More Notes on Limited Palettes

So, last post was about creating a palette and using it to create the graphics for your game. Apparently when I said "That's all I have to say about the issue" I was lying :P

This time, I'll be looking at how to take the palette we've already created and work with it to create a few characters that, whilst sharing a common colour scheme, look unique.

One big problem with using a fixed palette is that, if you tend to draw your characters with a similar style each time, they can start to look a little similar. There are a few ways around having a bunch of clones in your game - try and use different skin tones, body shapes, hairstyles and poses to give each character a unique look. You can dress up your character with so many extras - ties, hats, scarves, gloves, stockings, belts, jackets, vests - there are a ton of possibilities, so if you are worried that your characters look 'samey', dress them up!

Here's 6 characters I've drawn up with the palette from last time:



I've drawn three different characters in both male and female form, bringing us to six in all. As you can see, even though our palette is very small, we can use different hair colours, different clothing colours and different skin tones with a little bit of experimentation. This allows us to get some real variety in character types, which is definitely a bonus.

Combined with less human characters, like our chess piece from before, we really have an almost unlimited range of characters we can create. If you're running out of ideas, try going a bit silly with body shapes and see what you can come up with - I did this in Annie Android and the result seemed to work well. If you've got a set skin tone and a set clothing tone, try switching them around - blue skin and pink hair makes for a great alien.

I hope these few extra little notes and examples give you some more to think about!

Establishing a Palette

I've been meaning to post about using a set palette to illustrate a game for a while now. Although this is basically structured like a tutorial, please keep in mind that this is only one way of creating a palette for your game, and not only are there other ways, there are also much better ways (I am sure). Nevertheless, this is how I go about creating my palettes.

Before I get stuck in, you may ask "Why use a limited palette?". Well, there's a number of reasons - it makes your graphics all blend together very well, seeing as they share the same colours, it lets you draw your game rather quickly once you have established the colour set because you're not mixing new colour after this point, just using the ones you have already decided on. It also gives your game a unique look that some find desirable and others less so.

For me, the palette I use is mostly decided when I draw the very first character sprite. I very rarely mix any new colours once that first sprite is drawn and coloured, so it's very important to create a character that is going to allow for some variety in colours.



The first step here is to draw the character's outline. This character isn't too radical but should illustrate the process fairly nicely. Once I've come up with a character that I am fairly satisfied with, I begin to colour. The 7 little boxes there represent the palette I have eventually decided on. Note I have split them into two columns, the cool colours and the warm colours. I find a palette a lot easier to work with if I give myself both warm and cool colours as it allows for much greater diversity when shading. Note that when I mix the highlights I move the hues towards the warmer end of the spectrum and with the shadows I move the hues towards the cooler end of the spectrum. Saturation is very important, however I mostly just guess my way through it.

With my 7 colours decided on and refined, I proceed to colour the character. I find having an outline very important when I use limited palettes as it stops parts of the character from blending in with the background and looking strange. While I mostly stick to my ramps (ie, purple shade for blue tone, brown shade for peach tone) I sometimes mix it up if the situation requires it (notice the balloon has a purple shade on a brown tone - mixing your colour ramps like this is a good way of getting the most out of your palette).

Once I am satisfied with how the character looks and have tweaked the colours, I often draw a second character with a different appearance to make sure I've given myself enough variety to create my graphics with. Here I've done a chess piece, and the colours manage to represent the shine and colour of the knight quite well, so I am satisfied.

And that's it! I have 7 colours and I am ready to create all the graphics for a game! 7 colours isn't much, but because I've given myself heaps of variety with hue and brightness (whilst keeping them in a uniform looking palette) I should have no trouble creating an entire game with these tones.

Here's a quick sketch to give you an example of how I begin to create a background with these colours:



Something important to remember is that because we're using so few colours, you have to be creative with the colours you use. Here I've drawn peach coloured clouds, which certainly are not something you see in the sky, however they fit the palette and the illusion works well enough to give the suggestion of clouds. Note in certain places I've mixed my ramps - when I ran out of highlight colours for the light coloured squares on the chess board, I used the warm highlight on a cool colour, and it still shows what I want it to. On the fence I've mixed warm midtone with cool shadow tone to give a more metallic appearance. With further experimentation a palette can really be expanded to give a range of tones.

That's about all I have to say on the subject - I hope it's been of some help to those of you who wish to use limited palettes, and please feel free to share any comments you have about using this form of illustration in games!

Inside the Box

It's been nearly a week since Shifter's Box was released, and to my delight it has been incredibly well received. I had a feeling that this was a good game, but there always remains doubt when it comes to release time.

I believe it is important to move on and not dwell on the one release for too long, but I believe also that it is important to learn from past efforts, so I am going to share my thoughts with you about Shifter's Box. Please feel free to comment on any of these points or any of your own if you had something to add...

(a quick note - the following entry is long, dull, and contains spoilers for the game)

Shifter's Box started out very different to how it ended up. The first couple of rooms and the main character sprite I had laying around on my hard drive intended for a very simple, basically puzzle free game. I had set that idea aside (because it was pretty bad) and so the scenes were just sitting there, however I liked the concept I had built the story around (of a box that exists within itself) and decided to take another crack at it.

My first big decision when making the game was to use a completely different colour palette for every different location in the game. With Annie Android I pushed myself to adhering to a limit of just 6 colours - what this meant was that once I had the colour palette down, all the scenes were very quick to draw and animate. I basically got 80% of the graphics down for that game in one weekend.

