|
Processor milestones: Then to now ![]()
The Intel 8088 was the first chip to be used in PCs. This was not the best available CPU then. Intel's 8086 was, in fact, more powerful and had been released earlier. However, the 8088 was picked for economic reasons since its 8-bit data bus required less costly motherboards than the 16-bit 8086. Not only that, PCs then were designed for 8-bit use.
A pivotal sale of the 8088 to IBM's new personal computer division made the chip a success, propelling Intel into the ranks of the Fortune 500. Together with the Intel 8086, the 8088 would later introduce the x86 architecture to the world, which was considered to be the most commercially successful instruction set architecture in the history of personal computing. It has been implemented in processors from Intel to Cyrix, AMD, VIA, and many others since then. ![]()
This was a seminal year when Intel raised the PC to a new level with the release of its flagship Pentium processor. The first Pentium processor ran at a rapid 60MHz, had 3.3 million transistors, and performed 100 million instructions per second (MIPS). It was to be the first of four types of Pentium processors developed by Intel.
In 1997, the Pentium 2 emerged and was designed to run from 233MHz to 450MHz. About the same time, the Celeron processor was also unveiled. While both processors were identical, the Celeron had a smaller cache and slower bus speed that was targeted at budget/value PCs. By 1999, AMD stepped up with its Athlon processor at the time that Intel launched its 600MHz Pentium 3 processor. The Athlon showed superior in every benchmark against the Pentium III, running programs as well and with twice the bus speed of the Pentium 3. The Athlon also utilized a dual bus even though the clock speed (MHz rating) was the same as the Pentium III and Celeron. AMD, like Intel, also rolled out a lower-cost processor, the AMD Duron. In comparison, the Duron was cheaper and had a 200MHz bus, while the Celeron ran at a mere 66MHz. With the success of Athlon, AMD changed its processor architecture and in 2001 introduced a new line of Athlon processors: The Athlon XP. While still an Athlon processor, the major difference with the Athlon XP was that it did not use conventional MHz rating to depict its speed as AMD reckoned a MHz rating would undermine its true performance. It was half as cheap as the Pentium 4 which debuted a year earlier accompanied by a low-end Celeron processor (sometimes referred to as Celeron 4). ![]()
When it became apparent that keeping up with Moore's Law was becoming a bigger challenge due to the physical limitations of the technology, the industry looked for other ways to improve performance. One architecture rose to the forefront--the multi-core processor.
Simply put, a multi-core processor is a single chip containing more than one microprocessor core, which multiplies the potential performance with the number of cores. Some components, such as bus interface and second level cache, may be shared between cores. Because the cores are physically very close, they interface at much faster clock rates compared to discrete multiprocessor systems, improving overall system performance. The dual-core PC era began in April 2005 when Intel released the Pentium Extreme Edition 840 processor, a 3.2GHz, 90nm chip. Following this pricey Extreme Edition chip was "dual-core for the mainstream" in the form of the Pentium D 800 series. Intel's second-generation dual-core chips, the Pentium D 900 series, were released in early 2006 and saw the company move to the 65-nanometer process. During this time, AMD was garnering rave reviews with its dual-core, 90nm Athlon 64 X2 line, thanks in large part to its integrated memory controller. While Intel's chips still needed to shuttle data via the slower front-side bus to communicate with system memory, AMD's chips featured a memory controller on the die that operated at the same frequency as the processor itself. It wasn't until Intel introduced its Core technology in July 2006 that users were able to fully grasp the benefits of dual-core processing. The architecture behind Core 2 Duo chips not only brought about leaps in performance but also improved efficiency. While they didn't include an on-die memory controller, the chips did introduce a host of architectural improvements, most significantly a unified cache structure. Instead of a separate allotment of L2 cache dedicated to each of the two cores as was the case with Intel's previous dual-core chips and AMD's Athlon 64 X2 CPUs, Core 2 Duo chips had one large pool to pull from, which provides greater flexibility in allowing each core to access more cache as needed.
|
More Resources From CNET Asia
The multicore era is upon us
Power demands and heat concerns meant that AMD and Intel couldn't simply keep ramping up clock speeds. The answer: More processing cores in one chip.
AMD demos "Shanghai"
AMD recently held a demonstration of its new quad-core, 45nm Opteron processor, Shanghai, to show a live demo of the working product in front of the assembled press.
Intel Core i7-965 Extreme Edition review
Speed never comes cheap, however, and if you're willing to spend for it now, you'll find yourself in possession of the fastest CPU on the market.
Developers get boost for multicore apps
As Intel prepares for multiple cores in every machine, it is bringing new tools to the table for software developers.
Sponsored links
Make Your Brand Sing Out!
The Music Matters Advertising Forum in association with Media, Singapore, Dec 8
IPTV Forum Asia
Enhancing IP Networks for Optimal Video Service Delivery
Samsung Camera, Two LCDs.
Twice the fun of any camera. Dual LCD makes it easy.
NEW OLYMPUS PEN E-P1 Micro Four Thirds
Shoot SLR-quality images & HD videos without the bulk
CNET Asia is now on Facebook!
Be part of the most happening tech community in Asia on Facebook
CNET Asia TV
Watch gadget reviews, quick tips, movie trailers and more for FREE.




