 2006
* Clock rate: 1600, 1800, 2000, 2200 MHz
o 31W TDP:
+ TL-50: 1600 MHz (256 KiB L2-Cache per core)
+ TL-52: 1600 MHz (512 KiB L2-Cache per core)
o 33W TDP:
+ TL-56: 1800 MHz (512 KiB L2-Cache per core)
o 35W TDP:
+ TL-60: 2000 MHz (512 KiB L2-Cache per core)
+ TL-64: 2200 MHz (512 KiB L2-Cache per core)

Tyler (65 nm SOI)

* Dual AMD64 core
* L1 cache: 64 + 64 KiB (data + instructions) per core
* L2 cache: 512 KiB per core, fullspeed
* Memory controller: dual channel DDR2-800 MHz (12.8 GB/s full-duplex CPU/RAM bandwidth)
* 100 MHz granularity (Dynamic P-state Transitions)
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, PowerNow!, NX Bit, AMD-V
* Socket S1, HyperTransport (800 MHz / 1600 MT/s)
* Power consumption (TDP): 31, 35 watt max.
* First release: 2007
* Clock rate: 1700, 1800, 1900, 2000, 2200, 2300, 2400 MHz
o 31W TDP:
+ TK-53 1700 MHz (256 KiB L2-Cache per core) - ※Athlon 64 X2 Dual-Core for Notebooks
+ TK-55 1800 MHz (256 KiB L2-Cache per core) - ※Athlon 64 X2 Dual-Core for Notebooks
+ TL-56 1800 MHz (512 KiB L2-Cache per core)
+ TK-57 1900 MHz (256 KiB L2-Cache per core) - ※Athlon 64 X2 Dual-Core for Notebooks
+ TL-58 1900 MHz (512 KiB L2-Cache per core)
+ TL-60 2000 MHz (512 KiB L2-Cache per core)
o 35W TDP:
+ TL-62 2100 MHz (512 KiB L2-Cache per core)
+ TL-64 2200 MHz (512 KiB L2-Cache per core)
+ TL-66 2300 MHz (512 KiB L2-Cache per core)
+ TL-68 2400 MHz (512 KiB L2-Cache per core)
Turion X2 Lion (65 nm SOI)

* Dual AMD64 core
* L1 cache: 64 + 64 KiB (data + instructions) per core
* L2 cache: 512 KiB per core, fullspeed
* Memory controller: dual channel DDR2-800 MHz
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, PowerNow!, NX bit, AMD-V
* Socket S1, HyperTransport (1800 MHz, 3600 MT/s, 12.8 GB/s CPU-RAM + 14.4 GB/s CPU-I/O transfer rate)
* Power consumption (TDP): 31 watt max
* First release: June 4, 2008
* Clock rate: 2000 MHz
o 31W TDP:
+ RM-70: 2000 MHz

Turion 64

Turion 64 is the brand name AMD applies to its 64-bit low-consumption (mobile) processors codenamed K8L.[1] The Turion 64 and Turion 64 X2 processors compete with Intel's mobile processors, initially the Pentium M and currently the Intel Core and Intel Core 2 processors.

Earlier Turion 64 processors are compatible with AMD's Socket 754. The newer "Richmond" models are designed for AMD's Socket S1. They are equipped with 512 or 1024 KiB of L2 cache, a 64-bit single channel on-die memory controller, and an 800 MHz HyperTransport bus. Battery saving features, like PowerNow!, are central to the marketing and usefulness of these CPUs.

Features
Turion 64 "Lancaster" (90 nm)

All models support:

* MMX
* SSE
* SSE2
* SSE3
* Enhanced 3DNow!
* NX bit
* AMD64 instruction set
* PowerNow!

Turion 64 "Richmond" (90 nm)

The models support the same features available in Lancaster, plus AMD-V.
Model naming methodology
The model naming scheme does not make it obvious how to compare one Turion with another, or even an Athlon 64. The model name is two letters, a dash, and a two digit number (for example, ML-34). The two letters together designate a processor class, while the number represents a performance rating (PR). The first letter is M for single core processors and T for dual core Turion 64 X2 processors. The later in the alphabet that the second letter appears, the more the model has been designed for mobility (frugal power consumption). Take for instance, an MT-30 and an ML-34. Since the T in the MT-30 is later in the alphabet than the L in ML-34, the MT-30 consumes less power than the ML-34. But since 34 is greater than 30, the ML-34 is faster than the MT-30.

Cores
Lancaster (90 nm SOI)

* L1 cache: 64 + 64 KiB (data + instructions)
* L2 cache: 512 or 1024 KiB, fullspeed
* MMX, Enhanced 3DNow!, SSE, SSE2, SSE3, AMD64, PowerNow!, NX Bit
* Socket 754, HyperTransport (800 MHz, HT800)
* VCore: 1.00 V - 1.45 V
* Power consumption (TDP): 25/35 watt max
* First release: March 10, 2005
* Clock rate: 1600, 1800, 2000, 2200, 2400 MHz

o 25W TDP:
+ MT-28: 1600 MHz (512 KiB L2-Cache)
+ MT-30: 1600 MHz (1024 KiB L2-Cache)
+ MT-32: 1800 MHz (512 KiB L2-Cache)
+ MT-34: 1800 MHz (1024 KiB L2-Cache)
+ MT-37: 2000 MHz (1024 KiB L2-Cache)
+ MT-40: 2200 MHz (1024 KiB L2-Cache)
o 35W TDP:
+ ML-28: 1600 MHz (512 KiB L2-Cache)
+ ML-30: 1600 MHz (1024 KiB L2-Cache)
+ ML-32: 1800 MHz (512 KiB L2-Cache)
+ ML-34: 1800 MHz (1024 KiB L2-Cache)
+ ML-37: 2000 MHz (1024 KiB L2-Cache)
+ ML-40: 2200 MHz (1024 KiB L2-Cache)
+ ML-42: 2400 MHz (512 KiB L2-Cache)
+ ML-44: 2400 MHz (1024 KiB L2-Cache)

Richmond (90 nm SOI)

* L1 cache: 64 + 64 KiB (data + instructions)
* L2 cache: 512 KiB, fullspeed
* MMX, Enhanced 3DNow!, SSE, SSE2, SSE3, AMD64, PowerNow!, NX Bit
* Socket S1, HyperTransport (800 MHz, HT800)
* VCore: 1.00 V - 1.45 V
* Power consumption (TDP): 31 watt max
* First release: September 1, 2006
* Clock rate: 2000, 2200 MHz
o 31W TDP:
+ MK-36: 2000 MHz (512 KiB L2-Cache)
+ MK-38: 2200 MHz (512 KiB L2-Cache)

Phenom (processor)

Phenom (pronounced as IPA:/finɒm/, as in the word phenomenon) is the AMD desktop processor line based on the K10 (not "K10h") microarchitecture,[1] or Family 10h Processors, as AMD calls them. Triple-core versions (codenamed Toliman) belong to the Phenom 8000 series and quad cores (codenamed Agena) in the AMD Phenom X4 9000 series. AMD considers the quad core Phenoms to be the first "true" quad core design, as these processors are a monolithic multi-core design (all cores on the same piece of silicon wafer), unlike Intel's Core 2 Quad series which are a multi-chip module (MCM) design. The processors are on the Socket AM2+ platform.[2]

AMD has launched several models of the Phenom processor in 2007/2008

Change of model nomenclatures
The model numbers of the new line of processors were changed from the PR system used in its predecessors, the AMD Athlon 64 processor family. The new model numbering scheme, for later released Athlon X2 processors, is a four digit model number with a different family indicator as the first number.[7] Energy Efficient products end with the letter “e” after the model number (example Phenom 9350e) and some Sempron processors use the LE prefix (example Sempron LE-1200), as follows:

Translation Lookaside Buffer (TLB) Error

Before Phenom's release[citation needed], a flaw was discovered in the translation lookaside buffer that could cause a system lock-up in rare circumstances. Phenom processors up to and including stepping "B2" and "BA" are affected by this bug. BIOS and software workarounds disable the TLB, and typically incur a performance penalty of at least 10%.[9] This penalty was not accounted for in pre-release previews of Phenom, hence the performance of early Phenoms delivered to customers is expected to be less than the preview benchmarks. "B3" stepping Phenom processors were released March 27, 2008 without the TLB bug and with "xx50" model numbers.[10]

An AMD subsidiary has released a patch for the Linux Kernel,[11] which it said has received "minimal functional testing", to overcome this bug by software emulation of accessed- and dirty-bits causing little performance loss.
Phenom II

Phenom II is the updated brand name for AMD's 45nm multicore central processing units. Socket AM2+ version of Phenom II will be released in January 2009 while February will see Socket AM3 versions of the same along with 3 core processors [12].

The Phenom II range of CPUs are the first series of AMD CPUs to eliminate the "cold bug" (A physical phenomenon which causes the processor to cease functioning below a certain temperature. This bug prevents the use of "extreme" cooling methods such as Dry Ice or Liquid Nitrogen). With the elimination of this cold bug, these CPUs are expected to overclock to much higher levels than any other AMD CPU range.[13][14]

In a pre-release demonstration of the Phenom II's overclocking potential, Macci (a record breaking overclocker) used a Phenom II X4 940 and a Gigabyte MA790GP-DS4H with liquid nitrogen cooling to take the processor to a clock speed in excess of 5GHz[15].
Cores

Phenom X4
Agena (65 nm SOI)

* Four AMD K10 cores
* L1 cache: 64 KB + 64 KB[16] (data + instructions) per core
* L2 cache: 512 KB per core, full-speed
* L3 cache: 2 MB shared between all cores
* Memory controller: dual channel DDR2-1066 MHz with unganging option
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, SSE4a, AMD64, Cool'n'Quiet, NX bit, AMD-V
* Socket AM2+, HyperTransport with 1600 to 2000 MHz
* Power consumption (TDP): 65, 95, 125 and 140 Watt
* First release
o November 19, 2007 (B2 Stepping)
o March 27, 2008 (B3 Stepping)
* Clock rate: 1800 to 2600 MHz
* Models: Phenom X4 9100e to 9950

Phenom X3
Toliman (65 nm SOI)

* Three AMD K10 cores
* L1 cache: 64 KB + 64 KB (data + instructions) per core
* L2 cache: 512 KB per core, full-speed
* L3 cache: 2 MB shared between all cores
* Memory controller: dual channel DDR2-1066 MHz with unganging option
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, SSE4a, AMD64, Cool'n'Quiet, NX bit, AMD-V
* Socket AM2+, HyperTransport with 1600 to 1800 MHz
* Power consumption (TDP): 65 and 95 Watt
* First release
o March 27, 2008 (B2 Stepping)
o April 23, 2008 (B3 Stepping)
* Clock rate: 2100 to 2500 MHz
* Models: Phenom X3 8250e to 8850

Future models
Starting in the second half of 2008, AMD is expected to launch a series of 45 nm processors [17]. As of October 2007, only the codenames are known to the public. They are Deneb FX for Phenom FX, Deneb for quad-core processors, Heka for triple-core Phenom processor and Regor for Athlon X2. These processors are expected to be available in late 2008 to early 2009, with the support of DDR3 memory and are expected to have larger shared L3 cache (6 MBytes) as well as the implementation of Socket AM3 for single-processor systems, Socket F+ for dual-processor Quad FX platform[18].