Because I was doing the complete opposite for Shifter's Box, it took me a lot longer to produce the graphics. Some of the palette ideas I threw around were quite terrible and needed redoing, and some of the ideas I had for a world just didn't look any good when I drew them and so I had to start all over again. Rochelle was initially going to live in a giant tree, in a grassy place. Longblow Tower was initially without walls (just columns) and had a pink sky (which I ended up reusing). I had drawn one scene that was basically all bile coloured. Thankfully I took the time to redo these ideas ;).

Doing all these colour palettes was a big push for me as an illustrator and encouraged me to experiment a lot more. It also meant that I spent quite a bit of time sitting staring at a blank screen thinking "Where can I take the game to next!? I've run out of colours!".

The next big decision was to focus a lot more on puzzle design than I ever had before. I find most of the puzzles from the previous three games quite unsatisfying both in concept and execution, and wanted to put puzzles into the game that I felt were well designed. My two favourite puzzles (spoiler alert here)  are the island with steam pipes and the one with the lamps and the floating door, which seem to be the most popular puzzles with other people as well. I was quite proud of the cat + jellyfish puzzle as it took me some time to design and put together, but I feel it is a lesser puzzle than the other two for a couple of reasons. The first is that the clue to solving the puzzle is not as clear as with the other two, therefore it could be seen as a task of guessing and checking to get it right. The other is that it is quite illogical... who would expect a cat to jump up at a floating jellyfish?

Another thing I tried to do with the puzzle design was to keep a nice feel of diversity between the puzzles. Someone commented that there really wasn't anything to do in the tower, and that it was quite easy. This was something I purposefully strove to do - I wanted to throw a simple puzzle between two more complex ones to stop the player from feeling like they're just going from getting stuck to getting stuck. I feel it works quite well, however I can understand if this is up for debate. I also designed the beginning of the game around this concept. Before throwing you straight into the steam platform puzzle, and instead of making you watch Sally open the box, I thought it'd be fun just to have a really simple puzzle to get players used to the game mechanics and what they'd be doing through the game.

Story wise there really isn't much to the game, but I did try a few new things. Sebastian gave me the idea of using a narrator, which I took and completely ran away with, creating Colonel Slimbritches (who was thankfully renamed to Ebenezer Leary by Sebastian) mainly because I wanted the severed, framed head of a 19th century hobbyist scientist in the game. Writing his lines was enormous fun for me, and he really is my favourite character in the game. I also felt like he finally added some story to the game (he was one of the last things added) and added a bit of humour to an otherwise quite dull story.

Animation wise, I put all the basic animations in and realized that most of the worlds looked very lifeless. I wanted to give a nice feeling of life to all of them, and so I added a whole bunch of background animations in, which I think work quite well. There's quite a bit of animation in the game (I think the sprite count is somewhere around 830) and I feel that it was a very good decision.

When it comes to the game's story, I did something very different from my past efforts, and yet with hints of the others still in there. It shares a common theme with Man Boy and Shoot! of being in a strange place and needing to return home, but if you look at all of the games I've worked on in the past I've always created a very obvious villain character. When I wrote Shifter's Box (and I didn't realize this until all the story was in place) I didn't have a 'bad guy' - in fact, if anyone is to blame for events, it is the main character, who is certainly not a bad guy. Considering I had got to a point where I was worried that all of my games would always be 'Good vs Bad' I'm pretty happy with the way it turned out.

The last thing I'll analyse is the actual construction of the game. I've always been fairly poor at putting games together (if any experienced scripters ever saw the game source file for Man Boy vs Doctor Sock they'd probably weep bitter tears) but I did my best to cover as much scripting ground as I comfortably could without getting myself into a situation where I simply could not put something together. Things like the cat puzzle saw me writing a very clear list of things that needed to be done and ticking them off bit by bit when I'd completed each milestone. I also got to be creative in very small ways that help the game - when you ride the hoverboard over to Kadin, you have a shadow underneath which disappears when you are over the air, which I initially thought I couldn't do. Small things like this add to the game, even if nobody notices.

And yes, there are a number of bugs in the game. But I'm quite proud of the things I did manage to script, and hope to use the knowledge I gained doing this game to make even better games in the future.

As a small point of interest, when 2009 started and I looked at what I wanted to achieve for the year, one of the big ones was "I want to make a game that gets a 4 blue cup rating". 2 months in and I've managed to do this, which is very exciting for me.

The first week of Shifter's Box has been a very exciting time for me, and it's been awesome to see all the nice responses. Let's just hope I can get another game (or maybe two ;) ) out before the year is done!

Re: lease.

Dear readers,

Some few days ago I posted a screenshot of a game I promised to release soon. Soon is now. The game has been released.

I will post my thoughts and other such things about the game, but for now, here is a quick summary of the game features:

-Science fiction/fantasy/weirdo plot involving extra dimensions, a box and a jellyfish.

-More complex puzzles than any Ben304 game to date.

-More animation than any Ben304 game to date (around three and a half times as many sprites as Annie Android)

-A completely different palette for every area.

-Multi dimensional soundtrack by the esteemed Sebastian Pfaller.

-Wide array of audio effects by Mr David Pfaller.

-A panel of advisors including the aforementioned Sebastian, Ghost, Bicilotti and Pittiz.

-100% less lens flare than previously indicated.

..and some other things...

If you wish to play the game, you may do so here.