The subject may not reflect the issue I'm facing since I don't know how to describe it well in a short sentence.
Here is a little background and the question that I have.
I'm using ocean script to run stability analysis over corners. When RC extracted netlist is being used, spectre has difficulties in finding DC operation solution. To deal with this issue, I run a transient first and then run stability analysis using the transient final state "spectre.fc" as the nodeset for stability analysis.
The following is part of my script. My intention is to run transient analysis with one netlist where supply DC voltage is 0, and then run stability analysis with another netlist where DC voltage is back to normal. This part works fine.
My problem is that the second run in the for loop still does a transient analysis. That will take an extra five to ten minutes for nothing. I'm wondering if there is any way to eliminate the transient sim from the second run. Thank you in advance for your help!
for(i 0 1
desVar( "avdd" 0); Set variable avdd to 0 for transient analysis
analysis('tran ?anaName "tran" ?stop "20u" ?errpreset "conservative"
?write "spectre.ic" ?writefinal "spectre.fc" ?annotate "status" ?finalTimeOp t
?maxiters "5" ?threshold "0.0" ?detail "node" ?sort "name" )
desVar( "avdd" Va); Set variable avdd to Va for stability analysis
analysis('stb ?anaName "stb" ?start "1k" ?stop "10G"
?probe "/Idiff" ?usingProbeMode t
?stbProbeMode "differential" ?readns "spectre.fc"
?oppoint "rawfile" ?annotate "status")
); end if(i == 0)
); end for
You can use delete('analysis 'tran) to delete the transient analysis, or delete('analysis) to delete all analyses (I think, if my memory is correct).
In reply to Andrew Beckett: