floating-point operation
1
6.8
- noun
- /ˈfloʊtɪŋ pɔɪnt ˌɑpəˈreɪʃən/
- Specialized
:
operación de punto flotante
- Modern computers can perform trillions of floating-point operations every second, allowing them to solve complex mathematical problems quickly.
- perform floating-point operations
- number of floating-point operations
Examples
-
In some ways, the Java definition is tighter: Floating-point operations do not produce exceptions.
Academic text (1999) -
Nowadays, trillions of floating point operations per second (teraflops) are feasible using a single GPU processor.
Academic text (2018) -
Therefore, these operations can be done in parallel with the floating-point operations.
Academic text (2000) -
Consequently, it is important that all platforms provide consistent floating-point operations.
Academic text (1999) -
The performance of Java programs depends on the density of their floating-point operations.
Academic text (1999) -
By balance, we refer to the number of data values that can be loaded in a cycle relative to the number of floating-point operations that can be performed in a cycle.
Academic text (1997) -
All floating-point operations except for divide, square root, and extended multiply require only one cycle in pipelined mode.
Academic text (1997) -
Rendering is not a graphics operation, but rather a series of floating-point operations where the mathematics behind visual effects are computed.
Academic text (1993) -
Dual units enable the FPU to execute two double-precision multiply/add instructions every cycle, resulting in up to four floating-point operations per cycle.
Academic text (1994) -
For the TRSM and SYRK kernels, more floating-point operations are performed, which reduces the ratio of the number of accesses to the number of operations.
Academic text (2000)
Synonyms
19
8.2
A unit that shows how many decimal calculations a computer can do each second
Surface Forms
Morphology
floating-point + operation
The headword 'operation' clearly signals an arithmetic or computational procedure, so learners will grasp that this is a kind of calculation. However 'floating-point' encodes a technical way of representing numbers (a format for decimals/exponents) that is not fully predictable from 'floating' + 'point' for an average B1 learner; they may guess it involves decimal/non-integer numbers but would not derive the precise technical meaning without prior exposure.
Etymology
Floating-point operation comes from how computers write numbers: the point is the 'decimal point' and can move, or float, so a number can be very big or very small. So a floating-point operation is a calculation a computer does using these numbers with a moving 'point'.