The 45nm SSE4 capable processor

Today, Intel officially lifted the embargo on it's latest Core 2 Extreme Processor. This new processor named QX9650 is a quad core processor build on 45nm manufacturing process technology.

Clocked at 3GHz (1333MHz FSB), this new processor comes with a larger 12MB L2 Cache and the newer SSE4 instructions set. The QX9650 is also lead-free.

45nm

So how small is 45nm ? In a comparison with real life objects, we can see that 45nm is why too tiny, even smaller than a bacteria of 2K nm in size. The original transistor built by Bell Labs in 1947 could be held in your hand, while hundreds of Intel’s new 45nm transistors can fit on the surface of a single red blood cell.  You could fit more than 2,000 45nm transistor gates across the width of a human hair. That is how small 45nm is.

45nm Size Comparison
o A nail = 20 million nm
o A human hair = 90,000nm
o Ragweed pollen = 20,000nm
o Bacteria = 2,000nm
o Intel 45nm transistor = 45nm
o Rhinovirus = 20nm
o Silicon atom = 0.24nm

SSE4

From the early days of Floating point extensions in 8086 to the SIMD and Intel MMX technology. All these technologies helped to achieve increase parallelism in the P5 microarchitecture through the use of special instructions that operate on multiple pieces of data simultaneously. The MMX technology delivered better performance in media application such as graphics, video streaming and gaming by loading loading it into MMX technology registers.

In P6, Intel introduced SSE (Streaming SIMD Extensions) designed for P3. SSE extended MMX and allowed SIMD computations on 4 packed Single precision FP using 128 bits registers. With P4, Intel introduced SSE2 (and MMX). It allowed more simultaneous computation in parallel enabling support of 128 bit integers and packed double precision FP data types. In SSE2, 144 new instructions were added to increase performance across applications. The SSE2 help enhanced the performance of Mpeg2, mp3 and 3D graphics.

SSE3 in the P4 added 13 more instructions to improve thread synchronisation and x87 Floating point math capability.

In Penryn, SSE4 will offer dozens of new innovative instructions in two major categories:
• SSE4 Vectorizing Compiler and Media Accelerators
• SSE4 Efficient Accelerated String and Text Processing

Vectorizing Compiler and Media Accelerators

The new compiler vectorization primitives include improved integer
and floating-point operations, support for packed DWORD and QWORD
operations, new single precision FP operations, fast register operations, performance-optimized memory operations, and more.
Applications that will benefit include those involving image processing, graphics, video processing, 2-D/3-D generation, multimedia, gaming, memory-intensive workloads, HPC workloads, and more.

Efficient Accelerated String and Text Processing

SSE4 provides new string and text processing instructions that will
enhance the performance of string and text processing operations,
resulting in a performance boost for a wide variety of data processing,
search, and other text-based applications. These new instructions
will include advanced packed string comparison instructions
that can perform multiple compare and search operations in a single
instruction. In general, each of these new instructions has a rich set
of innovative string processing capabilities to replace operations
in which several instructions were required to deliver the same
functionality in the previous ISA.

Applications that will benefit include those involving databases,
text search, virus scanning, string process libraries like ZLIB, Token
parsing/recognizing applications like compilers, and state machine oriented applications.

Here are some of the specifications of this processor:

• 3.0GHz 1333MHz FSB Quad Core

• 45nm high K metal gate transistor technology

• 12MB L2 cache

• SSE4 instructions

• 820M transistors

• 214mm2 die size

• LGA775 Socket 130 TDP C-step

• Overspeed protection removed

• Supported by all Intel P35 or X38 chipset mainboards (with bios upgrades).

In the subsequent pages, we take a look at some of the benchmarks, comparing the new QX9650 with the older brother QX6700 (o/c to 3.0GHz at 333x9) clock to clock at 3GHz.

Next >>>