At their GPU Technology Conference in San Jose, NVIDIA revealed the roadmap for future Tegra-series processors, among other things. The roadmap highlights NVIDIA’s plans for 2013 and 2014 with two future processors, Logan and its successor, Parker, which continue NVIDIA’s tradition of naming them after the real last names of superheroes.
Logan will be the successor to Wayne, more commonly known as the Tegra 4 chipset, and will be appearing in devices in early 2014. It will feature a quad-core ARM Cortex A15 MPCore alongside a low power companion core. It will also be the first Tegra chip to include CUDA cores, NVIDIA’s solution for parallel computing on the GPU, and will be based on a Kepler architecture, similar to the desktop GPUs that NVIDIA makes.
Logan will also be introducing support for CUDA 5.0, the latest version of CUDA, as well as OpenGL 4.3, which brings new features such as tessellation, computer shaders and geometry shaders over OpenGL ES 3.0.
Next is Parker, which will use the upcoming Denver CPU with 64-bit support and the upcoming Maxwell series of GPU. All the advancements in Logan will be carried forward to Parker. Parker is expected to arrive in devices by 2015.
NVIDIA also showcased an mITX-like board called Kayla, which features a Tegra 3 SoC and a low power Kepler GPU, presumably from Logan. This board is designed primarily for developers to test their applications, specifically for the new CUDA functionality that will be introduced from Logan onwards. Not much details about Kayla are available at the moment.
No comments:
Post a Comment