EDA is not enough!
By Michel Tabusse
edadesignline.com (March 17, 2010)
Good EDA tools, even combined within well-automated flows, are not enough to produce quality designs, whatever those designs are for software, systems-on-chip (SoCs), integrated circuits (ICs), intellectual property (IP) or embedded systems. Why is quality so difficult to achieve? Here are some of the things we are finding:
- Quality is often not defined operationally, making measurement and reporting onerous.
- Tools may be used incorrectly.
- Quality reporting is often informal, not objective, or comprised of too much information to be actionable.
- Worldwide teams and concurrent IP/ SoC/ software design produce burdensome quality monitoring overhead.
- Quality compromises tend to be made in order to meet tight schedules.
How does one define quality measures so that they can be easily deployed and used? Every time there is a panel on quality, designers and design managers realize that a huge amount of question-and-answer time is spent on defining quality criteria. And that quality is not the same for every type of design or every company.
E-mail This Article | Printer-Friendly Page |
|
Related Articles
- SoC design: When is a network-on-chip (NoC) not enough?
- Agile Analog's Approach to Analog IP Design and Quality --- Why "Silicon Proven" is NOT What You Think
- An Unexpected IoT Problem: Not Enough Randomness
- It's Not My Fault! How to Run a Better Fault Campaign Using Formal
- Enough of the sideshows - it's time for some real advancement in functional verification!