Francesco Casella

I thought again why some parts of code generation scale as O(N^2). It could be due to the fact that the inductor coupling part is dense, so the Jacobians have...

This should also be fixed for the 1.19.0 release, it's not good to have options that don't work. @dietmarw are you working on Linux or Windows? This is not clear...

Or @lochel

Thanks @AnHeuermann! So, `chatteringInfo.numEventLimit` is defined as https://github.com/OpenModelica/OpenModelica/blob/1cb1c69e62808b086ac8d88ec6598835293b8faa/OMCompiler/SimulationRuntime/c/simulation/solver/model_help.c#L1184 As I understand from the [above-reported logic](https://github.com/OpenModelica/OpenModelica/blob/bc6c9619b24929efda3f71224bef3702619849f0/OMCompiler/SimulationRuntime/c/simulation/solver/events.c#L231-L241), if there are more than `numEventLimit` (i.e., 100) event within one `data->simulationInfo->stepSize` and `-abortSlowSimulation` is...

If that is the case, there are many `Buildings` simulations which last hundred of hours and have no `Interval` annotation, so they get the default 500 intervals, each possibly lasting...

Hmm, not sure this is always a good idea. Say your system gets to a steady state after sunset, and then starts having a lot of events at dawn on...

In fact, we could change the logic of chattering detection completely. Consider this MWE package ```modelica package TestChattering model Test1 "Chattering with variable step-size, gets stuck" Real x(start = 1.5,...

I don't think algebraic variables matter, the chattering is revealed on states. @phannebohm can you help me with the coding? I am not really able to work on the source...

@beutlich there are 2 weeks left until we branch off. Do you think you and @Harisankar-Allimangalath can fix it by then?

> Personally I can think of many sophisticated examples that are based on MSL but are lacking one or two special components. But that should not be a reason to...