AMD Embedded Discrete GPU E8860
On 25th Feb 2014, AMD has announced first Embedded Discrete GPU E8860 (codenamed as “Adelaar”) industry’s first Discrete Graphic cards based on Graphic Core Next(GCN) architecture designed specially for the visual growth and parallel processing capabilities of Embedded Applications. It give double the performance with the same power from it’s predecessors.
Once again the increase in available transistors provided the opportunity to integrate CPU with GPU into a single device called as “APU”. The close proximity of these two processor types offered the ability to more tightly couple the memory system to enable direct sharing of data structures. Faced with limitations in silicon scaling (maintaining Moore’s Law) and the driving need to improve performance and efficiency, the electronics industry is banding together to change the very nature of computing. This is the transition to “heterogeneous computing” where the various execution units are more tightly integrated and share system responsibility and resources. The critical part of this is elevating the other programmable execution units like the GPU to the same level of the CPU for memory access, queuing, and execution. In other words, rather than having a CPU and various co-processors, these various processor elements can be referred to in combination as “Compute Cores.” Kaveri is the first HSA enabled APU which is based on the concept of Compute cores.
HSAIL is the PORTABLE COMPILER IR FOR HSA. Heterogeneous system architecture is an open industry standard designed to support a large variety of data-parallel and task-parallel programming models. Many application processor vendors, including AMD, ARM, Imagination, MediaTek, Texas Instrument, Samsung and Qualcomm are members of the HSA Foundation. HSA Defined features such as
A multi-processor system contains two or more processors.All processors share access to common set of memory modules,I/O channels, and peripheral devices. Inter-processor communication can be done through the shared memory. hUMA is the answer to this sharing of CPU and GPU workloads.
Morphological Antialiasing or MLAA is a cutting edge technology that uses GPU-accelerated DirectCompute to apply fullscene anti-aliasing as a post processing effect. This process differs From more traditional methods like Supersample Anti-Aliasing (SSAA), which are applied during the rendering phase of the GPU pipeline.
As the name implies, heterogeneous computing refers to systems that use more than one kind of processor. These are multi-core systems that gain performance not just by adding cores, but also by incorporating specialized processing capabilities to handle particular tasks. simplifying the particular discussion, an APU is a heterogeneous system that incorporates DirectX 11 discrete-level GPU capabilities for graphics processing and other mathematically intensive computations on very large data sets, to handle visual tasks such as 3-D rendering as well as certain functions.