With PCI, texture maps are loaded from the hard drive to system memory, processed by the CPU and then loaded into the framebuffer of the graphics card.

Photo courtesy Intel Corporation

PCI Graphics Rendering: Wasting RAM

Speed is not the only area where AGP has bested its predecessor. It also streamlines the process of rendering graphics by using system memory more efficiently.

Any 3-D graphic you see on your computer is built by a texture map. Texture maps are like wrapping paper. Your computer takes a flat, 2-D image and wraps it around a set of parameters dictated by the graphics card to create the appearance of a 3-D image. Think of this as wrapping an invisible box with wrapping paper to show the size of the box. It is important to understand this because the creation and storage of texture maps is the main thing that drains the memory from both the graphics card and the system overall.

With a PCI-based graphics card, every texture map has to be stored twice. First, the texture map is loaded from the hard drive to the system memory (RAM) until it has to be used. Once it is needed, it is pulled from memory and sent to the CPU to be processed. Once processed, it is sent through the PCI bus to the graphics card, where it is stored again in the card's framebuffer. The framebuffer is where the graphics card holds the image in storage once it has been rendered so that it can be refreshed every time it is needed. All of this storing and sending between the system and the card is very draining to the overall performance of the computer.