en.unionpedia.org

Half-precision floating-point format, the Glossary

In computing, half precision (sometimes called FP16 or float16) is a binary floating-point computer number format that occupies 16 bits (two bytes in modern computers) in computer memory.^[1]

71 relations: ARM architecture family, Bandwidth (computing), Bfloat16 floating-point format, Binary file, C Sharp (programming language), C++/CLI, C++/CX, Cg (programming language), Computer graphics, Computer memory, Computer number format, Computing, Control register, D3DX, Data type, Digital image processing, Direct3D, Double-precision floating-point format, Dynamic range, Exponent bias, Exponentiation, F Sharp (programming language), F16C, Floating-point arithmetic, GeForce FX series, GIMP, IEEE 754, IEEE 754-2008 revision, Industrial Light & Magic, Infinity, Instruction set architecture, ISO/IEC 10967, JPEG XR, Machine learning, MATLAB, Microsoft, Minifloat, NaN, Neural network, Nvidia, Offset binary, OpenCL, OpenEXR, OpenGL, Polygon mesh, Power ISA, Primitive data type, Radix, RGBE image format, RISC-V, ... Expand index (21 more) »
Floating point types

ARM architecture family

ARM (stylised in lowercase as arm, formerly an acronym for Advanced RISC Machines and originally Acorn RISC Machine) is a family of RISC instruction set architectures (ISAs) for computer processors.

See Half-precision floating-point format and ARM architecture family

Bandwidth (computing)

In computing, bandwidth is the maximum rate of data transfer across a given path.

See Half-precision floating-point format and Bandwidth (computing)

Bfloat16 floating-point format

The bfloat16 (brain floating point) floating-point format is a computer number format occupying 16 bits in computer memory; it represents a wide dynamic range of numeric values by using a floating radix point. Half-precision floating-point format and bfloat16 floating-point format are binary arithmetic and floating point types.

See Half-precision floating-point format and Bfloat16 floating-point format

Binary file

A binary file is a computer file that is not a text file.

See Half-precision floating-point format and Binary file

C Sharp (programming language)

C# is a general-purpose high-level programming language supporting multiple paradigms.

See Half-precision floating-point format and C Sharp (programming language)

C++/CLI

C++/CLI is a variant of the C++ programming language, modified for Common Language Infrastructure.

See Half-precision floating-point format and C++/CLI

C++/CX

C++/CX (C++ component extensions) is a language projection for Microsoft's Windows Runtime platform.

See Half-precision floating-point format and C++/CX

Cg (programming language)

Cg (short for C for Graphics) and High-Level Shader Language (HLSL) are two names given to a high-level shading language developed by Nvidia and Microsoft for programming shaders.

See Half-precision floating-point format and Cg (programming language)

Computer graphics

Computer graphics deals with generating images and art with the aid of computers.

See Half-precision floating-point format and Computer graphics

Computer memory

Computer memory stores information, such as data and programs, for immediate use in the computer.

See Half-precision floating-point format and Computer memory

Computer number format

A computer number format is the internal representation of numeric values in digital device hardware and software, such as in programmable computers and calculators.

See Half-precision floating-point format and Computer number format

Computing

Computing is any goal-oriented activity requiring, benefiting from, or creating computing machinery.

See Half-precision floating-point format and Computing

Control register

A control register is a processor register that changes or controls the general behavior of a CPU or other digital device.

See Half-precision floating-point format and Control register

D3DX

In computing, D3DX (Direct3D Extension) is a high level API library which is written to supplement Microsoft's Direct3D graphics API.

See Half-precision floating-point format and D3DX

Data type

In computer science and computer programming, a data type (or simply type) is a collection or grouping of data values, usually specified by a set of possible values, a set of allowed operations on these values, and/or a representation of these values as machine types.

See Half-precision floating-point format and Data type

Digital image processing

Digital image processing is the use of a digital computer to process digital images through an algorithm.

See Half-precision floating-point format and Digital image processing

Direct3D

Direct3D is a graphics application programming interface (API) for Microsoft Windows.

