Why would I NOT want bi-directional Crossfiltering between tables?

[edhans]

I am used to working in apps like Microsoft Access where I can do inner joins and enforce referential integrity. Power Pivot in Excel is slightly frustrating to me as it only allows single direction crossfiltering. I think I have overcome this with the CROSSFILTER function, which is a bit of a pain, but still workable.

In Power BI though, I can enable bi-directional crossfiltering. Why would I NOT want to enable this by default with each relationship, unless I was trouble-shooting and wanted to find where a fact table had something a dim table did not - i.e. a left/right join?

Example is here, using the AdventureWorks data from "Learn to Write DAX"

[vanessafvg]

@edhans because sometimes depending on your model you might get odd results, i think its got a lot to do with people not necessarily understanding what they doing from a modellig perspective.   If you know what you doing shouldn't be a problem in my opinion.