All you ever heard about graphics anti-aliasing is wrong!

Juan Vuletich has been working for some time on Morphic 3, a research and development project aimed at building the next standard in 2D user interfaces. One of his aims is to do mathematically proved alias-free rendering. In order to achieve these objectives, he has been experimenting with several techniques and design features, some which are new and others are not, but have never been consistently applied to a 2D GUI.

Juan’s contention is that, although the theory behind sampling is about 80 years old, existing graphics software completely ignore the theory and that his quest for higher quality results has resulted in the idea of applying the Sampling Theory which allows for mathematically proved alias free rendering. He informed the squeak-dev mailing list of his latest post which makes the startling claim that “I developed new drawing algorithms that give better results than those in Cairo, AGG, etc.” and he has created some examples showing some of the problems with existing algorithms, and how his approach improves these issues.

Juan is preparing the algorithms for release, which will involve him publishing it in a  journal or as a Ph.D. thesis, and securing it for free use by either putting the code in the public domain or releasing it under the MIT licence.

Juan’s work on Morphic 3 is supported by ESUG’s Support Your Project programme.