Sky-rocketing chip prices? Not with embedded FPGA
Article By : Andy Jaros
eFPGA has become a significant factor in chip design, particularly as the industry moves to the more complicated and expensive process nodes.
Today’s chip design market has become extremely competitive and when you add on the rising cost of semiconductor manufacturing, it’s no wonder why the embedded FPGA (eFPGA) market is taking off. In fact, Allied Market Research projects that eFPGA license revenue in 2024 to grow to $300 million, which is a compound annual growth rate of more than 50% per year for 5 years. The reason? Simply stated, eFPGA has become a significant factor in chip design, particularly as the industry moves to the more complicated and expensive process nodes.
Using eFPGA, chip designers are no longer locked in once RTL is frozen, but rather have the flexibility to make changes at any point in the chip’s life span, even in the customers’ systems. This eliminates many expensive chip spins and enables designers to address many customers and applications with the same chips. It also extends the life of chips and systems because designers are now able to update their chips as protocols and standards change.
eFPGA has also become highly reliable and easy to integrate, which is why it’s becoming one of the most important tools in a chip designer’s toolbox. As an example, one recent company that we worked with was able to tape-out in less than two and half months on a FinFET process. And more importantly, the silicon worked the first time. It’s a vital factor in the chip design world because eFPGA can end up saving millions of dollars while delivering unprecedented flexibility.
FPGA or eFPGA?
Many people think that eFPGA is the same as traditional FPGAs such as those offered by Xilinx and Altera. This is not the case at all. While the technology is similar, eFPGA requires no SERDES and PHYs because on-chip signaling is very fast. Although density is also very similar, some eFPGA platforms are much better than others, so designers need to do their homework and shop around for the best platform. The real difference is the users. FPGA chips are used primarily by systems companies, with some in high volume. On the other hand, eFPGAs are used primarily by chip companies that need to integrate a small amount of FPGA-like flexibility into their chips.
An FPGA combines an array of programmable/reconfigurable logic blocks in a programmable interconnect fabric. In an FPGA chip, the outer rim of the chip consists of a combination of GPIO, SERDES and specialized PHYs such as DDR3/4. In advanced FPGAs, the I/O ring is roughly 1/4 of the chip and the “fabric” is roughly 3/4 of the chip. The “fabric” itself is mostly interconnect in today’s FPGA chips where 20% to 25% of the fabric area is programmable logic and 75% to 80% is programmable interconnect.
In contract, an eFPGA is an FPGA fabric without the surrounding ring of GPIO, SERDES and PHYs. Instead, an eFPGA connects to the rest of the chip using standard digital signaling, enabling very wide, very fast on-chip interconnects.
An eFPGA is entirely different from FPGA in terms of design fabric and end application. Source: Flex Logix
Because eFPGAs are scalable from hundreds to millions of look-up tables (LUTs), SoC designers can select exactly how much reconfigurability they need, and in many cases can distribute multiple eFPGAs throughout the chip, locating them where needed rather than in a single large block. In addition, some systems that use power-hungry FPGAs in their systems can now integrate them into SoCs for lower cost, lower power, and higher speed.
Use cases expanding
With eFPGA now available for a wide range of processes, including 12 nm, 16 nm, 22 nm, 28 nm and 40 nm nodes, it’s being adopted in many markets and applications.
- Defense: DARPA is using it for chip development and recently expanded the availability of eFPGA to its research teams. Likewise, another defense company is leveraging it for design obfuscation where it can put its secret circuits and code into the eFPGA. It’s also highly advantageous to commercial customers such as set-top box providers, which sell to multiple carriers such as Comcast or Dish that have their own proprietary transport de-multiplexer IP blocks. The set-top box manufacturer gets one chip to service both markets, which can save millions of dollars in new chip development costs alone. In addition, the customers they sell to, such as the carriers, can personalize their end systems without having to share their trade secrets with the MCU manufacturer.
- Networking: Companies are using it so that their accelerators can now reside in the eFPGA on the same chip as the data processing unit (DPU), which reduces BOM costs as well as system power and improves performance.
- Wireless communications: Companies are leveraging eFPGA to make their radio base stations more flexible in order to allow for customization, regionalization, and real-time updating of protocols and algorithms.
- Aerospace: Firms such as Boeing are using it to enable their aerospace systems to be smaller, lighter and consume less power.
One exciting area where eFPGA is being leveraged is vehicle-to-vehicle communications. By using eFPGA, the communication algorithms in wireless devices can be easily modified to ensure they have the highest level of security. Rather than have a closed system waiting to be hacked, these next generation wireless devices can constantly adapt and update their algorithms to always stay one step ahead of any potential threats or hacking.
Another popular use case for eFPGA is in the fabless semiconductor market. Case in point: Socionext is leveraging eFPGA in a 7-nm ASIC being developed for a major communication company’s 5G platform. By leveraging eFPGA, Socionext can deliver a reprogrammable ASIC that can be reconfigured after tape-out to adapt to new requirements and changing standards and protocols as needed. By integrating the FPGA, Socionext’s customer can improve performance and reduce power by eliminating one chip in their system. This also delivers personalization benefits to carriers who no longer need to share their proprietary design with the ASIC provider in order to have it added to the FPGA.
The future
eFPGAs have evolved to the point that they are now competitive in terms of density with traditional FPGAs, but offer lower cost and reduce the need to service different customers or markets with different chips as well as extend the chip and system’s lifespan. These are significant advantages in today’s competitive chip design market where costs have skyrocketed and chips have become extremely complex.
eFPGA is changing the way chips are designed, providing a level of flexibility and reprogrammability that never existed. Many chip developers are already using eFPGA, and many more are in design or evaluation. Once chip designers enjoy the flexibility that eFPGA offers, they will never want to go back to the old process of being locked-in with their RTL.
This article was originally published on EDN.
Andy Jaros vice president of IP sales and marketing at Flex Logix.
