Updated: 3/16/10, 4:56pm PDT
A couple of years ago, we at NVIDIA announced that the world was entering the Visual Computing era. We saw this coming as a result of the community's work with general purpose computation on graphics processing units or GPGPU. That is when we also began pushing for the Optimized PC, one that has the correct balance of CPU and GPU horsepower.
A small group of creative engineers began using the massively parallel architecture of the graphics processing unit (GPU) as a computational engine, to create GPGPU. The sequential part of the application runs on the CPU (ie: running an OS where a user is unpredictable and moving a mouse, going from application to application) and the computationally-intensive part of the application (large, more predictable, amounts of data) runs on the GPU. From the user’s perspective, the application just runs faster because it is using the high-performance of the GPU to boost performance. We have seen application performance jump from twice as fast to 300 or 400 times faster when the GPU is used for processing.
GPU computing began to explode when it got easier to program. The pioneer for this was NVIDIA’s CUDA C, a small set of extensions developers use to specify which functions will be executed on the GPU and how the parallel processing capabilities of the GPU will be used by an application. Now developers have even more options from programming for the GPU, with Microsoft’s DirectCompute and Open CL APIs. Today the GPU is used to accelerate consumer applications for changing video file formats, video enhancement, video editing, photo editing, face recognition, audio mixing and in-game physics.
Microsoft has always been a leader in visual computing. They were the first to make a visual computing application nearly everyone would use when they released Windows Vista. This was the first time an operating system was accelerated by a GPU. They expanded on that with Windows 7 and the introduction of the Direct Compute API.
There is a global movement underway to enhance the visual computing experience in every type of device. Internet Explorer 9 and its GPU hardware acceleration will kick that in to overdrive. It is truly a killer app for visual computing. It is a huge proof point for the optimized PC concept.
To get the best experience with today’s visual computing applications you need to make sure your PC has the right mix of CPU/GPU horsepower. Whether you’re using the latest operating systems, viewing or editing photos, finding directions, playing a game, or watching a Blu-ray movie, a balanced PC with an NVIDIA GeForce GPU yields the best experience…. even if you are just surfing the net.