Abstract

A method and computer graphics system capable of super-sampling and performing real-time convolution are disclosed. In one embodiment, the computer graphics system may comprise a graphics processor, a sample buffer, and a sample-to-pixel calculation unit. The graphics processor may be configured to generate a plurality of samples. The sample buffer, which is coupled to the graphics processor, may be configured to store the samples. The sample-to-pixel calculation unit is programmable to generate a first subset of pixels by filtering using the rendered samples and a second subset of the output pixels by interpolating using the first subset of pixels and/or the rendered samples. By interpolating a subset of the output pixels, the graphics system may be able to operate at higher resolutions and/or refresh rates since filtering of the samples is computationally intensive. Any ratio of the number of first subset of pixels to the number of the second subset of pixels may be used, and furthermore, a different ratio may be used for different regions of the display screen. A higher ratio of the number of first subset of pixels to the number of second subset of pixels may be used for regions of the screen where a higher quality image is desired and a lower ration may be used for regions where a lower quality image may be sufficient. For example, a higher quality image may be desired close to and around the foveation point where the eye of an observer is focused.