Why do semiconductor companies keep up with the pace of mergers and acquisitions?
Semiconductor market headwind cycle is still not over, the footsteps of corporate mergers and acquisitions have not stopped. After finishing the discovery of 2023 global semiconductor field mergers and acquisitions are mainly concentrated in the field of AI, EDA, wide bandwidth semiconductor, presenting the following characteristics.
Feature 1: EDA mergers and acquisitions in the field of concentration
In terms of quantity, in 2023, the acquisition of electronic system design companies represented by EDA was active, and most of them were horizontal acquisitions of companies in the same or adjacent fields. Cadence, the head of EDA, acquired four companies in one year, three of which were also engaged in electronic system design. Synopsys and Keysight all have software companies under their wings. Of course, there are also EDA companies were acquired, such as Insight and Diakopto were Siemens and ANSYS two engineering software companies in the bag. EDA companies actively expanding to horizontal acquisitions to deal with the challenges of advanced processes to increase the complexity of chip design, to become the main semiconductor M&A events in 2023.
Acquisition reason 1: advanced process to increase the complexity of chip design
EDA companies generally hope that through acquisitions to enhance the chip design automation experience, in order to deal with the advanced process to bring the layout, wiring, testing and other major design process challenges.
Typical A: Keysight acquires Cliosoft
Keysight acquired Cliosoft for its process and data management capabilities. Chip design using advanced processes involves more and more complex IP modules and data, and the management tools provided by companies such as Cliosoft can help users build efficient workflows.
Typical B: ANSYS Acquires Diakopto
ANSYS acquired Diakopto in response to the problem of parasitic effects. As the processes used in chips become more and more advanced, the problem of parasitic effects in chip design layouts is becoming more and more complex. Based on the acquisition, ANSYS hopes to detect parasitic effects early in the chip design and provide engineers with viable solutions to improve chip reliability and accelerate time-to-market.
Acquisition Reason 2: Follow up on Hot Technologies and Hot Areas
As usual, EDA companies have a strong interest in hot technologies and hot application areas. Keeping up with the times and not being left behind is the inexhaustible driving force for EDA companies to "buy and buy".
Typical A: Cadence acquires Pulsic
The main goal of Cadence's acquisition of Pulsic is to introduce artificial intelligence into chip design. Although the automation level of chip design has continued to improve in recent years, but the design of analog circuits still has a high degree of dependence on manual layout, wiring, Cadence and other EDA companies and Pulsic, Astrus and other chip design solutions companies, are based on intelligent technology to improve the efficiency of the layout of analog circuits. Previously, Cadence launched the Virtuoso Studio solution, which supports customized chip design through AI, cloud, 2.5D/3D packaging, etc. Cadence hopes that this acquisition will integrate Pulsic's automation technology into Virtuoso Studio, and use AI to reduce the difficulty of analog chip design.
Typical B: Synopsys Acquires PikeTec
Synopsys acquires PikeTec, aiming for software-defined automobiles. As automotive electronic and electrical architectures become more complex, software-defined systems consisting of software, sensors, computing, and mechatronics become larger, and testing and verification efforts increase. Synopsys plans to add PikeTec's test automation tools and services to its own product line to provide automotive OEMs with more comprehensive virtualization and testing solutions.
Feature 2: Artificial Intelligence-related M&A "ripple effect"
The explosive development of artificial intelligence has blown up not only the AI field, but also the semiconductor industry. If you want to find a "covenant" for the acquisition of semiconductors in different application fields and different industrial segments in 2023, artificial intelligence is the right choice. Because of the "meat and potatoes" of artificial intelligence, software and hardware companies have occurred between the "cross-border" acquisitions, involving NVIDIA, AMD, Microsoft and other leading technology companies, but also by the industry degree of attention.
Acquisition reason 1: hardware and software synergy
In 2023, there are both arithmetic chip enterprises to acquire software enterprises, there are also software enterprises to acquire arithmetic chip enterprises. The purpose is to build software and hardware integrated AI solutions, so that users can conveniently deploy AI services according to their own scenarios and needs.
Typical A: AMD acquires two AI software companies
AMD acquired two AI software companies, Mipsology and Nod.ai, to improve the AI software stack, expand software tools, databases and model ecosystems, so that users can more easily deploy and run AI models on AMD hardware.
Typical B: Microsoft acquires DPU chip company
Microsoft, a major investor in OpenAI, has hosted ChatGPT on its own cloud service Azure and is pushing for the integration of ChatGPT with search engine Bing. A robust data center infrastructure is essential to support the invocation of AI software, applications and tools by a wide range of users. Microsoft's acquisition of Fungible is a DPU supplier, and DPU is known as the third main chip in the data center after the CPU and GPU, which can make the calculation closer to the place where the data is generated, and reduce the burden on the CPU, so as to improve the overall computational efficiency of the system. At present, Amazon, Aliyun and other cloud vendors have self-developed DPUs, and Microsoft's choice to catch up through acquisitions also reflects the fact that cloud vendors have placed hardware and software coupling in an important position when responding to the data deluge triggered by AIGC.
Acquisition Reason 2: Grabbing Edge AI
With the advancement of model compression technology and the improvement of edge device arithmetic, AIGC is sinking to the edge side. Since edge devices do not need to transmit sensitive data to the cloud or store it in external servers, edge AI has advantages in terms of privacy protection and cost savings. Based on edge AI, developers can enable end-products to benefit from machine learning without having to compromise on key metrics such as user privacy, power consumption, cost and real-time performance.
Typical A: NVIDIA's acquisition of OmniML
NVIDIA's low-profile acquisition of OmniML focuses on providing machine learning application miniaturization technology that compresses the volume of AI models and enables them to run on edge AI devices. Currently, AI businesses such as AIGC and big models commonly use data centers as their main infrastructure. Perhaps edge devices will become the next battleground for AI, allowing some of the capabilities of AI and even big models to come to consumers.
Typical B: Infineon acquires Imagimob
Infineon's acquisition of Imagimob, a provider of machine learning technology for edge devices, is an effort to strengthen its edge AI solutions and grab a head start in the fast-growing edge AI market.
In the coming years, AI and machine learning will enter embedded applications on a large scale, enabling new functionalities while protecting user privacy and improving control and energy efficiency. With this acquisition, Infineon increases its presence in machine learning solutions and complements its AI product line, enabling machine learning algorithms to run in type of MCUs to support customer innovation. Infineon wants to make it easy for every developer - even novices - to use Edge AI's tools and develop inference models.
Feature 3: Broadband semiconductor-related M&A to look forward to
In 2023, the field of wide-band semiconductor ushered in two large acquisitions - the so-called "large", refers to the acquisition amount exceeded the main wide-band semiconductor market size of the acquirer, or the wide-band semiconductor in a certain area of the market size of this type of wide-band semiconductor. This also reflects the confidence of the acquirer for the development prospects of wide-band semiconductor.
If you have product need，please send the inquiry here or email us :sales@lantronchips.com
