The two runs were produced on different machines (ryzen-5950x-2 against ryzen-9950x), so the timings compare the hardware as much as the pull request. The phases a model reaches are still comparable.
1 of the 102 libraries were not tested in the same version, or not with the same configuration and reference files, in the two runs; see Library Changes below.
76 libraries of the baseline run were not tested by the pull request run: AixLib_latest, AixLib_master, Annex60, Annex60_1.0.0, Buildings, Buildings_1.6, Buildings_10, Buildings_2.0.0, Buildings_2.1.0, Buildings_3.0.0, Buildings_5.0.1, Buildings_6.0.0, Buildings_7.0.0, Buildings_8, Buildings_8.0.0, Buildings_9, Buildings_maint.7.0.x, Buildings_maint.8.0.x, Buildings_master, ConPNlib, Dynawo_1.6.1, EHPTexamples_main, EHPTlib_main, FCSys, IBPSA_master, IdealizedContact, ModelicaTestOverdetermined_3.2.2, ModelicaTest_3.1, ModelicaTest_3.1_cpp, ModelicaTest_3.2.1, ModelicaTest_3.2.1_cpp, ModelicaTest_3.2.2, ModelicaTest_3.2.2_cpp, ModelicaTest_3.2.2_fmi, ModelicaTest_3.2.2_fmi-cpp, ModelicaTest_3.2.2_fmi_cpp, ModelicaTest_3.2.3_cpp, ModelicaTest_4.1.0-beta.om, ModelicaTest_trunk_cpp, Modelica_3.1, Modelica_3.1_cpp, Modelica_3.2.1, Modelica_3.2.1_cpp, Modelica_3.2.2, Modelica_3.2.2_-O1, Modelica_3.2.2_-O2, Modelica_3.2.2__O1, Modelica_3.2.2__O2, Modelica_3.2.2_cpp, Modelica_3.2.2_fmi, Modelica_3.2.2_fmi-cpp, Modelica_3.2.2_fmi_cpp, Modelica_3.2.3_cpp, Modelica_4.1.0-beta.om, Modelica_LinearSystems2, Modelica_LinearSystems2_dev, Modelica_Noise, Modelica_StateGraph2, Modelica_Synchronous, Modelica_Synchronous_cpp, Modelica_Synchronous_fmi, Modelica_Synchronous_fmi-cpp, Modelica_Synchronous_fmi_cpp, Modelica_trunk_cpp, ObjectStab, OpenIPSL_release, PowerSystems_cpp, PowerSystems_fmi, PowerSystems_fmi-cpp, PowerSystems_fmi_cpp, PowerSystems_master, SiemensPower, SiemensPower_OMCtest, SolarTherm_alberto-merge, ThermofluidStream_OM_adaptions, iPSL.
The baseline is the newest run of master, not the commit the pull request is based on, so a difference can also come from something merged into master since the pull request was branched.
| Branch | Run | Compiler | Machine | |
|---|---|---|---|---|
| Baseline | master | 2026-10-02 13:47:38 | v1.28.0-dev.1162+gd836b49e16.cmake | ryzen-9950x |
| Pull request | pr/17076 | 2026-10-02 14:42:08 | v1.28.0-dev.1164+g284c6134de.cmake | ryzen-5950x-2 |
| Models compared | 20031 |
| Number of Improvements | 2 |
| Number of Regressions | 11 |
| Number of Performance Improvements | 0 |
| Number of Performance Regressions | 356 |
| Models only in the pull request run | 0 |
| Models only in the baseline run | 0 |
Time spent on the 20031 models both runs have. The last column counts only the 20018 that reached the same phase in both: a model that fails earlier stops paying for the phases it no longer reaches.
| Phase | master | pr/17076 | Change | Change, same phase |
|---|---|---|---|---|
| frontend | 1:50:46 | 2:31:24 | +36.7% | +36.7% |
| backend | 2:49:20 | 3:48:33 | +35.0% | +35.0% |
| simcode | 0:25:11 | 0:33:49 | +34.3% | +34.0% |
| templates | 1:15:45 | 1:39:31 | +31.4% | +31.4% |
| compile | 14:31:04 | 1 day, 1:54:31 | +78.5% | +79.9% |
| simulate | 6:42:03 | 9:05:39 | +35.7% | +37.0% |
| verify | 0:13:16 | 0:18:47 | +41.6% | +43.2% |
| total | 1 day, 21:12:55 | 2 days, 22:46:50 | +56.5% | +57.2% |
| Library | Change |
|---|---|
| NeuralNetwork_main | Configuration hash (OMC settings, the testing script or a reference file changed) |
| Library | Model | Where |
|---|