# Electronic Design Automation (EDA)
Electronic Design Automation (EDA), also known as electronic computer-aided design (ECAD), is a category of software tools for designing electronic systems such as integrated circuits and printed circuit boards. Today, it is impracticable—in fact, unthinkable—to design chips or boards without software tools, but it hasn't been always like this. The origins of EDA software are deeply intertwined with the evolution of computing and electronics, stretching back to the 1950s and 1960s, when the first rudimentary tools and programming languages began to take shape.
In the early days, the design of electronic circuits was a manual process, involving physical drawings and mock-ups (see figure below). As electronic devices became more complex, the need for automation in the design process became evident. This led to the development of the first computer programs aimed at aiding circuit design. These initial efforts were primarily focused on schematic capture, which allowed designers to create circuit diagrams electronically rather than on paper.

> [!Figure]
> _An Analog Devices engineer laying out a circuit board with tape by hand. Credit: Analog Devices_
The 1960s and 1970s saw the introduction of the first true EDA tools, with the development of software for the layout, simulation, and analysis of electronic circuits.
During the 1980s, the EDA industry experienced rapid growth, fueled by the burgeoning personal computer market and the increasing complexity of electronic devices. This period marked the emergence of commercial EDA companies, some of which are key players in the industry today, such as Cadence Design Systems, Mentor Graphics, and Synopsys. These companies began to offer a more comprehensive suite of tools, covering every aspect of the design process from conceptualization to verification.
The 1990s and 2000s were characterized by further consolidation in the EDA industry, with larger companies acquiring smaller ones to expand their range of offerings. During this time, EDA tools became increasingly sophisticated, incorporating advanced features like high-level synthesis, which allows designers to describe circuit behavior in a high-level programming language, and physical verification, to ensure that the design meets all manufacturing requirements.
Another significant development in EDA was the introduction of hardware description languages (HDLs) such as VHDL and Verilog. These languages allowed for the description of the structure and behavior of electronic circuits in a textual form, enabling a more efficient design process, particularly for complex integrated circuits and systems-on-chip.
In recent years, the focus of EDA has shifted towards supporting the design of systems that integrate electronic hardware with software, known as electronic system level (ESL) design. This reflects the growing complexity of modern electronic systems, which often include not just hardware components but also embedded software.
Furthermore, the advent of artificial intelligence and machine learning has begun to influence EDA, with tools incorporating these technologies to automate and optimize various aspects of the design process, such as layout optimization, signal integrity, and thermal and power consumption analysis.
Today, EDA software is an indispensable tool in the electronics industry, underpinning the development of virtually all modern electronic devices. It has evolved from simple schematic capture tools to comprehensive suites that support the entire design lifecycle, enabling the creation of complex electronic systems that are foundational to contemporary technological advancements.
EDA tools encompass the whole hierarchy of digital systems: from the chip level all the way to the board level and at cases at the system level.
## EDA Switching Barriers
EDA tools tend to show a strong vendor lock-in and high switching barriers. Switching barriers are the challenges or obstacles users may face when considering transitioning from one EDA tool to another. EDA users often become proficient in specific EDA tools over time. That means, switching to a new tool requires relearning processes, user interfaces, and commands, which can slow down productivity.
What is more, very often design flows and methodologies are tightly integrated with specific EDA tools: switching to a new tool may require significant customization and configuration to match the existing workflow, including script migration, setting up design environments, and integrating with other tools in the design flow.
EDA tools tend to use proprietary file formats for designs and projects. Changing tools may require converting or translating design files using dubious automatic converters which never work completely right, which can lead to data loss or high amounts of manual re-work.
To make things more difficult (intentionally or unintentionally), EDA tool vendors often offer complex tool suites composed of many different applications that are designed to work tightly together. Users cringe at the idea of switching tools due to the perceived risk of losing integration benefits and support services provided by a single vendor.
Transitioning to a new EDA tool requires training users on the new software, which involves time and costs money. Additionally, users may be concerned about the availability and quality of support services for the new tool. Existing designs and projects may be tied to specific EDA tools due to historical reasons or legacy support. Switching tools may require migrating or redesigning these projects, which can be time-consuming and costly.
EDA tools are often part of a larger ecosystem that includes third-party software, libraries, and IP cores. Users are typically concerned about the compatibility of these components with a new EDA tool.
All in all, EDA tools can significantly impact the decision-making process for users considering a transition.
> [!warning]
> This section is under #development
## Open Source EDA
The semiconductor industry is a complex sector that encompasses the design, manufacturing, and distribution of semiconductor devices. The market is rather concentrated with a few key players. Market share in semiconductor devices varies depending on the specific segment. For example, Intel has traditionally dominated the microprocessor market for personal computers, while Samsung and SK Hynix are major players in the memory chip market. TSMC dominates the foundry market, producing chips for various fabless semiconductor companies.
The market for EDA tools is dominated by a few key players, including Synopsys, Cadence Design Systems, and Mentor, a Siemens Business. These companies offer a wide range of tools for tasks such as logic synthesis, simulation, verification, and physical design. The market share among EDA tool vendors can fluctuate based on factors such as technology advancements, customer preferences, and mergers/acquisitions within the industry.
The semiconductor shortage and shifts in global politics have prompted significant changes in the strategy for semiconductors and chip design. For instance, the European Chips Act^[https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age/european-chips-act_en] aims to incentivize the establishment of new fabrication facilities and the development of leading-edge chip design capabilities in Europe. Universities play an important role in research and education for chip design, fostering innovation and training future designers. Open-source initiatives, particularly those based on the [[Semiconductors#RISC-V|RISC-V]] instruction set, have spurred research activities and innovation in Europe, leading to the rapid growth of a vibrant ecosystem.
The United States has also announced its plan to boost U.S. capabilities for advanced packaging, a key technology for manufacturing state-of-the-art semiconductors. The National Institute of Standards and Technology (NIST) has laid out how the U.S. will benefit from the CHIPS for America^[https://www.nist.gov/chips] program’s manufacturing incentives and research and development efforts. While the United States remains a global leader in semiconductor design and research and development, it has fallen behind in manufacturing and now accounts for only about 10 percent of global commercial production. Today, none of the most advanced logic and memory chips are manufactured at commercial scale in the United States. In addition, many elements of the semiconductor supply chain are geographically concentrated, leaving them vulnerable to disruption and endangering the global economy and U.S. national security.
To further advance chip design education and innovation, the adoption of open-source Electronic Design Automation (EDA) tools is of great importance. Open-source EDA tools significantly lower the barrier to entry for students and researchers, facilitate collaboration, and enable hands-on learning and innovation. Funding programs must increase their support for open-source [[Electronic Design Automation (EDA)|EDA tools]], recognizing their importance in making chip design more accessible, fostering independent design platforms, and filling gaps in the toolchain.
> [!warning]
> This section is under #development