Graphics Processing Unit History (2010 - Present)

2000 to 2010
 

By October 2002, with the introduction of the ATI Radeon 9700 (also called R300), the world's first Direct3D 9.0 accelerator, pixel and vertex shaders could implement looping and lengthy floating point math, and were quickly becoming as flexible as CPUs, yet orders of magnitude faster for image-array operations. Pixel shading is often used for bump mapping, which adds texture, to create an object appear shiny, dull, rough, or actually round or extruded.

Source: gpuhub cho thue render

With the introduction of the GeForce 8 series, that was produced by Nvidia, and new generic stream processing unit GPUs became a far more generalized computing device. Today, parallel GPUs possess begun producing computational inroads against the CPU, and a subfield of research, dubbed GPU Computing or GPGPU for General Purpose Computing on GPU, offers found its way into fields as diverse as machine learning, oil exploration, scientific image processing, linear algebra, statistics, 3D reconstruction and even commodity pricing dedication. GPGPU at that time was the precursor from what we have now call compute shaders (e.g. CUDA, OpenCL, DirectCompute) and also abused the hardware to a degree by treating the info passed to algorithms as texture maps and executing algorithms by drawing a triangle or quad with a proper pixel shader. This obviously entails some overheads since we involve units just like the Scan Converter where they aren't really needed (nor do we even value the triangles, except to invoke the pixel shader). Through the years, the energy consumption of GPUs provides increased also to manage it, a number of techniques have already been proposed.

Nvidia's CUDA platform, initial introduced in 2007, was the initial widely adopted programming model for GPU computing. Recently OpenCL is becoming broadly supported. OpenCL can be an open standard defined by the Khronos Group that allows for the development of code for both GPUs and CPUs with an focus on portability. OpenCL solutions are supported by Intel, AMD, Nvidia, and ARM, and according to a recently available report by Evan's Data, OpenCL may be the GPGPU development platform hottest by developers in both US and Asia Pacific.

2010 to present
 

This year 2010, Nvidia began a partnership with Audi to power their cars' dashboards. These Tegra GPUs had been powering the cars' dashboard, offering increased functionality to cars' navigation and entertainment systems. Advancements in GPU technology in cars offers helped push self-driving technology. AMD's Radeon HD 6000 Series cards were released this year 2010 and in 2011, AMD released their 6000M Series discrete GPUs to be utilized in cellular devices. The Kepler type of graphics cards by Nvidia arrived in 2012 and had been found in the Nvidia's 600 and 700 series cards. A fresh feature in this fresh GPU microarchitecture included GPU boost, a technology adjusts the clock-speed of a video card to improve or decrease it according to its power draw. The Kepler microarchitecture was manufactured on the 28 nm process.

The PS4 and Xbox One were released in 2013, they both use GPUs predicated on AMD's Radeon HD 7850 and 7790. Nvidia's Kepler type of GPUs was accompanied by the Maxwell line, manufactured on a single course of action. 28 nm chips by Nvidia were manufactured by TSMC, the Taiwan Semiconductor Manufacturing Company, that was manufacturing using the 28 nm procedure at the time. When compared to 40 nm technology from days gone by, this new manufacturing procedure allowed a 20 percent boost in performance while drawing less power. Virtual reality headsets have high system requirements. VR headset manufacturers recommended the GTX 970 and the R9 290X or better during their release. Pascal may be the following generation of consumer graphics cards by Nvidia released in 2016. The GeForce 10 group of cards are under this generation of graphics cards. They are created using the 16 nm manufacturing procedure which improves upon previous microarchitectures. Nvidia provides released one nonconsumer card beneath the new Volta architecture, the Titan V. Changes from the Titan XP, Pascal's high-end card, consist of a rise in the amount of CUDA cores, the addition of tensor cores, and HBM2. Tensor cores are cores specially created for deep learning, while high-bandwidth memory is on-die, stacked, lower-clocked memory that provides an exceptionally wide memory bus that's useful for the Titan V's intended purpose. To emphasize that the Titan V isn't a gaming card, Nvidia removed the "GeForce GTX" suffix it increases consumer gaming cards.

On August 20, 2018, Nvidia launched the RTX 20 series GPUs that add ray-tracing cores to GPUs, improving their performance on lights. Polaris 11 and Polaris 10 GPUs from AMD are fabricated a 14-nanometer procedure. Their release results in a considerable upsurge in the performance per watt of AMD video cards. AMD in addition has released the Vega GPUs series for the top quality marketplace as a competitor to Nvidia's top quality Pascal cards, also featuring HBM2 just like the Titan V.