Things to keep in mind when making a tileset for an RPG

So you want to make a tileset for your RPG.  "This shouldn't be too bad," you think, "there are all sorts of tutorials on the subject and doing art at that size shouldn't be hard."  STOP!  Go read Celianna's post and if you aren't overwhelmed by the sheer amount of work involved then you're entirely missing the point.  It's supposed to be overwhelming, I've been doing this for years and I'm still a bit overwhelmed by the massive amount of work that goes into making your game look unique.  If you're new to art as well expect that work load to be at least doubled, more likely quadrupled.  You've got a lot of learning to do on top of all that art you're expecting to draw.

The reason why I bring this up is simple.  You have to understand what you're getting into or you will fail.  I have run into so many budding pixel artists who make exactly one square of grass tile and think to themselves "Wow, that wasn't so bad.  I just need to do this a few more times and I've got a whole set!"  They then start working on the other tiles and by the time they reach the work load that they thought would be required to make the whole set they don't even have enough for a single screen of mockup and that half finished mockup doesn't even come close to what they saw in their heads.

This doesn't mean you can't do this or shouldn't try though.  Making art for an entire RPG is a huge endeavor but it also makes for a huge result.  Your game is now unique, interesting, and the art will be made specifically to suit it.  Because of it's rarity in independently made RPGs, original art makes for a lasting impression.

So, what exactly is involved in making a tileset for a game? Quite a bit actually.  Time, effort, artistry, consistency, and sometimes a bit of money to get the tools you need.  In other words, this is going to be a whole lot of work.  Here are some things to keep in mind when doing that work.




Understanding the Nature of Your Set 

Before you even start thinking about the visuals, your game play should be solid and you should have a really good outline for your story.  Games are an interactive medium, it doesn't matter how beautiful these fancy tiles are if your game doesn't match up.  The visuals should be a support, not the point of your game.  Yes, your visuals will get people to play but if that's all you've got, you're setting your players up for disappointment.  If you get to the point where you've got the visuals complete and then go back and try to make the game play work you will make decisions based on what won't require you to redraw anything instead of what works best for the game.  There's not really a way to do it so that it won't hurt your development.

The nature of your game will dictate quite a bit about the visuals.  Where and when does this game take place?  Is it a serious story or a silly one?  Moody or light?  Something inbetween?  Would serious visuals add to the mood or detract?  Would juxtaposing cutesy visuals with a disturbing plot help or hinder it?

Defining the style and setting of your visuals should be the very first thing you do before you even try to make that first grass tile.  You don't want to end up redoing work that you could have just done once, you have enough of that ahead of you as it is.  If your game is set in space, that work to get a good grass tile is wasted.  If you're doing a cute style then the beautiful rich texture with dark colors you just spent hours on is going to detract from your game instead of supporting it.

It's also important to have a very firm idea of what you're going for, and by that I'm talking about references.  Find images that match what you want in the game.  They don't have to include all the elements in one picture, in fact it's better to have a range of images.  Have a few that have the same artistic style, a few that depict the setting, a few that capture the mood and so on.  Finding references solidifies the ideas in your head and helps you make your concept into reality.

Positive and Negative Space and Visual Priority

Have you ever played a game where you can't figure out where the main character is?  What about a game that makes your eyes tired whenever you look at it?  A game where every aspect of the game looks like it's blending together?  Where you don't know where you can and can't walk?  These games likely haven't been done with the concepts of positive and negative space and visual priorities in mind.

Visual priority is the concept of making the most important things the center of attention.  This is mostly done by contrasting artistic elements from one another.  Detail is contrasted with emptiness, saturated colors with greyness, light with darkness, bold lines with thin ones, and so on.  In an RPG the priorities should be something like this:  Main Character, NPCs and interactive items, things closest to the "camera", non-walkable areas, walkable area.

