[SOLVED] How to extract a value from a self-defined plot(Cadence), maybe using ocean script?

[Alex Liao]

Hey,

I am using Cadence IC 6.1.6. The waveform or the plot is generated by
Virtuoso (R) Visualization & Analysis XL.
I have a plot which is gm/id (y-axis) Vs Id/(w/l) (x-axis) by doing a Parametric Analysis of sweeping Vgs.

I found in the result folder, it has lots of sub-folders named as Vg=0.111,..., Vg=0.555, Vg=0.666,..., Vg=0.999. Inside each of them, a /psf folder is there. This is reasonable as plenty of DC points are covered in the Parametric Analysis. Therefore all of those '/psf' result directories' are working together to compose a full plot by providing data and maybe do some interpolations.


I want to extract a value from this big plot and do not know how to do it.

Say, I can read when the X which is id/(w/l) = 5u, the Y which is gm/id = 10. But I want to get Y values by providing X values.

I am thinking, since the X-Y in the big plot are one-to-one corresponded, if all the data are summarized in one place as an uniformed waveform or maybe a big table, then giving a X and asking for a Y should be straightforward.

I am not quite sure if the waveVsWave will summarize all the data and provide a function to access it.
If ocean script can do this, that's best.

Any idea?

Regards,

Alex

 

[erikl]

Hi Alex,

I'd suggest to ask your question in a more specialized .

erikl

 

[Alex Liao]

Hi Alex,

I'd suggest to ask your question in a more specialized .

erikl

Hey Erikl,

Nice suggestion.
Got solved by professional Cadence people.

Thanks,
Alex

 

[erikl]

Would be nice if you'd show or link here the solution.

Thanks, erikl

 

[Alex Liao]

Just use the value() function it can do what I want.
It works for me when I use this formular:
value(waveVsWave(?y (pv("M0" "gm" ?result "dcOpInfo") / pv("M0" "id" ?result "dcOpInfo")) ?x (pv("M0" "id" ?result "dcOpInfo") / (pv("M0" "w" ?result "instance") / pv("M0" "l" ?result "instance"))) ) 10e-6 )

Please click the "helped me" if you real think so.

Thanks,
Alex