Wednesday, 18 October 2017

Gearing Myself Towards a Game Art Workflow

I am a 3D modeller hoping to specialise as an environment artist for video games.
I was taught how to use 3D modelling software during my three year Bachelor’s degree ending in 2015. Since completing my degree I found employment as an Architectural Visualizer, where I worked on realistic 3D renderings for clients.
 
My Archviz Workflow

When approaching a client’s brief, I had a lot of creative freedom in how I chose to carry out the project. Each member of our team was in charge of their own brief and we rarely had to use one another’s assets for our projects.

This meant that we had no consistent standard when creating scenes. For example the way I chose to light and texture my 3D scene, would be somewhat different from how my colleagues would be creating theirs. Because it ultimately all boiled down to creating a final rendered image, we didn’t have to worry that there were inconsistencies between our project files. The only thing that mattered was delivering the final images, so consequently we could use any approach to get the job done.
This is in stark contrast to how a game studio’s workflow operates. Keeping assets and naming conventions consistent is of the upmost importance, otherwise the game will not work or at best be very poorly optimised.

In his book, “3D Game Environments: Create Professional 3D Game Worlds” Luke Ahearn (2017) states that “…we can’t escape the fact that creating art for a game is more than just making pretty pictures. When you create art assets for a computer game, you need to create art that will accomplish several goals. Other than looking good, your art must work with the software being used to create the game you are working on. Your art needs to be as efficient as possible in that environment. This requires planning and creating your art to very specific guidelines. Those guidelines may change depending on many variables… In any case, you need to be very well aware of what those variables are and create accordingly. Making a game run at its very best is called optimizations. This optimization should be on everyone’s mind from the very beginning of development.” – Introduction Page 3

I bring all this up because since starting my Master’s Degree in Game Art and Design, my approach to creating 3D assets has had to change.

The Polycount:

In 3D art there exists a term known as the “polycount” or “polylimit”. This pertains to the total number of polygons found in a three-dimensional model. It is often preferable to aim for a low polycount so that the 3D scene runs smoothly, however the trade-off for better performance is that low poly models look a lot worse than high poly ones. 

This is why a 3D artist is always conscious of the polycount but depending on the final output, they may not have to worry about the poly limit as much. This is why, historically speaking, the visuals in 3D games tend to look worse than say, the visuals in 3D movies. 

One medium essentially doesn’t have to worry about a poly limit and can make the visuals look as beautiful as desired during post-production, and the other has to make sure all assets for the game meet the standards of the games hardware, otherwise it flat out won’t work.

Ahearn goes on to say “The balance between looking good and running good is always a tug-of-war. Bad optimization means lower quality art must be used to achieve the required performance of the game.”

As an artist I am obviously concerned with making the visuals for my games as good as possible. However I will also need to be aware of the hardware limitations for the game. When it comes to games looking good isn’t enough, they have to run good or it’s all ultimately pointless. These are some of the problems I will have to be aware of as I enter this industry.

No comments:

Post a Comment