Embedded FPGA design without hard barriers using OpenBus
By Rob Evans, Altium
Programmable Logic DesignLine (10/31/07, 10:38:00 AM EDT)
As the market-defining properties of today's electronic products continue to move into the software domain, a design's intellectual property (IP) increasingly is characterized by the unique 'intelligence' that is programmed into the product. In turn, the underlying physical hardware that supports this intelligence is playing a reduced role in defining a product's IP. This has changed the fundamental landscape of electronics design, allowing product developers who work in the software space to become the prime movers in defining the function and behavior of today's electronic products.
The explosion of programmable hardware components such as FPGAs over the last few years has allowed an even greater part of the design to move into the soft domain. Not only are these components becoming more capable, higher in capacity, and more feature-rich, they are becoming significantly cheaper. The availability of high-capacity, high-performance FPGAs at relatively low cost has opened up the possibility of using this technology to change the way designers interact with hardware and software by extending the concept of programmed device intelligence from just software to encompass both software and hardware.
With this approach, the system hardware itself can be defined in the soft realm – from mass logic through to high-performance microprocessors and matching system memory – allowing developers to create whole systems within the reprogrammable fabric of an FPGA. A much greater portion of the design process is performed in a soft domain, meaning that the defining value of that product has moved away from the unique properties of the physical platform. This hardware platform nevertheless supports the soft elements and forms the interface to the outside world, and therefore remains a large and crucial part of the electronic development process – and in the process, takes a proportionally large share of the product development time and cost.
Despite all of the revolutionary advances in electronics devices and how electronic products are designed, the process we use to develop and complete those products has generally failed to progress at the same pace. We still treat the design of the board-level hardware separately from the development of the software that runs on it, and programmable hardware design ends at the pins of the device – a separate discipline in its own right.
As more of a design is moved into a soft platform, the lines between the traditional design disciplines such as hardware, software, and FPGA design begin to blur. Dealing with these design elements independently and with separate tools becomes increasingly difficult and inefficient as design complexity increases and time-to-market cycles shrink.
The move to higher levels of abstraction within individual processes helps to cope with specific complexities, but at the same time increases the specialization required within each domain. Ultimately, of course, these individual elements of the design must be brought together to create a final product, but the increased specialization of each piece is making the final puzzle much harder to assemble. This blows out design times and limits product innovation.
Related Semiconductor IP
- TSMC 7nm 0V75 / 0V9 ESD Local Clamp – Low Cap
- TSMC 65nm 3V3 ESD Local Clamp – Rad Hard
- TSMC 5nm 1V8, 1.2V and 0.9V ESD Local Protection – Low Cap
- TSMC 3nm 3V3 ESD Local Clamp
- TSMC 3nm 1V2 ESD Local Clamp – Low Capacitance
Related Articles
- Control an FPGA bus without using the processor
- Add graphics without using a dedicated graphics controller
- Designing embedded SoCs using older resistive technologies
- Building a security-optimized embedded design using protected key storage
Latest Articles
- LACE: Large Language Model Aided Multi-Agent Framework for Agile RISC-V Instruction Extension
- A Process-Aware Hybrid Si/IGO Monolithic-3D 6T SRAM with BEOL Pass-Gates for the 2nm Node
- Automated Estimation of MBIST Area and Test Time in Heterogeneous Memory IPs via Stacked Ensemble Framework
- VIPER: Architecture-Aware Performance Modeling for Processing-in-Memory Design-Space Exploration
- CTTE: An Open Dual-Protocol RISC-V Trace Encoder for N-Trace and E-Trace