Polygon Rendering and Fill Rates

This data tells an important story. The very first statistic lists how many triangles per second the GeForce256 could render, marking a radical shift from previous graphics card design. The GeForce256 was the original "poly pusher", designed to process as many polygons as it could as quickly as possible. The third statistic is the next most important, as tells us the fill rate of the card. The fill rate determines how many pixels per second the card is able render to the frame buffer.

So why are polygon rendering and fill were rates so important? The answer lies in the way 3D games are created. When developing a 3D application, the programmer works in the theoretical 3D world. The graphics he draws have width, height and most importantly depth. Nevertheless, when these graphics are actually rendered to the frame buffer, they need to be transformed into 2D graphics. This is an exceptionally computationally expensive task, even for the latest of today's cards. It is so time consuming that it is actually impossible to transform an entire screen of pixels from 3D space to 2D space quick enough to produce a good-looking display. This is of course a major issue – after all, if pixels cannot be transformed quick enough to fill the screen, the whole exercise becomes pointless. To resolve this issue, developers began using triangles.

A triangle is the simplest geometrical shape that can exist in 3D space, because it only requires 3 points to describe its size, position and orientation. Now instead of transforming every single pixel, large areas of the 3D world the programmer has created are divided up into triangles. This gives an enormous speed increase over transforming individual pixels. Imagine a flat wall in a 3D game comprised of 4,000 pixels. This wall can then be represented by 2 triangles positioned adjacent to each other to form a square (called a “quad”), meaning that only 6 points have to be transformed (or less, depending on the tricks you use!) instead of 4,000 pixels. This is why the number of triangles a graphics card can render is important.

Just drawing triangles is not enough, however. Once the triangles have been transformed, they need to be filled with colored pixels that resemble the object they are representing, called “texture mapping”. In the wall example, we would take a bitmap file containing a photo of a brick wall, then texture map this bitmap onto the two triangles we created. This is exactly how 3D games are created. The number of pixels a graphics card can “fill in” is called the fill rate – the higher the fill rate, the better the quality of textures used enhancing the overall 3D experience.

Returning to the press release and most telling of all is the last line. The GeForce256 was the first consumer card with a hardware transformation and lighting pipeline designed to work with a 3D API. To understand why, we need to understand the role of OpenGL and DirectX in the 3D graphics world.

If you find CertiGuide.com useful, please consider making a small Paypal donation to help the site, using one of the buttons below. You can also donate a custom amount using the far right button (not less than \$1 please, or PayPal gets most/all of your money!) In lieu of a larger donation, you may wish to consider buying an inexpensive PDF equivalent of the CertiGuide to A+ (A+ 4 Real) from StudyExam4Less.com. Thanks for your support!
 Donate \$2
 Donate \$5
 Donate \$10
 Donate \$20
 Donate \$30
 Donate: \$