An executive from Advanced Micro Devices said that while both ATI, graphics product group of AMD, as well as Nvidia Corp. have been talking about general purpose computing on graphics processing units (GPGPU) for years, the capability will only become popular sometime after 2010, when more advanced compilers become available and after Microsoft releases its DirectX 11 in 2009.

Even though graphics processing units (GPUs) pack much more computing power than traditional central processing units (CPUs), in order to take advantage that performance, software must be specifically tailored for GPUs. As a result, only some programs can benefit from increased processing horse power of GPUs now due to the virtual absence of tools that help software developers to create programs that rely on GPUs.

However, this is going to change in three years, said Giuseppe Amato, sales & marketing technical director at AMD in EMEA region. In fact, besides tools that help to develop GPGPU software, the world’s second largest maker of x86 processors also expects the emerge of special real-time compilers (akin to those used in drivers for GPUs nowadays) that will be able to choose whether to compute on a CPU or a GPU cores.

“In the future, GPUs will become more easily usable for general purpose processing, thanks to the increasing availability of compilers designed for this environment. After 2010 we can assume that there will be a real-time compiler available that will choose what code to execute by checking which hardware resource is available, choosing from a CPU, GPU or specific accelerators. A dispatcher will check which resources are available and will compile the code in real-time to be used with the processor that at that moment is able to provide the most efficiency,” Mr. Amato said in an interview with Hardware Upgrade web-site.

But besides software tools that will help GPGPU to rise and shine, Microsoft’s new next-generation DirectX application programming interface will make GPGPU much easier than today, according to Mr. Amato.

“Eventually we will witness evolutions in our approach to GPGPU which will allow software developers to more easily use the full capacity of the GPU. Thanks to the availability of a higher number of registers, general purpose GPU computing will also be made much easier in 2009 when Microsoft releases the DirectX 11 API,” Mr. Amato claimed.