Intel Launching New 45nm CPU Lineup Monday

Intel Corp. plans to roll out its newest generation of processors Monday, flexing its manufacturing muscle with a sophisticated new process that crams up to 40 percent more transistors onto the company's chips. The world's largest semiconductor company expects to start shipping 16 new microprocessors — which also boast inventive new materials to stanch electricity loss — for use in servers and high-end gaming PCs. The most complex chips being launched Monday have 820 million transistors, compared with the 582 million transistors on the same chips built using the current standard technology. Intel's first chips, introduced in the early 1970s, had just 2,300 transistors.

Advances in chip technology occur as smaller and smaller lines are etched onto the chips. Intel's new chips shrink the width of those lines to an average of 45 nanometers, or 45 billionths of a meter, compared to 65 nanometers on the previous generation of chips. The smaller circuitry allows Intel to squeeze more transistors — the building blocks of computer chips — onto the same slice of silicon. That accelerates performance and drives down manufacturing costs. The transistors on the new chips are so small that more than 30 million of them can fit onto the head of a pin. Performance zooms ahead with smaller transistors because more of them are available, they twitch faster to process data and less energy is required to power them.

View: Full Story @ MSNBC

Report a problem with article
Previous Story

Vista's Biggest Problem Remains Windows XP, Survey Says

Next Story

Xbox Live Collects Eight Million Users in Five Years

15 Comments

Commenting is disabled on this article.

My wife and I are engineers at Intel in Chandler, AZ and to this day, the 45nm process still blows my mind away. It's amazing what they have done with the gate stack and how the Oregon R&D folks managed to get everything to work in high volume.

Most people don't realize that there are several prototypes of the latest type of transistors in labs all over the world but the trick is how do you get it to work in high volumen manufacturing... this is what Intel is very good at doing!

SSE 4.1 and architectural changes make these chips noticably faster then previous Core 2 labelled processors.

Look out for more processors based on these new processes and designs (....!....) in future as buyers will want to be aware of the extra facility afforded by these newer chips. Sometimes these kind of changes dont really get noticed by the end user as the brand labelling is not changed much (say like Northwood and Prescott used to be, or 25mhz FSB vs 40 :))

This is probably deliberate on the part of companies like Intel but I can't see why it would happen to be honest Maybe they are banking on the interestingness of their future products?

So.... any pricecuts associated with the new models introduction into the market?

That would be sweet....

Mr.Ed

So how do they know how to build these things, surely no mind would be able to know where 820 million transistors are, and the patterns they need to follow...

Intel probably has their own high level language for describing the layout, which is then converted into the actual CMOS design by a computer. That, or they provide actual schematics for it, which the computer then physically lays out. I'm not sure what they do on that kind of scale....I mean, obviously on the very low end, you can design them by hand (We had to design a CMOS adder in an EE class), but for anything with more than like 10 FETs (Transistors), no one does that.

Then, there's about 500 steps of various photoetches, and depositing layers of interconnects and laser boring holes for interconnects, ion implantation, etc.

It's a pretty complex procedure.

you have 30+ years experience design engineers, 30+ years experience process design engineers, 30+ years experience layout design engineers. They create block diagram models to break down each parts of the processor then connect together using lines.

Anyways, just think of it this way, layers after layers of masked patterns are placed on top of each other like hamburgers and connected togther on the z-plane. How do they make these things so small? Look up plasma etching a.k.a. dry etch and MOSFET.