Scan's TekSpek

Our Aim
To provide you with an overview on New And existing technologies, hopefully helping you understand the changes in the technology. Together with the overviews we hope to bring topical issues to light from a series of independent reviewers saving you the time And hassle of fact finding over the web.

We will over time provide you with quality content which you can browse and subscribe to at your leisure.

TekSpek CPUs
Intel’s 45nm technology

Intel’s 45nm technology


Date issued:

This TekSpek discusses Intel’s 45nm process technology. What it is, what it provides, and how it benefits the consumer.

What is a die shrink?

Every once in a while, a processor manufacturer will instigate a die shrink. The shrinking process, popular among semiconductor manufacturers, involves creating near-identical circuitry using a more advanced fabrication process.

In simple terms, it literally means making a processor smaller, whilst increasing performance and, potentially, reducing the power requirement. By moving on over to a smaller, more-efficient process, less current is required to drive the smaller transistors and, ceteris paribus, manufacturing costs are reduced due to more processor dies being manufactured on the same piece of silicon wafer. This in turn will result in increased revenue through greater number of products sold.

What is Intel’s 45nm process technology?

Intel describes its 45nm high-k silicon technology as one of the biggest advancements in fundamental transistor design in 40 years. What makes it so significant? Let’s take a look.

Intel’s hugely popular processor line-up, headed by the Intel Core 2 Duo and Intel Core 2 Quad, has been manufactured on a 65nm process ever since they launched in mid-2006. Now, a little less than two years later, Intel has made the move to a smaller 45nm manufacturing process.

Intel’s 45nm process technology, sometimes referred to as Intel’s high-k metal-gate silicon technology, uses new materials, including hafnium and new metalsAnd what is hafnium? Hafnium (Hf) is a chemical element on the periodic table. It’s silver-gray in color, highly ductile, corrosion resistant, and is chemically most similar to zirconium. Intel engineers discovered that introducing hafnium into silicon chips helps reduce electrical leakage enabling smaller, more energy-efficient and performance-packed proc, to ensure that current leakage doesn’t become a problem when shrinking the process down. Using these new materials, Intel reckons it is able to reduce leakage by a factor of up to 100x when compared to traditional silicon dioxide used in present processors. Keeping the all-important current in check allows Intel to pack nearly double the number of transistors into the same silicon space as its 65nm technology.

To put things into perspective, that’s more than 400 million transistors for dual-core processors and more than 800 million for quad-core. Intel’s 45nm technology provides up to 50 per cent larger L2 cache and, thanks to smaller transistors which require less power, superior energy efficiency to its older 65nm predecessor whilst delivering more than a 20 percent improvement in transistor switching speed.

What does Intel’s 45nm process mean to the consumer?

We’ve established that Intel’s 45nm process is hugely important to Intel, but how does it affect the consumer?

Well, a die shrink is beneficial to the consumer in numerous ways. It ultimately results in a product with less power consumption, lower heat output, increased performance, and lowered prices. Essentially, a processor that’s far more efficient.

Intel first debuted its desktop-based 45nm process, code-named Penryn, with the Core 2 Extreme QX9650. Penryn also introduced a number of other performance enhancements, including SSE4.1 and enhanced virtualization support, too.

Having launched Penryn in late 2007, Intel’s 45nm-based Core 2 Duo and Core 2 Quad desktop processors, codenamed Wolfdale and Yorkfield, respectively, have slowly but surely become readily available.

Wolfdale, Intel’s 45nm-based Core 2 Duo desktop processor, features two processor cores sharing 6 MiB of L2 cache, operating at 2.66 GHz and above. They utilise an FSB of 1333 MHz and includes the SSE4.1 media extensions.

Yorkfield, Intel’s 45nm-based Core 2 Quad desktop processor, features a dual-die, quad-core design with two unified 6 MiB L2 caches resulting in a total 12 MiB of L2 cache (Q9300 has 6MiB of L2 cache, though). They also feature a 1333 MHz FSB and were released in late March 2008.

What’s next for Intel’s die size?

In keeping with the semiconductor’s tick-tock process, Intel’s 45nm process will be carried over to its forthcoming Nehalem microarchitecture. Nehalem introduces an enhanced design with more performance per given frequency. Nehalem, however, is for another day, and another TekSpek.