Stats

  • Locked Locked
  • Replies 1
  • Subscribers 91
  • Views 3483
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

how to correlate the correct RC factor?(SoC encounter)

summer

1. As the spef file extracted by QRC extractor is different from the spef files extracted by native detail RC extractor,I want to know the suggestede factor by "generateRCFactor" command,but the result factor is  0.67 which is outside the range of  0.8-1.2, i doubt the accuracy of this factor very much, Is that reasonable? how to get the correct factor?  (the detailed RC is in typical corner)

 

2.when I "setRCFactor" before "optDesign",how to define this cap and res factor?Is that from generateRCFactor? And how to set the specific value when I optimize setup/hold timing respectively?

 

Thanks a lot:)

  • Hi Summer,

    Great questions!

    First, I agree with your assessment that a scale factor of 0.67 is cause for concern.  While the raw value of the scale factor isn't nearly as important as the percentage of nets that are within +/-10% (or +/-20%) seeing it so low makes us wonder whether we're using a capacitance table that was produced from the same source as the icecaps.tch file that is driving QRC extraction.  If you're sure the technology files are aligned, and that both extractors are running on the same detail routed database, the next thing I'd recommend is that you run Ostrich to get more insight into the mis-correlation.  Ostrich is shipped with the software, and if "encounter" is in your path, "ostrich" should be too.  If Ostrich reports the same scale factor (which it should if generateRCFactor and Ostrich were run consistently) then you could look to see what the distribution looks like in Ostrich's plots.  This might give you some insight.  It would also be good to check what percentage of the nets are within +/-10% and +/-20% with the recommended scale factor applied.  I like to see 90% of the nets within +/-20%.  There's a short video on using Ostrich posted here:

    http://www.cadence.com/Community/blogs/di/archive/2008/10/15/an-interview-with-global-timing-debug-architect-thad-mccracken.aspx?postID=11887

    Regarding "setRCFactor"- the cap and res factors are set with the -detcap and -res options (that come from generateRCFactor -or- Ostrich):

    Usage: setRCFactor [-defcap <scaleFactor>] [-detcap
                       <scaleFactor>] [-xcap <scaleFactor>]
                       [-res <scaleFactor>]

    Setting the scale factors differently for hold optimization (while maintaining the scale factors associated with the setup view) requires a move to the Multi-Mode/Multi-Corner (MMMC) infrastructure.  Within MMMC, you can use the "create_rc_corner" command to attach scale factors to each view:

    Usage: create_rc_corner -name <rcCornerName> [-cap_table
                            <capTableFile>] [-T <rcTemperatureValue>]
                            [-res_factor <float>] [-default_cap_factor
                            <float>] [-detailed_cap_factor
                            <float>] [-xcap_factor <float>]
                            [-qx_lib_file <fileName>] [-qx_tech_file
                            <fileName>] [-qx_conf_file <fileName>]

    Hope this helps,
    Bob

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.