Advanced Micro Devices may not be in position to release its new dual-processor enthusiast-class platform this year, however, it may be thinking about a different product, according to some unofficial information. Apparently, the company may be mulling about a triple-core microprocessor for desktops.
According to a news-story at Hard Tecs 4U web-site, AMD is working on a processor that would have three processing engines. The company reportedly told its partners about the AMD Phenom X3 chip at a closed meeting, explaining that the chipmaker plans to release such a product because of “market demand”.
The new triple-core microprocessors will feature its own design and will not be quad-core chips with one core disabled, according to the web-site. Nevertheless, the chips will still include 2MB of shared L3 cache and will take advantage of other K10 micro-architecture features, such as SSE4A instruction set, 128-bit floating point units (FPU) and so on. Obviously, the chips will also have advanced power management capabilities.
According to estimates by X-bit labs, each processing engine of quad-core AMD Opteron/Phenom processors takes about 13% of the die size. Given the whole die size of approximately 285mm² and about 218 chip candidates obtained from every 300mm wafer, X-bit labs believes that it is highly unlikely that AMD had decided to develop a separate tripe-core design with about 247mm² die size and 250 chip candidates obtained from a 300mm wafer unless the yields of the new chips are so low that the company needs a redundant third processing engine to create a dual-core product with sufficient yield.
Saturday, April 4, 2009
AMD’s Latest Opteron, Phenom Chips Already Feature DDR3 Memory Support
AMD’s Latest Opteron, Phenom Chips Already Feature DDR3 Memory Support
The new tenth-generation microprocessors from AMD promise to increase performance because of micro-architectural improvements and the increased amount of processing engines. However, it appears that the new chips do hide some additional capabilities under the hood. For example, DDR3 support.
Advanced Micro Devices has always been conservative about new memory types support as each type of memory requires a brand new platform family and having several platform families for one processor is something that AMD cannot afford. However, this does not mean that the company’s chips cannot support new types of memory. Apparently, already now AMD’s latest processors feature DDR3 memory controller.
Technical documents that cover AMD’s latest microprocessors, such as quad-core AMD Opteron or AMD Phenom and their derivatives, reveal that the new built-in memory controllers can support both DDR2 and DDR3 memory types, provided that processors are installed into mainboards with DDR3 memory slots and appropriate BIOS.
According to roadmaps that AMD shows to its partners, DDR3 memory support will only be available in microprocessors made using 45nm process technology and on AM3 platforms due out in the second half of 2008. However, given that AMD’s documents for BIOS developers already note both AM3 and DDR3, it is highly likely that even current versions of AMD’s tenth-generation microprocessors can support DDR3 memory in certain conditions.
If AMD’s K10 processors that AMD makes now support DDR3, then the company has some additional flexibility when it comes to transition to a new memory type. For example, the company decided not to align transition to DDR2 with processors made using 65nm process technology. Given AMD usually proceeds relatively slowly with manufacturing process change, it may certainly be a wise idea to switch to DDR3-supporting socket AM3/socket 1207 revision 2 ahead of fabrication technology change, as in this case AMD could plug even current chips to new platforms.
The new tenth-generation microprocessors from AMD promise to increase performance because of micro-architectural improvements and the increased amount of processing engines. However, it appears that the new chips do hide some additional capabilities under the hood. For example, DDR3 support.
Advanced Micro Devices has always been conservative about new memory types support as each type of memory requires a brand new platform family and having several platform families for one processor is something that AMD cannot afford. However, this does not mean that the company’s chips cannot support new types of memory. Apparently, already now AMD’s latest processors feature DDR3 memory controller.
Technical documents that cover AMD’s latest microprocessors, such as quad-core AMD Opteron or AMD Phenom and their derivatives, reveal that the new built-in memory controllers can support both DDR2 and DDR3 memory types, provided that processors are installed into mainboards with DDR3 memory slots and appropriate BIOS.
