What role does research play in your concept art workflow?

Research is the cornerstone to my artistic output. The mathematical forms that emanate from these investigations into the scientist's understanding of our sensory world - and concepts that are extrapolated or interpolated therefrom - is a personal response born of a profound uncertainty. Is this art, is this science, is it both or is it neither? But I do know that it has taken a lot of thought!

Concept art refers to artists' visual representation of ideas - usually for films and video games - before these ideas are finalized into products. Thus, a concept art workflow simply refers to the iterative process of visual display of ideas, such as a movie or a video game's characters and their costumes. Given these definitions, research plays a crucial role in the concept art workflow. Through research, designers can better understand the contexts (for example, historical context) of the characters and their costumes, lending credibility to the movie or video game. By conducting secondary research, for instance, concept artists can better portray video game characters and their costumes more accurately.

In my experience as an artist with a background in the sciences research is both a wonderful opportunity to indulge my own curiosity as well as gain a deeper understanding as to how to ground a concept in reality. I enjoy delving into any relevant topic, be it how a mythical creature's musculature might lay upon its bones or what traditions and customs an alien society could value. Finding a broad wealth of inspiration from legitimate and vetted resources is vital, as well as considering historical art or how a subject has been portrayed in more contemporary works. No concept exists in a vacuum, and what context a viewer or player may have must be understood.

In all seriousness I find the most helpful step is to set up a folder for the project and relevant subfolders before setting onto the internet to seek out inspirations and resources. After that I take care to avoid AI generated images and to clearly label which images (especially textures) are available for use in a concept art project. From there I will create a visual library folder by organizing the images into groups by subject or even color, and a resources folder for any images or textures that I could use to help render an iteration of a concept.

Over the years I have built up a small library of reliable sources for historic fashions, customs, and technologies to use as a starting point for gaining an accurate understanding of the look and ethos of certain time periods. Sometimes I even find it helpful to read works from those times if available, to better understand how people would express themselves in those times. Their linguistics play a role as well; which I would recommend doing some research into common phrases to gain a more grounded understanding. Otherwise its easy to miss both relevant folklore and attitudes from that time and place.

This is the step wherein it helps to be a shameless nerd with a genuine curiosity about how everything works. Whether its understanding the basics of how a type of engine works (or might work if it is speculative) or the anatomical features of a specific genus of plants there is so much opportunity for detail and life in a design. Even a rudimentary understanding can also make a concept seem logical within the context of the project as a whole. This is especially true for anything meant to be a real creature within the story; understanding the anatomy of how a subject moves is vital for a convincing concept and final asset.

Understanding the demographic of your audience is vital; even within one nation the variation in cultural context between age groups is enormous. If an artist doesn't properly understand the context of a potential viewer it is extremely difficult to have a concept read in the intended way in the final asset and project as a whole. This is where an uninformed artist can run the risk of creating something that comes across as disingenuous or condescending.