I wonder if these options are anyway different from each other. If you refer to the reply from Lars in the thread The SAP HANA Engines?, it is mentioned that
"during activation of the analytic views, the joins in the attribute views get 'flattened' and included in the analytic view run time object"
So, as per my interpretation, both the options you mentioned, should have the similar execution plan.
I think the alternative would be to filter the data as much as it is possible, with additional filter criteria. Also you can try partitioning as suggested by Krishna. Also check the feasibility (although quite less likely) of persisting the data with additional redundancy so that the join may not be required and you can get the data from 1 table only.
Regards.
Ravi