See Half-precision floating-point format and Direct3D

Double-precision floating-point format

Double-precision floating-point format (sometimes called FP64 or float64) is a floating-point number format, usually occupying 64 bits in computer memory; it represents a wide dynamic range of numeric values by using a floating radix point. Half-precision floating-point format and Double-precision floating-point format are binary arithmetic and floating point types.

See Half-precision floating-point format and Double-precision floating-point format

Dynamic range

Dynamic range (abbreviated DR, DNR, or DYR) is the ratio between the largest and smallest values that a certain quantity can assume.

See Half-precision floating-point format and Dynamic range

Exponent bias

In IEEE 754 floating-point numbers, the exponent is biased in the engineering sense of the word – the value stored is offset from the actual value by the exponent bias, also called a biased exponent.

See Half-precision floating-point format and Exponent bias

Exponentiation

In mathematics, exponentiation is an operation involving two numbers: the base and the exponent or power.

See Half-precision floating-point format and Exponentiation

F Sharp (programming language)

F# (pronounced F sharp) is a general-purpose, high-level, strongly typed, multi-paradigm programming language that encompasses functional, imperative, and object-oriented programming methods.

See Half-precision floating-point format and F Sharp (programming language)

F16C

The F16C (previously/informally known as CVT16) instruction set is an x86 instruction set architecture extension which provides support for converting between half-precision and standard IEEE single-precision floating-point formats.

See Half-precision floating-point format and F16C

Floating-point arithmetic

In computing, floating-point arithmetic (FP) is arithmetic that represents subsets of real numbers using an integer with a fixed precision, called the significand, scaled by an integer exponent of a fixed base.

See Half-precision floating-point format and Floating-point arithmetic

GeForce FX series

The GeForce FX or "GeForce 5" series (codenamed NV30) is a line of graphics processing units from the manufacturer Nvidia.

See Half-precision floating-point format and GeForce FX series

GIMP

GNU Image Manipulation Program, commonly known by its acronym GIMP, is a free and open-source raster graphics editor used for image manipulation (retouching) and image editing, free-form drawing, transcoding between different image file formats, and more specialized tasks.

See Half-precision floating-point format and GIMP

IEEE 754

The IEEE Standard for Floating-Point Arithmetic (IEEE 754) is a technical standard for floating-point arithmetic established in 1985 by the Institute of Electrical and Electronics Engineers (IEEE). Half-precision floating-point format and IEEE 754 are binary arithmetic and floating point types.

See Half-precision floating-point format and IEEE 754

IEEE 754-2008 revision

IEEE 754-2008 (previously known as IEEE 754r) is a revision of the IEEE 754 standard for floating-point arithmetic.

See Half-precision floating-point format and IEEE 754-2008 revision

Industrial Light & Magic

Industrial Light & Magic (ILM) is an American motion picture visual effects company that was founded on May 26, 1975 by George Lucas.

See Half-precision floating-point format and Industrial Light & Magic

Infinity

Infinity is something which is boundless, endless, or larger than any natural number.

See Half-precision floating-point format and Infinity

Instruction set architecture

In computer science, an instruction set architecture (ISA) is an abstract model that generally defines how software controls the CPU in a computer or a family of computers.

See Half-precision floating-point format and Instruction set architecture

ISO/IEC 10967

ISO/IEC 10967, Language independent arithmetic (LIA), is a series of standards on computer arithmetic.

See Half-precision floating-point format and ISO/IEC 10967

JPEG XR

JPEG XR (JPEG extended range) is an image compression standard for continuous tone photographic images, based on the HD Photo (formerly Windows Media Photo) specifications that Microsoft originally developed and patented.

See Half-precision floating-point format and JPEG XR

Machine learning

Machine learning (ML) is a field of study in artificial intelligence concerned with the development and study of statistical algorithms that can learn from data and generalize to unseen data and thus perform tasks without explicit instructions.

See Half-precision floating-point format and Machine learning

MATLAB

MATLAB (an abbreviation of "MATrix LABoratory") is a proprietary multi-paradigm programming language and numeric computing environment developed by MathWorks.

See Half-precision floating-point format and MATLAB

Microsoft

Microsoft Corporation is an American multinational corporation and technology company headquartered in Redmond, Washington.

