| Q: | What did AMD announce on September 10, 2007?
|
| A: | Today AMD is introducing the Quad-Core AMD Opteron™ processor, previously code-named “Barcelona,” the world’s most advanced x86 processor ever designed and manufactured and also the first native x86 quad-core microprocessor. Quad-Core AMD Opteron 2300 Series, the 2300 HE Series, the 8300 Series and the 8300 HE Series are now available.
In addition, AMD is announcing a new metric today called Average CPU Power (ACP) to better represent real-world processor power consumption to the IT managers responsible for power budgeting to help them make more informed buying decisions from the initial purchase to the lifetime cost of the server. This metric represents processor power consumption and will be provided for Quad-Core AMD Opteron processors, in addition to the thermal design power specification (or TDP) used by system designers. |
|
|
| Q: | What’s new about the Quad-Core AMD Opteron processor?
|
| A: | Designed from inception for the most demanding datacenters, the Quad-Core AMD Opteron processor brings significant enhancements to market in the most critical areas for today’s datacenter consumers: energy efficiency, investment protection, virtualization and high-performance:
Energy-Efficiency:
- AMD CoolCore™ technology can reduce energy consumption and heat generation by turning off unused parts of the processor.
Independent Dynamic Core Technology, an enhancement to AMD PowerNow! technology, allows each core to vary its clock frequency depending on the specific performance requirement of the applications it is supporting, helping to reduce power consumption.
- Dual Dynamic Power Management (DDPM) provides an independent power supply to the cores and to the memory controller, allowing the cores and memory controller to operate on different voltages, depending on usage.
Investment Protection:
- Quad-Core AMD Opteron processors maintain compatibility with the socket and thermal envelopes of Second-Generation AMD Opteron processors to enable a seamless customer upgrade path. AMD’s common core strategy and same socket technology are designed to empower customers to scale with one AMD64 core architecture from 1P all the way to 8P servers and workstations, to reduce management complexity and increase datacenter uptime and productivity.
Virtualization:
- Virtualization is memory intensive and the Quad-Core AMD Opteron processor is expected to continue delivering industry-leading performance for virtualized environments thanks to memory throughput through Direct Connect Architecture and its integrated memory controller.
- AMD Virtualization™ introduces Rapid Virtualization Indexing (formerly called “Nested Paging”) and Tagged TLBs. AMD’s Rapid Virtualization Indexing feature is designed to reduce the overhead penalty associated with virtualization technologies by moving the process of managing virtual memory from software to hardware, reducing the complexity of existing x86 virtualization solutions and enabling increased performance and efficiency for many virtual workloads, allowing for a higher performing, more flexible IT environment.
High-Performance:
- AMD Memory Optimizer Technology increases memory throughput by up to 50% compared to previous generations of the AMD Opteron processor.
- AMD Wide Floating Point Accelerator provides 128-bit SSE floating point capabilities, which enable each core to simultaneously execute up to four FLOPS per clock (four times the floating-point computations of previous AMD Opteron processors) for significantly improving performance in compute-intensive and workstation applications.
- AMD Balanced Smart Cache - significant cache enhancements with 512KB L2 cache per core and 2MB shared L3 cache across all four cores.
|
|
|
| Q: | What does this launch mean to AMD?
|
| A: | The introduction of Quad-Core AMD Opteron processors is a pinnacle of x86 design and customer-centric innovation. Today, AMD launches not only the world’s first native quad-core processor with the most advanced x86 architecture extending its leadership in server processor design; we also set the stage for phenomenal PC performance later this year based on this design. This launch demonstrates again that the best ideas are inspired by conversations with the most demanding customers.
Furthermore, with the support of enterprise customers and industry partners around the world and more than 50 socket-compatible AMD Opteron processor-based platforms already on the market, AMD is ready to continue its growth as the datacenter technology of choice. |
|
|
| Q: | What does this launch mean to the industry?
|
| A: | Quad-Core AMD Opteron processors – the first true native x86 quad-core server and workstation processors – represent a significant step forward in enterprise computing, especially for those whose priorities include energy efficiency, investment protection, virtualization and high performance. By delivering the most advanced x86 processors at compelling price points, AMD is helping to enable today’s enterprises to maximize their IT investments. |
|
|
| Q: | How does Quad-Core AMD Opteron differ from Dual-Core AMD Opteron?
|
| A: | Recognizing that platform stability and scalability are important to many customers, AMD is enabling significantly increased performance with its newly introduced quad-core processors without increasing power requirements by leveraging the same thermal and power envelopes as its dual-core processors. AMD has also taken its Quad-Core AMD Opteron processors to the next level by implementing significant enhancements in the most critical areas for today’s datacenter consumers such as hardware-enabled virtualization and energy-saving technologies. |
|
|
| Q: | What does “native” architecture mean?
|
| A: | AMD’s Quad-Core AMD Opteron processors go far beyond simply adding two more cores, but rather presents a “native” multi-core design where all four cores are on one piece of silicon. More than just four cores, Quad-Core AMD Opteron processors feature Direct Connect Architecture, which means you have processors directly connected to one another, and an I/O and Memory Controller directly connected to each processor to reduce bottlenecks for better performance. Furthermore, Quad-Core AMD Opteron processors are not only the industry’s first native x86 quad-core processors, they are also the first quad-core processors designed within the same thermal and power envelopes as AMD’s current generation of dual-core processors. |
|
|
| Q: | What’s the die size and how many transistors does it use?
