Stats

  • Locked Locked
  • Replies 3
  • Subscribers 126
  • Views 8206
  • 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

DC and STB analysis reuse in larger sweeps

jehh

Hi,

I have observed that running a DC but especially STB analysis will provide different results, depending on whether it is part of a larger PVT sweep or singled out corners.

When running larger sweeps, does Spectre reuse information, in such a way that it can affect the end result? Is there some flag / option that can disable this behaviour?

Or is this some completely different phenomena?

BR,

Christian

Version numbers:
Virtuoso IC.6.1.8-64b.500.23
Spectre 21.1.0.303.isr5

  • Hi Christian,

    yes, in sweeps Spectre is reusing the results of previous points to faster converge. So there can be a difference in the results depending on the analysis setup.  Here is a public solution which describes it and how to control it in Spectre.  This also can be controlled from within the ADE UI in the analog options or  the analysis setup:

    Article (20495347) Title: Understanding difference in results when parametric sweep is used with DC sweep w.r.t. standalone DC sweep
    URL: https://support.cadence.com/apex/ArticleAttachmentPortal?id=a1O3w00000AE7VIEA1

    Kind regards,

    Marc

  • in reply to Marc Heise

    Although I wouldn't expect this to be the case with PVT sweeps (Marc's response is definitely right for some kinds of sweeps, such as Monte Carlo or native sweeps inside the simulator) - those should end up being discrete simulations and (unless you try hard using saved node sets etc) wouldn't re-use anything from previous simulations.

    I suggest you contact support so that we can look at what is going on.

    Andrew

  • in reply to Andrew Beckett

    Hi Andrew,

    Adding the swpuseprevic to the spectre arguments most definetly did affect the outputs of the sweep, allthough by very a few hundred uV changes. This was using a standard corner setup in ADE Assembler.

    It is unfortunately not always consistent, and small changes to the schematic will affect which points fail. If I do end up with a minimal representative case, I will contact customer support. However, for now, I continue with the fix proposed by Marc.

    /Christian

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.