Athlon 64 X2

The Athlon 64 X2 is the first dual-core desktop CPU manufactured by AMD. It is essentially a processor consisting of two Athlon 64 cores joined together on one die with additional control logic. The cores share one dual-channel memory controller, are based on the E-stepping model of Athlon 64 and, depending on the model, have either 512 or 1024 KiB of L2 Cache per core. The Athlon 64 X2 is capable of decoding SSE3 instructions (except those few specific to Intel's architecture), so it can run and benefit from software optimizations that were previously only supported by Intel chips. This enhancement is not unique to the X2, and is also available in the Venice and San Diego single core Athlon 64s.

In June 2007, AMD released low-voltage variants of their low-end 65 nm Athlon 64 X2, named "Athlon X2".[1] The Athlon X2 processors feature reduced TDP of 45 W.[2]

Multithreading

The main benefit of dual-core processors like the X2 is their ability to process more software threads at the same time. The ability of processors to execute multiple threads simultaneously is called thread-level parallelism (TLP). By placing two cores on the same die, the X2 effectively doubles the TLP over a single-core Athlon 64 of the same speed. The need for TLP processing capability is dependent on the situation to a great degree, and certain situations benefit from it far more than others. Certain programs are currently only written with one thread, and are therefore unable to utilize the processing power of the second core.

Programs often written with multiple threads and capable of utilizing dual-cores include many music and video encoding applications, and especially professional rendering programs. High TLP applications currently correspond to server/workstation situations more than the typical desktop. These applications can realize almost twice the performance of a single-core Athlon 64 of the same specifications. Multi-tasking also runs a sizable number of threads; intense multi-tasking scenarios have actually shown improvements of considerably more than two times [2]. This is primarily due to the excessive overhead caused by constantly switching threads, and could potentially be improved by adjustments to operating system scheduling code.

In the consumer segment of the market as well, the X2 improves upon the performance of the original Athlon 64, especially for multi-threaded software applications. The overall increase in performance of the entry level Athlon 64 X2 chip (the Athlon 64 X2 3800+) over the single-core Athlon 64 3800+ chip is almost 10%. The spread between the latter and the Athlon 64 X2 5000+ is almost 40% [3]. One can interpret from these numbers that the majority of applications (at least in the benchmark test) are still largely single thread-dominated, hence the absence of a larger gap between the two 3800+ processors. As software programmers begin to take advantage of multi-core processing, the spread between single- and multi-core processors will increase.

Manufacturing costs

Having two cores, the Athlon 64 X2 has an increased number of transistors. The 1-MiB-L2-cache 90 nm Athlon 64 X2 processor is 219 mm² in size with 243 million transistors [3] whereas its 1-MiB-L2-cache 90 nm Athlon 64 counterpart is 103.1 mm² and has 164 million transistors [4]. The 65 nm Athlon 64 X2 with only 512 KiB L2 per Core reduced this to 118 mm² with 221 million transistors compared to the 65 nm Athlon 64 with 77.2 mm² and 122 million transistors. As a result, a larger area of silicon must be defect free. These size requirements necessitate a more complex fabrication process, which further adds to the production of fewer functional processors per single silicon wafer. This lower yield makes the X2 more expensive to produce than the single-core processor.

In the middle of June 2006 AMD stated that they would no longer make any non-FX Athlon 64 or Athlon 64 X2 models with 1-MiB L2 caches [4]. This led to only a small production number of the Socket-AM2 Athlon 64 X2 with 1 MiB L2 cache per core, known as 4000+, 4400+, 4800+, and 5200+. The Athlon 64 X2 with 512 KiB per core, known as 3800+, 4200+, 4600+, and 5000+, were produced in far greater numbers. The introduction of the F3 stepping then saw several models with 1 MiB L2 cache per core as production refinements resulted in an increased yield.

Manufacturing costs
Change of model nomenclatures

The model numbers of the new line of processors was apparently changed from the PR system used in its predecessors, the Athlon 64 X2. The new model numbering scheme, for later released Athlon X2 processors, is a four digit model number with different family indicator as the first number [5], while Sempron remained using the LE prefix, as follows:
CPU cores

Athlon 64 X2
Manchester (90 nm SOI)

* CPU-Stepping: E4
* L1-Cache: 64 + 64 KiB (Data + Instructions), per core
* L2-Cache: 256, 512 KiB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
* Socket 939, HyperTransport (1000 MHz, HT1000)
* VCore: 1.35 V - 1.4 V
* Power Consumption (TDP): 89 Watt
* First Release: 1 August 2005
* Clockrate: 2000 - 2400 MHz
o 256 KiB L2-Cache:
+ 3600+: 2000 MHz
o 512 KiB L2-Cache:
+ 3800+: 2000 MHz
+ 4200+: 2200 MHz
+ 4600+: 2400 MHz (110 Watt TDP)

Toledo (90 nm SOI)

* CPU-Stepping: E6
* L1-Cache: 64 + 64 KiB (Data + Instructions), per core
* L2-Cache: 512 or 1024 KiB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
* Socket 939, HyperTransport (1000 MHz, HT1000)
* VCore: 1.35 V - 1.4 V
* Power Consumption (TDP):
o 89 Watt: 3800+, 4200+ and 4400+
o 110 Watt: 4400+, 4600+ and 4800+
* First Release: 21 April 2005
* Clockrate:: 2000 - 2400 MHz
o 512 KiB L2-Cache:
+ 3800+: 2000 MHz
+ 4200+: 2200 MHz
+ 4600+: 2400 MHz
o 1024 KiB L2-Cache:
+ 4400+: 2200 MHz
+ 4800+: 2400 MHz

Windsor (90 nm SOI)

* CPU-Stepping: F2, F3
* L1-Cache: 64 + 64 KiB (Data + Instructions), per core
* L2-Cache: 256, 512 or 1024 KiB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket AM2, HyperTransport (1000 MHz, HT1000)
* VCore: 1.25 V - 1.35 V
* Power Consumption (TDP):
o 35 Watt (3800+ EE SFF)
o 65 Watt (3600+ to 5200+ EE)
o 89 Watt (3800+ to 6000+)
o 125 Watt (6000+ to 6400+)
* First Release: May 23, 2006
* Clockrate: 2000 MHz - 3200MHz
o 256 KiB L2-Cache:
+ 3600+: 2000 MHz
o 512 KiB L2-Cache: (often mislabeled as Brisbane core)
+ 3800+: 2000 MHz
+ 4200+: 2200 MHz
+ 4600+: 2400 MHz (F2&F3)
+ 5000+: 2600 MHz (F2&F3)
+ 5400+: 2800 MHz (F3)
o 1024 KiB L2-Cache:
+ 4000+: 2000 MHz
+ 4400+: 2200 MHz
+ 4800+: 2400 MHz
+ 5200+: 2600 MHz (F2&F3)
+ 5600+: 2800 MHz (F3)
+ 6000+: 3000 MHz (F3)
+ 6400+: 3200 MHz (F3)

Brisbane (65 nm SOI)

* CPU-Stepping: G1, G2
* L1-Cache: 64 + 64 KiB (Data + Instructions), per core
* L2-Cache: 512 KiB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket AM2, HyperTransport (1000 MHz, HT1000)
* VCore: 1.25 V - 1.35 V
* Die Size: 126 mm²
* Power Consumption (TDP): 65 Watt
* First Release: Dec 5, 2006
* Clockrate: 1900 MHz - 3100MHz
o 3600+: 1900 MHz
o 4000+: 2100 MHz
o 4200+: 2200 MHz (G1&G2)
o 4400+: 2300 MHz (G1&G2)
o 4600+: 2400 MHz (G2)
o 4800+: 2500 MHz (G1&G2)
o 5000+: 2600 MHz (G1&G2)
o 5200+: 2700 MHz (G1&G2)
o 5400+: 2800 MHz (G2)
o 5600+: 2900 MHz (G2)
o 5800+: 3000 MHz (G2)
o 6000+: 3100 MHz (G2)

Athlon X2

'64' was omitted from the name of the Brisbane 'BE' series; the 64-bit marketing campaign initiated by AMD became insignificant once essentially all consumer CPUs became 64-bit processors.

Brisbane (65 nm SOI)

* CPU-Stepping: G1, G2
* L1-Cache: 64 + 64 KiB (Data + Instructions), per core
* L2-Cache: 512 KiB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket AM2, HyperTransport (1000 MHz, HT1000)
* VCore: 1.15 V - 1.20 V
* Die Size: 118 mm²
* Power Consumption (TDP): 45 Watt
* First Release: June 1, 2007
* Clockrate: 1900 MHz - 2600 MHz
o BE-2300: 1900 MHz
o BE-2350: 2100 MHz (G1&G2)
o BE-2400: 2300 MHz (G2)
o 4050e: 2100 MHz (G2)
o 4450e: 2300 MHz (G2)
o 4850e: 2500 MHz (G2)
o 5050e: 2600 MHz (G2)

Athlon 64

The Athlon 64 is an eighth-generation, AMD64-architecture microprocessor produced by AMD, released on September 23, 2003.[1] It is the third processor to bear the name Athlon, and the immediate successor to the Athlon XP.[2] The second processor (after the Opteron) to implement AMD64 architecture and the first 64-bit processor targeted at the average consumer,[3] it was AMD's primary consumer microprocessor, and competes primarily with Intel's Pentium 4, especially the "Prescott" and "Cedar Mill" core revisions. It is AMD's first K8, eighth-generation processor core for desktop and mobile computers.[4] Despite being natively 64-bit, the AMD64 architecture is backward-compatible with 32-bit x86 instructions.[5] Athlon 64s have been produced for Socket 754, Socket 939, Socket 940, and Socket AM2.

