Stats

  • Locked Locked
  • Replies 3
  • Subscribers 126
  • Views 15929
  • 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

SPECTRE-18023: Parasitics RC reduction disabled (fast APS)

vivkr

Hi All,

I generally need to use fast APS (++aps) and parasitic reduction (++parasitics) options when running postlayout simulations with Spectre, in order to cut simulation times to manageable levels.

Now, I am suddenly getting the following warning in some cases:

WARNING (SPECTRE-18023): Parasitics RC reduction is disabled. Possible reason could be 'preserve_inst=all' is defined in the option statement


I also see that the simulation times are massively longer as a result. 

1) I am not specifying the option preserve_inst at al. Here's an extract out of my netlist:

simulatorOptions options reltol=1e-5 vabstol=1e-6 iabstol=100e-12 \
    temp=125.0 tnom=25 scalem=1.0 scale=1.0 macromodels=yes gmin=1e-12 \
    rforce=1 maxnotes=5 maxwarns=5 digits=5 cols=80 pivrel=1e-3 \
    sensfile="../psf/sens.output" checklimitdest=psf ignorezerovar=yes \
    colonasdelimiter=none

2) The situation doesn't seem to change even if I explicitly force preserve_int=none.
The above warning does not always seem to appear, but I don't know the exact cause/effect mechanism.

3) I am using MMSIM 15.10.284. Here's an excerpt from my spectre.out log showing the run options used

/cadence/mmsim/15.10.284/tools/bin/spectre  \
        -64 =log spectre.out ++aps ++parasitics=conservative +mt=8 =mdl  \
        mcarlo.mdl input.scs -raw psf -measure measure

4) I am running TRAN analysis with errpreset=conservative

5) The errpreset options, the standard simulator options are something I use as a default for everything that I run, and it seems as if parasitic reduction works for some cases and not for others, i.e. different netlists, different simulation setups.


Has anyone faced a similar issue? Any workarounds would be most welcome.

Thanks,

Vivek

Parents
  • vivkr

    Hi Andrew,

    Thanks for that. I am a bit late in responding but I had contacted our local customer support rep. In the meantime, I figured out the following (all non-representative mind you, since I am only looking at one testbench really):


    1) If I don't use spectremdl, but spectre, the warning goes away and reduction occurs.

    2) I can use spectremdl but remove the montecarlo analysis statement I was running inside it (moving it to my spectre base netlist), and reduction still occurs.

    So, it would seem somehow that trying to run montecarlo analysis out of spectremdl somehow interferes with parasitic reduction in this setup of mine. No idea why that ought to be the case, but there you are. I've passed on the info to our Cadence rep.

    Regards,

    Vivek

Reply
  • vivkr

    Hi Andrew,

    Thanks for that. I am a bit late in responding but I had contacted our local customer support rep. In the meantime, I figured out the following (all non-representative mind you, since I am only looking at one testbench really):


    1) If I don't use spectremdl, but spectre, the warning goes away and reduction occurs.

    2) I can use spectremdl but remove the montecarlo analysis statement I was running inside it (moving it to my spectre base netlist), and reduction still occurs.

    So, it would seem somehow that trying to run montecarlo analysis out of spectremdl somehow interferes with parasitic reduction in this setup of mine. No idea why that ought to be the case, but there you are. I've passed on the info to our Cadence rep.

    Regards,

    Vivek

Children
No Data

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.