Which data should be used for finding marker genes?

Hello all,
In the “Finding marker genes” part of PBMC3K tutorial, the authors mentioned that “For this, by default, the .raw attribute of AnnData is used in case it has been initialized before” and the fuction sc.tl.rank_genes_groups(adata, ‘leiden’, method=‘t-test’) setting use_raw=True as default.
I tried use_raw=False in this sc.tl.rank_genes_groups() function and found that the results of marker genes are quite different. So, which data is recommended for finding the marker genes, adata after scaling, which is use_raw=False, or adata.raw?