According to roadmaps that AMD shows to its partners, DDR3 memory support will only be available in microprocessors made using 45nm process technology and on AM3 platforms due out in the second half of 2008. However, given that AMD’s documents for BIOS developers already note both AM3 and DDR3, it is highly likely that even current versions of AMD’s tenth-generation microprocessors can support DDR3 memory in certain conditions.
If AMD’s K10 processors that AMD makes now support DDR3, then the company has some additional flexibility when it comes to transition to a new memory type. For example, the company decided not to align transition to DDR2 with processors made using 65nm process technology. Given AMD usually proceeds relatively slowly with manufacturing process change, it may certainly be a wise idea to switch to DDR3-supporting socket AM3/socket 1207 revision 2 ahead of fabrication technology change, as in this case AMD could plug even current chips to new platforms.
AMD Promises 2.5GHz Quad-Core Microprocessors
Advanced Micro Devices said during the formal unveiling of its quad-core server microprocessors that it would introduce 25% faster versions of the chip in about two months time. The move will allow the world’s second largest chipmaker not only to improve performance for servers, but also to present a competitive solution for desktops.
Randy Allen, corporate vice president of server and workstation division at AMD, said during the quad-core AMD Opteron launch conference that while initially AMD’s chips with four processing engines work only at up to 2.0GHz clock-speed, in December the company will be in position to offer 2.5GHz chips. But 25% speed increase will bring only up to 15% performance improvement, said Mr. Allen.
“AMD today shows the path to better performing design for demanding data center requirements, again. In fact, here is a proof to you of our 2.50GHz [quad-core] processor that is going to be available in December and what it is going to deliver to the market. […] This chart shows incremental performance that is going to be delivered on top of that existing [2.0GHz] product: you can see 15% performance benefit, 15% performance increase that will be delivered [by] our 2.50GHz product,” said Randy Allen.
In addition to clock-speed increase of server processors late this year, the chipmaker also promised to deliver the first AMD Phenom processors for desktops in December, though, remained tight-lipped regarding their working frequencies and other peculiarities.
Earlier it was reported that the chipmaker only plans to unveil its AMD Phenom FX-80 processor, which is designed for single-processor configurations, this year, whereas AMD Phenom FX-90 and FX-91 products are set to be released sometime in the first quarter of 2007. The move basically delays the introduction of AMD’s FASN8 platform, which employs two quad-core central processing units and puts AMD’s success in the field of computer enthusiasts under question.
AMD Phenom FX-80 quad-core processor is projected to operate at 2.20GHz – 2.40GHz clock-speed, have 2MB of L2 cache (1MB per core), 2MB shared L3 cache, dual-channel PC2-8500 (DDR2 1066MHz) memory controller and AM2+ form-factor with HyperTransport 3.0 bus support. The AMD Phenom FX-90 and FX-91 models are expected to operate in 2.20GHz – 2.60GHz clock-speed ranges and support dual-processor configurations.
Randy Allen, corporate vice president of server and workstation division at AMD, said during the quad-core AMD Opteron launch conference that while initially AMD’s chips with four processing engines work only at up to 2.0GHz clock-speed, in December the company will be in position to offer 2.5GHz chips. But 25% speed increase will bring only up to 15% performance improvement, said Mr. Allen.
“AMD today shows the path to better performing design for demanding data center requirements, again. In fact, here is a proof to you of our 2.50GHz [quad-core] processor that is going to be available in December and what it is going to deliver to the market. […] This chart shows incremental performance that is going to be delivered on top of that existing [2.0GHz] product: you can see 15% performance benefit, 15% performance increase that will be delivered [by] our 2.50GHz product,” said Randy Allen.
In addition to clock-speed increase of server processors late this year, the chipmaker also promised to deliver the first AMD Phenom processors for desktops in December, though, remained tight-lipped regarding their working frequencies and other peculiarities.
Earlier it was reported that the chipmaker only plans to unveil its AMD Phenom FX-80 processor, which is designed for single-processor configurations, this year, whereas AMD Phenom FX-90 and FX-91 products are set to be released sometime in the first quarter of 2007. The move basically delays the introduction of AMD’s FASN8 platform, which employs two quad-core central processing units and puts AMD’s success in the field of computer enthusiasts under question.
