Semiconductor and electronics companies are trying to make more use of data from the chip development process. NEITHER offers a helping hand.
Designing a modern chip is an extremely complicated process that can take years, requiring hundreds or even thousands of engineers to test the chip at every stage of the development process, from concept to mass production. Testing a modern chip creates large amounts of data that can be used to troubleshoot the device or improve its performance. But industry experts say companies are struggling to take full advantage of it.
According to Ritu Favre, vice president and general manager of the semiconductor and electronics business unit at NI, “The past few decades have focused on design and test automation, which generates a large volume of data. Our clients want to leverage this data to enable better project visibility and better decision making. »
To accelerate the pace of wireless, semiconductor, and electronics innovation, NI is releasing its latest analysis software called DataStudio. It is intended to bridge data from the early stages of chip design through production testing, providing customers with deeper insights and greater visibility into the development lifecycle.
Featured at its annual NI Connect event in Austin, Texas, DataStudio connects data from software tools and other test equipment used at various stages of the development process, from product definition, chip design, and verification to laboratory validation and production testing on the factory floor.
David Hall, semiconductor and electronics marketing manager at NI, said the new DataStudio software will help customers reduce their time to market, even as they struggle with levels of complexity. unprecedented. “They not only want to better understand their data, but they also want to improve collaboration between engineers at different stages of the process.”
Gaurav Verma, senior director of engineering at Qualcomm, said the current pace of chip development is faster than ever. Ten years ago, it took the smartphone chip giant at least 24 months to go from prototype to customer sample. Shipping one million units would take up to eight months on top of that. But market pressures have pushed the company to the point where it needs to complete the process in just six months.
“One of the biggest challenges is the sheer complexity that has moved from 3G to 5G,” Verma said. NI works closely with Qualcomm and other major chip players such as Analog Devices and NVIDIA.
There are several important steps in the chip development process. In the pre-silicon phase, companies use electronic design automation (EDA) tools from Cadence, Synopsys, and other vendors to run simulations that help their engineers design, model, test, and verify chips before they do not go into mass production. This gives them the ability to evaluate a chip’s performance and identify any issues before building physical prototypes.
Once a company has passed design and verification, it moves into the validation phase. At this stage, he rigorously tests prototypes in the lab with hardware (such as NI’s PXI-based modular test equipment) and other systems.
The last step in the process is production testing, where engineers test thousands to millions of chips inside the factories where they are made. According to NI, each step in the process generates huge amounts of data.
Engineers are typically assigned to different departments overseeing a single stage of the development process, using test equipment and other tools from different vendors and widely incompatible data formats.
“With today’s simulators, it’s hard to tell where the problem is with prototype semiconductors if they don’t meet all specifications. Is there a particular part that is defective? Was it something in the manufacturing process that went wrong? Is this a fundamental flaw in the design? says Favre of NI. “To debug this, engineers have to work against these different silos, using different tools from different vendors with different datasets.”
As a result, data that could be used to improve the performance of chip designs gets lost when translating from one stage of the development process to another, said George Zafiropoulos, vice president of solutions marketing at NI.
“Very rarely do we find someone trying to hook everything up,” he explained. “You need more infrastructure to put the data together; otherwise, it can get out of control and become very ineffective.
Offered in on-premises and cloud deployments, DataStudio catalogs data from the entire product development lifecycle so that engineers from different departments can use it to collaborate on projects.
NI is also introducing a set of software applications that run on top of DataStudio. The first offering in NI’s DataStudio family, called Specification Compliance Manager (SCM), is designed to test various chips and other electronic devices against their original specifications.
Every chip design starts with a company creating a list of specifications for the final product, including the characteristics of the underlying hardware, the interfaces used by the software to control and communicate with it the performance of the part at a specific temperature. In other words, a pre-datasheet datasheet. But due to unforeseen challenges or pressure from competitors, companies frequently change their specifications.
DataStudio saves test results and other data to a central repository while managing device specifications and creating compliance reports. NI said this gives semiconductor and electronics companies more visibility into where products meet the rules and where they fall short of specifications. Customers can leverage information from DataStudio SCM to identify areas for improvement in the final design.
Not only does this help improve productivity and reduce the manual labor required during chip development, but it also allows companies to more closely track the progress of engineering projects over time, NI said.
“You can track exactly what’s going on with every aspect of the part, whether it’s 50 specs or 5,000,” Zafiropoulos said. “It just depends on the complexity of the device.”
According to NI, DataStudio will begin by supporting Cadence’s analog, mixed-signal, and RF simulation software. For validation, DataStudio will support data from the NI test system validation bench and any hardware test platform that can support DataStudio’s simple data format. This way, NI customers or anyone else using hardware controlled by the company’s test software environment can save their data in the correct format.
Additionally, NI deploys DataStudio Bench Data Connector (BDC). The Validation Bench Test Library provides customers with a standard way to convert test equipment data that is automatically compatible with the DataStudio SCM.
For the production test, the company said it is linking to standard data formats that will be released in the future.
NI said that while giving companies more visibility into the chip development process, the new DataStudio software is not designed to pinpoint areas for improvement in a chip’s design or identify the cause of a problem.
There are plans to upgrade DataStudio software with machine learning to help identify specific problems in the design of a chip or other electronic device.