Five Secrets of Game Art Direction

This is a reproduction of the original post at UpYourGame.

When I was just starting out directing art and experience for games, I was assigned a big, high-priority game project. It was a wonderful opportunity to showcase my team's capability and make a name for myself, and I was quite excited. The theme of the game was Irish, and the art style was cartoony. I spent a whole week researching art styles, colour palettes and creating mood boards; of the six artists in my team, I chose the most experienced and had a long kickoff meeting where we discussed the project at length and pored over all the research material I had carried out.

A few days later, as I looked at the first sketches, they didn't feel right. I spent another day looking trying to figure out why it just wasn't right, and another few hours with the artist. The next day, as I saw the updated sketches, it wasn't looking any closer to the vision I had of the final game art. I was desperate.

Just then, I happened to pass by the desk of the youngest artist in our team, a lad just out of art school a few months ago. He had his Artstation page open, and I happened to see something that almost made me spill my coffee. It was a Leprechaun, a character done in a Cartoony style that was EXACTLY what I had in mind for the Big Project. I asked him to show me more, and I had found what I was looking for.

I assigned the project to the young lad, thanking the senior artist for his time and effort. Long story short, that game went on to become the biggest hit the studio had for the last five years. I recommended that young artist for a 70% pay rise at the next review, and he was promoted to Senior Game Artist within six months of release. It was the game that made my reputation.

When it comes to making a successful game, great Art is only one of the several pieces of the puzzle that need to fall into place, but it is a very important one. In this post, I will tell you about five hard-earned lessons I have learned from making dozens of games over a period of more than a decade.

It doesn't matter if you are a game artist, game designer, producer or any other role; these lesson can be applied universally to the game-making process.

#1: UNDERSTAND THE ARTIST'S STRENGTHS AND MOTIVATION

This follows from the story I just narrated; every game Artist is unique in what he or she is good at, and more importantly what he or she aspires to do. The closer a project is aligned with the artist's personal goals and preferences, the better it is likely to go. Asking an artist to create game art in a style that is far removed from their preferred one is pretty risky; it takes hundreds or maybe even thousands of hours of practice to develop a particular style, and most artists will struggle to work at short notice in a style that is alien to them.

This does mean that you are kind of limited when it means exploring art styles for your game; there are a couple of things you can do to change this. Firstly, plan well in advance and allow the artist sufficient time and space to practice and develop the required style, and secondly, find an artist who can do that art style! This is one of the reasons why many game studios doing lots of small projects work with art outsourcing studios instead of hiring full-time artists.

Personally, I like to maintain an art team with a wide variety of skills. I have one artist who does great concept art for environments, another does cartoony style really well, one old-school artist who can make super-realistic poster style art and a few more generalist who can do a bit of everything among them.

#2: THOROUGH PRE-PRODUCTION AND PROTOTYPING

This is absolutely vital; I have seen many, many projects crash and burn because the team just did not take the time and make the effort to ask the three questions: WHY are we making this game? WHAT are we making? HOW will we make it? Game art needs to closely support the game mechanics, narrative and player emotional journey, and to achieve this, it is vital that the team (or individual) goes carries out a systematic pre-production before the production art phase of the game (creation of assets) begins.

From an art standpoint, the following questions need to be asked (and answered)-

> What are our resources for creating the game art (this means number of artists available, their experience level, availability in terms of hours per day etc.)? > Will this art style support the game mechanics?

"If we want to do a side-scrolling platformer, is Voxel art really the best choice?"

> How many hours (approximately) will it take to create the game art? This is where game asset lists come in really handy! It is a big help to the game project if you have an idea (however rough) about how much time it will take to create each art asset, as there will either be a programming, world building or design task dependent on it!

> Can we actually pull this off? You might think a steampunk-inspired game world with fifty different characters would be awesome, but do you actually have the expertise to create everything AND make it look as good as it does inside your head?

Prototyping is a fantastic way of reducing the risk of your project. If there is a game mechanic in there that involves very specific art or visuals, build a prototype at the pre-production stage! If it works well, you've just completed a key part of your game and you can then build the game around it. If it doesn't, then you just saved months of work that would have gone down the drain. Prototyping unproven mechanics is a win-win!