AMD Phenom FX-80 quad-core processor is projected to operate at 2.20GHz – 2.40GHz clock-speed, have 2MB of L2 cache (1MB per core), 2MB shared L3 cache, dual-channel PC2-8500 (DDR2 1066MHz) memory controller and AM2+ form-factor with HyperTransport 3.0 bus support. The AMD Phenom FX-90 and FX-91 models are expected to operate in 2.20GHz – 2.60GHz clock-speed ranges and support dual-processor configurations.
AMD Confirms Triple-Core Microprocessors in 2008.
Advanced Micro Devices on Monday officially confirmed that it would release triple-core central processing units (CPUs) for desktops in Q1 2008. The AMD Phenom X3 chips will help the company to offer a unique solution for personal computers (PCs), however, it is hardly going to be easy for AMD to position the new chips keeping in mind the current condition of the market.
“With our advanced multi-core architecture, AMD is in a unique position to enable a wider range of premium desktop solutions, providing a smarter choice for customers and end users. This innovation is a direct result of our development of the industry’s first true, native quad-core design, coupled with AMD’s manufacturing flexibility, to create multi-core processors in two, three, and four computational core configurations on a single die of silicon,” said Greg White, vice president and general manager, desktop division, AMD.
AMD did not unveil specifications of the new AMD Phenom triple-core microprocessors, but said that the chips will feature shared level three cache, built-in memory controller as well as HyperTransport 3.0 bus, just like the rest desktop CPUs based on the new code-named K10 micro-architecture.
Nevertheless, the company revealed that triple core processors can provide “significant performance advantages over similar dual-core AMD processors” in benchmarks like SYSmark 2007 and 3DMark 2006, as well as “ over similar quad-core AMD processors” in certain gaming and digital content creation scenarios. While the company did not elaborate over the nature of “similarity” between dual-, triple- and quad-core processors, it is highly likely that clock-speeds of triple core CPUs will be higher than those of quad-core products, but a bit lower compared to dual-core offerings from AMD.
AMD also did not disclose whether the company’s triple-core offerings will feature quad-core silicon with one core disabled, or the chipmaker will develop a separate design with only three cores available.
But while AMD’s triple-core microprocessors will be rather unique on the market, even the current Intel’s lineup of dual-core and quad-core microprocessors may pose a threat AMD’s triple-core offering, as the price of quad-core chips begins only at $266, the price of the most advanced dual-core chip from Intel.
“With our advanced multi-core architecture, AMD is in a unique position to enable a wider range of premium desktop solutions, providing a smarter choice for customers and end users. This innovation is a direct result of our development of the industry’s first true, native quad-core design, coupled with AMD’s manufacturing flexibility, to create multi-core processors in two, three, and four computational core configurations on a single die of silicon,” said Greg White, vice president and general manager, desktop division, AMD.
AMD did not unveil specifications of the new AMD Phenom triple-core microprocessors, but said that the chips will feature shared level three cache, built-in memory controller as well as HyperTransport 3.0 bus, just like the rest desktop CPUs based on the new code-named K10 micro-architecture.
Nevertheless, the company revealed that triple core processors can provide “significant performance advantages over similar dual-core AMD processors” in benchmarks like SYSmark 2007 and 3DMark 2006, as well as “ over similar quad-core AMD processors” in certain gaming and digital content creation scenarios. While the company did not elaborate over the nature of “similarity” between dual-, triple- and quad-core processors, it is highly likely that clock-speeds of triple core CPUs will be higher than those of quad-core products, but a bit lower compared to dual-core offerings from AMD.
AMD also did not disclose whether the company’s triple-core offerings will feature quad-core silicon with one core disabled, or the chipmaker will develop a separate design with only three cores available.
But while AMD’s triple-core microprocessors will be rather unique on the market, even the current Intel’s lineup of dual-core and quad-core microprocessors may pose a threat AMD’s triple-core offering, as the price of quad-core chips begins only at $266, the price of the most advanced dual-core chip from Intel.
