I am genuinely interested in the process though, and the production issues that sometimes interfere with artistic vision. If it's not against policy, or fear that it may affect how the Reaper brand is viewed or anything like that, I'd love to hear the hows and whys of some of the specific decisions--why cavities are usually filled, some of the special constriction of working with the new material as opposed to traditional metal or resin, things that it has allowed that were not feasible with metal, etc. Any interesting tidbits for the million ways the manufacturing affects aesthetic choice that haven't been mentioned or even noticed, etc.
I doubt that I'm alone, and wouldn't that be a great way to transform a petty argument into an edifying and educational experience? I sincerely think this would be awesome.
Probably the number one reason for any sort of filled in space on any mini: undercut. Undercut is the woof-mistress of all molding process, regardless of whether it is RTV, vulcanized rubber spin-cast, or injection molded plastic. Fortunately, the first two mediums have some play, and can be a little bit forgiving, so long as the undercut is minor. Injection molding? Now it is the material that has to give; if the material cast in it is too brittle (i.e Resin-type plastic), you'll never get it out of the mold without breaking the pieces. If the material is too soft, it will be deformed by the extraction. The Bones material walks a very fine line of being slightly flexible and elastic, while also still being reasonably rigid. That's why it works pretty well as a medium to produce miniatures from. Still, the flebility and elasticity of the material has its limits, and in many ways its limits are more rigid than casting metal figures in a vulcanized rubber mold (which has a surprising amount of give, especially once you get the molds warmed up after several passes of hot metal through them).
The other solution to undercut is making the mold multipart to cast it properly. The problem with this is, of course, cost. More cost for molds, more cost for casting time (you are now doing the work ~twice for a model - or more, if it is even more multipart than just 2 pieces), and more time spent in QC, packing, etc.
Take a look at some multipart models you may already own. Find the mold lines on the parts - this will tell you the plane that the model was placed in the mold. All the empty spaces around the model are filled with rubber. Now imagine separating the mold to remove the fully assembled model from it. You should get a pretty good picture of why models are molded the way they are. A voided space inside Kaladrax' chest cavatiy, for example, could never be removed from the mold if it were cast as one piece.
Vince has pretty much hit it. To understand the reasons you might fill in a cavity or undercut area, you must imagine the molds themselves, and how they bend (or don't bend), how the figures are removed, etc.
In casting a metal object, we use a soft rubber mold, so when there is an undercut or cavity, the rubber will give, flex, and bend out of the way, allowing us to remove the metal without deforming the final product. In working with injection-molded plastic, the molds are solid steel, and it is the Bonesium™ that flexes. Small amounts of flexing are fine, because it is a resilient piece, and much more pliable when hot.
However, larger pieces that are asked to flex might be seriously distorted by flexing so shortly after molding. These machines crank out a human-sized figure every 13 seconds, so the figures are only barely cooled below their freezing point when the mold is opened and they are popped out. In this case, if what is currently the body segment, including ribs, neck, spine, etc. were cast with less fill-in, ie deeper cavities, we increase the chance of two things - 1. that it might stick in the mold, as the friction of the extruded ribs holds tightly to the steel mold. If this happens, then when the mold goes to close again (remember, this is all done by an automated machine that repeats a cycle every 13-30 seconds) it might squish the figure inside, resulting in catastrophic failure, requiring a technician to clean and potentially repair something. So deeper cavities is not a good option. And thing number 2 that deeper cavities might result in is additional deformity. These figures are cast at over 500 degrees, and freeze near that point, between 400 and 500. When they come out, they are most vulnerable to being misshapen, and larger figures even more so, because they retain their heat longer. So these larger (depthwise) rib membranes are hot and very flexible. Every 30 seconds one is dropped from the machine into a receptacle. That drop, on a soft and unsupported feature is likely to cause distortion if the Bonesium™ is soft (ie hot) enough. On a sword or spear, which is very small and has very little heat, we see some flex even now - imagine how much worse that gets when we're holding more heat because we have a larger object.
Ok, so that's why the fill isn't "smaller". But why not cast the ribs as two or more pieces instead of as just one? The most intuitive option would be to cut the piece into two pieces, a left half and a right half. Each would join at the spine and sternum, and might contain portions of the spine and sternum in themselves. Imagine, if you will then, that the left half looks like ( and the right like ). when you join the halves, you get a hollow interior (). Well, we see now that deformity is STILL a problem - it is with tiny swords and spears, and is going to still be an issue with the ribs. It's really only not a problem, in fact, on solid chunks. But the biggest reason this deformity is a problem here is because of the need to reassemble the two halves. Distorted halves might not fit together well, and this would be a customer service nightmare for us, since unlike metal, we do not yet have the ability to simply make another piece and try for a good fit.
This does not even factor in that a mold has a fixed size, and so there's a maximum polygonal area a part can take up. Each half, in cross section, is identical to the whole, so two halves takes up twice the square cm are of the mold - increasing the number of molds needed to make it, and as the mold is the most expensive part of the process, drastically increasing the cost. This in contrast to metal, where the material is the expensive part, so hollow and partial shapes are preferred, as they reduce material used. Were this cast in metal, I would rather it be broken into five or 6 pieces, to reduce the metal involved, even at the expense of 5 new molds.
Then, there's stability. Bonesium™ *IS* much lighter than metal, but even so, it does have some heft. Trust me, I have to move boxes filled with hundreds of great worms, and those are actually pretty heavy. Kaladrax's wings are MASSIVE and quite heavy (relatively). The weight of those wings is focused entirely on one contact point - the entire left wing balances on the scapula attached to the left ribcage, likewise for the right. When you have the entire mass of the wing focused on a small contact point, that's a lot of pressure on that rib pair, and only that rib pair. A hollow bit, especially one with a weak or poorly attached join at the spine (because of small deformities, see above) is highly likely to experience further compression - and distort even more. To resolve this, the part had to be engineered with a less flexible ribcage, and the limits of the materials and the molds, as described above, suggested a fill-in as the best option of those available to us.
Now, we recognize the fill-in is not particularly attractive, and this is one of the few times that we have violated protocol. We got the pictures and the prototype, and were so excited, we showed them. But at the same time, we have changes we are asking for from the sculptor. We do not normally show prototypes of greens that are "unfinished", but in this case we felt the coolness of the piece trumped that protocol.
"What changes?" you ask. Well, to begin with, we want an open mouth. That way you can choose as you desire to cut the jaw and close the mouth, but if it is sculpted closed, choosing to open it is much harder. Not only can a skilled modeler close an open mouth with more ease than the inverse, but we feel an open mouth is more menacing and more intimidating on the table, especially when it is so clearly able to swallow you whole. Another change we have requested is to have the fill-in with additional texture, as of rotting tissue, organs, skin, and muscle.
That's a lot of text. Thank you for reading it, and feel free to ask for more data.