Sumo is expected to yield the same results on all platforms when run with the same inputs and random seed. The random seed is fixed unless you explicitly set option --random. There are known platform differences but they are mainly due to the 'proj' library and they mostly affect 'netconvert' and possibly traci scripts that rely on coordinate transformation. There is also a known issue with the EIDM model ( https://sumo.dlr.de/userdoc/Car-Following-Models/EIDM.html) Of course you have to make sure that it's the exact same sumo version on all platforms. - What is the minimal toolchain to achieve different results? - Are you calling any python tools in your process or just running sumo? - What sumo options are you using?
regards, Jakob Am Mo., 27. Juni 2022 um 17:55 Uhr schrieb [email protected] < [email protected]>: > Hello, > > I am using sumo mainly inside docker containers but need to compare/check > the results with a windows installation. > > But even when I use completely identical configurations and parameters I > get slightly different results. I suspect it comes from different sources > of randomness (like the vehicle creation in a flow). > > > > Is it possible to generate identical simulation results on different > operating systems? > > > > Regards, > > Peter > _______________________________________________ > sumo-user mailing list > [email protected] > To unsubscribe from this list, visit > https://www.eclipse.org/mailman/listinfo/sumo-user >
_______________________________________________ sumo-user mailing list [email protected] To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/sumo-user
