Another option would be to use two seperate render targets. You render everything to one target that doesn't require any special treatment, then you copy the contents of that render target to another render target. You then use one as the input and one as the output.

Although it can be done as Yourself mentions, note that it's for a single format - DXGI_FORMAT_R32_UINT - which may not be compatible for color sampling.

Although it can be done as Yourself mentions, note that it's for a single format - DXGI_FORMAT_R32_UINT - which may not be compatible for color sampling.

Indeed, which is why he mentioned having to do manual packing and unpacking. Another possible workaround is to use a StructuredBuffer instead of a texture.

I am working on a project where I need to sample the pixel on the render target before the pixel shader returns so I can determine what blend operation I want to use on it (I have several that are color based). Is there any way to sample the render target like this?

I don't quite understand the use case, but generally you don't need to sample the render target itself. Since you're in the process of overwriting the value in the render target in your pixel shader, you should already have the (pre-blending) color you're about to write out.

If you need the value after the blending operation is done, this isn't possible. The pixel shader runs before blending (some hardware may interleave the operations, but this isn't guaranteed to happen by the spec), so there's no possible way to have the result.

The best way to achieve this effect is to use multiple passes. In your first pass, write out the new color values. In the second pass, read in the generated texture bound to the first pass's render target and use that to write out to a second render target.

You can also consider making use of the stencil buffer instead of color, as this may be faster for many hardware implementations. Write out a different number to the stencil buffer for each desired blend operation, then do a pass over the input to the output for each blend predicated on the corresponding stencil value. This both lets the hardware do early outs for pixel shader evaluation as well as make it easier to use the built-in blending hardware (I'm not sure what you're trying to do so I don't know if that matters or not).

For a lot of hardware, multiple passes with simple shaders can be more efficient than trying to make a single complex shader that does everything in one pass. It depends on target resolution and color accuracy and the hardware's shader execution cores vs its memory bandwidth.