History
The Athlon 64 was originally codenamed ClawHammer by AMD,[3] and was referred to as such internally and in press releases. The first Athlon 64 FX was based on the first Opteron core, SledgeHammer. Both cores, produced on a 130 nanometer process, were first introduced on September 23, 2003. The models first available were the FX-51, fitting Socket 940, and the 3200+, fitting Socket 754.[6] Like the Opteron, on which it was based, the Athlon FX-51 required buffered RAM, increasing the final cost of an upgrade.[7] The week of the Athlon 64's launch, Intel released the Pentium 4 Extreme Edition, a CPU designed to compete with the Athlon 64 FX.[8] The Extreme Edition was widely considered a marketing ploy to draw publicity away from AMD, and was quickly nicknamed among some circles the "Emergency Edition".[9] Despite a very strong demand for the chip, AMD was plagued by early manufacturing difficulties that made it difficult to deliver Athlon 64s in quantity. In the early months of the Athlon 64 lifespan, AMD could only produce one hundred thousand chips per month.[10] However, it was very competitive in terms of performance to the Pentium 4, with magazine PC World calling it the "fastest yet".[11] "Newcastle" was released soon after ClawHammer, with half the Level 2 cache.[12]

On June 1, 2004, AMD released new versions of both the ClawHammer and Newcastle core revisions for the newly-introduced Socket 939, an altered Socket 940 without the need for buffered memory.[13] Socket 939 offered two main improvements over Socket 754: the memory controller was altered with dual-channel architecture,[14] doubling peak memory bandwidth, and the HyperTransport bus was increased in speed from 800 MHz to 1000 MHz.[15] Socket 939 also was introduced in the FX series in the form of the FX-55.[16] At the same time, AMD also began to ship the "Winchester" core, based on a 90 nanometer process.

Core revisions "Venice" and "San Diego" succeeded all previous revisions on April 15, 2005. Venice, the lower-end part, was produced for both Sockets 754 and 939, and included 512 KB of L2 cache.[17] San Diego, the higher-end chip, was produced only for Socket 939 and doubled Venice's L2 cache to one MB.[18] Both were produced on the 90 nm fabrication process.[19] Both also included support for the SSE3 instruction set,[20] a new feature that had been included in the rival Pentium 4 since the release of the Prescott core in February 2004.[21] In addition, AMD overhauled the memory controller for this revision, resulting in performance improvements as well as support for newer DDR RAM.[22]

Dual-core Athlon 64
On April 21, 2005, less than a week after the release of Venice and San Diego, AMD announced its next addition to the Athlon 64 line, the Athlon 64 X2.[23] Released on May 31, 2005,[24] it also initially had two different core revisions available to the public, Manchester and Toledo, the only appreciable difference between them being the amount of L2 cache.[25] Both were released only for Socket 939.[26] A response to Intel's dual core Pentium D, the Athlon 64 X2 was received very well by reviewers and the general public, with a general consensus emerging that AMD's implementation of multi-core was superior to that of the Pentium D.[27][28] Some felt initially that the X2 would cause market confusion with regard to price points since the new processor was targeted at the same "enthusiast," US$350 and above market[29] already occupied by AMD's existing socket 939 Athlon 64s.[30] AMD's official breakdown of the chips placed the Athlon X2 aimed at a segment they called the "prosumer", along with digital media fans.[24] The Athlon 64 was targeted at the mainstream consumer, and the Athlon FX at gamers. The Sempron budget processor was targeted at value-conscious consumers.[31]
DDR2

The Athlon 64 had been maligned by some critics for some time because of its lack of support for DDR2 SDRAM, an emerging technology that had been adopted much earlier by Intel.[32] AMD's official position was that the CAS latency on DDR2 had not progressed to a point where it would be advantageous for the consumer to adopt it.[33] AMD finally remedied this gap with the "Orleans" core revision, the first Athlon 64 to fit Socket AM2, released on May 23, 2006.[34] "Windsor", an Athlon 64 X2 revision for Socket AM2, was released concurrently. Both Orleans and Windsor have either 512KB or 1MB of L2 cache per core.[35] The Athlon 64 FX-62 was also released concurrently on the Socket AM2 platform.[36] Socket AM2 also consumes less power than previous platforms, and supports AMD-V.[37]

The memory controller used in all DDR2 SDRAM capable processors (Socket AM2), has extended column address range of 11 columns instead of conventional 10 columns, and the support of 16 KB page size, with at most 2048 individual entries supported. An OCZ unbuffered DDR2 kit, optimized for 64-bit operating systems, was released to exploit the functionality provided by the memory controller in socket AM2 processors, allowing the memory controller to stay longer on the same page, thus benefitting graphics intensive applications.[38]

Features
There are four variants: Athlon 64, Athlon 64 FX, Mobile Athlon 64 (later renamed "Turion 64") and the dual-core Athlon 64 X2.[39] Common among the Athlon 64 line are a variety of instruction sets including MMX, 3DNow!, SSE, SSE2, and SSE3.[40] All Athlon 64s also support the NX bit, a security feature named "Enhanced Virus Protection" by AMD.[41] And as implementations of the AMD64 architecture, all Athlon 64 variants are able to run 16 bit, 32 bit x86, and AMD64 code, through two different modes the processor can run in: "Legacy mode" and "long mode". Legacy mode runs 16-bit and 32-bit programs natively, and long mode runs 64-bit programs natively, but also allows for 32-bit programs running inside a 64-bit operating system.[42] All Athlon 64 processors feature 128 Kilobytes of level 1 cache, and at least 512 KB of level 2 cache.[40]

The Athlon 64 features an on-die memory controller,[5] a feature not previously seen on x86 CPUs. Not only does this mean the controller runs at the same clock rate as the CPU itself, it also means the electrical signals have a shorter physical distance to travel compared to the old northbridge interfaces.[43] The result is a significant reduction in latency (response time) for access requests to main memory.[44] The lower latency is often cited as one of the advantages of the Athlon 64's architecture over those of its competitors.[45]

Translation Lookaside Buffers (TLBs) have also been enlarged (40 4k/2M/4M entries in L1 cache, 512 4k entries),[46] with reduced latencies and improved branch prediction, with four times the number of bimodal counters in the global history counter.[42] This and other architectural enhancements, especially as regards SSE implementation, improve instruction per cycle (IPC) performance over the previous Athlon XP generation.[42] To make this easier for consumers to understand, AMD has chosen to market the Athlon 64 using a PR (Performance Rating) system, where the numbers roughly map to Pentium 4 performance equivalents, rather than actual clock speed.[47]

Athlon 64 also features CPU speed throttling technology branded Cool'n'Quiet, a feature similar to Intel's SpeedStep that can throttle the processor's clock speed back to facilitate lower power consumption and heat production.[48] When the user is running undemanding applications and the load on the processor is light, the processor's clock speed and voltage are reduced. This in turn reduces its peak power consumption (max TDP set at 89 W by AMD) to as low as 32 W (stepping C0, clock speed reduced to 800 MHz) or 22W (stepping CG, clock speed reduced to 1 GHz). The Athlon 64 also has an Integrated Heat Spreader (IHS) which prevents the CPU die from accidentally being damaged when mounting and unmounting cooling solutions. With prior AMD CPUs a CPU shim could be used by people worried about damaging the die.

The No Execute bit (NX bit) supported by Windows Vista, Windows XP Service Pack 2,[49] Windows XP Professional x64 Edition, Windows Server 2003 x64 Edition, and Linux 2.6.8 and higher is also included, for improved protection from malicious buffer overflow security threats. Hardware-set permission levels make it much more difficult for malicious code to take control of the system. It is intended to make 64-bit computing a more secure environment.

The Athlon 64 CPUs have been produced with 130 nm and 90 nm SOI process technologies.[50] All of the latest chips (Winchester, Venice and San Diego models) are on 90 nm. The Venice and San Diego models also incorporate dual stress liner technology[51] (an amalgam of strained silicon and 'squeezed silicon', the latter of which is not actually a technology) co-developed with IBM.[52]

As the memory controller is integrated onto the CPU die, there is no FSB for the system memory to base its speed upon.[53] Instead, system memory speed is obtained by using the following formula (using the ceiling function):[54]

In simpler terms, the memory is always running at a set fraction of the CPU speed, with the divisor being a whole number. A 'FSB' figure is still used to determine the CPU speed, but the RAM speed is no longer directly related to this 'FSB' figure (known otherwise as the LDT).

To summarize, the Athlon 64 architecture features two buses from the CPU. One is the HT bus to the northbridge connecting the CPU to the chipset and device attachment bus (PCIe, AGP, PCI) and the other is the memory bus which connects the on-board memory controller to the bank of either DDR or DDR2 DRAM.

Athlon 64 FX
The Athlon 64 FX is positioned as a hardware enthusiast product, marketed by AMD especially toward gamers.[55] Unlike the standard Athlon 64, all of the Athlon 64 FX processors have their multipliers completely unlocked.[56] The FX line is now dual-core, starting with the FX-60.[57] The FX always has the highest clock speed of all Athlons at its release.[58] From FX-70 onwards, the line of processors will also support dual-processor setup with NUMA, named AMD Quad FX platform.

Athlon 64 X2
The Athlon 64 X2 is the first dual-core desktop CPU manufactured by AMD. In 2007, AMD released two final Athlon 64 X2 versions: the AMD Athlon 64 X2 6400+ and 5000+ Black Editions. Both processors feature an unlocked multiplier, which allows for a large range of overclocked settings. The 6400+ is based on a 90nm Windsor core (3.2GHz, 2x1MB L2, 125W TDP) while the 5000+ is based on a 65nm Brisbane core (2.6GHz, 2x512KB L2, 65W TDP). These Black Edition processors are available at retail, but AMD does not include heatsinks in the retail package.

Turion 64 (formerly Mobile Athlon 64)

Previously introduced as "Mobile Athlon 64", Turion 64 is now the brand name AMD applies to its 64-bit low-consumption (mobile) processors codenamed K8L.[59] The Turion 64 and Turion 64 X2 processors compete with Intel's mobile processors, initially the Pentium M and currently the Intel Core and Intel Core 2 processors.

