Unlock instant, AI-driven research and patent intelligence for your innovation.

Processing using compact arithmetic processing elements

A technology of floating-point calculation and execution unit, which is applied in the direction of calculation using number system representation, digital data processing parts, electrical components, etc., and can solve problems such as limitations

Active Publication Date: 2016-03-02
SINGULAR COMPUTING
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In other words, although the theoretical computing power of computing hardware has grown exponentially, for anything approaching the hardware's theoretical maximum computing power, the interface required by software to access the hardware limits the software's ability to use the hardware to perform computations

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Processing using compact arithmetic processing elements
  • Processing using compact arithmetic processing elements
  • Processing using compact arithmetic processing elements

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] As mentioned above, today's CPU chips don't fully utilize their transistors. For example, a conventional CPU chip containing a billion transistors might enable software to perform operations only a few times per clock cycle. While this is highly inefficient, those of ordinary skill in the art design CPUs in this manner for well-accepted valid reasons. For example, such a design meets the (often necessary) requirement for software compatibility with earlier designs. In addition, they provide tremendous precision, performing exact arithmetic with generally 32- or 64-bit long integers, and fairly accurate and widely normalized arithmetic with 32- and 64-bit floating-point numbers. Many applications require this precision. As a result, conventional CPUs are generally designed to perform arithmetic operations using transistors on the order of millions to provide such precision.

[0019] There are many economically important applications, however, which are not particularl...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention relates to processors or other devices, such as programmable and / or massively parallel processors or other devices, including those designed to perform arithmetic operations on low precision but high dynamic range (LPHDR arithmetic) values ​​(possibly but not necessarily including processing elements such as one or more of addition, multiplication, subtraction, and division). Such a processor or other device may be implemented, for example, on a single chip. Whether implemented on a single chip or not, the number of LPHDR arithmetic elements in a processor or other device in some embodiments of the invention significantly exceeds (for example, by a factor of 3 or at least 20) that in a processor or other device The number of arithmetic elements (if any) in that are designed to perform conventional precision high dynamic range operations (such as 32-bit or 64-bit floating-point operations).

Description

Background of the invention [0001] The ability to calculate quickly has become very important to humans. Weather and climate forecasting, medical applications (such as drug design and non-invasive imaging), national defense, geographic exploration, financial modeling, Internet search, network communications, scientific research in a variety of fields, and even the design of new computing hardware, each Both have become dependent on the ability to perform massive calculations quickly. Future advances in seeing, hearing, and understanding, such as computer-aided design of complex nanoscale systems or the development of consumer products, will require even greater computing power to be economically available. [0002] Gordon Moore's prediction that computing performance per dollar will double every two years has proven valid over the past 30 years, and looks likely to continue in some form. But despite this rapid exponential increase, the fact is that the inherent computing pow...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F9/38G06F9/46G06F13/14
CPCG06F7/4833G06F7/5235H03K19/17728G06F9/38G06F9/46G06F7/38G06F7/483
Inventor 约瑟夫·贝茨
Owner SINGULAR COMPUTING
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com