Intel will also add the Core 2 Duo E7300 at 2.66GHz and US$133 to its entry-level line and to drop the price of its Core 2 Duo E7200 from US$133 to US$113.

With the updated pricing, motherboard makers and channel vendors both expect increased demand in the third quarter, with motherboard makers predicting an average sequential growth of 15% in third-quarter revenues.

Intel’s Penryn family of Core 2 CPUs

As you no doubt know by now, Penryn is the codename for Intel’s upcoming successor to today’s Core 2 CPUs. Penryn represents a wide family of products, ranging from mobile processing to replacing today’s Conroe-based and Kentsfield-based Core 2 CPUs on the desktop. Penryn CPUs are based on Intel’s upcoming 45-nm high k+ metal gate process technology, allowing Intel to cram more transistors into the processor’s die without significantly increasing its size. According to Intel, the new 45-nm process gives them twice the transistor budget, this allows them to add performance enhancing features such as larger L2 caches while still delivering a cost effective die size. Intel’s quad-core variant of Penryn for instance will contain approximately 820 million transistors, while the dual-core variant of Penryn will have a die size of just 107 square millimeters.

Intel has incorporated a number of architectural enhancements into Penryn that are designed to deliver clock-for-clock performance enhancements over today’s Core 2 CPUs at a given clock speed. One key new technology Intel has incorporated into Penryn is their Fast Radix-16 divider. Intel’s Radix-16 divider is a new divider technique providing double the divider speed over previous processors when handling math computations (both floating-point and integer operations): 4-bits processed per cycle in Penryn versus 2-bits per cycle in today’s processors.

Penryn will also support Intel’s SSE4 instructions. This should improve the processor’s performance when dealing with multimedia apps (such as photo imaging, video encoding, etc) and games that have been designed with SSE4 in mind. Incorporating alongside this is Penryn’s new super shuffle engine. This is a 128-bit wide, single-pass shuffle unit that will improve Penryn’s performance with SSE2, SSE3, and SSE4 instructions that have shuffle-like operations. Penryn processors can perform these operations in a single cycle.

Penryn also features Intel’s enhanced virtualization technology. Intel claims machine transition times have been improved from 25-75%.

For mobile Penryn processors, Intel has developed their enhanced dynamic acceleration technology for single-threaded apps. The idea is to boost performance for software applications that aren’t multi-threaded. In these apps, the second core is left idling; when this occurs, Intel’s dynamic acceleration technology kicks in, bumping up the clock frequency (beyond the stock CPU speed) on the processor core that is being used while the second core idles. Intel has also added a new power management state for mobile Penryn processors to reduce power consumption when the system is idle.

Larger caches, faster FSB
Penryn processors will feature considerably larger, more associative caches than today’s Core 2 CPUs. Dual-core Penryn CPUs will ship with up to 6MB of L2 cache while quad-core processors will contain up to 12MB of L2 cache.

On the front-side bus, Intel is also expected to crank up the speed. During today’s press briefing, Intel confirmed a 1600MHz FSB speed for Xeon processors, while desktop Penryn processors will be outfitted with a 1333MHz FSB. Overall clock speeds will also be significantly higher than today’s Core 2 CPUs. While Intel wouldn’t outline anything specific, they did confirm that desktop Penryn CPUs will ship at speeds greater than 3GHz at launch.

Intel plans to deliver a total of six Penryn processors including dual and quad-core desktop processors and a dual core mobile processor, all which will be sold Core processor brand name. Four of these processors will be in production by the second half of 2008.

Nehalem: Intel’s next brand new architecture

After Penryn, Intel plans to ship their next-gen architecture codenamed Nehalem, which will begin production in 2008. Not as much was disclosed about Nehalem. Arguably Intel’s most significant announcement was that Nehalem will offer an enhanced version of their Hyper-Threading technology, allowing Nehalem to process up “1-16+ threads” while utilizing just “1-8+” cores. This will allow Nehalem to deliver enhanced performance without dramatically increased power consumption. Intel also disclosed that Nehalem will incorporate an integrated memory controller and a multi-level shared cache architecture. This is similar to AMD’s plans for Barcelona, with the L3 cache shared among the processing cores.

Intel even plans to provide an integrated graphics option of Nehalem, incorporating the graphics engine onto the CPU itself, just as AMD intends to do with their “Fusion” line of processors. With so many similarities, it will be interesting to see how the two stack up against one another.

Final thoughts

Intel’s roadmap appears to be on schedule and with the performance additions inside Penryn, they’ll have a solid successor to today’s Core 2 processors, widely acknowledged by all to be the fastest processors on the market today. During today’s press briefing, Intel disclosed a 20% performance improvement for Penryn in games (this is comparing a Core 2 X6800 to a 3.2GHz Penryn CPU) and 40% improvement in video encoding when SSE4 is used.

How this will compare to AMD’s upcoming Barcelona processors is anyone’s guess at this point, but it appears that the two competing architectures will be quite competitive with one another. Right now it appears AMD may get Barcelona out the door ahead of Intel, but to be honest, we really won’t see volume shipments of either until sometime in 2008. In the meantime, expect more price drops and product introductions of today’s existing architectures.

“It points the way to the near future when Teraflop-capable designs will be commonplace and will reshape what we can all expect from our computers and the Internet at home and in the office.”

One of the breakthroughs achieved with this processor is that this 80-core research chip achieves a teraflop of performance while consuming only 62 watts – less than many single-core processors today, and way less than the ASCI Red Supercomputer built by Intel for the Sandia National Laboratory back in 1996. That computer took up more than 2000 square feet, was powered by nearly 10,000 Pentium Pro processors, and consumed over 500 kilowatts of electricity.

This Intel teraflop research chip achieves is not much larger than the size of a finger nail.

Intel also revealed that this 80-core processor features an innovative tile design in which smaller cores are replicated as “tiles,” making it easier to design a chip with many cores.

The Teraflop chip also features a mesh-like “network-on-a-chip” architecture allowing super high bandwidth communications between the cores, and capable of moving Terabits of data per second inside the chip. The research also investigated methods to power cores on and off independently, so only the ones needed to complete a task are used, providing more energy efficiency.

Obviously, this processor can only be useful if there’s software that can take advantage of it. Programmers are starting to take advantage of multithreading so it will take a while for applications to learn how to use all this computational power.

In the near future, Intel will focus on the addition of 3-D stacked memory to the chip as well as developing more sophisticated research prototypes with many general-purpose Intel Architecture-based cores.