Earlier Turion 64 processors are compatible with AMD's Socket 754. The newer "Richmond" models are designed for AMD's Socket S1. They are equipped with 512 or 1024 KB of L2 cache, a 64-bit single channel on-die memory controller, and an 800 MHz HyperTransport bus. Battery saving features, like PowerNow!, are central to the marketing and usefulness of these CPUs.

Model naming methodology

The model naming scheme does not make it obvious how to compare one Turion with another, or even an Athlon 64. The model name is two letters, a dash, and a two digit number (for example, ML-34). The two letters together designate a processor class, while the number represents a PR rating. The first letter is M for single core processors and T for dual core Turion 64 X2 processors. The later in the alphabet that the second letter appears, the more the model has been designed for mobility (frugal power consumption). Take for instance, an MT-30 and an ML-34. Since the T in the MT-30 is later in the alphabet than the L in ML-34, the MT-30 consumes less power than the ML-34. But since 34 is greater than 30, the ML-34 is faster than the MT-30.

Sockets

* Socket 754: The Athlon 64 value/budget line, 64-bit memory interface (Single-Channel)
* Socket 939: Athlon 64 performance line, Athlon 64 X2s, and newer Athlon 64 FXs, Opteron, 128-bit memory interface (Dual-channel)
* Socket 940: Opteron and old Athlon 64 FX, 128-bit memory interface - requires registered DDR memory
* Socket AM2: Athlon 64/Athlon 64 FX/Athlon 64 X2/Sempron, 940 Pins (Not compatible with Socket 940); the first AMD socket to use DDR2 SDRAM.
* Socket F: Opteron, 1207 Pins
* Socket F (1207 FX): Athlon 64 FX on AMD Quad FX platform, also compatible for dual-processor Opteron 2200 series[60]

At the introduction of Athlon 64 in September 2003, only Socket 754 and Socket 940 (Opteron) were ready and available. The onboard memory controller was not capable of running unbuffered (non-registered) memory in dual-channel mode at the time of release; as a stopgap measure, they introduced the Athlon 64 on Socket 754, and brought out a non-multiprocessor version of the Opteron called the Athlon 64 FX, as a multiplier unlocked enthusiast part for Socket 940, comparable to Intel's Pentium 4 Extreme Edition for the high end market.

In June 2004, AMD released Socket 939 as the mainstream Athlon 64 with dual-channel memory interface, leaving Socket 940 solely for the server market (Opterons), and relegating Socket 754 as a value/budget line, for Semprons and slower versions of the Athlon 64. Eventually Socket 754 replaced Socket A for Semprons.

In May 2006, AMD released Socket AM2, which provided support for the DDR2 memory interface. Also, this marked the release of AMD-V.

In August 2006, AMD released Socket F for Opteron server CPU which uses the LGA chip form factor.

In November 2006, AMD released a specialized version of Socket F, called 1207 FX, for dual-socket, dual-core Athlon FX processors on the Quad FX platform. While Socket F Opterons already allowed for four processor cores, Quad FX allowed unbuffered RAM and expanded CPU/chipset configuration in the BIOS. Consequentially, Socket F and F 1207 FX are incompatible and require different processors, chipsets, and motherboards.
Athlon 64 FX models

Sledgehammer (130 nm SOI)

* CPU-Stepping: C0, CG
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 1024 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, AMD64
* Socket 940, 800 MHz HyperTransport (HT800)
* Registered DDR-SDRAM required
* VCore: 1.50/1.55 V
* Power Consumption (TDP): 89 Watt max
* First Release: September 23, 2003
* Clockrate: 2200 MHz (FX-51, C0), 2400 MHz (FX-53, C0 and CG)

Clawhammer (130 nm SOI)

* CPU-Stepping: CG
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 1024 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, AMD64
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.50 V
* Power Consumption (TDP): 89 Watt (FX-55:104 Watt)
* First Release: June 1, 2004
* Clockrate: 2400 MHz (FX-53), 2600 MHz (FX-55)

San Diego (90 nm SOI)

* CPU-Stepping: E4, E6
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 1024 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.35 V or 1.40 V
* Power Consumption (TDP): 104 Watt max
* First Release: April 15, 2005
* Clockrate: 2600 MHz (FX-55), 2800 MHz (FX-57)

Toledo (90 nm SOI)

Dual-core CPU

* CPU-Stepping: E6
* L1-Cache: 64 + 64 KB (Data + Instructions), per core
* L2-Cache: 1024 KB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.30 V - 1.35 V
* Power Consumption (TDP): 110 Watt max
* First Release: January 10, 2006
* Clockrate: 2600 MHz (FX-60)

Windsor (90 nm SOI)

Dual-core CPU

* CPU-Stepping: F2, F3
* L1-Cache: 64 + 64 KB (Data + Instructions), per core
* L2-Cache: 512 - 1024 KB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket AM2, 1000 MHz HyperTransport (HT1000)
* VCore: 1.30 V - 1.40 V
* Power Consumption (TDP): 125 Watt max
* First Release: May 23, 2006
* Clockrate: 2000 - 3200 MHz (6400+)

Windsor (90 nm SOI) - Quad FX platform

Main article: AMD Quad FX platform

Dual-core, dual CPUs (four cores total)

* CPU-Stepping: F3
* L1-Cache: 64 + 64 KB (Data + Instructions), per core
* L2-Cache: 1024 KB fullspeed, per core
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket F (1207 FX), 2000 MHz HyperTransport (HT2000)
* VCore: 1.35 V - 1.40 V
* Power Consumption (TDP): 125 Watt max per CPU
* First Release: November 30, 2006
* Clockrate: 2600 MHz (FX-70), 2800 MHz (FX-72), 3000 MHz (FX-74)

Athlon 64 models
Clawhammer (130 nm SOI)

* CPU-Stepping: C0, CG
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 1024 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, AMD64, Cool'n'Quiet, NX Bit (only CG)
* Socket 754, 800 MHz HyperTransport (HT800)
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.50 V
* Power Consumption (TDP): 89 Watt max
* First Release: September 23, 2003
* Clockrate: 2000–2600 MHz

Newcastle (130 nm SOI)

Also possible: ClawHammer-512 (Clawhammer with partially disabled L2-Cache)

* CPU-Stepping: CG
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 512 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, AMD64, Cool'n'Quiet, NX Bit
* Socket 754, 800 MHz HyperTransport (HT800)
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.50 V
* Power Consumption (TDP): 89 Watt max
* First Release: 2004
* Clockrate: 1800–2400 MHz

Winchester (90 nm SOI)

* CPU-Stepping: D0
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 512 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, AMD64, Cool'n'Quiet, NX Bit
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.40 V
* Power Consumption (TDP): 67 Watt max
* First Release: 2004
* Clockrate: 1800–2200 MHz

Venice (90 nm SOI)

* CPU-Stepping: E3, E6
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 512 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
* Socket 754, 800 MHz HyperTransport (HT800)
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.35 V or 1.40 V
* Power Consumption (TDP): 89 Watt max
* First Release: April 4, 2005
* Clockrate: 1800–2400 MHz

San Diego (90 nm SOI)

* CPU-Stepping: E4, E6
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 1024 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit
* Socket 939, 1000 MHz HyperTransport (HT1000)
* VCore: 1.35 V or 1.40 V
* Power Consumption (TDP): 89 Watt max
* First Release: April 15, 2005
* Clockrate: 2200–2600 MHz

Orleans (90 nm SOI)

* CPU-Stepping: F2, F3
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 512 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket AM2, 1000 MHz HyperTransport (HT1000)
* VCore: 1.25 V or 1.40 V
* Power Consumption (TDP): 62 Watt max
* First Release: May 23, 2006
* Clockrate: 1800–2600 MHz

Lima (65 nm SOI)

* CPU-Stepping: G1
* L1-Cache: 64 + 64 KB (Data + Instructions)
* L2-Cache: 512 KB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX Bit, AMD-V
* Socket AM2, 1000 MHz HyperTransport (HT1000)
* VCore: 1.25/1.35/1.40V
* Power Consumption (TDP): 45 Watt max
* First Release: February 20, 2007
* Clockrate: 2000–2800 MHz

Successors

The Athlon 64 line is expected to continue to evolve. In particular, new models scheduled to be launched starting from the third quarter of 2007 are to be based on the "K10" microarchitecture. The initial offerings are expected to be based on the Agena (quad-core, 2 MB L3 cache), and Kuma (dual-core, 2 MB L3 cache) cores. These processors will be packaged in Socket AM2+ form factors, but are expected to function in Socket AM2 motherboards as well, with the loss of HyperTransport 3.0 enhancements, which will only be available with Socket AM2+ motherboards.

Sempron

Sempron has been the marketing name used by AMD for several different entry level desktop CPUs, using several different technologies and CPU socket formats.

The Sempron replaced the AMD Duron processor and competes against Intel's Celeron D processor.

AMD coined the name from the Latin semper, which means "always", to suggest the Sempron is suitable for "daily use, practical, and part of everyday life"[1].

History and features

The first Sempron CPUs were based on the Athlon XP architecture using the Thoroughbred or Thorton core. These models were equipped with the Socket A interface, 256 KiB L2 cache and 166 MHz Front side bus (FSB 333). Thoroughbred cores natively had 256 KiB L2 cache, but Thortons had 512 KiB L2 cache, half of which was disabled and could sometimes be reactivated by bridge modification. Later, AMD introduced the Sempron 3000+ CPU, based on the Barton core with 512 KiB L2 cache. From a hardware and user standpoint, the Socket A Sempron CPUs were essentially identical to Athlon XP desktop CPUs with a new brand name. AMD has ceased production of all Socket A Sempron CPUs.

The second generation (Paris/Palermo core) was based on the architecture of the Socket 754 Athlon 64. Some differences from Athlon 64 processors include a reduced cache size (either 128 or 256 KiB L2), and the absence of AMD64 support in earlier models. Apart from these differences, the Socket 754 Sempron CPUs share most features with the more powerful Athlon 64, including an integrated (on-die) memory controller, the HyperTransport link, and AMD's "NX bit" feature.

