Introduction

This is a course project for 18645 How to Write Fast Code.

SSAA

Supersampling anti-aliasing (SSAA) is a spatial anti-aliasing method to remove aliasing (jagged and pixelated edges, colloquially known as "jaggies") from images rendered in computer games or other computer programs that generate imagery. Unlike real-world objects, which have continuous smooth curves and lines, a computer screen shows the viewer a large number of small squares. These pixels all have the same size, and each one has a single color. A line can only be shown as a collection of pixels, and therefore appears jagged unless it is perfectly horizontal or vertical. The aim of supersampling is to reduce this effect. Color samples are taken at several instances inside the pixel (not just at the center as normal), and an average color value is calculated. This is achieved by rendering the image at a much higher resolution than the one being displayed, then shrinking it to the desired size, using the extra pixels for calculation. The result is a downsampled image with smoother transitions from one line of pixels to another along the edges of objects.

SIMD

Single instruction, multiple data (SIMD) is a type of parallel processing in Flynn's taxonomy. SIMD can be internal (part of the hardware design) and it can be directly accessible through an instruction set architecture (ISA). SIMD describes computers with multiple processing elements that perform the same operation on multiple data points simultaneously.

Such machines exploit data level parallelism, but not concurrency: there are simultaneous (parallel) computations, but each unit performs the exact same instruction at any given moment (just with different data). SIMD is particularly applicable to common tasks such as adjusting the contrast in a digital image or adjusting the volume of digital audio. Most modern CPU designs include SIMD instructions to improve the performance of multimedia use.

Objectives

• Practice design methods for architecture-specified computing kernel
• Practice theoretical peak and bottleneck analysis, and their improving methods.
• Familiar with the use of SIMD instructions

Team & Work division

This project is conducted by a team of 3. (Chang Liu, Wenlin Mao, Mitchell Yang)

I am responsible for the development of the circle drawing core.

Design

SSAA - Draw circle

The SSAA algorithm for circle kernel follows the steps:

1. For each pixel in picture, supersample it to 16x16 supersampled target.
2. Draw circle on supersampled target.