The style of the University of Rochester library (and other original campus buildings) has been described as Greek Revival, otherwise known as Free Classical. The style’s classical proportions and details are readily apparent in the facade. But when it comes to the tower and lantern, well, ancient Greek structures did not have towers, so modern architects “imaginatively” adapted ancient freestanding structures and “placed them atop the temple-like buildings they designed.”
“Imaginatively” is the operative term here.
The tower that houses the carillon – the lantern – consists of two concentric rings of columns – 12 cylindrical columns surrounding 12 square columns. They rest on bases carved in the shape of large scrolls. In turn, the columns support platforms on which rest stylized candlesticks and more scrolls. Leaf forms adorn the capitals of the columns and also jut out above the candlesticks.
A concave dome rises above the candlesticks, and above the dome is a spire with a sphere, an intricate bronze compass and, finally, a bronze weathervane in the shape of a flag with the university’s seal.
Metal screens cover all of the openings to protect the carillon from the elements. This is upstate New York, and the winters can be harsh. Some exterior damage, in fact, is clearly visible: a missing leaf-form and a broken candlestick on the front side.
Below the lantern, ringing the base of the main, copper-sheathed dome, are twelve carved owls. There are more carvings on the lower levels, including griffins and, at one time, sphinxes (now located elsewhere on campus).
All in all, it’s a glorious, inspired bit of work that seems as much baroque as classical. This will be fun.
Why a model?
Why not just use a photograph to illustrate how the carillon works, or a hand-drawn illustration? Why make a 3D model in the first place?
First of all, it’s impossible to make a photograph that clearly shows how the carillon is constructed and how it works. The bells and supports occupy almost all of the space inside the lantern. It’s crowded, and there is no angle that shows the carillon in its entirety. The ideal point of view is located outside the lantern, suspended in mid-air. Even if you could position a camera there, the lantern’s columns and screens would obscure everything inside.
Hand-drawn illustrations can be beautiful and, in this case, could work well. But once the drawing is started, the point of view is fixed and can’t be changed. It can be difficult to repurpose the finished illustration for other publication platforms.
A well-rendered 3D model can be nearly as beautiful as a hand-drawn illustration. You can freely select any point of view and change it at any time. Virtual cameras aren’t subject to gravity (or any other real-world restrictions), so you can place them anywhere.
Finally, once the model is finished you can render it any number of times, from various angles, to suit the needs of print, web, or animation. Creating a model is a lot of work (though maybe not as much as creating a detailed hand-drawn illustration). But the payoff can be well worth the investment.
We have quite a few resources to draw on: detailed elevation drawings of the library and lantern, floor plans and photos of the lantern’s interior, and exterior shots taken from ground level. (The architect’s drawings were provided by the university library; all of the photos are my own.)
The lantern is built on the plan of a regular dodecagon, with one of the columns located at each vertex. This greatly simplifies our work, since we have to model only one-twelfth of the structure. (In fact, since each of the 12 slices is bilaterally symmetrical, we need to model only half of each slice, or 1/24th of the structure.) Once the slice is completed, we’ll duplicate it along the central vertical axis to create the final model.
The old model used several shapes to create the lantern, but my plan this time is to do as much as possible as a single high-resolution mesh.
The slice that we’ll model will include the base scroll, the interior and exterior columns, the candlestick and top scroll, and possibily a section of the dome. We’ll start with the base scroll, as indicated by the arrow in the photo above. It seems natural to start at the bottom and build up. The scroll is also one of the most complex details, so it will be nice to get it out of the way early.
The scene is set up with two construction planes, one containing the front elevation of the library, and the other, one of the lantern cross sections. The model will be created at 1:1 scale, so one inch in Maya will equal one inch in the real world.