NoC Silicon IP for RISC-V based chips supporting the TileLink protocol
Platforms key to cost reduction: Rhines
EE Times: Latest News Platforms key to cost reduction: Rhines | |
Brian Fuller (05/05/2004 7:00 AM EDT) URL: http://www.eetimes.com/showArticle.jhtml?articleID=19502233 | |
PRAGUE, Czech Republic The only way to continue the march of electronics innovation in the midst of sometimes bedeviling complexity is for the industry to focus on developing platform-based design methodologies using robust tools, the chief executive of Mentor Graphics Corp. said Wednesday (May 5, 2004). Wally Rhines, speaking at an event here, said the industry is splitting into two distinct groups, those that create design platforms and those that use them to design systems. "What's really needed is an entire design platform where you have available to you embedded microprocessors, digital signal processors, standard busses, tightly coupled peripherals, loosely couple peripherals," he said. "And what you'd like as a system designer is to be able to plug and play with those components without all the detail without having to design each gate individually. In fact that exists today. It's not widely used." Rhines, in framing Moore's Law as an empirical observation rather than a law, noted that baseline estimates for designing an ASIC today start at $20 million, far too expensive a development cost for most markets. Structured ASICs, in which a handful of masked layers are left open to programmability late in the design process, are still struggling for acceptance. Some companies have moved to a higher level of abstraction, releasing designs at the register transfer level (RTL) stage or compiling straight from C and C++, he noted. "Ultimately we will have to address a change in ASIC metholdogy," he said. "It's really only with a (platform-based) method like this that we can go from that $20 million down to $2 million or $3 million that enable an ASIC design world that can attack lower volume applications," Rhines said. Pounding away at other segments of the industry, Rhines noted that both lithography and test costs are soaring. Lithography machines now cost tens of millions of dollars, and while the manufacturing cost per transistor has decreased over time, the test cost per transistor has not. "This cannot continue indefinitely," he said. The EDA industry has responded to a degree by offering relatively new design-for-manufacturing and design-for-yield tools to help address some of these problems, he noted.
| |
All material on this site Copyright © 2005 CMP Media LLC. All rights reserved. Privacy Statement | Your California Privacy Rights | Terms of Service | |
Related News
- Real Wireless Research Shows One Third Reduction for Private Network Infrastructure Cost Using AccelerComm 5G physical layer IP solution
- Arm battens down the spending hatches
- Accusonus Achieves 60 Percent Reduction in Computational Cost of Speech Enhancement Software for Cadence Tensilica HiFi DSPs
- Consumer Electronics Application Processor Hits the Performance Cost Reduction Sweet Spot
- Xilinx and Sumitomo Electric Collaborate on Smarter Solutions that Reduce Network CapEx and OpEx Costs
Breaking News
- HPC customer engages Sondrel for high end chip design
- Ubitium Debuts First Universal RISC-V Processor to Enable AI at No Additional Cost, as It Raises $3.7M
- TSMC drives A16, 3D process technology
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- TSMC drives A16, 3D process technology
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
E-mail This Article | Printer-Friendly Page |