Jen-Hsun Huang, Nvidia’s co-founder and CEO, announced the roadmap for the next few generations of his company’s graphics platforms this morning at the GPU Technology Conference in San Jose, CA.
During a keynote address that showed the potential of parallel processing using GPUs, Huang revealed the code names for the next two architectures, and their relative power. After Fermi, Nvidia’s next GPU will be called Kepler (based on 28nm technology), which will come out in 2011, and have three to four times the performance per watt as the current platform. After that, Maxwell will debut in 2013, and have about 16 times the performance per watt as Tesla.
In terms of consumer devices, new, more powerful GPUs will have the greatest impact when it comes to mobile devices, said Huang. Mobile computing (in the form of smart phones and the like) will no longer be just about having a powerful processor in your hands, but a computer equipped with all kinds of sensors–cameras, GPS, and so forth–will be a context aware, situation-aware device, making for a “magical, unexpected, delightful experience,” he said. “It’s the next personal computer revolution.”
Huang then brought a bunch of different presenters on stage to show the benefit of graphics-based parallel computing, such as modeling waves, dust, and molecules, to even performing heart surgery in three dimensions. One of the most intriguing applications, as far as consumers are concerned, is a technology being developed by Adobe using what’s called a plenoptic lens.
By attaching a number of small lenses to a camera, users can take a picture that looks, at first, like something an insect might see or a Chuck Close painting – a bunch of small squares, each displaying a small part of the entire picture. However, using an algorithm Adobe invented, the multitudes of little images can be resolved into a single photo.
Also, because the complete image is made up of individual shots, all taken with their own viewpoint, you can change the point of focus on the fly. For example, you can have the foreground in focus, or the background in focus, or both. No more blurry shots! Even better, the Adobe software can convert 2D images to 3D instantaneously.
Right now, the technology is still in the development stage–the camera lens is pretty bulky, and the software requires a good deal of muscle–but it could herald the next step in digital photography.