AMD's 64-bit technology includes the x86-64-bit architecture, which is a 64-bit extension of the x86-architecture. The x86-64 architecture supports 64-bit computing while maintaining full-speed compatibility with legacy 16-bit and 32-bit applications. It provides recompiled 64-bit applications with 64 bit addressing, 64-bit data types, 8 addtional general purpose registers and 8 additional SIMD registers. The need for a 64-bit x86 architecture is driven by applications that address large amounts of physical and virtual memory, such as high-performance servers, data base systems, and CAD tools. Many x86 server and workstation users are now facing the dilemma of how to transition to 64-bit computing and yet maintain compatibility with their existing code base, stick with existing tools and keep porting and development costs under control. By extending the x86 architecture to 64-bits, rather than forcing users to embark on yet another instruction set/architecture adventure, users can maintain transition to 64-bits at their own pace and developers protect their investment in their existing tool chain and knowledge base. My talk presents an overview of AMD's 64-bit x86 extensions, its design goals and tradeoffs.

Taipei, Taiwan, April 3 2003 - AMD has just announced their latest 333 FSB CPUs, the Barton 2500+, 2800+, 2900+ and 3000+, as well as the 400 MHz FSB Barton 3000+ and 3200+. ABIT's ABIT Engineered AMD motherboard line is ready for these industry leading high-performance CPUs. The NF7 Series has already been supporting AMD's new 3000+ Barton processors feature 640KB of total on-chip cache, an increase of nearly 70 percent, which leads the industry for desktop PC processors. More cache allows additional information to be stored closer to the processor, helping to speed up the retrieval of information that benefits applications such as those used in editing video, creating digital media effects and encoding music, as well as typical office productivity and gaming applications.

Later this month Advanced Micro Devices will launch its 64-bit microprocessor in a bold move to try and succeed in a market where its much larger rival Intel is struggling.

AMD will officially launch its Opteron 64-bit microprocessor on April 22nd and will announce the first computer manufacturers that will introduce Opteron-based systems. These first customers could include industry leader International Business Machines and Sun Microsystems, both of which said they have been evaluating the chips.

The launch for AMD is a plucky one but it is also a critical bid to turn round its fortunes as it reels from heavy losses. Sales of its PC microprocessors have suffered from weak PC demand and stiff price competition from Intel. In its most recent quarter it reported a loss of $854m, which included a large restructuring cost as it attempts to reduce operating expenses.

Chip maker AMD claims to have become the first in the semiconductor industry to achieve critical research milestones for next-generation transistor development.

In lab work to be fully unveiled in June, AMD researchers say they have created and demonstrated a high-performance transistor which is up to 30 per cent faster than the best available today.

The company's Fully Depleted Silicon-on-Insulator technology is expected to play a crucial role in semiconductor manufacturing in the second half of this decade.

In related research, AMD has also demonstrated a strained silicon transistor achieving 20-25 per cent higher performance than conventional strained silicon devices through the successful use of metal gates.

VIA have been on the forefront of HTPC design and recently the EPIA V series was released. Using the Mini-ITX form factor, the EPIA V9000 is the cost effective brother of the M9000, released last September. The Mini-ITX factor is quoted as being the worlds smallest to date, measuring in at a miniscule 17cm squared (Seen next to an Asus K7M Motherboard below). The tiny boards haven't gone unnoticed by the modding croud either, there are some very crazy and wacky case mods out there for these motherboards!

Before proceeding I feel that it is necessary to point out that the EPIA V is NOT designed for gamers or heavy users. It is designed for situations where a low power consumption, quiet device is prudent. It would also be suited for low level productivity and Internet tasks (surfing or acting as a gateway/router).



Board Specs:

Motheboard Picture	EPIA V9000 Specifications
	Processor:	Cyrix 3 933Mhz
	Chipset:	VIA PLE133 North Bridge & VT8231 South Bridge
	Memory:	2x PC133 DIMM (1Gb Max)
	Video:	Integrated Trident Chipset
	Expansion:	1x PCI
	Onboard Peripherals	2x UDMA 66/100 IDE 1x FDD 1x 10/100BaseT LAN 2x USB 1.1 ports 1x Parallel 2x PS2 1x Line In/1x Out/1x Mic 1x VGA 1x S-Video Out 1x Composite (RCA)Out
	Misc.	WOL/WOR SIR CIR Chassis Intrusion

#1: The Swiss Spaghetti Harvest
In 1957 the respected BBC news show Panorama announced that thanks to a very mild winter and the virtual elimination of the dreaded spaghetti weevil, Swiss farmers were enjoying a bumper spaghetti crop. It accompanied this announcement with footage of Swiss peasants pulling strands of spaghetti down from trees. Huge numbers of viewers were taken in, and many called up wanting to know how they could grow their own spaghetti trees. To this question, the BBC diplomatically replied that they should "place a sprig of spaghetti in a tin of tomato sauce and hope for the best." Check out the actual broadcast archived on the BBC's website (You need the RealVideo player installed to see it, and it usually loads very slowly).