INTEL core 2 duo and Quadcore Processors price and availability : INDIA
Intel Pentium E2140 1.6 GHz Rs.2,600
Intel Pentium E2160 1.8 GHz Rs.2,900
Intel Pentium E2180 2.0 GHz Rs.3,200
Intel Core 2 Duo E4600 2.4 GHz Rs.5,300
Intel Core 2 Duo E6550 2.33 GHz Rs.7,200
Intel Core 2 Duo E6750 2.66 GHz Rs.7,500
Intel Core 2 Duo E7200 2.53 GHz Rs.5,500
Intel Core 2 Duo E8200 2.66GHz Rs.7,650
Intel Core 2 Duo E8400 2.66GHz Rs.8,500
Intel Core 2 Quad Q6600 2.4GHz Rs.9,400
Intel Core 2 Quad Q9300 2.53GHz Rs.11,500
Intel Core 2 Quad Q9450 2.5 GHz Rs.15,200
Intel Core 2 Quad Q9550 2.83 GHz Rs.25,900
Intel Core 2 Extreme QX9650 Rs.48,500
Among these latest processors CORE 2 QUAD q9300 running at 2.53 would certainly be the best buy as it offers unmatchable performance at the cost of almost nothing, If you are searching for which processors to buy, trust me you wont be disappointed with this chip.
Intel Pentium E2160 1.8 GHz Rs.2,900
Intel Pentium E2180 2.0 GHz Rs.3,200
Intel Core 2 Duo E4600 2.4 GHz Rs.5,300
Intel Core 2 Duo E6550 2.33 GHz Rs.7,200
Intel Core 2 Duo E6750 2.66 GHz Rs.7,500
Intel Core 2 Duo E7200 2.53 GHz Rs.5,500
Intel Core 2 Duo E8200 2.66GHz Rs.7,650
Intel Core 2 Duo E8400 2.66GHz Rs.8,500
Intel Core 2 Quad Q6600 2.4GHz Rs.9,400
Intel Core 2 Quad Q9300 2.53GHz Rs.11,500
Intel Core 2 Quad Q9450 2.5 GHz Rs.15,200
Intel Core 2 Quad Q9550 2.83 GHz Rs.25,900
Intel Core 2 Extreme QX9650 Rs.48,500
Among these latest processors CORE 2 QUAD q9300 running at 2.53 would certainly be the best buy as it offers unmatchable performance at the cost of almost nothing, If you are searching for which processors to buy, trust me you wont be disappointed with this chip.
Fujitsu will integrate in the hybrid supercomputer three architecture
Welcome, Visitor. Subscribe to our RSS Feed and consider adding this article/site to your favorite social bookmark site if you find it useful. Thank you!
Fujitsu Limited Company has received an order to build a new supercomputer to a Japanese university.
A feature of a supercomputer will be a hybrid structure that includes three different architectures: UNIX-Server SPARC Enterprise M9000, a server for technical computing HX600 high-performance server and the FX1. The total theoretical peak performance of all three systems is 60 teraflop (trillion floating-point operations per second).
The first system is a UNIX-Server SPARC Enterprise M9000, a three-node (96 processors, 384 cores, 3 TB of memory). The system peak performance is 3.84 teraflop successor model PRIMEPOWER HPC2500. It uses Quad-core SPARC64 VII, developed by Fujitsu.
The second system - a distributed Linux-server HX600 (160 nodes, 640 CPUs, 2560 cores, 10 TB of RAM). A cluster of general-purpose peak performance is 25.6 teraflopa x86-compatible architecture and is designed for a wide range of tasks.
The third component of the supercomputer is a high-performance distributed UNIX-server FX1 (768 nodes, 768 processors, the kernel 3072, 24 TB of RAM). This server provides half of the theoretical performance supercomputer - 30.72 teraflop. A feature of FX1 company calls using a set of system logic with a large-capacity memory subsystem that is optimized to work with processors, SPARC64 VII.
To merge three disparate systems into one, the company intends to use the Fujitsu Parallelnavi HPC, HPC Portal and Management Portal. Disk Subsystem supercomputer will consist of an array ETERNUS2000 volume of 1.15 SDS.