The main character gets the most attention when doing this.  The player sprite has to have high contrast.  It often will have a different pallet than everything else, and use more saturated colors.  It will also likely have a bolder outline than anything else.  To make this sprite stand out against the grass tile, that grass tile has to look much less important.  The colors may need to be more muted, it may need less detail or maybe less contrast. Maybe more detail is the way to go if your sprite has large flat sections of color.

A common mistake I see in beginning tile artists is to make that grass tile the priority.  Blades of grass are defined boldly, the tile has lots of contrast, and the palette is brighter there than anywhere else.  Even the best sprites blend right into it and it becomes impossible to tell one from the other and the resulting set is headache inducing to look at.

To put it another way, the walkable areas of your set need to all be Negative Space.  They aren't the parts that grab attention, they are the parts that allow your eyes to rest and lead the eye around the image.  Sprites and interactive objects are Positive Space, they're the parts that say "Look at me!"  Keeping these concepts in mind is an extremely important way to keep your tiles cohesive and working well with each other.

If you'd like to read more about this concept when applied to tile making Adam 'Atomic' Saltsman wrote a very interesting tutorial that covered this subject.

Working with the Engine

All RPG engines that I know of have very specific rules for the tilesets.  These rules are the things that allow the engine to know exactly what it is you want those images to do while you're mapping with them.  Having a clear understanding of the resource standards for your engine is going to make a huge difference in the time and effort you spend on trying to get the tiles to work the way you want them to.  It can also help you make better decisions in what you should and shouldn't make.  Read the definition of those standards in the help file of your engine.  This is not optional.  It's also why I'm not going to go into more detail on this point, I'd just be repeating everything that help file says.

If you aren't working with an engine you'll need to at least format things consistently so that your programmer can make those definitions easily and without rearranging everything you just did.  It's important to work with your programmer on things like this, some things that you think should be easy really aren't and some that you think should be hard are simple.  Try to be someone that a programmer will enjoy working with, it's better for everyone in the long run.

Drawing Small Means Implying

There are some people who think that by not having to work at a large size all that loss of information means there's less work to do.  Nothing could be further from the truth.  Tiles are a bit like poetry because the economy of pixels means that each pixel matters even more.  You can no longer define things as what they are, you have to discover what makes the thing you're drawing look like itself when you can't actually draw it.  You have to imply the grass without actually drawing each blade.

One of the ways to do this is to focus on the silhouette.  If you get the general shape of something right, the insides can often be something completely different and it will still be recognizable.  A tree made of bubbles for a canopy and candy canes for a trunk is going to still look like a tree because of the silhouette.

Another way is to do the light and shadow correctly.  If you watch a really good pixel artist at work you'll notice that they almost never begin with an outline.  They block in the basic shapes, highlights and shadows and refine as they work.  It's a bit like a sculptor cutting away the things that don't look like the sculpture.  It becomes recognizable almost immediately and needs no refinement for you to know what it is.  You don't have to work in pixels to make a tileset, but the concept is the same.  Getting the highlights and shadows in the right places will define things better than an outline ever could.

When working small, you have to get the generals right, details won't save you.  In fact, details get in the way.

Ease of Use

At the end of all that work you're going to be making actual maps with this stuff.  How hard is that going to be?  Is that something you're going to do or is it something that has to be easy for someone other than you to understand and use?  Usually it's better to make the tiles as if someone else is going to use them without you, it makes it easier for you to use as well.  Sometimes that means careful naming of the files, sometimes that means adding a little bit of wall tile behind the decorative wall moldings, sometimes it means making tiles that are already layered together, and sometimes it means making each iteration of a 32x32 tile pieced together from what once was a series of 16x16 squares.

Testing out your tiles is an essential part of this, especially if you've never made a set before. 
Keep track of the quirks you find as you're testing and don't be afraid to change things so that they work better together.


I hope these things help you create your own tilesets, or perhaps better understand the effort involved in making those things to begin with.  Making your own set is tough, but seeing those tiles in a game, getting to see how it interacts with other elements of game making, watching people interacting with your art is exciting and lots of fun.  It's hard, but so very worth it in the end.