Hi guys,
Long time lurker first time poster.
My sincerest apologies for putting all the images at the bottom, but as a first-time user I’m not allowed more than one image per post, so I’ve had to put them all togeher into one. I ask for your patience.
There have been a number of threads over the years discussing workarounds and tricks for simulating a mirrored surface natively in SketchUp - i.e. without using a rendering plugin.
The most popular solution currently is to componentiate your geometry, copy it, and reflect it along your ‘mirror’ surface (typically just a low-opacity white surface). You can see an example here:
Now, for rectangular rooms like the one shown in the example above, this works pretty well.
But try using it on geometries where your mirror plane pokes into the space…
To get an idea of what I mean, imagine a floor plan like the one in Fig. A, where your mirrored surface is represented by the red line.
The obvious problem you run into using this trick is that your reflected, ‘mirror world’ copy of the room is going to overlap with the ‘real’ room, as shown in Fig. B.
Thus your space is just going to end up looking like Fig. C. Which is obviously very far from the intended result.
My question, then, for all you veteran SketchUp users and plugin developers, is this: What would be the feasibility of building something that basically generates a mirrored proxy of your geometry - a mirror-world version of it, as it were - and only displays that proxy within the frame of your mirror?
Again to provide a crappy example, if Fig. D is our intended result, can we get there by having a mirrored proxy geometry of the whole space (blue), and then only displaying the blue mirror-world geometry within the ‘clipping plane’ of the red mirror frame? See Fig. E
Sorry for the world’s shittiest illustrations of basic optics… But I hope someone is willing to explore the topic.
You obviously don’t want to bother mirroring the parts of your model that aren’t even visible in the mirror (i.e. geometries that are behind walls, floors, etc.) so would it possible to implement some kind of selection criteria for determining what surfaces get proxied and which ones are left out of mirror-world? Or would such a selection process itself end up being too computationally expensive…
Anyways, guys, any thoughts are appreciated! Just looking to brainstorm here.
Cheers,
Mati