See Half-precision floating-point format and Microsoft

Minifloat

In computing, minifloats are floating-point values represented with very few bits. Half-precision floating-point format and minifloat are floating point types.

See Half-precision floating-point format and Minifloat

NaN

In computing, NaN, standing for Not a Number, is a particular value of a numeric data type (often a floating-point number) which is undefined as a number, such as the result of 0/0.

See Half-precision floating-point format and NaN

Neural network

A neural network is a group of interconnected units called neurons that send signals to one another.

See Half-precision floating-point format and Neural network

Nvidia

Nvidia Corporation is an American multinational corporation and technology company headquartered in Santa Clara, California, and incorporated in Delaware.

See Half-precision floating-point format and Nvidia

Offset binary

Offset binary, also referred to as excess-K, excess-N, excess-e, excess code or biased representation, is a method for signed number representation where a signed number n is represented by the bit pattern corresponding to the unsigned number n+K, K being the biasing value or offset. Half-precision floating-point format and offset binary are binary arithmetic.

See Half-precision floating-point format and Offset binary

OpenCL

OpenCL (Open Computing Language) is a framework for writing programs that execute across heterogeneous platforms consisting of central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), field-programmable gate arrays (FPGAs) and other processors or hardware accelerators.

See Half-precision floating-point format and OpenCL

OpenEXR

OpenEXR is a high-dynamic range, multi-channel raster file format, released as an open standard along with a set of software tools created by Industrial Light & Magic (ILM), under a free software license similar to the BSD license.

See Half-precision floating-point format and OpenEXR

OpenGL

OpenGL (Open Graphics Library) is a cross-language, cross-platform application programming interface (API) for rendering 2D and 3D vector graphics.

See Half-precision floating-point format and OpenGL

Polygon mesh

In 3D computer graphics and solid modeling, a polygon mesh is a collection of, s and s that defines the shape of a polyhedral object.

See Half-precision floating-point format and Polygon mesh

Power ISA

Power ISA is a reduced instruction set computer (RISC) instruction set architecture (ISA) currently developed by the OpenPOWER Foundation, led by IBM.

See Half-precision floating-point format and Power ISA

Primitive data type

In computer science, primitive data types are a set of basic data types from which all other data types are constructed.

See Half-precision floating-point format and Primitive data type

Radix

In a positional numeral system, the radix (radices) or base is the number of unique digits, including the digit zero, used to represent numbers.

See Half-precision floating-point format and Radix

RGBE image format

RGBE or Radiance HDR is an image format invented by Gregory Ward Larson for the Radiance rendering system.

See Half-precision floating-point format and RGBE image format

RISC-V

RISC-V (pronounced "risk-five") is an open standard instruction set architecture (ISA) based on established reduced instruction set computer (RISC) principles.

See Half-precision floating-point format and RISC-V

Rust (programming language)

Rust is a general-purpose programming language emphasizing performance, type safety, and concurrency.

See Half-precision floating-point format and Rust (programming language)

Sapphire Rapids

Sapphire Rapids is a codename for Intel's server (fourth generation Xeon Scalable) and workstation (Xeon W-2400 and Xeon W-3400) processors based on the Golden Cove microarchitecture and produced using Intel 7.

See Half-precision floating-point format and Sapphire Rapids

Shadow

A shadow is a dark area where light from a light source is blocked by an object.

See Half-precision floating-point format and Shadow

SIGGRAPH

SIGGRAPH (Special Interest Group on Computer Graphics and Interactive Techniques) is an annual conference centered around computer graphics organized by ACM, starting in 1974.

See Half-precision floating-point format and SIGGRAPH

Sign bit

In computer science, the sign bit is a bit in a signed number representation that indicates the sign of a number. Half-precision floating-point format and sign bit are binary arithmetic.

See Half-precision floating-point format and Sign bit

Significand

The significand (also coefficient, sometimes argument, or more ambiguously mantissa, fraction, or characteristic) is the first (left) part of a number in scientific notation or related concepts in floating-point representation, consisting of its significant digits.

