Linked from

The 43 pages that link to Graphics processing unit, each with the reason it gives.

• Central processing unitCompared with: Its parallel design contrasts with the CPU’s broad role in sequential control and varied tasks.

• MicroprocessorCompared with: GPUs complement microprocessors by handling workloads suited to massive parallelism.

• System on a chipRelated: An integrated GPU handles graphics without requiring a separate graphics chip.

• Floating-point arithmeticRelated: Its floating-point throughput and supported formats shape many large computations.

• Parallel computingRelated: GPUs provide large numbers of processing units for data-parallel computations.

• Field-programmable gate arrayCompared with: A GPU offers fixed parallel execution resources, while an FPGA can customize its circuit structure.

• High-performance computingRelated: GPUs accelerate HPC workloads with large amounts of parallel arithmetic.

• Computer architectureCompared with: Its throughput-oriented organization differs from general-purpose central processors.

• Semiconductor industryBroader topic: GPU demand has made advanced computing a prominent source of chip sales.

• Display resolutionRelated: Rendering more pixels generally requires the GPU to process more image data.

• Video game consoleRelated: It renders the images that consoles send to displays.

• Application-specific integrated circuitCompared with: GPUs offer broad parallel programmability instead of ASIC-level specialization.

• Deep learningRelated: GPU parallelism made the large matrix computations of deep-network training faster.

• Direct memory accessRelated: GPUs exchange commands and data with system memory through DMA-capable interfaces.

• Computer monitorNarrower topic: The GPU commonly generates the frames that a monitor displays.

• Refresh rateRelated: It generates frames whose delivery rate interacts with display refresh.

• SupercomputerRelated: GPUs provide many arithmetic units for tasks that can be parallelized.

• Large language modelRelated: Parallel numerical operations make GPUs useful for training and running large models.

• Digital signal processorCompared with: GPUs exploit large-scale parallelism, whereas DSPs often target predictable, low-latency signal streams.

• CUDANarrower topic: CUDA targets GPUs as parallel processors rather than only graphics devices.

• Home video game consoleRelated: It renders the scenes that consoles display during gameplay.

• NvidiaNarrower topic: Nvidia’s business began with graphics processors and now centers on their parallel-computing capabilities.

• MotherboardRelated: Motherboard slots and interfaces determine how discrete GPUs connect to a computer.

• Arithmetic logic unitRelated: GPU execution units contain arithmetic and logical circuits adapted to parallel workloads.

• Real-time renderingRelated: Its parallel execution makes high-throughput image generation possible.

• Vector processorCompared with: GPUs use many parallel execution units rather than a classic vector pipeline.

• Computer hardwareBroader topic: Its parallel design makes it useful for graphics, machine learning, and scientific computing.

• PlayStation 4Narrower topic: The PS4 combines its CPU and graphics processor on one custom chip.

• Floating-point unitRelated: GPUs contain extensive floating-point hardware for graphics and scientific computation.

• Graphics cardBroader topic: The GPU performs most of a graphics card’s image-processing work.

• Hardware accelerationRelated: GPUs accelerate workloads that can be divided into many simultaneous operations.

• Jensen HuangNarrower topic: Huang built Nvidia around graphics processors before they became general-purpose computing engines.

• OpenGLNarrower topic: OpenGL commands commonly execute on a GPU, though implementations can use other processors.

• Output deviceRelated: The GPU generates image data that displays present.

• Direct3DNarrower topic: Direct3D commands are executed primarily by the GPU.

• Clock generatorRelated: GPU timing circuits supply clocks to processing and memory domains.

• Lisa SuNarrower topic: AMD’s GPU business is a major part of its product portfolio.

• PC gameRelated: Its performance and features set limits on visual effects and frame rates.

• AI boomRelated: GPU clusters supply much of the computing power needed to train large models.

• CoprocessorBroader topic: It is the most familiar coprocessor for rendering images and accelerating parallel workloads.

• Gaming computerRelated: The GPU renders most game visuals and often determines graphical performance.

• Multi-core processorCompared with: GPUs use many simpler execution units for parallel workloads, unlike general-purpose multi-core CPUs.

• Network on a chipRelated: Its many execution units and memory blocks exchange data across on-chip fabrics.