Stats

  • Locked Locked
  • Replies 4
  • Subscribers 125
  • Views 12010
  • 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

spectre alter in MTS environment

drdanmc

Is there any way to tell an alter statement to look for a parameter in the same scope as where the alter statement appears?  For example, suppose I have a really basic model file that includes some alter statements like this:

// myfile.scs
section params
    parameters VDD=3.3
endsection params

section nom
    include "myfile.scs" section=params
    ALT_VDD alter param=VDD value=3.3
    ALT_temp alter param=temp value=27
endsection nom

section slow
    include "myfile.scs" section=params
    ALT_VDD alter param=VDD value=3.0
    ALT_temp alter param=temp value=125
endsection slow

When I include this in a non-MTS simulation the alter statements work.  But now if I use this as an MTS model file in something like:

// MTS block
subckt tech1_cell IN OUT
    include "myfile.scs" section=nom
    // rest of technology #1 netlist
ends tech1_cell

// MTS block
subckt tech2_cell IN OUT
    include "myothermodels.scs" section=nom
    // rest of technology #2 netlist
ends tech2_cell

ITECH1A (T1INA T1OUTA) tech1_cell

ITECH1B (T1INB T1OUTB) tech1_cell

ITECH2 (T2IN T2OUT) tech2_cell

// rest of technology independent netlist
// spectre analysis statements

I have a problem which is the alter statements in the "myfile.scs" model file run in the top level scope and look for a top level parameter called "VDD" instead of the parameter that is local to the "tech1_cell" scope.  In addition, the temperature change is global and not to the two instances of tech1_cell.  I can't use "sub=ITECH1A" on those alter statements because the model file is not supposed to need to know details about the circuit that uses it and on top of that in this case I instantiated 2 instances of "tech1_cell".  

Am I missing some useful keyword that will make it all just work or is this one of the things that MTS sims just don't really support at this time?

Parents

  • Hi Dan,

    This is a very odd way of trying to achieve this, and not how MTS is intended to work. I don't think there's any expectation that alter (I had a discussion a few years back about altergroup being used in a similar way because of a peculiarity of one customer's models, and that just fails by saying that altergroup can't be used in subckts). 

    It could just define the parameter - or set the temperature as an option - rather than doing an alter (you might need to set redefinedparams to allow the redefinition).

    I suggest you contact customer support so that it can be investigated more closely.

    Andrew

Reply

  • Hi Dan,

    This is a very odd way of trying to achieve this, and not how MTS is intended to work. I don't think there's any expectation that alter (I had a discussion a few years back about altergroup being used in a similar way because of a peculiarity of one customer's models, and that just fails by saying that altergroup can't be used in subckts). 

    It could just define the parameter - or set the temperature as an option - rather than doing an alter (you might need to set redefinedparams to allow the redefinition).

    I suggest you contact customer support so that it can be investigated more closely.

    Andrew

Children
  • in reply to Andrew Beckett

    The dificulty of something like:

    OPT_temp options temp=110

    instead of the alter is that it appears I have to give numeric values everywhere.  So for example, I can't give names to my supply and temperature limits in one section and refer to those everywhere.  

    section globals

    parameters temp_hot=125

    parameters temp_nom=40

    parameters temp_cold=-20

    // supply values to follow

    endsection globals

    section nom

    include "myfile.scs" section=globals

    // fails claiming unknown parameter "temp_hot"

    OPT_temp options temp=temp_hot

    // works outside of the MTS context

    ALT_temp alter param=temp value=temp_hot

    // rest of models

    endsection nom

    In a somewhat ironic result, it appears that the scoped options try to resolve the parameter name "temp_hot" in the global context!  Meaning if I define "temp_hot" as a design variable that ends up as a parameter at the top level of the netlist, I can pass that down.  So I could hard code the numbers but it is very nice to have the numeric values contained to a single spot.  Again it appears that the options approach would work outside of MTS but not inside of MTS.

  • in reply to drdanmc

    The other challenge associated with using `options` instead of `alter` is that in a non-MTS setting, any `options` statements that set temperature in a model file are likely to be overwritten by the `options` statement generated by the netlister.  

    I've opened a case with customer support and see if they can dig deeper with me.  Thanks.

    -Dan

  • in reply to drdanmc

    Thanks for taking this up with support. I'd love to have tried this out, to make sure I give you a reasoned answer, but unfortunately I don't have the bandwidth or head space to be able to do that this week (or next), so support is the best approach here.

    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.