In the second half of 2005, AMD added 64-bit support (AMD64) to the Sempron line. Some journalists (but not AMD) often refer to this revision of chips as "Sempron 64" to distinguish it from the previous revision. AMD's intent in releasing 64-bit entry-level processors was to extend the market for 64-bit processors, which at the time of Sempron 64's first release, was a niche market.

In 2006, AMD announced the Socket AM2 and Socket S1 line of Sempron processors. These are functionally equivalent to the previous generation, except they have a dual-channel DDR2 SDRAM memory controller which replaces the single-channel DDR SDRAM version. The TDP of the standard version remains at 62 W (watts), while the new "Energy Efficient Small Form Factor" version has a reduced 35 W TDP. The Socket AM2 version also does not require a minimum voltage of 1.1 volts to operate, whereas all socket 754 Semprons with Cool'n'Quiet did. In 2006, AMD was selling both Socket 754 and Socket AM2 Sempron CPUs concurrently. In the middle of 2007 AMD appears to have dropped the 754 line and is shipping AM2 and S1 Semprons.

Models for Socket A Thoroughbred B/Thorton (130 nm)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 256 KiB, fullspeed
* MMX, 3DNow!, SSE
* Socket A (EV6)
* Front side bus: 166 MHz (FSB 333)
* VCore: 1.6 V
* First release: July 28, 2004
* Clockrate: 1500 MHz - 2000 MHz (2200+ to 2800+)

Barton (130 nm)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 512 KiB, fullspeed
* MMX, 3DNow!, SSE
* Socket A (EV6)
* Front side bus: 166 MHz - 200 MHz (FSB 333 - 400)
* VCore: 1.6 - 1.65 V
* First release: September 17, 2004
* Clockrate: 2000–2200 MHz (Sempron 3000+, Sempron 3300+)

Models for Socket 754 Paris (130 nm SOI)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 256 KiB, fullspeed
* MMX, 3DNow!, SSE, SSE2
* Enhanced Virus Protection (NX bit)
* Integrated 72-bit(Single channel, ECC capable) DDR memory controller
* Socket 754, 800 MHz HyperTransport
* VCore: 1.4 V
* First release: July 28, 2004
* Clockrate: 1800 MHz (3100+)
* Stepping: CG (Part No.: *AX)

Palermo (90 nm SOI)

* Early models (stepping D0) are downlabeled "Oakville" mobile Athlon64
* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 128/256 KiB, fullspeed
* MMX, 3DNow!, SSE, SSE2
* SSE3 support on E3 and E6 steppings
* AMD64 on E6 stepping

#
# Cool'n'Quiet (Sempron 3000+ and higher)
# Enhanced Virus Protection (NX bit)
# Integrated 72-bit(Single channel, ECC capable) DDR memory controller
# Socket 754, 800 MHz HyperTransport
# VCore: 1.4 V
# First release: February 2005
# Clockrate: 1400–2000 MHz

* 128 KiB L2-Cache (Sempron 2600+, 3000+, 3300+)
* 256 KiB L2-Cache (Sempron 2500+, 2800+, 3100+, 3400+)

# Steppings: D0 (Part No.: *BA), E3 (Part No.: *BO), E6 (Part No.: *BX)

Models for Socket 939
Palermo (90 nm SOI)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 128/256 KiB, fullspeed
* MMX, 3DNow!, SSE, SSE2, SSE3, AMD64 (E6 Steppings Only), Cool'n'Quiet, NX bit
* Integrated 144-bit(Dual channel, ECC capable) DDR memory controller
* Socket 939, 800 MHz HyperTransport
* VCore: 1.35/1.4 V
* First release: October 2005
* Clockrate: 1800–2000 MHz
o 128 KiB L2-Cache (Sempron 3000+, 3400+)
o 256 KiB L2-Cache (Sempron 3200+, 3500+)
* Steppings: E3 (Part No.: *BP), E6 (Part No.: *BW)
Models for Socket AM2 Manila (90 nm SOI)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 128/256 KiB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX bit
* Integrated 128-bit(Dual channel) DDR2 memory controller
* Socket AM2, 800 MHz HyperTransport
* VCore: 1.25/1.35/1.40 V (1.20/1.25 V for Energy Efficient SFF version)
* First release: May 23, 2006
* Clockrate: 1600–2000 MHz
o 128 KiB L2-Cache (Sempron 2800+, 3200+, 3500+)
o 256 KiB L2-Cache (Sempron 3000+, 3400+, 3600+, 3800+)
* Stepping: F2 (Part No.: *CN, *CW

Sparta (65 nm SOI)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 256/512 KiB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX bit
* Integrated 128-bit(Dual channel) DDR2 memory controller
* Socket AM2, 800 MHz HyperTransport
* VCore: 1.20/1.40 V
* First release: August 20, 2007
* Clockrate: 1900–2300 MHz
o 256 KiB L2-Cache (Sempron LE-1100, LE-1150)
o 512 KiB L2-Cache (Sempron LE-1200, LE-1250, LE-1300)
* Stepping: G1 (Part No.: *DE), G2 (Part No.: *DP)

Models for Socket S1 (638) Keene (90 nm SOI)

* L1-Cache: 64 + 64 KiB (Data + Instructions)
* L2-Cache: 256 or 512 KiB, fullspeed
* MMX, Extended 3DNow!, SSE, SSE2, SSE3, AMD64, Cool'n'Quiet, NX bit
* Integrated 128-bit(Dual channel) DDR2 memory controller
* Socket S1, 800 MHz HyperTransport
* VCore: 0.950-1.25 V
* First release: May 17, 2006
* Clockrate: 1000–2000 MHz
o 256 KiB L2-Cache (Sempron 2100+, 3400+)
o 512 KiB L2-Cache (Sempron 3200+, 3500+, 3600+)
* Stepping: F2 (Part No.: *CM)
Future plans
In 2008, Sempron-branded implementations of the Stars microarchitecture are expected to become available, based on the Rana core. These are expected to be dual-core processors without L3 cache. Initial clock rates will be between 2.1 GHz and 2.3 GHz. The Rana Semprons will feature HyperTransport 3.0 support and will be packaged for the Socket AM2+ form factor, although they are expected to function in Socket AM2 motherboards, albeit without support for HyperTransport 3.0 enhancements.[2]

Geode (processor)

Geode is a series of x86-compatible System-on-a-chip microprocessors and I/O companions produced by AMD targeted at the embedded computing market.

The series was originally launched by National Semiconductor as the Geode family in 1999. The original Geode processor core itself is derived from the Cyrix MediaGX platform, which was acquired in National's merger with Cyrix in 1997. AMD bought the Geode business from National in August 2003 to augment its existing line of embedded x86 processor products. AMD expanded the Geode series to two classes of processor: the MediaGX-derived Geode GX and LX, and the modern Athlon-derived Geode NX.

Geode processors are optimized for low power consumption and low cost while still remaining compatible with software written for the x86 platform. The MediaGX-derived processors lack modern features such as SSE and a large on-die L1 cache but these are offered on the more recent Athlon-derived Geode NX. Geode processors tightly integrate some of the functions normally provided by a separate chipset, such as the northbridge. Whilst the processor family is best suited for thin client, set top box and embedded computing applications it can be found in unusual applications such as the Nao robot

The One Laptop per Child project originally used the GX series Geode processor in the OLPC XO; but has since moved to the Geode LX. The Linutop is also based on the Geode LX. 3Com Audrey was powered by a 200 MHz Geode GX1.

The SCxxxx range of Geode devices are a single-chip version, comparable to the SiS 552 or VIA CoreFusion, which integrate the CPU, memory controller, graphics and I/O devices into one package. Single processor boards based on these processors are manufactured by Artec Group, PC Engines (WRAP) and Soekris.

These processors are named after geodes.

National Semiconductor Geode
Geode GXm

Cyrix MediaGXm clone. Returns "CyrixInstead" on CPUID.

* MediaGX-derived core
* 0.35 µm four layer metal CMOS
* MMX instructions
* 3.3 V I/O, 2.9 V core
* 16 Kb write-back unified L1 cache
* PCI controller
* 64-bit SDRAM memory
* CS5530 companion chip (implements sound and video functions)
* VSA architecture
* 1280x1024x8 or 1024x768x16 display

Geode GXLV

* MediaGX-derived core
* 0.25 µm four layer metal CMOS
* 3.3 V I/O
* 2.2 V, 2.5 V, 2.9 V core
* 16 kb write-back unified L1 cache
* Fully static design
* 1.0 W @2.2 V/166 MHz, 2.5 W @2.9 V/266 MHz

Geode GX1

* MediaGX-derived core
* 0.18 µm CMOS
* 200 - 333 MHz
* 1.6 - 2.2 V core
* 16 kB (16 KiB) L1 cache
* 0.8 W - 1.2 W typical
* SDRAM memory 111 MHz
* CS5530A companion chip
* 85 Hz VGA refresh rate

National Semiconductor/AMD SC1100 is based on the Cyrix GX1 core and the CS5530 support chip.
Geode GX2

Announced by National Semiconductor Corporation October, 2001 at Microprocessor Forum. First demonstration at COMPUTEX Taiwan, June, 2002.

* 0.15 µm process technology
* MMX and 3DNow! instructions
* 16 kB Instruction and 16 kB Data caches
* GeodeLink architecture, 6 GB/s on-chip bandwidth, up to 2 GB/s memory bandwidth
* Integrated 64-bit PC133 SDRAM and DDR266 controller
* Clockrate: 266, 333 and 400 MHz
* 3 PCI masters supported
* 1600x1200x24 bit display with video scaling
* CRT DACs and an UMA DSTN/TFT controller.
* Geode CS5535 companion chip

AMD Geode

In 2002, AMD introduced the Geode GX series, which was a re-branding of the National Semiconductor GX2. This was quickly followed by the Geode LX, running up to 667 MHz. LX brought many improvements, such as higher speed DDR, a re-designed instruction pipe, and a more powerful display controller. The upgrade from the CS5535 I/O Companion to the CS5536 brought higher speed USB.

Geode GX and LX processors are typically found in devices such as thin clients and industrial control systems. However they have come under competitive pressure from VIA on the x86 side, and ARM and XScale taking much of the low-end business.

Because of the relatively poor performance of the GX and LX core design, AMD introduced the Geode NX, which is an embedded version of the highly-successful Athlon processor, K7. Geode NX uses the Thoroughbred core and is quite similar to the Athlon XP-M that use this core. The Geode NX includes 256KB of Level 2 cache, and runs fanless at up to 1 GHz in the NX1500@6W version. The NX2001 part runs at 1.8 GHz, the NX1750 part runs at 1.4 GHz, and the NX1250 runs at 667 MHz.

