Sandy Bridge is the first GPU tightly integrated with an x86 through a shared L3 cache. Graphics performance has doubled, thanks to new shader cores and more powerful fixed functions. Sadly, there is no OpenCL or DirectX11 support till Ivy Bridge. Multimedia is superb, with full hardware decoding and accelerated encoding exposed through an API. The new design is a huge advance, but much work remains for future generations.
AMD has a grand vision for software and physical integration of CPUs and GPUs. The first Fusion generation focused on time to market, but created a solid foundation. Llano is a surprisingly attractive mid-range and value notebook product, due to the vastly enhanced power management. Future Fusion products will upgrade the CPU, GPU and media hardware and move towards a more tightly integrated computing model.
Enthusiasts and engineers know cooling is vital; it raises frequency and dramatically lowers power by reducing CPU or GPU temperatures. The world’s fastest supercomputer shows that thermal management can increase CPU performance/watt by 20% and cooling is critical for 3D integration and Moore’s Law.
Memory bandwidth is a critical to feeding the shader arrays in programmable GPUs. We show that memory is an integral part of a good performance model and can impact graphics by 40% or more. The implications are important for upcoming integrated graphics, such as AMD’s Llano and Intel’s Ivy Bridge – as the bandwidth constraints will play a key role in determining overall performance.
Intel’s Sandy Bridge ISSCC paper discusses a number of challenges they will eventually impact most vendors. The novel architectural choices and circuit design solutions that they describe give insight into current and future products from Intel, but also the general direction of the industry. The overarching theme is taking advantage of Moore’s Law at 32nm and beyond, which entails considerable attention to design complexity, process variation, power efficiency and validation.
Modern graphics processors are incredibly complex, but understanding their performance is essential, as they become an increasingly important component of computer systems. In this report, we use a set of benchmark results to build accurate performance models for AMD and Nvidia GPUs. We verify that our model can predict performance within roughly 6-8% for many desktop graphics cards and show how Nvidia’s microarchitecture and drivers achieve roughly 2X higher utilization than AMD’s VLIW5 design.
The major trend in graphics is programmability and targeting highly parallel, general-purpose workloads. Historically, AMD has focused on gaming performance. However, DirectCompute and OpenCL are beginning to take hold and create the seeds of a software ecosystem. AMD’s new Cayman architecture is a gradual and evolutionary step towards more general purpose hardware and a cautious embrace of GPU computing. While primarily a graphics processor, Cayman has made some fundamental microarchitecture changes to improve programmability and performance. In this article, we explore the Cayman architecture including the new VLIW4 SIMD, dynamic power management and other enhancements. Our report concludes with a preliminary assessment of the Radeon 6970 and 9650 graphics cards and projections for frequency, power and performance of future compute products.
A critical question for GPU computing is how programmers will interface with the underlying hardware. Users have the choice between three APIs: Nvidia’s proprietary CUDA, Microsoft’s DirectCompute and OpenCL. Of the three, OpenCL has garnered the most enthusiasm across the PC ecosystem (e.g. AMD, IBM, Intel and Nvidia) and the mobile and embedded market (e.g. ARM and Imagination Technologies). While still a nascent technology, OpenCL is very popular because it is an open, industry standard that promises compatibility on a huge variety of hardware. This article explores aspects of OpenCL, including the early development efforts at Apple and the standard itself, including the execution and memory model.
PhysX is a key application that Nvidia uses to showcase the advantages of GPU computing (GPGPU) for consumers. PhysX executing on an Nvidia GPU an improve performance by 2-4X compared to running on a CPU from Intel or AMD. We investigated and discovered that CPU PhysX exclusively uses x87 rather than the faster SSE instructions. This hobbles the performance of CPUs, calling into question the real benefits of PhysX on a GPU.
Larrabee is Intel’s unique architecture for a family of throughput processors, developed for the graphics and HPC markets. We have recently learned that graphics products based on Larrabee 1, the first implementation, have been canceled and that it will instead be used as a software development vehicle. Larrabee’s troubles lay in software, and now the question is what lies ahead in the future for Larrabee and Intel’s graphics products.