|
| A: | The die size is 285mm2. The processor uses approximately 463 million transistors. |
|
|
| Q: | Will AMD be providing benchmarks for Quad-Core AMD Opteron processors and when will they be available?
|
| A: | Continuing the legacy of the AMD Opteron processor family, Quad-Core AMD Opteron processors introduce several new ground-breaking technologies enabling impressive results across a suite of benchmarks. Within comparable thermal bands, Quad-Core AMD Opteron processors outperform the competition on several industry-standard benchmarks including: SPECfp®_rate2006, SPECint®_rate2006, SPECompM®2001Base, STREAM, Fluent and LS-DYNA. For more details on the performance of Quad-Core AMD Opteron processors, please visit
http://www.amd.com/opteronperformance.
|
|
|
| Q: | What motherboard support is available for these processors?
|
| A: | More than 50 different socket-compatible motherboards will be offered by leading vendors. |
|
|
| Q: | What chipset support is available for these processors?
|
| A: | Since Quad-Core AMD Opteron processors are designed to be compatible with Second-Generation AMD Opteron processors, all chipsets that currently support the dual-core processors will also support the quad-core processors. |
|
|
| Q: | Why should customers invest in this first version of Quad-Core AMD Opteron given the performance will dramatically increase with subsequent versions?
|
| A: | For the first time in its history, AMD is including its High Efficiency (HE) models at launch, based on the growing emphasis on performance-per-watt and the popularity of this processor line. While AMD expects its native quad-core processors to scale to higher frequencies in Q407 in both standard and SE versions, we believe customers looking for increased performance within the same thermal and power envelopes as current generation AMD Opteron processors will be compelled by the significant performance-per-watt, performance and virtualization enhancements delivered today. |
|
|
| Q: | What is ACP?
|
| A: | ACP is a metric that offers a relevant estimation of the power consumption of the Quad-Core AMD Opteron processors. ACP is determined by breaking down multiple components of power consumed within the processor, including the power dedicated to the cores, to the integrated memory controller, and to HyperTransport™ technology links.
ACP and TDP are both indicators of processor power. TDP refers to the thermal design power processors are capable of consuming that is the specification system designers follow. AMD has referenced processor power consumption based off of TDP wattages to date. ACP reflects a relevant wattage that reflects power consumption while running server-class enterprise workloads. ACP is a useful metric for data center operators to use when estimating power budgets to size their datacenters. Thermal design power or TDP is more useful and relevant to system designers. |
|
|
| Q: | How does ACP line up against Thermal Design Power (TDP)?
|
| A: | In the simplest terms, ACP conveys what a user’s CPU power experience may be like while running workloads under typical data center environmental and operating conditions. In contrast, AMD’s TDP specification is the condition under which the processor is designed to function. The two can be quite different as a result. AMD’s TDP specification is intended for use by system designers who need to know what thermal envelope to design a system for. AMD’s TDP and ACP values are intended for completely different uses, one for system designers, the other for those curious what the CPU power may be like under real world conditions. The ACP value can have greater relevance for people in job roles that require them to estimate power consumption for the purpose of power budgeting. AMD will continue to provide thermal design power (TDP) specifications to platform designers and industry partners in the AMD thermal and power datasheets. |
|
|
| Q: | How does ACP relate to thermal design power or TDP?
|
| A: | In the simplest terms, ACP conveys what a user’s CPU power experience may be like while running workloads under typical data center environmental and operating conditions. In contrast, AMD’s TDP specification is the condition under which the processor is designed to function. The two can be quite different as a result. AMD’s TDP specification is intended for use by system designers who need to know what thermal envelope to which to design a system. AMD’s TDP and ACP values are intended for completely different uses, one for system designers, the other for those curious what the CPU power may be like under real-world conditions. The ACP value can have greater relevance for people in job roles that require them to estimate power consumption for the purpose of power budgeting. AMD will continue to provide thermal design power (TDP) specifications to platform designers and industry partners in the AMD thermal and power datasheets. |
|
|
| Q: | Is ACP replacing TDP?
|
| A: | No. ACP and TDP will coexist. |
|
|
| Q: | What difference will customers see?
|
| A: | Customers will not see a difference, other than the fact that ACP is intended to more closely mirror their actual levels of processor power consumption under real-world workloads. |
|
|
| Q: | Why is AMD introducing ACP during the launch of the Quad-Core AMD Opteron™ processor?
|
| A: | The Quad-Core AMD Opteron™ processor architecture contains a number of key energy efficiency improvements to reduce the amount of power consumed by our processors across a wide variety of usage scenarios and workloads. A new feature called Dual Dynamic Power Management (DDPM) provides the processors with additional power rails which are dedicated to the processor cores, and to the HyperTransport™ technology links and internal memory controller. Separating the power rails of the cores from the internal memory controller allows each of the cores to independently adjust frequency for the given workload while also allowing for voltage changes to all the cores. This technology enables considerable power savings at the processor level during non-peak workloads. The ACP values for each power band include the power for the cores, integrated memory controller and HyperTransport™ technology links. |
|
|