The Geode NX, with its strong FPU, is particularly suited for embedded devices with graphical performance requirements, such as information kiosks and casino gaming machines, such as video slots.

However, it was reported that the specific design team for Geode processors in Longmont, Colorado, has been closed, and 75 employees are being relocated to the new development facility in Fort Collins, Colorado. It is expected that the Geode line of processors will be updated less frequently due to the closure of the Geode design center [1].
Geode GX

1. Geode GX 466@0.9 W: clock speed: 333 MHz
2. Geode GX 500@1.0 W: clock speed: 366 MHz
3. Geode GX 533@1.1 W: clock speed: 400 MHz

Geode LX
AMD Geode LX 800 (500MHz) CPU.

1. LX 700@0.8 W: clock speed: 433 MHz, with power consumption: 1.3 watts. (TDP 3.1 W)
2. LX 800@0.9 W: clock speed: 500 MHz, with power consumption: 1.8 watts. (TDP 3.6 W)
3. LX 900@1.5 W: clock speed: 600 MHz, with power consumption: 2.6 watts. (TDP 5.1 W)

Features:

* Low power.
* Full x86 compatibility.
* Processor functional blocks:
o CPU Core
o GeodeLink Control Processor
o GeodeLink Interface Units
o GeodeLink Memory Controller
o Graphics Processor
o Display Controller
o Video Processor
o Video Input Port
o GeodeLink PCI Bridge
o Security Block
+ 128-Bit Advanced Encryption Standard (AES) - (CBC/ECB)
+ True Random Number Generator

Specification:

* Processor frequency up to 600 MHz (LX900), 500 MHz (LX800) and 433 MHz (LX700).
* Power management: ACPI, lower power, wakeup on SMI/INTR.
* 64K Instruction / 64K Data L1 cache and 128K L2 cache
* Split Instruction/Data cache/TLB.
* DDR Memory 400 MHz (LX 800), 333 MHz (LX 700)
* Integrated FPU with MMX and 3DNow!
* 9 GB/s internal GeodeLink Interface Unit (GLIU)
* Simultaneous, high-res CRT and TFT (High and standard definition). VESA 1.1 and 2.0 VIP/VDA support
* Manufactured at a 0.13 micrometre process
* 481-terminal PBGA (Plastic Ball grid array)
* GeodeLink active hardware power management

Geode NX

1. NX 1250@6W: Clock speed: 667 MHz, power consumption: 6 watts (1.0 volts core operating voltage).
2. NX 1500@6W: Clock speed: 1 GHz, power consumption: 9 watts (1.1 volts core operating voltage).
3. NX 1750@14W: Clock speed: 1.4 GHz, power consumption: 25 watts (1.25 volts core operating voltage).

Features:

* 7th generation core (based on Mobile Athlon XP-M).
* Power management: AMD PowerNow!, ACPI 1.0b and ACPI 2.0.
* 128 KB L1 cache.
* 256 KB L2 cache with hardware data prefetch
* 133 MHz Front Side Bus (FSB)
* 3DNow!, MMX and SSE instruction sets
* 0.13 µm (130 nm) fabrication process
* Pin compatibility between all NX family processors.
* OS support: Linux, Windows CE, MS Windows XP.
* Compatible with Socket A motherboards

Geode NX 2001

In 2007, there was a Geode NX 2001 model on sale, which in fact was a relabelled Athlon XP 2200+ Thoroughbred. The processors, with part numbers AANXA2001FKC3G or ANXA2001FKC3D, their specifications are 1.8 GHz clock speed, and 1.65 volt core operating voltage, the power consumption is not specified. There are no official references to this processor except officials explaining that the batch of CPUs were "being shipped to specific customers", though it is clear it has no relation with the other Geode NX CPUs other than sharing the same CPU socket (Socket A). [2]
Chipsets for Geode

1. AMD Geode CS5530A Southbridge for Geode GX1.
2. AMD Geode CS5535 Southbridge for Geode GX and Geode LX (USB 1.1). Integrates four USB ports, one ATA-66 UDMA controller, one Infrared communication port, one AC97 controller, one SMBUS controller, one LPC port, as well as GPIO, Power Management, and legacy functional blocks.
3. AMD Geode CS5536 Southbridge for Geode GX and Geode LX (USB 2.0). Power consumption: 1.9 W (433 MHz) and 2.4 W (500 MHz). This chipset is also used on PowerPC board (Amy'05).
4. Geode NX processors are "100 percent socket and chipset compatible" with AMD's Socket A Athlon XP processors: SIS741CX Northbridge and SIS 964 Southbridge, VIA KM400 Northbridge and VIA VT8235 Southbridge, VIA KN400A Northbridge and VIA VT8237R Southbridge and other Socket A chipsets.

AMD (Advanced Micro Devices)

Advanced Micro Devices, Inc. (AMD) (NYSE: AMD) is an American multinational semiconductor company based in Sunnyvale, California, that develops computer processors and related technologies for commercial and consumer markets. Its main products include microprocessors, motherboard chipsets, embedded processors and graphics processors for servers, workstations and personal computers, and processor technologies for handheld devices, digital television, and game consoles.

AMD is the second-largest global supplier of microprocessors based on the x86 architecture after Intel Corporation, and the third-largest supplier of graphics processing units. It also owns 21 percent of Spansion, a supplier of non-volatile flash memory. In 2007, AMD ranked eleventh among semiconductor manufacturers in terms of revenue.[1]

Corporate history

Advanced Micro Devices was founded on May 1, 1969, by a group of former executives from Fairchild Semiconductor, including Jerry Sanders, III, Ed Turney, John Carey, Sven Simonsen, Jack Gifford and three members from Gifford's team, Frank Botte, Jim Giles, and Larry Stenger. The company began as a producer of logic chips, then entered the RAM chip business in 1975. That same year, it introduced a reverse-engineered clone of the Intel 8080 microprocessor. During this period, AMD also designed and produced a series of bit-slice processor elements (Am2900, Am29116, Am293xx) which were used in various minicomputer designs.

During this time, AMD attempted to embrace the perceived shift towards RISC with their own AMD 29K processor, and they attempted to diversify into graphics and audio devices as well as EPROM memory. It had some success in the mid-80s with the AMD7910 and AMD7911 "World Chip" FSK modem, one of the first multistandard devices that covered both Bell and CCITT tones at up to 1200 baud half duplex or 300/300 full duplex. While the AMD 29K survived as an embedded processor and AMD spinoff Spansion continues to make industry leading flash memory, AMD was not as successful with its other endeavors. AMD decided to switch gears and concentrate solely on Intel-compatible microprocessors and flash memory. This put them in direct competition with Intel for x86 compatible processors and their flash memory secondary markets.

AMD announced a merger with ATI Technologies on July 24, 2006. AMD paid $4.3 billion in cash and 58 million shares of its stock for a total of US$5.4 billion. The merger completed on October 25, 2006[2] and ATI is now part of AMD.

It has been reported in December 2006 that AMD along with its main rival in the graphics industry nVidia, received subpoenas from the Justice Department regarding possible antitrust violations in the graphics card industry, including the act of fixing prices.[3]

In October 2008, AMD stated that it is going to spin off their manufacturing operations in the form of a multibillion-dollar joint venture with Advanced Technology Investment Co., an investment company formed by the government of Abu Dhabi. The new venture is called Foundry Co. This will allow AMD to focus solely on chip design.[4]

Processor market history

IBM PC and the x86 architecture

In February 1982, AMD signed a contract with Intel, becoming a licensed second-source manufacturer of 8086 and 8088 processors. IBM wanted to use the Intel 8088 in its IBM PC, but IBM's policy at the time was to require at least two sources for its chips. AMD later produced the Am286 under the same arrangement, but Intel canceled the agreement in 1986 and refused to convey technical details of the i386 part. AMD challenged Intel's decision to cancel the agreement and won in arbitration, but Intel disputed this decision. A long legal dispute followed, ending in 1994 when the Supreme Court of California sided with AMD. Subsequent legal disputes centered on whether AMD had legal rights to use derivatives of Intel's microcode. In the face of uncertainty, AMD was forced to develop "clean room" versions of Intel code.

In 1991, AMD released the Am386, its clone of the Intel 386 processor. It took less than a year for the company to sell a million units. Later, the Am486 was used by a number of large original equipment manufacturers, including Compaq, and proved popular. Another Am486-based product, the Am5x86, continued AMD's success as a low-price alternative. However, as product cycles shortened in the PC industry, the process of reverse engineering Intel's products became an ever less viable strategy for AMD.

AMD's first in-house x86 processor was the K5 which was launched in 1996.[5] The "K" was a reference to "Kryptonite", which from comic book lore, was the only substance that could harm Superman, with a clear reference to Intel, which dominated in the market at the time, as "Superman".[6]

In 1996, AMD purchased NexGen specifically for the rights to their Nx series of x86-compatible processors. AMD gave the NexGen design team their own building, left them alone, and gave them time and money to rework the Nx686. The result was the K6 processor, introduced in 1997.

K5, K6 and Athlon

The K7 was AMD's seventh generation x86 processor, making its debut on June 23, 1999, under the brand name Athlon. On October 9, 2001 the Athlon XP was released, followed by the Athlon XP with 512KB L2 Cache on February 10, 2003.[7]

Athlon 64, Opteron and Phenom

The K8 was a major revision of the K7 architecture, with the most notable features being the addition of a 64-bit extension to the x86 instruction set (officially called AMD64), the incorporation of an on-chip memory controller, and the implementation of an extremely high performance point-to-point interconnect called HyperTransport, as part of the Direct Connect Architecture. The technology was initially launched as the Opteron server-oriented processor.[8] Shortly thereafter it was incorporated into a product for desktop PCs, branded Athlon 64.[9]

AMD released the first dual core Opteron, an x86-based server CPU, on April 21, 2005.[10] The first desktop-based dual core processor family — the Athlon 64 X2 — came a month later.[11] In early May 2007, AMD had abandoned the string "64" in its dual-core desktop product branding, becoming Athlon X2, downplaying the significance of 64-bit computing in its processors while upcoming updates involves some of the improvements to the microarchitecture, and a shift of target market from mainstream desktop systems to value dual-core desktop systems. AMD has also started to release dual-core Sempron processors in early 2008 exclusively in China, branded as Sempron 2000 series, with lower HyperTransport speed and smaller L2 cache, thus the firm completes its dual-core product portfolio for each market segment.

