The value of selecting IP based on a platform
Sales of semiconductor IP, both hard IP and soft, continue to grow at a rate faster than that of the overall semiconductor industry. Comparing 2006 semiconductor sales to 2005, the Semiconductor Industry Association (SIA) recently reported 8.9% growth worldwide. According to Warren Savage, president and CEO of IPextreme, an IP licensing facilitation company, “the IP market generated approximately $1.6B in sales last year.” And according to Henke Associates, the top seven publicly traded companies in the IP space grew at a rate of 18.2% in 2006, nearly twice the rate of the semiconductor market as a whole.
This accelerated growth in the IP business is a result of shrinking semiconductor process geometries. At each successively smaller process node, significantly more functional content can be achieved on a single ASIC or FPGA. What was once a large multi-chip system can now be integrated onto a single chip. With markets demanding higher and higher levels of integration, it is no longer feasible for a development team to design all of the function needed by a large system on chip (SoC). Instead, development teams have turned to third party IP for the solution. IP providers, by offering both cores and application-specific domain expertise, enable product development teams to keep pace with the pressure to deliver larger SoCs on very rapid timetables.
IP for a particular function is often available from dozens of sources and packaged in as many different forms. However, whether the IP is provided by a large public company or a small firm specializing in a particular application domain, the primary needs of a customer are the same.
The desire to get to market fast (and before competition) is one of the primary motivators for using IP developed outside the company. Many companies consider the option of making IP in-house rather than buying it. In the end, however, most companies agree that tapping into the application and domain experience of the company providing the IP is perhaps even more important than the physical embodiment of the IP itself. Cost of the IP, ease of integration, support, and reliability are also key attributes companies look for when choosing an IP partner.
Dozens of the IP offerings available on the market today are offered as application-specific individual cores or as a suite of application-specific cores with different configurations that provide engineers with performance and area trade-offs. A simple search for MPEG IP on the Design and Reuse web site, for instance, results in 133 different products offered by 34 different companies. It’s a daunting task for a designer in need of an IP core to sort out the various costs, area, and performance trade-offs with so many individual component suppliers available.
IP companies with a core expertise or application domain experience have a clear advantage when it comes to IP selection. Given the fact that a company’s IP selection prioritizes domain experience ahead of the configuration metrics of the IP core itself, it is essential that IP providers identify their core expertise to the market. The majority of IP providers address this by offering more cores and more configurations for the same function. This approach falls short of providing real end-to-end solutions that many companies need to truly achieve faster time to market.
A better overall strategy for conveying core domain expertise is provided by those IP suppliers that offer IP cores in the context of an integrated platform. The platform often includes both software and hardware components. IP suppliers with comprehensive platforms not only provide peace of mind by conveying their application domain expertise, but also provide a true framework that enables design engineers to build end-to-end solutions that meet their needs.
Providing customers with solutions rather than discrete blocks is not a new concept in the semiconductor industry. Standard product chip providers moved to a model such as this in the early 2000s and it’s becoming evident that the same vertical level of integration is important in the semiconductor IP market as well. This is evidenced by Mentor Graphics’ recent announcement of a change in their IP go-to-market strategy. Mentor highlights their new IP subsystems implemented as a combination of hardware IP and software. Mentor’s core expertise is PCI, Ethernet and USB and they plan to offer integrated solution-based subsystems in these domains. Other companies (such as CebaTech and Ceva) with focused domain expertise are also entering the market with integrated IP platforms. CebaTech’s domain expertise is in data networking and storage; the area of strength for Ceva is DSP processing for the wireless consumer and multi-media markets.
There are several distinct advantages to selecting an IP partner that embraces a platform- based approach to IP. An IP platform provides a solution-based integrated framework that allows companies to use the hardware IP and software elements of the platform directly or to use the platform as a template for their own solution integration efforts.
IP partners with integrated platforms clearly understand the end market for their IP core offerings, how to use their cores in the context of the end market application, and how to support customers going to market with products targeting these end markets. This is in stark contrast to IP suppliers offering merely components, where little more than the component boundary is the extent of domain experience and application support.
IP platforms also can offer customers an implied roadmap for growth. The platform represents an end-to-end solution that can be embraced initially at the component level by companies that already have most of their solution in place. As the market and standards change, the platform approach allows these companies to quickly move to deeper levels of integration on the platform while preserving their initial component investment as a starting point for integration.
Indeed, the chasm between the early IP adopters and the pragmatic mainstream users appears to have been crossed. The value of using qualified third party IP has proven itself to be a tangible bottom line benefit. We can expect future year-to-year increases in the use of IP. And, with the increase in IC complexity, we also can expect leading-edge companies to meet their time-to-market pressures with a platform approach to IP rather than trying to fit together one-size-fits-all pieces of individual IP.
Joe Rash has over 18 years of experience designing and bringing to market complex ASIC and standard product ICs for the data networking industry. He has a BSEE from the University of Central Florida and holds seven patents in the field of chip design and data networking.
Related Semiconductor IP
- JESD204D Transmitter and Receiver IP
- 100G UDP IP Stack
- Frequency Synthesizer
- Temperature Sensor IP
- LVDS Driver/Buffer
Related White Papers
- Capitalizing on the Architectural Flexibility of FPGAs with RISC-V and a Simplified Programming Flow
- Understanding the Deployment of Deep Learning algorithms on Embedded Platforms
- Context Based Clock Gating Technique For Low Power Designs of IoT Applications - A DesignWare IP Case Study
- VLSI Physical Design Methodology for ASIC Development with a Flavor of IP Hardening
Latest White Papers
- New Realities Demand a New Approach to System Verification and Validation
- How silicon and circuit optimizations help FPGAs offer lower size, power and cost in video bridging applications
- Sustainable Hardware Specialization
- PCIe IP With Enhanced Security For The Automotive Market
- Top 5 Reasons why CPU is the Best Processor for AI Inference