We ran a case with electric and magnetic dipoles as point sources with a layered dielectric medium above it with relative permittivity=1 using the RL-GO solver method. Between two versions of FEKO, v2017 and v2018.2.1, there were vastly different 2D gain patterns in the UN (xz-plane) direction with the same setup for both versions. In other cases of permittivity that we tested, the versions produce the same results. I don't see an option to include pictures otherwise I would, apologies.
To solve this, you can use er=1.0001 or similar if needed to have the dielectric sheet similar to free space.