#3: REVIEW PROGRESS REGULARLY

There's nothing worse than spending a hundred hours making an art asset and then having it rejected out of hand (and needing to start over). So if there are other stakeholders in the game project (such as a client, an Art Lead or a Producer) that have a say in how the game looks, regular reviews of the art should take place (ideally every day, or even twice a day) so that if the artist takes a wrong turn, he/she doesn't proceed too far down that path before being corrected.

As a corollary to this- if you are the one reviewing the art, learn to look at Work In Progress art and imagine what the finished product will look like. Use your imagination and try to picture what the finished art asset will be (after five or six more hours of work); it takes some doing, but it is very much possible.

Even though I'm game art director, I get artwork approved by my own boss (the studio head) and sometimes domain experts on particular games. On some projects that have a very particular ask, such as an art style that we've never tried before, I don't wait for my artists to finish the artwork; I keep sending my boss sketches as well to track that my team is on track.

Sending these through e-mail and getting feedback can be a bit tiresome; this where collaboration software like Trello comes in very handy. I use it for the projects where there are multiple stakeholders like myself, an external art team, product managers and technical team (developers) involved.

#4: GAME ART LOOKS DIFFERENT WHEN YOU PUT IT INTO THE GAME

Art assets (characters, props, buildings) may look good when seen in isolation, but that doesn't mean that they will look as good once you put them into the game; this is because the game has an aesthetic of it's own and there is usually a "Hierarchy of Attention" for art assets that depends on what you want the player to do, look at, and pay attention to.

For example, if you want the player to play a level where he/she follows a path through the forest guided (subconsciously )by a particular tree/leaf colour, you won't be able to tell how clear the path is, unless you put trees of different colours and types together into the game and play it. This will take several iterations of leaf size and colour, tree height and probably texture/colour of the ground as well. It may well take a dozen tries to get right, so it is probably a good idea to hold off on final polishing of the tree mesh/texture until you understand which one works best.

Therefore, it is a good idea to keep deploying the assets into the game before 'clearing' them; while they look good all by themselves, they may not actually work as well when they are deployed into the game build due to how they work with the other game elements.

#5: CONCEPT ART IS THE BEST THING EVER!

So there's something called a "Game Art Pipeline"; it's called that because to maintain an efficient workflow, in most cases a particular sequence needs to be followed as regards the game art production process. Two important points to remember here as regards the game development process-

a) Game projects can have many stakeholders (designers, artists, developers, producers, product team , marketing team, company execs) that have wildly different ideas about how a game should look, and

b) A game project may have more than one artist working on it, where each artist has a unique and particular art style.

These two have the potential to cause massive conflicts (and disruption) to your game. I've seen it happen many, many times- there is no agreement on what style the game should follow, and the artists end up doing rework upon rework as a result of trying to make everyone happy, which never happens. Or, different artists create assets in their own style which, when integrated into a game, give the art a disjointed, mis-matched feel.

The key to preventing this, is to build a game art pipeline that starts with concept art! During the pre-production process, the concept artist starts making sketches of what the visuals of the game look like-characters, props, environments and user interface among others. These sketches should then be shown to the stakeholders, and any changes that need to be made, happen at this stage. This prevents conflicts, takes into account the concerns (and opinions!) of the different parties, and shortens the time taken to create the final art of the game. It also prevents fatigue, burnout and loss of motivation among the artists!

If there are different artists working on the same game, the concept art acts as a style guide to what the production art needs to looks like, so that their output is in sync and matches the overall aesthetic of the game. Trust me, this is a very big deal- without concept art to guide them, all artists default to their individual styles and it's a big mess.

Now I do realize that good concept artists are hard to come by, and are quite expensive to hire. If you can't afford to hire one, I strongly recommend getting a freelancer to help you out. It's well worth the money, time and effort!

And to wind it up, here's a 'Bonus' secret thrown in: Your art is only as good as the sounds that accompany it. Great music and sound effects are the secret sauce that bind game art and experience together, so don't neglect your game's sounds and music.

Thanks for reading, I would love to hear about your experiences with making games and game art; do comment and let me know all about it.