The new version integrates shaders with atomic counters as well as load, store and atomic read-modify-write operations to a single level of a texture. Additionally, there is support for GPU-tessellated geometry, the modification of an arbitrary subset of a compressed texture without having to re-download the whole texture to the GPU for significant performance improvements, and the ability to pack multiple 8 and 16 bit values into a single 32-bit value for efficient shader processing.
“OpenGL 4.2 has integrated feedback from developers that are shipping significant OpenGL-based applications and games, making for a faster, more capable API which will continue to evolve to meet market needs,” said Barthold Lichtenbelt, working group chair of the OpenGL ARB and director of Tegra graphics at Nvidia.
Both AMD and Nvidia said that their products will support OpenGL 4.2 in the immediate future.