See Half-precision floating-point format and Significand

Significant figures

Significant figures, also referred to as significant digits or sig figs, are specific digits within a number written in positional notation that carry both reliability and necessity in conveying a particular quantity.

See Half-precision floating-point format and Significant figures

Silicon Graphics

Silicon Graphics, Inc. (stylized as SiliconGraphics before 1999, later rebranded SGI, historically known as Silicon Graphics Computer Systems or SGCS) was an American high-performance computing manufacturer, producing computer hardware and software.

See Half-precision floating-point format and Silicon Graphics

Single instruction, multiple data

Single instruction, multiple data (SIMD) is a type of parallel processing in Flynn's taxonomy.

See Half-precision floating-point format and Single instruction, multiple data

Single-precision floating-point format

Single-precision floating-point format (sometimes called FP32 or float32) is a computer number format, usually occupying 32 bits in computer memory; it represents a wide dynamic range of numeric values by using a floating radix point. Half-precision floating-point format and Single-precision floating-point format are binary arithmetic and floating point types.

See Half-precision floating-point format and Single-precision floating-point format

Subnormal number

In computer science, subnormal numbers are the subset of denormalized numbers (sometimes called denormals) that fill the underflow gap around zero in floating-point arithmetic.

See Half-precision floating-point format and Subnormal number

Swift (programming language)

Swift is a high-level general-purpose, multi-paradigm, compiled programming language created by Chris Lattner in 2010 for Apple Inc. and maintained by the open-source community.

See Half-precision floating-point format and Swift (programming language)

Unit in the last place

In computer science and numerical analysis, unit in the last place or unit of least precision (ulp) is the spacing between two consecutive floating-point numbers, i.e., the value the least significant digit (rightmost digit) represents if it is 1.

See Half-precision floating-point format and Unit in the last place

Visual Basic (.NET)

Visual Basic (VB), originally called Visual Basic.NET (VB.NET), is a multi-paradigm, object-oriented programming language, implemented on.NET, Mono, and the.NET Framework.

See Half-precision floating-point format and Visual Basic (.NET)

Vulkan

Vulkan is a low-level, low-overhead cross-platform API and open standard for 3D graphics and computing.

See Half-precision floating-point format and Vulkan

X86

x86 (also known as 80x86 or the 8086 family) is a family of complex instruction set computer (CISC) instruction set architectures initially developed by Intel based on the 8086 microprocessor and its 8-bit-external-bus variant, the 8088.

See Half-precision floating-point format and X86

Zig (programming language)

Zig is an imperative, general-purpose, statically typed, compiled system programming language designed by Andrew Kelley.

See Half-precision floating-point format and Zig (programming language)

.NET

The.NET platform (pronounced as "dot net") is a free and open-source, managed computer software framework for Windows, Linux, and macOS operating systems.

See Half-precision floating-point format and .NET

0

0 (zero) is a number representing an empty quantity.

See Half-precision floating-point format and 0

16-bit computing

16-bit microcomputers are microcomputers that use 16-bit microprocessors.

See Half-precision floating-point format and 16-bit computing

3dfx

3dfx Interactive, Inc. was an American computer hardware company headquartered in San Jose, California, founded in 1994, that specialized in the manufacturing of 3D graphics processing units, and later, video cards.

See Half-precision floating-point format and 3dfx

References

[1] https://en.wikipedia.org/wiki/Half-precision_floating-point_format

Also known as 16-bit binary floating-point format, 16-bit floating point format, 16-bit floating-point format, Binary16, FP16, Float16, Half float, Half precision, Half precision floating point, Half precision floating-point, Half precision floating-point format, Half-float, Half-precision, IEEE 754-2008 half precision.

, Rust (programming language), Sapphire Rapids, Shadow, SIGGRAPH, Sign bit, Significand, Significant figures, Silicon Graphics, Single instruction, multiple data, Single-precision floating-point format, Subnormal number, Swift (programming language), Unit in the last place, Visual Basic (.NET), Vulkan, X86, Zig (programming language), .NET, 0, 16-bit computing, 3dfx.

Half-precision floating-point format, the Glossary

Table of Contents

See also

Floating point types

References