Access to a computer will have the research institutions and commercial companies. Writing a supercomputer in order to be implemented in two phases: the first will be completed in May this year, and the second - in October.
Recall, first place in the TOP500 list now belongs to the hybrid system - it took a supercomputer Roadrunner, built on AMD Opteron processors and the IBM PowerXCell.
Source: www.fujitsu.com
Prolimatech prepares three models for the cooler giants CPU
Welcome, Visitor. Subscribe to our RSS Feed and consider adding this article/site to your favorite social bookmark site if you find it useful. Thank you!
The range of cooling devices for processors from companies Prolimatech created three new models: «Genesis», «Jericho» and «The Mammoth».
Prolimatech Megahalems
These large cooling devices have their origin in the model Megahalems (pictured above) and will be produced under the brand of DFI LANParty.
Drawings of models show that Jericho will be cooler medium device with aluminum radiator, fan blows, and Mammoth, and Genesis - solutions to the special form of columns of radiators.
Processor cooler Mammoth will be abundantly supplied with heat pipes coming from the contact block, directly adjacent to the surface of the processor. Then this heat is transferred to a pair of aluminum radiators, which are also forced ventilation 120-mm fan. Tests productivity has not yet been given, but perhaps the design will be sufficiently effective.
Genesis is a cooler with a similar design, but one of its units, radiator fan is perpendicular to the motherboard.
Source: www.pcgameshardware.de
NVIDIA GTX260 during in an era of crisis becoming cheaper due to the simplification
Welcome, Visitor. Subscribe to our RSS Feed and consider adding this article/site to your favorite social bookmark site if you find it useful. Thank you!
End of the past, in 2008, was noteworthy, for example, that the series was the first NVIDIA GTX200, which became the 55-nm GPU. These maps are based on the board R654, which was worth slightly less than R651. In such a construction, as already known, the number of PCB layers reduced from 14 to 10. also to reduce the cost, it was decided to get rid of the chip Volterra.
But NVIDIA want to make a third-generation GTX260, even more simple and cheap. These cards will be constructed on the board code P897/D10U-20.
Product Colorful new design GTX260
Reference design P897, sent to the companies-partners NVIDIA, said that the new boosters for the GeForce GTX260 will be easier and cheaper. PCB of these solutions there are not 10, and 8 sectors, the food is presented formula 4 +2 NVVDD with power regulator ADP4100 (without I2C interface for control voltage). Food chain FBVDDQ reduced from 2 to phase 1. Also reduction referred to the unit mosfetov, a group which now represents a DPAK, not LFPAK.
The length of the printed circuit board remain the same, only reduced its height - 1,5 see also the GeForce GTX260 third generation will be reduced BIOS EEPROM - from 1 Mbps to 512 kilobits and establish a cheap DVI-connectors.
All these changes will reduce the cost of production of ready-accelerator ... at 10-15 dollars, which should also lead to a cheaper product for the end user.
Source: Expreview.com
Coolermaster presents the system of 20 core processor PC
Welcome, Visitor. Subscribe to our RSS Feed and consider adding this article/site to your favorite social bookmark site if you find it useful. Thank you!
Near the Coolermaster stand at CeBIT, will closely watch it with your own eyes on the system, consisting of twenty-processor cores, many will want.
Five motherboards form factor Mini-ITX processors Core 2 Quad Q9400s producer appropriate in one case ATCS 840.
Meals provided kilowatt power bloom Real Power M1000W with six +12 V supply lines (one for each system and one for optical drives and hard drive).
Apart from the above fit within the shell 10 GB of memory, three 200-mm fans, one 120-mm, and five waterblock.
Of course, Coolermaster not aimed to speed records, and focused on the cooling of the strange exhibits in what has succeeded. After five hours of continuous load of the hottest CPU temperature was only 66 degrees with an ambient temperature of 30 degrees.
Near the Coolermaster stand at CeBIT, will closely watch it with your own eyes on the system, consisting of twenty-processor cores, many will want.
Five motherboards form factor Mini-ITX processors Core 2 Quad Q9400s producer appropriate in one case ATCS 840.
