-------------------------------------------------------------------- -------------------------------------------------------------------- SEMCAD Matterhorn V16.2.4 Release Notes -------------------------------------------------------------------- -------------------------------------------------------------------- Release Date: 20.03.2017 SEMCAD Matterhorn V16.2.4 is a release based on SEMCAD Matterhorn V16.2.0. SEMCAD Matterhorn V16.2.4 contains the novel adaptive subgridding feature, a new automated parameter sweep engine (preview), powerful toolsets for 5G as well as various enhancements and bug fixes. This document summarizes the features, modules, enhancements, known bugs, workarounds, and other important issues to be aware of in this release. -------------------------------------------------------------------- -------------------------------------------------------------------- TABLE OF CONTENTS -------------------------------------------------------------------- 1. Installation Requirements 2. Supported OS 3. New Features & Enhancements 3.1. Licensing & Installation 3.2. Modeling 3.3. Simulation 3.4. Analysis 3.5. ARES 3.6. Python 4. List of Main Bugs (Selection) Fixed Since Last Release 5. Known Issues -------------------------------------------------------------------- -------------------------------------------------------------------- 1. Installation Requirements -------------------------------------------------------------------- - Administrator rights are required during the installation process. - NVIDIA graphics card, with support for OpenGL versions higher than v3.0. -------------------------------------------------------------------- 2. Supported OS -------------------------------------------------------------------- - Windows Server 2012 R2 - Windows 7 64 bit - Windows 8 64 bit - Windows 8.1 64 bit - Windows 10 64 bit SEMCAD Matterhorn V16.2.4 uses Acceleware's Napier library version 11.1.11. Please get the latest NVIDIA driver for your hardware from the Acceleware website: http://www.acceleware.com/fdtd-11.1.11. Please note that only NVIDIA GPUs with CUDA Compute Capability 1.3 or higher are supported by the S4L CUDA solver (e.g., Tesla C10XX). The AXE solver supports only CUDA Compute Capability >= 2.0. -------------------------------------------------------------------- 3. New Features & Enhancements -------------------------------------------------------------------- 3.1. Licensing & Installation - Advanced options after successful license installation now allow: - Multiple license file/server install. - Redundant and/or triad license server install. - Improved license feature information. - Improved log of license operations. - Automatic opening/close of firewall for at install/uninstall time for ARES and main application processes. 3.2. Modeling - Various surface mesh related modeling tools have been added. - The mesh doctor allows repairing and reconstructing surface meshes. - New modeling tools for phased arrays. 3.3. Simulation - Support for overlapping subgrids in FDTD simulations. - Adaptive placement of (overlapping) subgrids. - Significant speed improvements to the LF-MQS solver. - The filtering for user-defined source in the FDTD simulator can be disabled. - Support for phased arrays in the FDTD simulator. - Supporting the latest NVIDIA Pascal architecture for Tesla and GTX/Titan series GPUs. 3.4. Analysis - Cleanup and improvements in viewer options. - Improved support for masked regions in slice field and surface viewers. - Stability improvements when working with pipelines which are being evaluated. - Support for phased arrays in the simulation combiner and far-field combiner. - Support for import of measurement data from SPEAG's DASY6 scanner. - Matching network tool supports import from and export to Optenni Lab. - Matching network tool support generic import and export via Touchstone files. - Poynting vector based evaluator for 5G compliance (preview). - Improved viewer for tabular result data. - Movie export for viewers with animation support. 3.5. ARES - Job priorities can be changed after submission. 3.6. Python - Various improvements to the python API. - Ability to create script from analysis pipelines in the GUI. (Note that this currently does not support all algorithms.) 3.7. Interface - Orbit & pane modes to better support systems which have no mouse attached. - Readability & usability improvements to the distance tool. - Voxel related actions are available irrespective of whether the voxel folder is selected in a rectilinear simulation. 3.8 Parametrization - Improved and enhanced support for parametrization of model entities, simulation parameters and post-pro algorithms. - The analysis environment allows to build pipelines from simulations without results. Simulations will run upon pipeline evaluation. - PREVIEW: New automated parameter sweep engine to automatically evaluate parametrized simulations. - PREVIEW: New post-processing algorithms and viewers for sweep results. -------------------------------------------------------------------- 4. List of Main Bugs (Selection) Fixed Since V16.0.2 Release -------------------------------------------------------------------- - Case 79528: Fixed issue with the input power calculation in simulations with waveguide sources - Case 79733: Fixed an issue where field sensors recording regions in subgrids (non-harmonic excitations) recorded with too low sampling rate - Case 80866: Fixed issue of point sensor normalization - Case 81253: Fixed issue related to surface voxeling in Topological Voxeler - Case 81414: Fixed grid computation when User-Defined Signal used - Case 82735: Fixed issue in To-Python when Huygens source used - Case 82816: Fixed issue of slow multiport simulation extraction when the number of ports is large - Case 82975: Fixed issue in dragging planes to assign as boundary conditions in MS Vector Potential - Case 84109: Fixed some inconsistencies in Far-Field details report - Case 85867: Included error message when trying to run simulation of plane-wave sources with zero padding - Case 85879: Fixed interval settings in Surface-J Vector Field Viewer - Case 86416: Fixed connectivity check for subgrid - Case 87575: Fixed licensing issue/inquiry of AXE/subgrid - Case 87745: Fixed plotting of arrows in LF which hangs for coils made of several segments - Case 87829: Fixed FDTD user-defined didn't accept non-uniform signals - Case 88295: Fixed item with Simulation combiner after MATCH - Various fixes in 2D plots - Multiple causes for crashes have been addressed -------------------------------------------------------------------- 4.1 Update: List of Bugs (Selection) Fixed Since V16.2.2 Release -------------------------------------------------------------------- - Case 74833: Correctly store material parameters (eps,mu) in Waveguide mode selector - Case 79733: Fixed issue with Overall Field sensor normalization when an edge source is inside a subgrid region - Case 87840: Fixed bug which caused boxed subgrids to be wrongly merged and prevented the user from properly creating multiple boxed subgrids - Case 87944: Fixed issue that prevented extracting external sources from MQS solver - Case 88511: Fixed backward-compatibility issue in Python API (SliceFieldViewer, Slice) - Case 88758: Fixed consistency check preventing users from running MQS simulation with external Vector Potential - Case 88966: Fixed serialization issue in SAPD evaluator - Case 89011: Fixed occasional GUI crash in Worst-Case SAR Optimizer - Case 89082: Fixed occasional GUI crash when updating grid - Case 89450: Fixed issue in current sensor if 'Interpolate to exact model location' is checked - Case 90140: Fixed GUI issue when Coax Port Dimension was less than 0.5 mm -------------------------------------------------------------------- 5. Known Issues -------------------------------------------------------------------- - Case 32807: Instability for some rare dispersive material configurations - Case 40481: Python memory usage occasionally still occupied after resetting analysis - Case 47035: Random problems in visualization of Dispersive Viewer - Case 54000: Random element appearing in Gridder - Case 54179: Occasional under-performance of CUDA solver compared to SEMCAD X 14.8 - Case 58852: Artwork executable detected by Antivirus - Case 61881: Rare issue with distribution of the simulation job to the GPUs when using CUDA - Case 66068: Rounding precision issues makes partial/total radiated power appear larger than input power in some FDTD simulations with few or no losses