Reciprocal Estimation Circuitry Segmentation for Processing Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data processing systems face performance bottlenecks when performing reciprocal estimation operations due to the combination of floating-point addition and reciprocal estimation operations in a single unit, which slows down both processes.
Innovation Solution
Implementing physically distinct reciprocal estimation circuitry separate from floating-point add circuitry allows for faster execution of both operations by reducing the number of processing cycles required for each, utilizing significand and exponent generating circuits with lookup tables for efficient estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If floating-point addition and reciprocal estimation operations are combined in a single unit, then device complexity is reduced, but processing speed decreases for both operations
Solution Approach 1:
The patent divides the data processing unit into separate floating-point add circuitry and reciprocal estimation circuitry. This segmentation allows each circuit to be optimized independently for its specific function, with the reciprocal estimation circuitry using dedicated lookup tables and the floating-point add circuitry using separate addition logic, thereby improving processing speed for both operations while maintaining manageable device complexity through modular design.
2Ease of manufacture
If reciprocal estimation operation is performed using the same circuitry as floating-point addition, then ease of manufacture is improved, but productivity decreases
Solution Approach 1:
The patent segments the processing functions into distinct circuitry blocks that can be manufactured separately and then integrated. The reciprocal estimation circuitry with its lookup tables and the floating-point add circuitry with its addition logic are designed as separate modules, improving productivity by allowing parallel processing and independent optimization while remaining manufacturable through standard semiconductor fabrication processes.
Solution Approach 2:
The patent creates a data processing apparatus where separate circuitry blocks serve universal purposes within their domains - the reciprocal estimation circuitry handles all reciprocal estimation operations efficiently through lookup tables, while the floating-point add circuitry handles all floating-point addition operations, providing multi-functional capability that improves overall system productivity.
3Area of stationary object
If a single circuitry performs both floating-point addition and reciprocal estimation, then device area is reduced, but processing precision may be compromised
Solution Approach 1:
The patent uses segmentation to create dedicated reciprocal estimation circuitry with specialized lookup tables that store pre-calculated reciprocal values. This dedicated circuitry ensures high estimation precision by using purpose-built hardware optimized for reciprocal calculations, while the separate floating-point add circuitry maintains addition precision independently, with both circuits occupying manageable areas through efficient layout and modular design.
Data Source
AI summary
A data processing apparatus has floating-point add circuitry for performing a floating-point add operation for adding or subtracting two floating-point operands. The apparatus also has reciprocal estimation circuitry for performing a reciprocal estimation operation on a first operand to generate a reciprocal estimate value which represents an estimate of a reciprocal of a first operand or an estimate or a reciprocal of the square root of the first operand. The reciprocal estimation circuitry is physically distinct from the floating-point adder circuitry, which allows both the reciprocal estimate and the add operations to be faster.