Go to playground home.

Performance Improvements


V15


AXIEM

  • Faster and more robust DC simulations with a true DC solver.
  • Faster simulations and lower memory requirements with thick metal simulations.

Analyst

  • Major reduction in peak memory usage when asking for fields at a lot of frequencies.
  • Simulate larger Analyst structures than previously possible on Linux simulation farms.
  • Distribute the solution for a single frequency to many Linux simulation nodes.

APLAC

  • Improved initial guess for multi-tone Harmonic Balance simulations.
  • Faster and more robust convergence algorithm.
  • Sparse matrix preconditioner improvements.
  • Faster and more accurate Fourier Transform implementation.
  • More memory-efficient transient analysis.
  • Faster model evaluation.
  • Improved netlist read-in performance.

While all performance improvements are problem-dependent, the items listed are those that R&D feels are significant.


V14


General

  • Better performance for extremely large VSS System Diagrams.
  • Faster layout rendering of complicated structures.

AXIEM

  • Dramatically improved performance for complex multi-layer boards.
  • More intelligent and easier to use geometry simplification for PCBs.
  • Improved iterative solver convergence at low frequencies above DC.

Analyst

  • New and improved hierarchical matrix solver.
  • Utilize resources better with improved parallelization.
  • Reduce simulation time with per-shape meshing and solver options.

Analyst Multi-physics

  • Annotation refresh 10-20x faster.

APLAC

  • Faster simulations with extremely large MDIF files.
  • Improved DC, Harmonic Balance, and Circuit Envelope, and Transient performance.
    • Faster DC convergence due to improved dynamic voltage limiting.
    • New HB coupled frequency pre-conditioner.
    • Faster TAHB using HBAT.
    • Improved HB piece-wise linear.
    • Envelope analysis supports trapezoidal integration.
    • Improvements Transient iteration, time stepping, and convolution.
    • Improved Transient sparse solver and faster pivoting sparse solver available.
    • Faster matrix reordering for large circuits.

Note that while all performance improvements are problem-dependent, the items mentioned are the ones that R&D feels are significant


V13


General

  • Improved Knowledge Base and Interactive Content make online help easier than ever
  • Dramatically faster simulated load pull
  • Improved computer resource allocation with new capability to pause remote jobs

APLAC

  • Faster simulations with huge s-parameter files
  • Improved linear simulation speed with multi-threading
  • Convert frequency domain models to the time domain more quickly and cache the answers for improved time domain simulation performance

Analyst

  • Faster meshing, matrix solve, and scaling
  • Solve only the area you need to with new Spherical PML for antennas
  • Compute far field without needing PML boundary condition
  • Create efficient meshes with re-sizeable wave ports

AXIEM

  • Get answers more quickly with enhanced parallelization in the iterative solver
  • Solve fewer frequency points using improved AFS algorithm

Note that while all performance improvements are problem-dependent, the items mentioned are the ones that R&D feels are significant