Meals provided kilowatt power bloom Real Power M1000W with six +12 V supply lines (one for each system and one for optical drives and hard drive).
Apart from the above fit within the shell 10 GB of memory, three 200-mm fans, one 120-mm, and five waterblock.
Of course, Coolermaster not aimed to speed records, and focused on the cooling of the strange exhibits in what has succeeded. After five hours of continuous load of the hottest CPU temperature was only 66 degrees with an ambient temperature of 30 degrees.
Intel is preparing to release new processor Atom Z5xx, including special models PT
Welcome, Visitor. Subscribe to our RSS Feed and consider adding this article/site to your favorite social bookmark site if you find it useful. Thank you!
According to a source from the company Intel in the near future, in April, the intention is to release a pair of new Atom processor series Z5xx.
Model Z550 will be the fastest atom - the frequency of the processor will reach 2.0 GHz. This decision falls within the 2.4 W TDP.
Second Processor - Z515, as a possible frequencies are called the values of 800 and 1200 MHz. One of them is chosen as a result of the technology burst mode performance. TDP of the CPU - just 0.6 Watts.
From the issuance of these decisions would benefit the users of devices MID (mobile internet device), users UMPC (ultra-mobile PC) and notebook. Also, speed at the expense of new processors will increase and the COM-modules based on Atom.
Unlike today, the popular Atom N270, Atom series processors Z5xx more compact - 14 x 13 mm to 22 x 22 mm and are paired with the system logic Intel SCH US15W (Z550) or UL11L (Z515). This increases the degree of miniaturization and energy efficiency of end solutions.
In addition, Intel will produce more processors Z5xx with the device at the end of the name «P» or «PT». This family in the fifth series will differ from conventional Z5xx increased until 22h22 mm in size for optimal cooling, while working CPU Z530P, Z530PT, Z510P and Z510PT be paired with increased to 37,5 x 37,5 mm chip Poulsbo. Intel will position these processors as a solution able to transfer heat-resistant temperature of up to +85ºC.
According to a source from the company Intel in the near future, in April, the intention is to release a pair of new Atom processor series Z5xx.
Model Z550 will be the fastest atom - the frequency of the processor will reach 2.0 GHz. This decision falls within the 2.4 W TDP.
Second Processor - Z515, as a possible frequencies are called the values of 800 and 1200 MHz. One of them is chosen as a result of the technology burst mode performance. TDP of the CPU - just 0.6 Watts.
From the issuance of these decisions would benefit the users of devices MID (mobile internet device), users UMPC (ultra-mobile PC) and notebook. Also, speed at the expense of new processors will increase and the COM-modules based on Atom.
Unlike today, the popular Atom N270, Atom series processors Z5xx more compact - 14 x 13 mm to 22 x 22 mm and are paired with the system logic Intel SCH US15W (Z550) or UL11L (Z515). This increases the degree of miniaturization and energy efficiency of end solutions.
In addition, Intel will produce more processors Z5xx with the device at the end of the name «P» or «PT». This family in the fifth series will differ from conventional Z5xx increased until 22h22 mm in size for optimal cooling, while working CPU Z530P, Z530PT, Z510P and Z510PT be paired with increased to 37,5 x 37,5 mm chip Poulsbo. Intel will position these processors as a solution able to transfer heat-resistant temperature of up to +85ºC.
Intel 17 Latest Intel Xeon processors
This is latest 17 Xeon processors are showing best conceptual future from another processor and this processor is very unique and providing best new technology. And this Intel today announced 17 new Intel Xeon processors based on Intel Core micro-architecture, codenamed Nehalem before. The processors are designed for use in servers, workstations and embedded segment planned. Main features of the processors are the automatic adaptability of the energy according to the requirement of the particular application and the speed with the highest performance levels for challenging transactions in the computing and large database queries. In addition, Intel presented today with the Intel 82599 10 Gigabit Ethernet Controller a new network chip. The new chips are the biggest advancement in the field of server CPUs, Intel has been 14 years ago with the Intel Pentium ® Pro in this market segment is entered. Intel expects that by 2015 more than 15 billion devices connected to the Internet will be and provide the necessary infrastructure to ensure that the new Xeon processors with their adaptability, efficiency and versatility also make a significant contribution. The most intelligent energy management contributes significantly to operating costs at a minimum: after about eight months, the cost of the Xeon 5500er processor paid off.
