directx11 – About presumably utilizing Depthstencil 2D array in multiplerendertarget mode

I am rendering in ahead mode my scene with its depth buffer and since I am utilizing animated characters I am plannig to discover a manner to make use of a number of rendertargets to do the directional shadow go on the similar time utilizing a geometry shader. The rationale for that is additionally that I’ve organized my shadowmap measurement to be the identical because the display measurement. Furthermore I am doing “clear coloured shadows” impressed from right here that requires for my wants an additional colour channel and an extra depth data. The impact is cool however has some impression on the efficiency so doing the animation solely as soon as needs to be attention-grabbing.
So my plan is :

with geometry shader:

ship to RTV0+Depth0 the scene

ship to RTV1+RTV2+Depth1 shadow coloured(RTV1)+extra depth layer(RTV2) and common shadowmap (depth1)

at the moment I’ve these two rendering made in two passes (scene then shadows).

I’ve already used depthmap dice for pointlight shadow so I’m wondering if we are able to use 2Darray depth with a number of targets.

If utilizing 2DARRAY depth with 2Darray rendertarget is feasible I am questioning how this have to be organized.
Do we want the variety of slices for the goal 2Darray to match the variety of slice for the depth 2DArray or is it potential to make use of much less depth buffer than the variety of rendertarget? Or can the variety of goal per depth not be the identical? Is the machine managing this robotically and the way?