Intel Ships New 45nm Penryn Chip
Gordon Moore’s Law will remain in effect for the foreseeable future. Intel Corporation’s new 45nm Penryn microprocessor relies on a new recipe that combines the element Hafnium and metal gate technology to increase performance and significantly reduce eco-unfriendly, wasteful electricity leaks.
The challenge for Intel to create a “new generation” of technology every two years — a challenge laid out by Co-Founder Gordon Moore in the 1960s — faces very real physical limitations. In fact, Moore himself has predicted the end of the rhythmic advances (more than once, too). The new process technology and the efficiency and energy-savings that it brings, is the subject of a video demo at the Intel site.
In recognition of the 45nm technology (but also for the innovation that will allow Intel to continue doubling, and doubling and doubling every two years), Penryn has already joined Apple’s iPhone and other game-changing gadgets as a member of the elite group of Time Magazine’s Best Inventions of the Year, and just today, Dean Takahashi said, “Intel is about to hit one out of the park.”
In his Sept., 2007 article in the New York Times, G. Pascal Zachary noted that the hafnium-and-gate innovations are at the heart of Intel’s ability to deliver increasingly speedy chips that won’t, for example, explode into flames. Zachary’s article shines the spotlight on Hafnium and Mark T. Bohr, the Intel physicist who oversaw its introduction into the process, since consumers will mostly be treated to the what-does-it-mean-to-me messaging that accompanies new technology — not the where-did-it-come-from messaging that actually explains the developments.
While consumers might not hear much about the element that replaces silicon oxide as the insulator in their new chips, they’ll soon have a clear idea of how they’ll benefit from using them.
It may be a closely-guarded secret where Intel’s new Penryn chips will end up, but recent HP and Lenovo announcements indicate that pick-up in the industry will be swift. (As always, Mac rumors abound, too.)
Trumpeting energy savings and higher speeds for intensive projects like video encoding and multithreaded gaming, the new chips will soon re-set the industry standard, and of course the industry will be powering on behind the scenes.
Intel is already deep into development on its next generation 22nm microprocessors, and if the speed of that effort is any indication, the technology behind computing is developing at a faster clip than in previous years (Intel says it may already be several months ahead of schedule when compared with earlier generations).
Intel’s new eco-friendly facility in Chandler, Ariz. was built specifically for the new 45nm production. Construction on Fab 32 began in August 2005, and you can observe pretty much the whole process in this video.
Related Stories: IntelMooresLaw
Some photos in video courtesy of: nick wiesner, Fouad Bechwati, Jason Upton and Chris Vick via Creative Commons Attribution-No Derivative Works 2.0 Generic; Windell H. Oskay, Maui Äina, Stevan Sheets, John Tregoning, Roland Tanglo, Taz (sporkist) and Erik Charlton, via Creative Commons Attribution 2.0 Generic; Suzi Duke, Rick Kimpel, Christian Fleschhutm, Maks D., Martin Kliehm and _e.t, via Creative Commons Attribution-Share Alike 2.0 Generic
Host: Intel has reinvented the brain inside your computer — again. It’s code-named Penryn. It uses Intel’s new 45 nanometer technology. It’s cooked up using their new recipe, designed to be faster, smaller, and even greener.
So, what does it mean?
To understand what a big deal it is, we need to go back 60 years to the invention of the first transistor. What does a transistor do? It turns on and off. Big deal. But it was small, smaller than the vacuum tubes it replaced, and if you put a bunch in a row, you could control the on’s and off’s and, voila — a computer!
So, what does it mean?
Jump ahead to the late 60s and to Gordon Moore, Intel co-founder, who predicted that the number of transistors on a computer chip would continue to double every two years. This seemingly offhand remark became a mandate for the industry, and with Intel leading the way and in the face of many who said it couldn’t, it kept doubling and doubling, and doubling, so in 1971 the Intel 4004 processor had 2000 transistors. Today, the Penryn chip can have 800 million. To make this happen, the transistor had to be very, very, very, very small, 45 nanometers small to be as exact, or 45 billionths of a meter. It’s almost impossible. I mean, Intel had to employ new materials like high-k dialectric hafnium-based stuff and metal gates, and I don’t know, some kind of magic that allows them to make these things so small and at the same time work at an incredible speed and — bonus — use less power doing so.
So, what does it mean?
It means that not too long ago a cellphone like this would have needed to be size of this to do what it does and this new breed of chip is bound to make this one seem downright bulky. It means the backbone of information technology, datacenters that handle the vast array of information in banking, manufacturing, stock markets — not to mention your YouTube video of Uncle Phil at the family reunion — this backbone will not only become more powerful but at the same time use less energy doing it.
Less energy means longer battery life for your laptop, your PDA, your robotic vacuum — anything that relies on a microprocessor — and these are just the things we’ll see right now. This tiny technology is so advanced that even Intel can’t predict how the world will take advantage of these processors to make mobile workspaces, advance telecommuting options and other stuff that people haven’t even dreamed off yet.
The technology that we rely on to keep us moving is getting faster, smaller, and greener. That’s what it means.
For more information on Intel IT Best Practices, please visit intel.com/IT
Posted in: CES Las Vegas - Consumer Electronics Show, Connected Social Media, Corporate, Information Technology, InfoWorld, Intel, Intel CES, Intel IT, Intel Moore's Law, Intel vPro, Intel-OpenPort, IT@Intel