PRESCOTT INFO UP ON INTEL.COM!
http://www.intel.com/labs/features/mi02031.htm?iid=ihc+cmt_fot_miniaturization&
Some of this is known, but some interesting bits...
Doing More, Faster
As processor components continue to get smaller, the challenges to improving performance grow. For instance, you either engineer the processor to handle more instructions per clock cycle (doing more work per cycle), increase frequency (the time it takes for a cycle to complete itself), or perform some combination of both.
Scaling theory — shrinking the dimensions and energy required by a processor's components — has been the primary performance-enhancing strategy for over 30 years. The result has been greater transistor density, faster transistors, and higher performance.
When Making it Smaller Is no Longer Enough
The challenge today is that scaling improvements have slowed from 30 percent to 15 percent. Intel is seeking new solutions, especially in architecture. These solutions have included:
Creating shorter interconnects (the layers of wiring used to connect a processor's transistors to each other).
Making more-efficient uses of memory structures with on-die caches instead of relying on logic alone.
Prescott, the next-generation IA-32 processor, uses these and other ideas to overcome many of the current issues in scaling. Prescott's innovations include:
An updated Intel NetBurst Microarchitecture yielding higher frequency and raw performance. This microarchitecture includes larger on-die caches. For instance, a 1MB Level 2 Advanced Transfer Cache allows the transfer of more data on each core clock, delivering a much higher data throughput channel. This is especially important in a HT Technology-enabled processor like Prescott. In these cases, two threads are running on one processor and consequently need more cache space.
Improved Hyper-Threading Technology, including important enhancements to this innovation that make a single processor act like multiple processors to the operating system.
13 new processor instructions designed to improve performance for special application areas such as media and gaming. These instructions are grouped into five areas: floating point to integer conversions, complex arithmetic, video encoding, SIMD floating point using AOS format and thread synchronization.
Better clock distribution up to four times better than the previous desktop platform based on the Intel NetBurst architecture. This in part helps Prescott scale to the 4-5GHz range.
A new 90nm process using strained silicon technology that enables faster gate length scaling as well as faster electron flow through the gates.
Seven layers of low-k copper interconnect, featuring a new carbon-doped oxide dielectric material that reduces wire-to-wire capacitance, enabling Intel to increase signal speed inside the chip and reduce chip power consumption.
AND at the end:
In summary, Prescott represents advancement in many ways. Furthermore, there are several new ideas Intel is exploring for the next generation to allow processors to reach 15-20 GHz by 2010.
News 
Market Data 
Discover 
AI
