Stats

  • Locked Locked
  • Replies 15
  • Subscribers 145
  • Views 20711
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Question on OCEAN standalone evaluation

trdastidar

Hi,

Got a question on OCEAN scripts (same applies to artil as well)

I have an ADE L state, from which I save the OCEAN script, and then run it. It runs fine.

Suppose, there is a calculator function in this OCEAN script, which looks like this:

gain_1 = value(dB20(mag(v("/outp" ?result "ac-ac"))) 1)

plot( gain_1 ?expr '( "gain_1" ) )

Now, suppose I want to run the calculator function again (with possibly some modifications). Since the simulation is already run, I remove the entire top part of the OCEAN script, and replace it with a call to openResults, like below:

openResults( "results.psf" ) ; Replace "results.psf" with the actual path of the PSF directory

I run the OCEAN script, but it fails, with the following error:

> WARNING (OCN-6054): The output '/outp' you selected does not exist. Type outputs() to see the list

Now comes the funny part. I change the calculator expression to the following:

gain_1 = value(dB20(mag(v("outp" ?result "ac-ac"))) 1)

That is, I removed the "/" character in the signal name. The script now runs fine!

Could anyone tell me why this happens? I know the workaround for it, but would like to understand why that workaround is necessary at all.

  • I think this tells why they are not the same ---

    WARNING (SPECTRE-8281): `I0.rxcpdqrx.ctlep<5>' is not a node nor an instance name. 

     And I am already in touch with customer support (no resolution so far). Lets see... 

  • Well, the slashes in the save statement would clearly be wrong, because spectre doesn't understand names in Virtuoso's namespace (i.e. "schematic names") that I mentioned before.

    It ought to work with the dots and backslashes though. What version of spectre are you using? For example, I used this hand written netlist:

    //

    subckt myblock (a b)
    r1 (a bus\<1\>) resistor r=1k
    r2 (bus\<1\> bus\<2\>) resistor r=1k
    r3 (bus\<2\> b) resistor r=1k
    ends myblock

    b1 (n1 0) myblock
    v1 (n1 0) vsource type=sine freq=1M ampl=1
    save b1.bus\<1\> b1.bus\<2\>
    tran tran stop=1u

    and it worked fine. In your case I can't see the subckt definition, so can't tell if there's a mismatch or whether it's an issue with the spectre version you're using.

    Kind Regards,

    Andrew.

  • Thanks. Actually, you are right. Apart from nodes 'a' and 'b', there are add-on nodes 'bus1', 'bus2', etc. So b1.a, b1.bus1, etc. all carry finite meanings. I was initially trying to differentiate between nets (as used in the layout domain) and nodes (as used in the simulation domain). Thanks a lot for all the quick back and forth help. 

  • Hi everybody.
    I'm facing a similar problem as the thread opener but was not able to solve it with the hints given here.
    In ADE-XL I defined my outputs e.g. as follows:
    pvr('pss "/OutP" "/OutM" 200 '(1))
    pvi('pss "/VDD" 0 "/V0/PLUS" 0 '(0))
    Single runs and variable sweeps in ADE-XL are running without problems, but evaluation of monte carlo simulation fails due to missing outputs:
    WARNING (OCN-6054): The output '/OutP' you selected does not exist...
    WARNING (OCN-6054): The output '/OutM' you selected does not exist...
    ...
    When going to the individual result directories of the mc runs and checking the saved outputs using 'outputs()' I see that the relevant nets and currents should have been saved actually, but every mc run fails to access them.
    However, removing the '/' from the nets in the equations solved the problem but only for the expressions with voltage signals since I don't know how to proceed with the currents : e.g. /V0/PLUS
    Anyway, I would appreciate a solution without manually removing the '/' from the expressions which are actually working for simulations different than mc.

    The problem is not only limited to pss but also occurs for other analyses. I'm using IC6.1.6.
    Thanks for you help

  • You shouldn't need to remove the slashes in this case. I'm guessing it must be a specific problem with the subversion of IC616 you're using and maybe the combination of MMSIM version you're using. I suggest you contact customer support so that we can explore this in more detail.

    You should be able to find the netlist names by using outputs(?map nil) - and you should find that the currents through the voltage sources will be "V0:p". However, that's a workaround rather than a real solution to your problem.

    Regards,

    Andrew.

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.