It's probably about the same, using Rhino Inside or Dynamo in terms of what gets put into Revit. Dynamo might even be better at handling the proxy geometry, not sure. In general, I prefer Rhino.Inside for most things; it's just a more stable piece of software and requires a few less components to do the same things, in my opinion. The main reason for choosing Rhino here was that the blocks were in Rhino to start with. I'm not exactly sure how we would have done it with Dynamo, but, like you outlined above, we could write all the block locations to a csv from rhino/gh, then use dynamo to reference that csv- something along those lines. That type of multistep process might be more stable any way. We talked about doing this entire process with a c# script too, which would probably make the whole thing quicker and more reliable, so we may do that at some point too.