Ship Design - #5 Model Technical Details
Lord of Rigel’s 3d models are built using polygonal modeling, a technique used in the vast majority of today’s games. What this essentially means is that each model is made of small surfaces called polygons – each defined by 3 vertexes and a normal, determining from which side the polygon is visible. The amount of detail that can be put in each model is directly tied to the amount of polygons each object will have.
Determining a good number of polygons is always a balancing act between efficiency and looks; luckily, modern processors and graphics cards are generally able to work with a large amount of polygons, allowing us to create detailed looking fleets. This doesn’t mean that the polygon count is irrelevant, though; since a single scene can still contain many ships, everything from big capital ships to small fighters and missiles. As with everything, a good balance is always needed.
It is also important to keep in mind while building models that the model will need to be textured later on. Some features need to be planned up front, such as which parts will be mirrored or repeat on the texture, or which will be uniquely mapped, but I will cover the specifics of those another time. In general, planning ahead for the texturing phase saves time and makes things overall easier.
These methods slightly differ across races; it’s harder to mirror parts of the ship on a species such as the Katraxi, which utilize very asymmetric designs in their entire fleet. The Humans, on the other hand, use completely symmetrical designs, but have wide, full bodies where mirroring on the top and bottom of the ship would be quite visible when weathering (dirt, grime, paint chips) is applied. For this reason I’d mirror the sides on the Human vessels, but keep the entire top and bottom unique.
Other species, such as the Xantus, have very little subtle weathering allowing for easier mirroring in the modeling phase. The same goes for the Selach, while the Yalkai’s specific hull design allows for it to some extent as well.
Another thing that I find very important is scale; by modeling in 1:1 scale it is not only possible to avoid a lot of scaling issues later on, it also allows for much better planning of details, which in turn allows for much better look and feel of all ships, scale wise; this is why, during modeling, I always consider which species I’m designing the ships for, how big are they physically, so I can know how tall the decks would be (the exception here being the Xantus, which, being shape shifters, have curved decks of variable heights).
When extruding a deck, I always know how tall it’s going to be and I know if, for example, it’s tall enough to place escape pods on, or how many lines of windows I’ll have, each representing one deck. Since these are spaceships, there aren’t nearby reference point that can be helpful in determining size visually, such as trees, cars, humans or any other details that, for example, first person shooters contain. It is therefore all the more important to get the details correctly scaled on each ship – particularly the windows; when we see a frigate with a single deck of large windows, we’ll know it’s a small ship. Likewise, on kilometer + long battleships windows turn into a lot of small, bright dots, helping us feel the size visually.
Read the next article in the series as I discuss the Tulock fleet design!