Goal


The purpose of this 3 day exercise was to try an approach to design a cultural style by going to 3d as early as possible in order to speed up workflow. The plan was to create a number of mood concepts on an architectural style. The goal was to end up with a series of reasonably high quality initial visualisations of the style as well as a versatile asset set as a base for further iterations


What follows is a walkthrough and conclusions of the experiment.

 

Research

My preliminary idea was some kind of bronze age people, in a mediterranean type setting. Buildings would have somewhat round and "humble" forms overall.

Sketch

I started by sketching out various ideas, searching for interesting shapes and patterns. These would form the backbone of the visual style and make sure that everything has cohesion, creating a well established look for the culture.

 

"Elongated egg"- shaped towers seemed like a promising direction to branch out of shape wise. I imagined they act perhaps like "cooling towers" that were used in middle eastern architecture. Or perhaps they are used to store grain in a well ventilated area where they could hang high up away from rodent’s teeth.

 

I was also planning the future steps of the project itself, taking notes on possible workflows or assets that could be especially useful.


Painting

In this phase I got to know the shapes a bit better to gain a better idea of what I’m after.

3D blockout

I made the base assets focusing on the largest elements first. Some quick tests to see the repeatability of them.

 

For the basic house models, I created one house that presented a good variety of shapes I was after. While the original house itself is intentionally a bit cluttered, the variations that were easily generatable from it each highlighted different parts of the design in a more balanced way.


Final concepts

I blocked out a variety of scenes and made additional assets based on paintovers done on early renders. The new assets were then used in further scenes, continually refining the look as well as keeping the style cohesive.

Conclusions

I managed to produce a set of varied concepts of the culture that took less time than images with similarly defined details would without 3d. The style has a certain look, but it isn't quite as defined as I would like it to be. There are many ideas that got left out from the initial ideation, which would get taken along in next rounds of the process


I am also left with an asset, texture and a tool set that continues to develop as the look tightens, and can used as base/reference for final models, thus easing the workload of the later stages of production.

Critique

Some things I would do differently:


1. I noticed the first paintings quickly became somewhat unnecessary, not helped by the fact that my main concern at the phase was to test the overall direction, which was easily checkable in the next phase. One basic house concept would have done just as well.


2. I would spend more time on the initial sketching phase, not only finding interesting elements, but also solving how to use them in the architecture. This would have allowed me to incorporate them earlier in the base models. Even if the initial building would look very cluttered, the unnecessary elements would be quicker to paint away when doing paintovers, compared to adding them to the scenes later. Philosophy would be: Prepare the base models for maximum amount of happy accidents, and deal the unhappy ones later.


3. Another time saver: I would spend a bit more time on the models that were on the generic side (tables, pots, people…), as the possible design changes in them wouldn't affect the scene much, but did now produce extra cleanup work later.

 

4. My initial plan for final images was too vague: "Get a feeling of the culture's look". Rather, I would set specific scenes that would test the needed variety of locations for the culture.


5. Pay more attention to scale in the beginning. Don't rush: measure twice, boolean once.

Mika Harju


mikaharjuart@gmail.com