I am not sure if I understand the question.
I guess you are referring to the part of the video where the author is showing how he changes material properties in Corona and fstorm with interactive rendering and in Corona he needs to wait some time for the rendering to refresh while in fstorm it is almost instant. We would first need to understand how exactly he did it. For example, maybe his Corona scene is complex (has displacement, other pre-rendering effects to parse) and in fstorm he only has a single object isolated? In that case, the solution would be to isolate the object in Corona and only then tweak its materials.
Another option is that he is using the UHD Cache in Corona. It needs some time to re-calculate when using IR. I don't know if there is some cached GI solver in fstorm. If not, then to compare those two you would need to switch to path tracing only in Corona and then compare.
Generally, we would need more specific details what is happening and then we could come up with some solutions.
It is easy to create comparisons which favor one render engine or another. It is not so easy to create fair comparisons.