The latest AMD microprocessor architecture, known as K10, became the successor to the K8 microarchitecture. The first processors released on this architecture were introduced on September 10, 2007 consisting of nine quad-core Third Generation Opteron processors. This was followed by the Phenom processor for desktop. K10 processors will come in dual, triple-core,[12] and quad-core versions with all cores on one single die.

Fusion

After the merger between AMD and ATI, an initiative codenamed Fusion was announced that merges a CPU and GPU on one chip, including a minimum 16 lane PCI Express link to accommodate external PCI Express peripherals, thereby eliminating the requirement of a northbridge chip completely from the motherboard. AMD will move to a modular design methodology named "M-SPACE", where two new processor cores, codenamed "Bulldozer" and "Bobcat" will be released in the 2009 timeframe.

While very little preliminary information exists even in AMD's Technology Analyst Day 2007, both cores are to be built from the ground up. The Bulldozer core focused on 10 watt to 100 watt products, with optimizations for performance-per-watt ratios and HPC applications and includes newly announced SSE5 instructions, while the Bobcat core will focus on 1 watt to 10 watt products, given that the core is a simplified x86 core to reduce power draw. Both of the cores will be able to incorporate full DirectX compatible GPU core(s) under the Fusion label, or as standalone products as a general purpose CPU.

Other platforms and technologies

AMD chipsets

Before the launch of Athlon 64 processors in 2003, AMD designed chipsets for their processors spanning the K6 and K7 processor generations. The chipsets include the AMD-640, AMD-751 and the AMD-761 chipsets. The situation changed in 2003 with the release of Athlon 64 processors, and AMD chose not to further design its own chipsets for its desktop processors while opening the desktop platform to allow other firms to design chipsets. This is the "Open Platform Initiative". The initiative was proven to be a success, with many firms such as Nvidia, ATI, VIA and SiS developing their own chipset for Athlon 64 processors and later Athlon 64 X2 and Athlon 64 FX processors, including the Quad FX platform chipset from Nvidia.

The initiative went further with the release of Opteron server processors as AMD stopped the design of server chipsets in 2004 after releasing the AMD-8111 chipset, and again opened the server platform for firms to develop chipsets for Opteron processors. As of today, Nvidia and Broadcom are the sole designing firms of server chipsets for Opteron processors.

As the company completed the acquisition of ATI Technologies in 2006, the firm gained the ATI design team for chipsets which previously designed the Radeon Xpress 200 and the Radeon Xpress 3200 chipsets. AMD then renamed the chipsets for AMD processors under AMD branding (for instance, the CrossFire Xpress 3200 chipset was renamed as AMD 580X CrossFire chipset). In February 2007, AMD announced the first AMD-branded chipset since 2004 with the release of the AMD 690G chipset (previously under the development codename RS690), targeted at mainstream IGP computing. It was the industry's first to implement a HDMI 1.2 port on motherboards, shipping for more than a million units. While ATI had aimed at releasing an Intel IGP chipset, the plan was scrapped and the inventories of Radeon Xpress 1250 (codenamed RS600, sold under ATI brand) was sold to two OEMs, Abit and AsRock. Although AMD states the firm will still produce Intel chipsets, Intel had not granted the license of 1333 MHz FSB to ATI. Considering the rivalry between AMD and Intel, AMD is less likely to release more Intel chipset designs in the foreseeable future.

On November 15, 2007, AMD has announced a new chipset series portfolio, the AMD 7-Series chipsets, covering from enthusiast multi-graphics segment to value IGP segment, to replace the AMD 480/570/580 chipsets and AMD 690 series chipsets. Marking AMD's first enthusiast multi-graphics chipset. Discrete graphics chipsets were launched on November 15, 2007 as part of the codenamed Spider desktop platform, and IGP chipsets were launched at a later time in Spring 2008 as part of the codenamed Cartwheel platform.

AMD will also return to the server chipsets market with the next-generation AMD 800S series server chipsets, scheduled to be released in 2009 timeframe.

AMD Live!

AMD LIVE! is a platform marketing initiative focusing the consumer electronics segment, with a recently announced Active TV initiative for streaming Internet videos from web video services such as YouTube, into AMD Live! PC as well as connected digital TVs, together with a scheme for an ecosystem of certified peripherals for the ease of customers to identify peripherals for AMD Live! systems for digital home experience, called "AMD Live! Ready".[13]

AMD Quad FX platform

The AMD Quad FX platform, being an extreme enthusiast platform, allows two processors to connect through HyperTransport, which is a similar setup to dual-processor (2P) servers, excluding the use of buffered memory/registered memory DIMM modules, and a server motherboard, the current setup includes two Athlon 64 FX FX-70 series processors and a special motherboard. AMD pushed the platform for the surging demands for what AMD calls "megatasking" for true enthusiasts,[14] the ability to do more tasks on one single system. The platform refreshes with the introduction of Phenom FX processors and the next-generation RD790 chipset, codenamed "FASN8".

Commercial platform

The first AMD server/workstation platform after ATI acquisition is scheduled to be released on 2009 timeframe. Codenamed Fiorano, AMD's first multi-processor server platform after ATI acquisition consists of AMD SR5690 + SP5100 server chipsets, supporting 45 nm, codenamed Shanghai Socket F+ processors and registered DDR2 memory. Future update include the Maranello platform supporting 45 nm, codenamed Istanbul, Socket G34 processors with DDR3 memory. On single-processor platform, the codenamed Catalunya platform consists of codenamed Suzuka 45 nm quad-core processor with AMD SR5580 + SP5100 chipset and DDR3 support. [15]

AMD's virtualization extension to the 64-bit x86 architecture is named AMD Virtualization, also known by the abbreviation AMD-V, and is sometimes referred to by the code name "Pacifica". AMD processors using Socket AM2, Socket S1, and Socket F include AMD Virtualization support. AMD Virtualization is also supported by release two (8200, 2200 and 1200 series) of the Opteron processors. The third generation (8300 and 2300 series) of Opteron processors will see an update in virtualization technology, specifically the Rapid Virtualization Indexing (also known by the development name Nested Page Tables), alongside the Tagged TLB and Device Exclusion Vector (DEV).

AMD also promotes the "AMD I/O Virtualization Technology" (also known as IOMMU) for I/O virtualization.[16] The AMD IOMMU specification has been updated to version 1.2.[17] The specification describes the use of a HyperTransport architecture.

AMD's commercial initiatives include the following:

    * AMD Trinity, provides support for virtualization, security and management. Key features include AMD-V technology, codenamed Presidio trusted computing platform technology, I/O Virtualization and Open Management Partition.[18]

    * AMD Raiden, future clients similar to the Jack PC[19] to be connected through network to a blade server for central management, to reduce client form factor sizes with AMD Trinity features.

    * Torrenza, coprocessors support through interconnects such as HyperTransport as PCI Express (though more focus was at HyperTransport enabled coprocessors), also opening processor socket architecture to other manufacturers, Sun and IBM are among the supporting consortium, with rumoured POWER7 processors would be socket-compatible to future Opteron processors. The move made rival Intel responded with the open of Front Side Bus (FSB) architecture as well as Geneseo,[20] a collaboration project with IBM for coprocessors connected through PCI Express. Note that AMD positioned Torrenza for commercial segment, whilst Intel positioned Geneseo for all segments including consumer desktop segments[citation needed].

    * Various certified systems programs and platforms: AMD Commercial Stable Image Platform (CSIP), together with AMD Validated Server program, AMD True Server Solutions, AMD Thermally Tested Barebones Platforms and AMD Validated Server Program, providing certified systems for business from AMD.

Desktop platforms

Starting in 2007, AMD, following Intel, began using codenames for its desktop platforms such as Spider. The platforms, unlike Intel's approach, will refresh every year, putting focus on platform specialization. The platform includes components as AMD processors, chipsets, ATI graphics and other features, but continued to the open platform approach, and welcome components from other vendors such as VIA, SiS, and Nvidia, as well as wireless product vendors.

Updates to the platform includes the implemtation of IOMMU I/O Virtualization with 45 nm generation of processors, and the AMD 800 chipset series in 2009.[21]

Embedded systems

In February 2002, AMD acquired Alchemy Semiconductor and continued its line of processor in MIPS architecture processors, targets the hand-held and Portable media player markets. On June 13, 2006, AMD officially announced that the Alchemy processor line was transferred to Raza Microelectronics Inc.[22]

In August 2003, AMD also purchased the Geode business which was originally the Cyrix MediaGX from National Semiconductor to augment its existing line of embedded x86 processor products. During the second quarter of 2004, it launched new low-power Geode NX processors based on the K7 Thoroughbred architecture with speeds of fanless processors 667 MHz and 1 GHz, and 1.4 GHz processor with fan, of TDP 25 W. This technology is used in a variety of embedded systems (Casino slot machines and customer kiosks for instance), several UMPC designs in Asia markets, as well as the OLPC XO-1 computer, an inexpensive laptop computer intended to be distributed to children in developing countries around the world.

For the past couple of years AMD has been introducing 64-bit processors into its embedded product line starting with the AMD Opteron processor. Leveraging the high throughput enabled through HyperTransport and the Direct Connect Architecture these server class processors have been targeted at high end telecom and storage applications. In 2006 AMD added the AMD Athlon, AMD Turion and Mobile AMD Sempron processors to its embedded product line. Leveraging the same 64-bit instruction set and Direct Connect Architecture as the AMD Opteron but at lower power levels and in smaller footprint packages[citation needed], these processors were well suited to a variety of traditional embedded applications. Throughout 2007 and into 2008 AMD has continued to add both single-core Mobile AMD Sempron and AMD Athlon processors and dual-core AMD Athlon X2 and AMD Turion processors to its embedded product line and now offers embedded 64-bit solutions starting with 8W TDP Mobile AMD Sempron and AMD Athlon processors for fan-less designs up to multi-processor systems leveraging multi-core AMD Opteron processors all supporting longer than standard availability.