AT&T Wireless Backtracks on New, Video- (and More) Unfriendly TOS
AT&T Wireless has backtracked on changes to its Terms of Service (TOS), which were modified earlier in the week in such a way as to pretty much gimp anything from SlingPlayer to Skype, and more.
As noted by the Public Knowledge Policy Blog, the changes (in bold) were:
So, while you bask in the fact that public outcry got that change reversed, how does that last thought grab you?
As noted by the Public Knowledge Policy Blog, the changes (in bold) were:
This means, by way of example only, that checking email, surfing the Internet, downloading legally acquired songs, and/or visiting corporate intranets is permitted, but downloading movies using P2P file sharing services, customer initiated redirection of television or other video or audio signals via any technology from a fixed location to a mobile device, web broadcasting, and/or for the operation of servers, telemetry devices and/or Supervisory Control and Data Acquisition devices is prohibited.However, on Friday AT&T backtracked, issuing the following statement:
“The language added on March 30 to AT&T’s wireless data service Terms and Conditions was done in error. It was brought to our attention and we have since removed it. We apologize for any inconvenience this may have caused.”It's nice to see the furor the report on the changes started has resulted in AT&T removing the new TOS text. What's unsettling, however, is the fact is that if they inserted that text, even if done erroneously, it means they are most likely considering the change in the future.
So, while you bask in the fact that public outcry got that change reversed, how does that last thought grab you?
Intel Details Accelerated 32nm Processor Roadmap
On Tuesday in Washington, D.C., Intel President and CEO Paul Otellini announced a $7 billion investment in U.S.-based manufacturing facilities. Those factories, in New Mexico, Arizona, and Oregon will be upgraded to manufacture using 32nm manufacturing technology. At the same time, in San Francisco, Intel held a news briefing detailing, and demoing, the 32nm Westmere family of processors.
Westmere doesn't feature a new microarchitecture; it's just Nehalem migrated to a 32nm process. The company appears pretty confident in its 32nm process, however, as it said it will accelerate 32nm production with Westmere production in 2009 and a "de-emphasis" of the 45nm Auburndale chips that Intel planned to let co-exist with Westmere.
Instead Intel will overlap the 32nm dual-core chips with the upcoming of 45nm quad-core processors, Lynnfield (desktop) and Clarksfield (mobile).
Key points from the news conference and presentation:
Hard to imagine Intel leapfrogging into 32nm so quickly; AMD isn't expected to intro 32nm chips until 2011. This definitely sets up Intel again to grab more market share from AMD; anyone remember the days when AMD was the gamer's choice?
Westmere doesn't feature a new microarchitecture; it's just Nehalem migrated to a 32nm process. The company appears pretty confident in its 32nm process, however, as it said it will accelerate 32nm production with Westmere production in 2009 and a "de-emphasis" of the 45nm Auburndale chips that Intel planned to let co-exist with Westmere.
Instead Intel will overlap the 32nm dual-core chips with the upcoming of 45nm quad-core processors, Lynnfield (desktop) and Clarksfield (mobile).Key points from the news conference and presentation:
- Intel has developed a 32nm logic technology with industry-leading features
- Intel is first to demonstrate working 32nm processors
- Intel’s 32nm process is on track for production readiness in Q4 ’09
- Both CPU and SoC versions of this 32nm process will be available
- Intel’s strength as an integrated device manufacturer allows us to continue to deliver new generations of advanced process technology on a 2 year cadence
Hard to imagine Intel leapfrogging into 32nm so quickly; AMD isn't expected to intro 32nm chips until 2011. This definitely sets up Intel again to grab more market share from AMD; anyone remember the days when AMD was the gamer's choice?
Subscribe to:
Posts (Atom)