Opteron (WS/Server)
• "SledgeHammer" : 1MB L2 Cache, 0.13u SOI, Q2 2003
• "Athens" : 1MB L2 Cache, 0.09u SOI, H1 2004
Desktop Athlon 64
• "ClawHammer" : 1MB L2 Cache, 0.13u SOI, 2800+/3100+/3400+(1.6/1.8/2Ghz) Q2 2003
  
  3700+ (2.2Ghz) Q3 2003
  4000+ (2.4Ghz) Q4 2003
• "San Diego" : 1MB L2 Cache, 0.09u SOI, H1 2004
• "Paris" : 256KB L2 Cache, 0.13u SOI, 3200+(2Ghz?) Q3 2003
  
  3500+ Q4 2003
• "Victoria" : 256KB L2 Cache, 0.09u SOI, H1 2004
Mobile Athlon 64
• "ClawHammer" : 1MB L2 Cache, 0.13u SOI, H2 2003
• "Odessa" : 1MB L2 Cache, 0.09u SOI, H1 2004
Desktop Athlon XP
• "Barton" : 400Mhz FSB, 512KB L2 Cache, 0.13, June 2003
• "Thorton" : 333Mhz FSB, 256KB L2 Cache, 0.13, Mid 2003
Mobile Athlon XP-M
• "Barton" : 512KB L2 Cache, 0.13, Now

Roadmap Image

AMD and Fujitsu, which have had a joint venture called Fujitsu AMD Semiconductor Ltd. (FASL) for about a decade, said the new company -- to be named FASL LLC and headquartered in AMD's home turf of Sunnyvale, California -- will begin operations in the third quarter of this year.

FASL LLC will have about 7,000 employees and a net book value of more than $2.5 billion, they said, declining to further comment on the financial aspects of the deal.

AMD will own 60 percent and Fujitsu the remaining 40 percent.

"We expect the new company to be the strong number two player in the flash market," Hector Ruiz, AMD chief executive, said in a conference call. "Integrated operations will enable economies of scale and include improved profitability" of the flash business.

• Socket A with FSB200/266/333 MHz
• Athlon XP(Barton CPU Ready!), Athlon, Duron Processor
• NVIDIA nForce2 IGP/MCP (Integrated Graphics Engine & 10/100Mbps Ethernet LAN
• AGP 8X
• Dual-Channel DDR400/333
• (Memory capacity up to 2 GB using Dual Channels, 3 GB using Single Channel)
• 6 ports USB 2.0 (4 by cable)
• AC’97 6 channel Audio
• 3 PCI Slots
• 2 ATA133 IDE

Picture of Albatron KM18G

• 0.13 micron process
• DDR not DDR2 memory because of the lower price of memory and PCB, more stability and availability in high volumes
• 256-bit DDR1 memory interface and there will be cards with 256MB of memory
• 400Mhz Memory and probably same core clock as NV30
• Same features set
• Twice or more faster in raw shader power for 1.4 and 2.0 pixel shaders
• Shows from 1.5 up to 2.0 times performance of NV30 depending on the tasks and the final frequency specs
• Definitely faster than RADEON 9800 PRO
• 400Mhz RAMDAC

System for detecting over-clocking uses a reference signal thereafter preventing over-clocking by reducing clock rate. An over-clock deterrent mechanism of a chipset which comprises an over-clock detection circuit for detecting over-clocking of a system (processor) clock signal based on comparison of ratio of the system (processor) clock signal which is likely to be over-clocked and a fixed, stable reference clock signal which is highly unlikely to be over-clocked, and an over-clock prevention (thwarting) circuit for deterring such an over-clocking by either disabling operations of a computer system or significantly undermining key operations of a computer system

The states of Massachusetts and Texas are preparing to consider bills that apparently are intended to extend the national Digital Millennium Copyright Act. The bills are obviously related to each other somehow, since they are textually similar. Here is one example of the far-reaching harmful effects of these bills. Both bills would flatly ban the possession, sale, or use of technologies that "conceal from a communication service provider ... the existence or place of origin or destination of any communication". Your ISP is a communcation service provider, so anything that concealed the origin or destination of any communication from your ISP would be illegal -- with no exceptions