Previous Blogs

May 20, 2025
Dell Showcases Silicon Diversity in AI Server and PC

May 19, 2025
Microsoft Brings AI Agents to Life

May 14, 2025
Google Ups Privacy and Intelligence Ante with Latest Android Updates

April 30, 2025
Intel Pushes Foundry Business Forward

April 29, 2025
Chip Design Hits AI Crossover Point

April 24, 2025
Adobe Broadens Firefly’s Creative AI Reach

April 9, 2025
Google Sets the Stage for Hybrid AI with Cloud Next Announcements

April 1, 2025
New Intel CEO Lays out Company Vision

March 21, 2025
Nvidia Positions Itself as AI Infrastructure Provider

March 13, 2025
Enterprise AI Will Go Nowhere Without Training

February 18, 2025
The Rapid Rise of On-Device AI

February 12, 2025
Adobe Reimagines Generative Video with Latest Firefly

January 22, 2025
Samsung Cracks the AI Puzzle with Galaxy S25

January 8, 2025
Nvidia Brings GenAI to the Physical World with Cosmos

2024 Blogs

2023 Blogs

2022 Blogs

2021 Blogs

2020 Blogs

2019 Blogs

2018 Blogs

2017 Blogs

2016 Blogs

2015 Blogs

2014 Blogs

2013 Blogs


















TECHnalysis Research Blog

June 4, 2025
Arm Brings Compute Platform Designs to Automotive Market

By Bob O'Donnell

While people have been talking about the automotive market as the next big tech device opportunity for some time now, in truth, today’s cars haven’t really advanced as far as many people hoped. Sure, we’ve got bigger, nicer screens and some level of assisted driving features, but we still haven’t gotten to the era of smartphone-level impacts for most of the vehicles we’re now buying or driving.

A big part of the problem is that the automotive development process continues to be long and tedious, in large part because of how technology gets adopted into modern vehicle platforms. Traditionally, car makers would wait until a given chip was design was completed, then integrate individual elements into more complete designs for their various car platforms and only then could they start writing software for those finished hardware designs.

Last year Arm took a big step towards reducing the standard 3- or 4-year development cycles by introducing two important innovations. First, they unveiled a full range of new automotive grade processor and other chip IP (intellectual property) that can be used in automotive platforms. Second, they debuted new cloud-based simulation and virtualization platforms in conjunction with major EDA (Electronic Design Automation) software providers Cadence, Synopsys and Siemens as well as Amazon’s AWS. These tools allow automakers and their partners to start writing software before final chips using Arm-based designs become available. The result is savings of up to two years on the automotive software development process, allowing significant time reductions in getting advanced capabilities to market. (It will be interesting to see how the recent export limitations on EDA software to China impact some of these developments for Chinese automakers.)

This year the company finished the process by integrating the different individual automotive cores into more complete system designs that can reduce both hardware design and software creation phases. In fact, the new offerings allow those steps to occur simultaneously, shaving up to another 12 months off the typical car platform development timeframe. Specifically, under the company’s new Zena brand for Arm-powered automotive silicon designs, the company unveiled an implementation of their Compute Subsystem (CSS) strategy for the vehicular market. As with other markets that Arm first brought CSS to—including datacenter, PC and smartphones—the idea is to integrate multiple components around Arm’s latest CPU, GPU and other core designs, thereby reducing the complexity, costs and time associated with designing these higher-level systems.

In the case of the Zena automotive options, CSS enables a variety of different system-on-chip (SOC) or “chiplet” designs that provide different combinations of CPUs, GPUs, ISPs (Image Signal Processors), security cores, I/O (Input/Output) components, inter-chip connectivity and more. Specifically, the new Zena provides support for a 16-core v9 Cortex-A720AE CPU, coherent chip-to-chip connections via the CMN S3AE, a real-time processing-capable safety island through the Cortex-R82AE, options to connect multiple Zena designs together through UCIe, and more.

The net result is multiple complete options from which to begin hardware designs and to which software can be written. The systems offer plenty of room for customization by individual semiconductor or automotive makers through the open hooks they enable to connect different combinations of AI accelerators, image processors, graphics processors and much more. Plus, different designs can be used to enable different types of applications from enhanced user interactions to more advanced in-vehicle infotainment (IVI) systems to assisted and autonomous driving features.

In fact, Arm believes it can drive the creation of what they’re calling an AI-defined car, which builds on the current concept of software-defined vehicles (SDVs) and brings more advanced AI processing into the experience. Given how rapidly AI has been integrated into other computing platforms, that seems a logical move—and the safety-focused capabilities Arm has integrated into the core Zena system will likely prove to be even more important here.

On a practical level, another key benefit of Zena is design efficiency for car makers. Importantly, it enables savings of not only time, but also the resources necessary to create vehicle platforms. This, in turn, lets automakers and Tier 1 suppliers create more platforms with less people (and in less time). In fact, Arm believes Zena CSS platforms can reduce the development costs/resources by as much as 20%.

For car buyers like us, that translates directly into getting later generations of technology into new car designs instead of buying cars that typically had semiconductor capabilities that were typically 2-3 years behind smartphones and other digital devices. For carmakers, this speeds time to market and allows companies to reuse certain hardware and software efforts across multiple product lines—instead of having to create new chip system designs and software for each range of models. Given how critical—and brand-defining—the digital capabilities of cars have become, this makes these development advances extremely important.

For Arm, this latest announcement is strategically important because it helps both extend and tie together the company’s effort in moving beyond individual chip IP into more complete system designs. This work to move up the value chain with CSS obviously allows the company to charge more for their offerings, but more importantly, it increases the significance of Arm platforms across all the markets in which it participates, including now automotive. Given that the company has increased its shipments into the car market by greater than 300% over the last 5 years and supports 94% of global automakers, it also seems clearly aligned with market demands as well, making it a win-win proposition.

In fact, as important as CSS platforms have proven to be in other markets where Arm has introduced the concept, the automotive market is the one that will most likely see the biggest benefit of it. The notoriously slow development times for car technology have been a challenge for many automakers and their major suppliers and the dramatic potential speed-up that Arm is bringing to them with Zena should prove to be a real difference maker.

Here's a link to the original column: https://www.linkedin.com/pulse/arm-brings-compute-platform-designs-automotive-market-bob-o-donnell-rg23c

Bob O’Donnell is the president and chief analyst of TECHnalysis Research, LLC a market research firm that provides strategic consulting and market research services to the technology industry and professional financial community. You can follow him on LinkedIn at Bob O’Donnell or on Twitter @bobodtech.