Supersampling

    ReddIt
    Twitter
    Facebook
    Pinterest

    Supersampling, abbreviated SSAA (Supersampling Anti-Aliasing), is an advanced method of spatial anti-aliasing. It’s commonly used to resolve jagged and pixelated edges known as aliasing from frames rendered in computer programs, such as games. Supersampling is also known as “pure” anti-aliasing.

    How does Supersampling work?

    Supersampling calculates the average color value of groups of pixels running at a higher resolution and downsamples (or downscales) each group into a single pixel to the native resolution of the monitor. This produces much smoother and more visually appealing images.

    In computer games, you will often find the ratio of calculated to displayed pixels as a number (4xSSAA would mean there’s an average of 4 calculated pixels per group). There are two main methods – Ordered Grid (OGSSAA) and Sparse Grid (SGSSAA). Ordered Grid will sample in each square of the grid, while Sparse Grid only samples in some of the squares. Depending on the game, one can be better than the other.

    Advantages of Supersampling (compared to MSAA and PPAA)

    When compared to other anti-aliasing methods such as multisample anti-aliasing (MSAA) and post-processing anti-aliasing (PPAA), SSAA provides the best image quality.

    Disadvantages of Supersampling

    Even though it has the best possible image quality, SSAA is very slow and requires a lot of power to run well. For example, if you had 8xSSAA on a 1920×1080 monitor, the pixel rendering stage actually needs to calculate the same number of pixels as if you were working with a resolution of 7680×2160. Which is a resource-intensive task that negative effects performance.

    « Back to Definition Index