Stats

  • Replies 3
  • Subscribers 19
  • Views 376
  • Members are here 0

How do I choose the right simulator in AWR?

AnalogMonk

Hi all,

When I try to run a simple linear simulation, I see multiple simulator options like:

  • Default Linear
  • APLAC Linear
  • Spectre Linear

At first glance, they all seem to do the same thing (just linear simulation), but clearly they’re not identical.

My questions:

  1. Is Default Linear always enough, or should I be using APLAC/Spectre?
  2. Are these just different engines for the same math, or do they behave differently in practice?
  3. In what scenarios do you switch from one to another?
Parents
  • OscPn

    Hi  AnalogMonk,

    For most cases, the default Linear simulator is the fastest and best option.

    In some cases, APLAC Linear subcircuit caching can improve simulation time, so APLAC may be preferred.

    The Spectre solver is available only on Linux.

    Other solver engines are also available, such as APLAC HB in Microwave Office, and VSS has its own separate solvers.

    Depending on the application, you can switch between solvers, and each measurement in AWR automatically enables the appropriate solver.

  • AnalogMonk
    in reply to OscPn

    Thanks for sharing the details. I have few more questions related to this:

    1. Can you share a practical example where switching from the default linear simulator to APLAC resulted in a noticeable improvement in results or performances?

    2. When I select the spectre simulator on my windows machine it throws an error. However, I noticed that it generates a netlist (.scs). How did it generates without that solve? - snap attached 

    3. What is the purpose of 'configuration' button in the simulator settings? Are there any scenario or measurement where we can change that?

Reply
  • AnalogMonk
    in reply to OscPn

    Thanks for sharing the details. I have few more questions related to this:

    1. Can you share a practical example where switching from the default linear simulator to APLAC resulted in a noticeable improvement in results or performances?

    2. When I select the spectre simulator on my windows machine it throws an error. However, I noticed that it generates a netlist (.scs). How did it generates without that solve? - snap attached 

    3. What is the purpose of 'configuration' button in the simulator settings? Are there any scenario or measurement where we can change that?

Children
  • CurtisAWR
    in reply to AnalogMonk

    1. I don’t know off-hand of any specific examples where switching between Default Linear and APLAC Linear shows a noticeable difference in results / performance. However, you can reference this link from the relevant section of the Help Manual to see where APLAC Linear Subcircuit Caching can specifically result in performance improvements.


    2. Note that Spectre will yield a simulation in Linux, and I think you said you are in Windows. MWO in Windows will indeed generate the netlist, but you would need the solver to generate the data simulation, and this is a Linux requirement.


    3. The Configuration menu reflects a list of Simulation Configurations available (either as the Default configuration, or specific Switch Lists, defined by the user). The best way for you to learn about them is to click the “Help” menu on that dialog, and on the resulting Help page, to click again on the Switch View Concepts link in the Configuration row of the table. See this link for more information on how -- and why you might -- set up a Switch List.

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.