The same consortium behind standards like OpenGL, Khronos Group
, has announced the formation of the Compute Working Group to implement Apple's proposed Open Computing Language. OpenCL will incorporate "royalty-free, open standards for programming heterogeneous data and task parallel computing across GPUs and CPUs." Apple announced at WWDC last week that OpenCL support will be an important part
of the architectural foundation being laid in Snow Leopard. The idea behind OpenCL is to create a standard—and easily accessible—way for developers to tap the latent power of all the numerous cores that reside in modern PCs' CPU and GPU. To date, programming for multicore processors has been a tough task, and one that traditional programming techniques don't readily address very well. OpenCL should, in theory, "enable any application to tap into the vast gigaflops of GPU and CPU resources through an approachable C-based language."
"Highly-accelerated parallel computation across GPUs and CPUs is essential to many emerging rich consumer applications that will transform the computing experience of diverse users,” said Khronos Group president Neil Trevett in a statement. Being able to tap multiple cores in a straightforward and useful way has been a sort of holy grail for some time. With Intel (and others) stuffing more and more cores into their processors, Apple believes OpenCL will be an important piece in the multicore puzzle.
Besides Apple, the initial working group is comprised of a variety of industry players (and one university), including: 3Dlabs, AMD, ARM, Codeplay, Ericsson, Freescale, Graphic Remedy, IBM, Imagination Technologies, Intel, Nokia, NVIDIA, Motorola, QNX, Qualcomm, Samsung, Seaweed, TI, and Umeå University.