Lead Verification Engineer

Synopsys and Cadence Design Systems dominate the chip design software market, yet both stocks have fallen over the past year despite strong semiconductor sector performance. Synopsys generated $9.42 billion in revenue, growing 42.4% last quarter largely through its Ansys acquisition. However, the deal brought significant costs: operating margin sits at 14.52%, total debt reached $10.84 billion, and return on equity fell to 3.66%. Cadence achieved 24.2% organic revenue growth to $5.84 billion without major acquisitions. It maintains a 28.58% operating margin and carries just $2.65 billion in debt. The valuation gap reflects these differences. Cadence trades at over 15 times revenue versus Synopsys's 8.5 times. On trailing earnings, Synopsys appears more expensive at 73 times versus 65 times for Cadence. Both companies benefit from customer lock-in, as switching design tools requires costly revalidation and retraining. AI chip designers have strengthened demand for both vendors. At Q2 2026, 85 hedge funds held Synopsys shares worth $4.1 billion, whilst 69 held Cadence positions valued at $3.2 billion.

Cadence BETA CAE announces the release of 2025.2.3 version of its software suite. About this release. Cadence Design Systems, Inc. officially announce the third hot fix release of 2025.2 for the Cadence BETA CAE ANSA Pre-Processor, EPILYSIS FEA Solver, META Post-Processor, FATIQ Platform, SPDRM Software and KOMVOS SDM Console with critical bug fixes. Known Issues Resolved in ANSA Pre-Processor. Focusing on "Isolate flanges proximity" functionality, unmeshed faces thickness would wrongly be disregarded, and solid thickness value would be used instead. Batch Meshing Certain CFD anisotropic mesh cases could occasionally lead to an unexpected termination when setup was taking place via ANSA Parameters. Volume Meshing Abrupt changes in layer heights could occur in corners where layers would connect to inclined sides. Connections and Assembly Realizing a Y-JOINT-SHELL with 'Preserve mesh' options could produce wrong results. In addition, after executing 'Erase FE' of a SPIDER2 connection realized on the 'Top Area' of a stamp feature, the elements filling the hole would be added to the 'Rest Area' of the stamp feature, causing a subsequent realize action to fail. Compare Tool Opening the 'Compare Tool' unexpectedly would cause tabbed 'Models List' to disappear. Decks - General Unexpected software abortion would also take place when using the '[STRIKEOUT:>' syntax to get the ID of a material used in a material reference field in a PART_COMPOSITE property list (i.e. MID1]>MID). Focusing on material DB and "Copy to List" command, the unmatched materials would wrongly get other IDs, while no overlap would exist. The calculation of the resultant force would be incorrect when the 'Resultant Force' tool was applied to DLOADs with load type Hydrostatic Pressure (HP) and the 'Magnitude > Loads' option was deactivated in the presentation parameters. The 'Compress' function would wrongly delete DEFINE_BOX entities ignoring the fact that they were used in the model. A contact entity consisting of a flex body and a rigid body having several geometries would erroneously be exported as several contact definitions with the same name and that would create a conflict in ADAMS. Focusing on flex bodies, while exporting models in ADAMS Solver Dataset (.adm) format, the .mtx data of flex bodies would not be written out on Windows O/S. Optimization In the 'Optimization' tool, the functions 'RSM Improve' and 'RSM Optimization' would occasionally not recognize the ML toolkit environment. Also, the 'Optimization' tool would falsely output integer design variable values (e.g. MID, PID) as float type, when symbolic variables option was activated. Focusing on OptiStruct, DTPG card would not be correctly output as expected. Task Manager ANSA Pre-Processor would terminate unexpectedly when long comments were added to the task manager. Known Issues Resolved in META Post-Processor Focusing on 2D plots, META Post-Processor would abort when double-clicking the X-coordinate of a curve point created by 'Bore Distortion Analysis.' In addition, unexpected termination could occur when deleting META variables. Curves from .h5 files were not read properly. Abaqus fastener coupling visibility actions were not working as expected. Results on transformed includes were not displayed correctly on the corresponding transformed geometry. Also, thickness and MID information were neither read, nor displayed properly in META Post-Processor. Math Operations on Field Data Calculations with tensor results for 'Strain Gauges' would not take large translations/rotations into account. Project Files and METADB Script function 'results.PrimaryResultsList' would not return the correct information for METADBs converted from d3plot. Images and Video Handling Scale points for 'Plot Tracked Features' could not be picked on the video and the model. Read Results When loading a curve from a column ASCII format file using the 'Bar Plot from Column' or 'Bar Plot from Row' format options, the corresponding command recorded in the session was not automatically recognized as a valid filename. META Post-Processor would freeze when selecting a results file through file path recommendations using only the keyboard. META Post-Processor was unable to recognize distinct subcases in H3D files during result selection. Managing Curve Data META Post-Processor would terminate unexpectedly in cases of binout files with rbdout section. Wrong vector fringe would be generated in Internal type boundaries of Fidelity Flow cases. Annotations Annotations on selection variables were not properly evaluated if 'Append Pointers' option was active. Also, creating annotations on picked entities from screen could possibly lead to abnormal termination. Pressing Ctrl+N following export to JT format would lead to an unexpected termination. Regarding Spreadsheet Editor, extending LOG10 function to several cells led to corruption of the function. The same corrupted behavior was noticed while loading XLSX spreadsheet with extended LOG10 formulas. In addition, when saving a property's HTML Laminate Report, the report was saved without the associated images in the output folder. Known Issues Resolved in FATIQ Platform Calculations - Deterministic Loads Multiple subcases in deterministic loads would not be handled properly. Known Issues Resolved in KOMVOS SDM Console The KOMVOS SDM Console profile name would not be updated in the main panel upon reconnecting to SPDRM Software. Data Management - Lifecycle Business Rules The evaluation of conflict resolution policy through API now operates as expected. Known Issues Resolved in SPDRM Software Performance This release introduces significant enhancements to the SPDRM Software server, boosting its ability to handle high concurrency and demanding workloads. Key improvements include: * Enhanced stability: Advanced mechanisms ensure consistent and reliable operation, even under extreme load conditions. * Greater resilience: Optimized handling of concurrent requests maintains responsiveness during peak demand. The efficiency in processing rapid concurrent uploads, queries and exports has increased by 50%. Furthermore, the new release showcases efficient resource utilization, optimized I/O handling, and robust concurrency control, which are the key factors for scaling distributed systems, specifically: * 40,000+ structured CAE objects (metadata and file) saved in 15 minutes * 350 concurrent workers executing read/write operations with consistent throughput * This stress test scenario is equivalent to and ensures that 1,000+ active users can effectively perform daily business tasks connected to SPDRM Software Where to Download From You can find this release on Cadence Downloads, dated as of September 22, 2026. Customers who are served by a local business agent should contact the local support channel for software distribution details. Visit Cadence Softwae Downloads.