Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

[SOLVED] Process parameters of particular iteration in Monte Carlo Analysis (Spectre)

Status
Not open for further replies.
P

panchkapunch

Newbie level 6
Joined
Oct 16, 2010
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,364
Hi,
Suppose I run a Monte Carlo analysis in Cadence Spectre (both process and mismatch) for 100 runs.Now I want to replicate the data of the 10th
iteration(say) without monte Carlo analysis .How do I vary my model files to replicate the particular corner shown in Monte Carlo? That is:

1. where do I get the varied process (or mismatch) data corresponding to the particular run?
2.How do i modify my model file accordingly?
 

1. You should get this info from the MC sim. Output Log
2. If you have a single model file, copy it to a new model name and change the parameters accordingly. Then simulate this new model.

If you get process & mismatch information from a central .scs model file with several corner sections (typ, min, max, ss, fs, sf, ff) (or separate .scs files for separate corners), copy one of the corner sections (or files) to a new name (e.g. my_corner) and call this one for your simulation.
 
Ok...so when i open the logfile (in /psf directory) , i see it calls files like mc1-010_designParamvals.info and i see some values for process parameters there ..But if i look in the
correspoding process.scs file (called from central.scs model file), i see some different parameters , but i can see the parameters in mc1-010_designParamvaols.info in the file
statistics.scs (for Monte Carlo stats).How do i find the correlation between the 2 sets of data?
 

panchkapunch said:
... it calls files like mc1-010_designParamvals.info and i see some values for process parameters there ..But if i look in the correspoding process.scs file (called from central.scs model file), i see some different parameters
Click to expand...
I think the process.scs file just states the corner parameters; the MC analysis statistically selects parameter sets between b.c. and w.c. parameter sets.

panchkapunch said:
but i can see the parameters in mc1-010_designParamvaols.info in the file
statistics.scs (for Monte Carlo stats).How do i find the correlation between the 2 sets of data?
Click to expand...

The parameter set which you find in the mc1-010_designParamvals.info file (resp. in the associated (part of the) statistics.scs file) is that one you want to save/repeat.
 

I don't understand the term "b.c. and w.c. parameter sets. "

---------- Post added at 17:28 ---------- Previous post was at 16:59 ----------

ok..u mean best case and worst case ...got it now
but process.scs is not for corners, it is a simple model file for the process parameters (such as tox etc of MOS) it is the .scs file used for nominal run (no monte carlo)
 

panchkapunch said:
... but process.scs is not for corners, it is a simple model file for the process parameters (such as tox etc of MOS) it is the .scs file used for nominal run (no monte carlo)
Click to expand...

Ok, in this case the process.scs file fixes the nominal (or typ.) parameter set, and the statistics vary parameters (for process or mismatch) determine the variation range, see here: View attachment Cadence_Monte_Carlo_simulation_tutorial_p33.pdf .

There are so many different set-ups ...
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top