Moore's law stems from an observation Intel co-founder Gordon Moore made back in 1965: The number of transistors on a 1-inch (2.5-centimeter) silicon chip tends to double every couple of years. While there's no universal law that dictates this must be so, technology companies like Intel have spent countless hours and billions of dollars on research and development to keep pace with Moore's law.
But Intel's strategy goes beyond finding ways to shrink components to tinier scales in order to boost power. The company has what it calls a tick-tock strategy. It develops chip technologies in two phases. The tick phase involves finding a way to shrink elements down to a smaller size. The tock phase is all about arranging the shrunken elements in the most efficient configuration to increase efficiency.
Intel's Sandy Bridge chip is an example of a tock technology. The previous tock chip, codenamed Nehalem, arranged 45-nanometer transistors in a way that allowed data multithreading, looping and branching, which made it a more powerful processor than the preceding 45-nanometer Penryn microprocessors.
After Nehalem came the next tick: the Westmere family of microprocessors. While they have the same configuration as the Nehalem family of chips, Intel engineered Westmere's components down to 32 nanometers. Following Westmere is the tock of Sandy Bridge.
The Sandy Bridge family of chips introduces some new features into Intel's bag of tricks. One of the most talked about is Intel's decision to dedicate part of the microprocessor to handling graphics processing -- a task often handled by a dedicated graphics processor. You might think Intel is firing a warning shot across the bow of companies that produce graphics processing units (GPUs).
Before we learn more about Sandy Bridge, it's important to understand how things work at this incredibly small scale.