You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
(from a discsussion at the FMI Design meeting with @pmai and @klausschuch )
There are CS FMUs, where the outputs are directly calculated from the input (independent of the the timestep) and without internal states.
Such a direct feedthrough behavior can be implemented with ME FMUs, but many users and tool vendors prefer the simpler CS interface.
FMI forbids calling fmi3doStep with communicationStepSize = 0, but an importer that is informed about the pure algebraic nature of such an FMU could call fmi3DoStep with any timestep and use the result in a suitable way.
For FMI 3.0 (and 2.0.4) we could mark such FMUs with a minimalistic layered standard, defining only a manifest file in the extra directory.
The text was updated successfully, but these errors were encountered:
Would it not be simpler to make an addition to allow output to be computed wihtout dostep?
This can also already be done with and fmu that allow event mode, just go into event mode compute your output->done.
(from a discsussion at the FMI Design meeting with @pmai and @klausschuch )
There are CS FMUs, where the outputs are directly calculated from the input (independent of the the timestep) and without internal states.
Such a direct feedthrough behavior can be implemented with ME FMUs, but many users and tool vendors prefer the simpler CS interface.
FMI forbids calling fmi3doStep with communicationStepSize = 0, but an importer that is informed about the pure algebraic nature of such an FMU could call fmi3DoStep with any timestep and use the result in a suitable way.
For FMI 3.0 (and 2.0.4) we could mark such FMUs with a minimalistic layered standard, defining only a manifest file in the extra directory.
The text was updated successfully, but these errors were encountered: