Slides for Richard - FCRP Review, Washington DC - PowerPoint PPT Presentation

1 / 4
About This Presentation
Title:

Slides for Richard - FCRP Review, Washington DC

Description:

E.g., shortage of qualified (analog) designers, and the issue of VLSI design education ... Become THE source of benchmark design examples and algorithm data ... – PowerPoint PPT presentation

Number of Views:23
Avg rating:3.0/5.0
Slides: 5
Provided by: Andre524
Learn more at: http://vlsicad.ucsd.edu
Category:

less

Transcript and Presenter's Notes

Title: Slides for Richard - FCRP Review, Washington DC


1
Calibrating Achievable Design Roundtable
Discussion June 9, 2002
Wayne Dai, Andrew Kahng, Tsu-Jae King, Wojciech
Maly, Igor Markov, Herman Schmit, Dennis Sylvester
Facilitator Bill Joyner, IBM/SRC
2
Attendees
  • Faculty
  • Herman Schmit, Wojciech Maly, Igor Markov, Dennis
    Sylvester, Andrew Kahng
  • Students
  • Michael Wang, Shuo Zhou, Bao Liu,, Stefanus
    Mantik, Yu Chen, Mike Oliver
  • Industry
  • Bill Joyner, Takahide Inoue, Simon Thomas, James
    Duley
  • Other Industry feedback
  • Sani Nassif

3
Discussion
  • Whats the big problem?
  • Cost tradeoffs need to be understood and modeled!
  • Comments 5 cents / mm2 cost with 90nm
    technology and 300mm wafers. Reference fab may
    be constructed and available possibly available
    to GSRC?
  • Variability
  • What variability is tolerable? (This may be a
    sub-text to cost.)
  • Technology utilization
  • E.g., How to best use 100nm
  • Comment we need a utilization vision, not just
    a technology vision (power? pin-limited?)
  • Sharing of information (across industry, across
    academia, across industry-academia)
  • E.g., shortage of qualified (analog) designers,
    and the issue of VLSI design education
  • Comment Sometimes design is limited by design
    talent, but the product still gets out (available
    talent is applied as best it can)
  • Is the solution education? (Re-education of old
    analog designers, education of digital designers
    in analog issues)

4
Discussion
  • Whats being done well?
  • Knowledge management for CAD and other concepts
  • Seeding and promoting ideas (collaboration,
    metrics, open-source) for the participants and
    within the GSRC and other centers
  • Motivating and justifying research directions for
    the other themes.
  • What could be improved?
  • Improved avenues for sharing data (esp. fab data)
    in addition to algorithms
  • Pushing others to use and adopt the Bookshelf
  • Become THE source of benchmark design examples
    and algorithm data
  • Explore and identify the future applications that
    will drive silicon consumption
Write a Comment
User Comments (0)
About PowerShow.com