Graphics Beginners' Guide, Part 2: Graphics Technology

Graphics Processor Architecture: Technology

Manufacturing Process

This term refers to the structural size and precision of the manufacturing process used to create an integrated circuit. The smaller the size, the smaller and more advanced the manufacturing process. For example, the 0.18 µm process produces processors that are larger, and not as efficient, as the 0.13 µm process, because smaller transistors usually require less voltage to operate properly. Less voltage means less thermal resistance, which causes less heat dissipation. The smaller process also means that the distances between operational units are also shorter and the data transmissions take less time. This shortening of distances, lowered voltages and other advantages allow smaller process products to have higher clock frequency speeds.

To further complicate matters, the terms 'micron' and 'nanometer' are both used to describe process sizes. Since 1 nanometer equals 0.001 microns, the term "0.09-micron process" is exactly the same as the term "90-nanometer process". As stated before, smaller processes are generally associated with higher clock speeds. For example, when comparing a card with a processor made on a 0.18-micron process, and one on a 0.09- micron (90-nanometer) process, it is reasonable to assume that the card with a 0.09-micron processor will have higher clock speeds.

Graphics Processor Clock Speed

Graphics processor clock speed is measured in Megahertz (MHz), which can be described as 'millions of cycles per second'.

The clock speed has a direct effect on the performance of the graphics processor. The faster it runs, the more work it does per second. In the first example, let's consider the Nvidia GeForce 6600 and 6600 GT: The 6600 GT has a graphics processor speed of 500 MHz, but the regular 6600 has a clock speed of 400 MHz. Because the processor is technically identical, the 20% clock speed boost of the 6600 GT translates into higher performance.

Clock speed isn't everything, however. You need to keep in mind that the architecture has a large impact on performance as well. In a second example, let's consider the GeForce 6600 GT and GeForce 6800 GT. The 6600 GT has a graphics processor clock speed of 500 MHz, but the 6800 GT runs at only 350 MHz. This does not tell the whole story, as the 6800 GT is a 16-pipeline architecture, while the 6600 GT has eight pipelines. In a sense, the 6800 GT with 16 pipelines at 350 MHz would offer roughly the same performance with eight pipelines at twice the speed (700 MHz). This is a very simplified comparison, but it can be used as a reasonable performance indicator.