In April 2007, AMD announced the release of the M690T integrated graphics chipset for embedded designs. This enabled AMD to offer complete processor and chipset solutions targeted at embedded applications requiring high performance 3D and video such as emerging digital signage, kiosk and Point of Sale applications. The M690T was followed by the M690E specifically for embedded applications which removed the TV output, which required Macrovision licensing for OEMs, and enabled native support for dual TMDS outputs, enabling dual independent DVI interfaces.

Flash technology

While less visible to the general public than its CPU business, AMD is also a global leader in flash memory. In 1993, AMD established a 50-50 partnership with Fujitsu called FASL, and merged into a new company called FASL LLC in 2003. The joint venture firm went public under ticker symbol SPSN in December 2005, with AMD shares drop to 37%.

AMD no longer directly participates in the Flash memory devices market now as AMD entered into a non-competition agreement, as of December 21, 2005, with Fujitsu and Spansion, pursuant to which it agreed not to directly or indirectly engage in a business that manufactures or supplies standalone semiconductor devices (including single chip, multiple chip or system devices) containing only Flash memory.[23]

Mobile platforms

AMD started a platform in 2003 aimed at mobile computing, but, with fewer advertisements and promotional schemes, very little was known about the platform. The platform used mobile Athlon 64 or mobile Sempron processors.

As part of the "Better by design" initiative, the open mobile platform, announced February 2007 with announcement of general availability in May 2007, comes together with 65 nm fabrication process Turion 64 X2, and consists of three major components: an AMD processor, graphics from either Nvidia or ATI Technologies which also includes integrated graphics (IGP), and wireless connectivity solutions from Atheros, Broadcom, Marvell, Qualcomm or Realtek.

The Puma platform and Turion Ultra processor was released on June 4, 2008. In the future, AMD plans quad-core processors with 3D graphics capabilities (Fusion) to be launched in 2009 as the Eagle platform.

Other initiatives

    * 50x15, digital inclusion, with targeted 50% of world population to be connected through Internet via affordable computers by the year of 2015.

    * The Green Grid,[24] founded by AMD together with other founders, such as IBM, Sun and Microsoft, to seek lower power consumption for grids. Intel was notably absent from the consortium when it was founded, and finally joined in early 2007.[25]

    * Codenamed SIMFIRE interoperability testing tool for the Desktop and mobile Architecture for System Hardware (DASH) open architecture.

Software

Extensions for software parallelism (xSP), aimed at speeding up programs to enable multi-threaded and multi-core processing, announced in Technology Analyst Day 2007. One of the initiative being discussed since August 2007 is the Light Weight Profiling (LWP), providing internal hardware monitor with runtimes, to observe information about executing process and help the re-design of software to be optimized with multi-core and even multi-threaded programs. Another one is the extension of Streaming SIMD Extension (SSE) instruction set, the SSE5.

Technologies from ATI

After the takeover of ATI, AMD restructured some of the product lineups from both companies. Some products were being rebranded under the AMD brand, including the Imageon for mobile phones and handheld devices, the Xilleon for consumer electronics (digital TV sets), ATI Xpress chipsets (to AMD chipsets) for AMD processors platform and GPGPU computing line-up FireStream, previously known as AMD Stream Processor. Some others retained the use of ATI branding, including the Radeon line of graphics, and chipsets for Intel processors.

Production and fabrication

AMD produces their own processors in wholly owned semiconductor Fabrication Plants, called "FABs". AMD uses a "FAB x" naming convention for their production facilities, where "x" is the number of years that have passed between the founding of AMD and the date the FAB opened.

At their Fabrication facilities, AMD utilizes a system called Automated Precision Manufacturing (APM). APM is a collection of manufacturing technologies AMD has developed over their history (many of which AMD holds patents for), which are designed to enhance the microprocessor production process, primarily in terms of yield. Much of APM is related to removing the "human equation" from the manufacturing process by isolating in-process wafers in containers that are only exposed to clean room facilities. AMD claims that the technologies that combine to make APM are unique to the industry and make it the foremost semiconductor manufacturer in the world - a fact which is lent some credence by their current agreement with Chartered Semiconductor Manufacturing based in Singapore.

AMD currently has a production agreement with foundry Chartered Semiconductor Manufacturing which allows Chartered access to AMD Automated Precision Manufacturing (APM) process technology, in exchange for which Chartered will act as extra production capacity for AMD.

Through the acquisition of ATI, AMD also has manufacturing agreements with TSMC to produce ATI's lines of graphics and chipset processors. It is currently unclear how much of ATI's manufacturing needs will be moved to AMD's own fabs and how much will remain outsourced to other foundry companies, but AMD has announced plans for future processors to be outsourced to TSMC, while coincidently TSMC had announced it had received orders to fabricate x86 processors.

AMD's main microprocessor manufacturing and design facilities are located in Dresden, Germany. Additionally, highly integrated microprocessors are manufactured in Taiwan made by third-party manufacturers under strict license from AMD. Between 2003 and 2005, they constructed a second manufacturing plant (300 mm 90 nm process SOI) in the same complex in order to increase the number of chips they can produce, thus becoming more competitive with Intel. The new plant has been named "Fab 36", in recognition of AMD's 36 years of operation, and is expected to reach full production in mid-2007. AMD recently announced that they have just completed the conversion of Fab 36 from 90 nm to 65 nm and have now shifted their focus to the 45 nm conversion.[26]

AMD has planned expansions in their production capacity. In addition to the completion of Fab 36 in Dresden, AMD is planning to upgrade Fab 30 (adjacent to Fab 36) in Dresden from 200 mm 90 nm process SOI to a 300 mm 65 nm process SOI facility and rename it Fab 38. Originally, Fab 30 was supposed to begin 65 nm production in late 2007 but AMD recently announced they would slow down the upgrade to reduce capital expenditures.[27]

Packaging and testing facilities for its microprocessor products are located in Singapore, Malaysia and China.

Furthermore, AMD announced plans to open a new $3.2 billion facility at the Luther Forest Technology Campus across the towns of Malta and Stillwater in Saratoga County, New York. This new Fab 4x will likely produce 300 mm 32 nm process SOI production, with construction taking place from 2009 to 2010. Some speculation exists as to whether this facility will use high-K/metal gate technology that AMD obtained from IBM.[28]

AMD has also invested $3billion to build a chip fabrication plant in India. Currently, AMD is manufacturing chips in India as a result of their partnership with SemIndia, a group of investors aiming at building a wafer fab, as well as assembly and test operation centers. "AMD ponders over new chip plant in India"

In June 2006, Chartered Semiconductor began shipments of manufactured AMD microprocessors, many of which are shipped from Singapore to Taiwanese and Chinese OEM/ODM manufacturing companies that build computers for companies like Lenovo and Dell.[29]

AMD maintains major design facilities in Fort Collins, CO, Sunnyvale, CA, Austin, TX, Boxborough, MA and Bangalore, India. With the acquisition of ATI Technologies, the company gained ownership over major design facilities in Markham, ON and Santa Clara, CA.

Corporate affairs

Partnerships

AMD utilizes strategic industry partnerships to further its business interests as well as to tackle Intel's dominance and resources. Notably Nvidia's nForce2 chipset generated substantial revenues for Nvidia as a popular enthusiast part.

A partnership between AMD and Alpha Processor Inc. developed HyperTransport, a point-to-point interconnect standard which was turned over to an industry standards body for finalization. It is now used in modern AMD processor compatible motherboards.

AMD also formed a strategic partnership with IBM, under which AMD gained silicon on insulator (SOI) manufacturing technology, and detailed advice on 90 nm implementation, the partnership was announced by AMD to be extended to 2011 for 32 nm and 22 nm fabrication related technologies.[30] Further, AMD is loosely partnered with end-user companies such as HP, Compaq, ASUS, Alienware, Acer, Evesham Technology, Dell and several others to facilitate processor distribution and sales.

On May 18, 2006, Dell announced that it would roll out new servers based on AMD's Opteron chips by years end, thus ending an exclusive relationship with Intel. Dell also began offering AMD Athlon X2 chips in their desktop line-up in September 2006.

AMD is also a sponsor of the Scuderia Ferrari Marlboro F1 Team since 2002 and the Discovery Channel Pro Cycling Team since 2004.

Litigation with Intel

AMD has a long history of litigation with former partner and x86 creator Intel.[31][32][33]

    * In 1986 Intel broke an agreement it had with AMD to allow them to produce Intel's micro-chips for IBM; AMD filed for arbitration in 1987 and the arbitrator decided in AMD's favor in 1992. Intel disputed this, and the case ended up in the Supreme Court of California. In 1994, that court upheld the arbitrator's decision and awarded damages for breach of contract.

    * In 1990, Intel brought a copyright infringement action alleging illegal use of its 287 microcode. The case ended in 1994 with a jury finding for AMD and its right to use Intel's microcode in its microprocessors through the 486 generation.

    * In 1997, Intel filed suit against AMD and Cyrix Corp. for misuse of the term MMX. AMD and Intel settled, with AMD acknowledging MMX as a trademark owned by Intel, and with Intel granting AMD rights to market the AMD K6 MMX processor.

    * In 2005, following an investigation, the Japan Federal Trade Commission found Intel guilty on a number of violations. On June 27, 2005, AMD won an antitrust suit against Intel in Japan, and on the same day, AMD filed a broad antitrust complaint against Intel in the U.S. Federal District Court in Delaware. The complaint alleges systematic use of secret rebates, special discounts, threats, and other means used by Intel to lock AMD processors out of the global market. Since the start of this action, The Court has issued subpoenas to major computer manufacturers including Acer, Dell, Lenovo, HP and Toshiba.

Events and publications

Although AMD frequently refuses to provide information about upcoming products and plans, it does hold annual Technology Analyst Days (often shortened to "Analyst Day") to reveal and explain key future technologies, and to present official technology roadmaps. The event held in mid-year is referred to as "Technology Analyst Day", with its main focus on upcoming technologies and trends[34]. The end-of-year event is referred to as "Financial Analyst Day" and focuses on the financial performance of the company through the previous year[35]

In addition to these events, AMD also publishes printed media. Publications include the AMD Accelerate and the discontinued AMDEdge. The AMD Accelerate magazine, originally published through Ziff Davis Media, focuses on SME and business applications, while AMD Edge focused on overall technologies from AMD. Since Ziff Davis Media filed for Chapter 11 bankruptcy protection, the AMD Accelerate magazine has been published through IDG. AMD also has electronic newsletters to promote its server-oriented Opteron processors and related business solutions.