Choosing the art style for a game, pt. 1

The initial purpose for this blog was to document the creation of my very own graphical point & click adventure game. So here we go!

I’ve had this underlying dream of creating my own game since like forever. Only lately I’ve realized that I actually might possess the required skills to do something. I have a long history with music creation. I’ve already posted a couple songs here and I’ll be posting more when I get something done. I’m also quite competent in creating 3D graphics. This is something I started almost 15 years ago with 3D Studio Max (version 4, if I remember correctly?). Now that I’ve graduated (and gotten employed) with a degree in ICT software development I have the programming part nailed also.

For me, the most important factors when playing games are the story, emotions, immersion and atmosphere. I won’t delve deeper in this matter now, but definitely later cause there are a lot of things to address.

So – point & click gaming genre. Quite simple in a programming perspective. Also my favorite genre since childhood. I started with the genre legend – Indiana Jones and The Fate of Atlantis. I was eight or nine at the time so highly suggestible also. (Weirdly though, the Indy movies never were that special for me.)

It’s a genre that relies heavily on the following aspects:

  • Story
  • Intriguing puzzles
  • Pretty art
  • Atmospheric audio

The narrative must be very, very, very superbly crafted! This I must handle in another blog post. Puzzles and audio too. Because I’m eager to post some screenshots and crappy graphics, this time I’ll focus on background art.

I have this personality trait of being overly pedantic and painstakinly precise with everything I do. There’s a good amount of OCD behaviour mixed in also. This can be quite gruelling but eventually leads to somewhat satisfactory results. So the bar is high and most of the time I dislike anything I achieve.

So before I started to create anything I had to benchmark my skills and analyze what I was going to do graphics-wise.

I had these rules for the background art in the game:

  1. 2D all the way.
  2. The art would be slightly painterly but with realistic overtones. (Or the other way around?)

As I am just a single person I realize that I have other limitations too. There is a saying that goes something like this: “High quality, low cost and quick results – choose two of these”. I agree that in a broad general way this is true. If I want high-class results fast, it is gonna cost. And if I want good results with low cost, it’s gonna take a lot of time. The third option would be fast and quick, but with a low quality. Such a shitty dilemma!

Fortunately there are gray areas and things you can compromise on.

As I’ve explained it is difficult for me to lower the requirements for the things I do. So I have to learn to allow myself to produce something that is not “perfect”. And here is a non-secret; I’m not a gifted painter either. By employing this approach I should gain some time savings then.

But the real key here is optimizing the workflow! This is something I constantly do when I’m working, or even when mixing music. This is contradictory with creating art and expressing oneself but it’s all about finding the balance. When the gloves come off, it’s not always pretty. We do not always have all the time in the world.

The aim is to achieve something pretty, something painterly but also something realistic. Something like the art in Broken Sword adventure games. I know, quite high aspirations. I wondered; could I (2D-)paint something? I started the whole process with searching something in Google Earth. I found this lovely little (Midsomer Murders -esque) village in the UK. I took a screenshot of the street view image and then used it as a reference in Blender. I modelled this atrocity quickly:

This truly is fucking horrible, sorry.
This truly is fucking horrible, sorry.

There was an ugly “HDR” dome acting as a light source in the sky and some quick repeating texture just for the reference. I then imported this picture to Photoshop and started painting over it using my Wacom Bamboo tablet. Although it was fun and educational, it took FOREVER. Like forever-forever. Here was the painting after many hours of work and an emerging carpal tunnel syndrome.

The guy is there for measuring reasons. And I like Alfas.
The guy is there for measuring reasons. And I like Alfas. Even though one almost destroyed me financially.

And it still isn’t finished as can be seen.

What I learned was that painting can be fun but I’m just too slow. And due to lack of experience, I don’t really have my own art style sot the result is clinical and dry.

What about retro-style-pixelated graphics then? I think it was Dave Gilbert from Wadjet Eye Games who said that they utilize lower resolution art so they can produce their graphics faster. This was definitely something that was worth looking into.

Well it appears that pixel art isn’t any faster to create, even if the resolution is super low. There are only so many pixels to use and pretty much all of them must be placed wisely.

Again I started with Google Earth! This time I found a nice intersection near Paris. Then I created a rough approximate of it in Blender.

Not nearly enough geometry but this was something I learned afterwards.
Not nearly enough geometry but this was something I didn’t consider at the time.


Then I spent too much time creating the sweet line art of the image and then adding some additional details by hand.

There's something in simple black and white line art that I just adore.
There’s something in simple black and white line art that I just adore.


Then it was time to paint over. I used some simple colors and slight dithering. I didn’t get to do any shadows or exiting details because once again – I was so slow!

Kind of like it but it's too cartoony, too childish somehow.
Kind of like it but it’s too cartoony, too childish somehow.


So yeah, pixel art was a no-go either. What I learned from all of these was that however I wanted to create 2D graphics, Blender was the way to go. Both of the pictures above suffered from poor lighting from the beginning. It is difficult (for me) to paint over something with such a bland lighting with an overcast sky. In the next post I will discuss how I further developed my Blender+paintover combination and were able to create a workflow with which I could produce decent graphics with minimal time. And in 1080p resolution!


Leave a Reply

Your email address will not be published. Required fields are marked *