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.

how to print all mos opertion point to a file with ocean

Status
Not open for further replies.

xianweng

Member level 1
Joined
Aug 26, 2011
Messages
33
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
1,535
hi eyeryone,

i am using IC 615 and MMSIM13.1. i am going to save all mos opertion point to a file with ocean.I know that i could save single mos op with a expression like op("/M0" "vds"),but there are a lot of mos in my schematic.So which expression i should use? i have tried op("/*" "vds"),but it is does not work .could you help me figure this out?

thanks you
 

In IC 5.1.41 I think we used
Code:
analysis('dc ?saveOppoint t )
Try it.
 

erikl ,thank you for you response .sorry,maybe i do not explain my problem clear.my code is blow,


Code dot - [expand]
1
2
3
4
5
6
7
8
9
ocnWaveformTool('wavescan)
simulator('spectre)
design("~/netlist")
resultDir("~/schematic")
modelFile('("~/eda/sms.scs" "tt"))
analysis('dc ?saveOppoint t)
temp(27)
run()
plot(OP("/M0" "gm"))



for example ,my schematic has hundreds of mosfet named M0,M1 ,M2,M3,N__3,N__79,I__N__5........,i want to plot or save gm of all mosfet,which expression should i use?is there any wildcard to represent all mosfet?
 
Last edited by a moderator:

Hi, if yoi know name structure of your MOS transistors in cell and maximum number in instance name, then you could use smthg like that:
Code:
for(i 1 11
a = i
a = sprintf(a "%d" a)
Gm = pv(strcat("I1.MN" a) "gm" ?result "dcOpInfo")
printf("GM of MN%L" i)
printf("is %L \r\n" Gm)
i = i+1
)
 

...my code is blow,

Code dot - [expand]
1
plot(OP("/M0" "gm"))


Instead you need:

Code dot - [expand]
1
2
selectResult( 'dc )
plot(getData("/M0") )



If it works for all nodes when you do it manually, you could save an ocean file from this run.
 

Oh, for the good old days of cdsSpice and

open 1 OPresults.txt
probe.1 op *
close 1

Not that hassling with debugging Ocean script for a
day or three isn't just as efficient. Yeah.
 

sarge,thank you for your response ,but the name structure of mos transistors is irregular.I guess dick_freebird's answer is great,maybe i could use hspice to do it

- - - Updated - - -

- - - Updated - - -

- - - Updated - - -

Instead you need:

Code dot - [expand]
1
2
selectResult( 'dc )
plot(getData("/M0") )



If it works for all nodes when you do it manually, you could save an ocean file from this run.

erikl,thanks again for your reply ,but i do not get it. i guess dick_freebird's answer is great ,maybe i coulde use hpsice to do it rather than ocean

- - - Updated - - -

dick_freebird,thank you !!!!
i guess your answer is great ,i do not have to debug ocean for a couple of days.Maybe i could use hspice,but do you happen to know do it with hpsice?
 

sarge,thank you for your response ,but the name structure of mos transistors is irregular.I guess dick_freebird's answer is great,maybe i could use hspice to do it

You can replace it for any structure you want. This is just example.
Anyway, you already have some other advices from dick_freebird and erikl, you can choose any case, which will be more suitable for you :)
 
If it works for all nodes when you do it manually, you could save an ocean file from this run.
erikl, ... , i do not get it.

Run your ADE analysis manually and select "all" nodes. If this succeeds to save all nodes' operation points, you can save an OCEAN file from this run (and perhaps change it manually, if necessary).
 
You might find there is a switch setting that prints all OP
info somewhere (whether you get to specify, or just in one
of the output / report files. I do not know, never got that
deep into Ocean scripting). But I'm thinking this may be in
the Spectre environment / setup files.

If you tell it to save OP, it has to be done somewhere.
Questions are whether, where, and how human-legible.

But I'm about a year west of my last Cadence session
and this is largely foggy conjecture at this point. There
could be some tutorials that could be searched, I imagine.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top