Computational optimization mechanism

DE602018080896T2Active Publication Date: 2025-04-09INTEL CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE602018080896
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2017-04-24
Filing Date
2018-03-02
Publication Date
2025-04-09
Estimated Expiration
2038-03-02

AI Technical Summary

Technical Problem

Current graphics processing units (GPUs) face challenges in optimizing computing efficiency due to the need for sorting and processing threads with varying bit depths and precision requirements within the graphics pipeline.

Method used

The proposed solution involves a compute mechanism that includes sorting logic to group threads based on the bit depth of floating-point operations, floating-point logic to process threads with higher bit depths, and variable precision support in math instructions, enabling efficient handling of mixed precision computations.

Benefits of technology

This approach enhances computing efficiency by optimizing thread grouping and precision handling, leading to improved performance in graphics processing and other compute-intensive tasks.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art
No text content released.