With the limited amount of information, this is difficult to answer (I admit, I don't know what I would need to give a good answer, anyway)

I would approach this in an iterative way, you can play with minimum seed length and bandwith, z-dropoff, then dive into match and mismatch scores and gap penalties. After all, much depends on how similar the genomes are. You will always align some xy percentage, but does that help you?

complementing Carambakaracho comment :

it's always advisable to start with the default settings and take it from there.

In principle, it's a bad idea to map to a different genome, but as suggested, start with the defaults and see how much can be used.