[Ngspice-users] Monte carlo issue solved

[John R. P <joh...@vi...>]

Dear Holger,
     Thank you for taking time to correct my monte carlo loop.
It worked....we were able to see 5 transient simulations with MOSFET
parameter vth varied in each transient simulation.We had to make a small
correction though.....i.e to put a *reset* command at the end of do-while
loop.
-------------------
dowhile count <= mc_runs
    altermod @TN[vth0]=gauss($n1vth0, 0.05, 2)
    altermod @TP[vth0]=gauss($p1vth0, 0.05, 2)
    print @TN[vth0]
    print @TP[vth0]
    tran 0.01n 1.2n
    write testMcarlo.raw
    if count = 1
        set appendwrite
    end
    let count= count +1
*reset*
end

*data are stored in 5 plots tran1 ... tran5
plot tran1.allv tran2.allv tran3.allv tran4.allv tran5.allv
.endc
.end
-------------------------------
 we observed a strange this while solving this: Only the first transient
simulation was correct and the remaining 4 were totally irrelevant.We
thought some variables were not re-assigned from the second do-while
looping.So we put "reset" command.Maybe it flushed all the variables of
first run.
Thank you for providing ngspice support.

-- 
Regards,
John