Reskinning A Game
..and it’s flaming cold, again. Don’t you just love the weather in the UK! 🙂
I haven’t had chance to update my blog lately – things have been a bit chaotic for me both at home and at work. I’ve had to put my Unity 3D Pool game on hold in order to raise some much needed funds. For the first time in 23 years I’ve been doing proper job – one that (sadly) doesn’t involve the creation of spaceships. I’ve been working for a company called Pearson / Edexcel. They are responsible for the mammoth task of orchestrating the marking of the UK’s exam papers over the Summer. I can tell you they do an amazing job – the logistics are truly mind-blowing. But my time there is coming to an end now, so it’s time get back to the serious task of building spaceships again…
Don’t do Drugs….
No kids – don’t do drugs, unless of course its alcohol or caffeine, both of which are mandatory if you work with computers. Prior to doing this work at Pearson, I had the opportunity to work with an old friend of mine on a game he had developed for iOS and Android. It’s a match-3 game that features a lots of drugs. Kind of like Candy Crush, but with added Viagra. The game was already complete and fully playable. I was asked to re-skin the product to a high standard, replacing the ‘stock’ content that was already in there.
So I came out of artist retirement to do the work, despite my vow never to do any artwork under contract again. And, while the task itself went very smoothly, I ran into the same old freelance payment problems (more on that at a later date). This is definitely the last art contract I will ever do. Which is kind of nice really, as I will have done it with the same coder and musician that I worked with on my first game, 23 year years ago… (Speedy Gonzales on Game Boy!)
I decided to use Trello.com to organise the work, something I’ve been using for my own Unity SDK projects. I wasn’t sure if the coder I was working with would take to it, but I think he started to warm to it a few weeks into the project. All the art to be replaced was uploaded there, along with all the reference material. We had panels for ‘work in progress’ art – where we could discuss the suitability of various elements.
Essentially, the assets ‘flowed’ from the original elements on the left, through an approvals area, then finally into a ‘game ready’ section at the end for slotting directly into the product. A lot of the work was ‘like for like’ replacements – however it wasn’t practical to replace much of the User Interface in this way, so the coder re-implemented new layouts to suit the new artwork.
The onus really, was on getting the work done as quickly (therefore as cheaply) as possible. But, the truth is – art is something you can’t rush. If you skimp on time, it shows in the end result. And people are very quick to point out when something doesn’t look right, so short-cuts were a bit thin on the ground.
There were a few surprises along the way – not least the complexity of the tile sets underlying the UI and game map. As is the case with a lot of game assets, things are created, added to, removed, changed over the course of development. So getting in there and replacing a very cryptic looking set of tiles required a fair bit of reverse-engineering. This is something to bear in mind if you take on a re-skinning job (and frankly, I underestimated that side of things!)
In the end, the game art was created using a combination of 3D Studio Max, Adobe Illustrator, Adobe Photoshop – and a new piece of software, Anime Studio Debut. It was impractical to re-model and re-render many of the game assets, so a lot of the game elements are plain, unmapped 3D models, heavily re-worked and textured in Illustrator or Photoshop.
Anime Studio Debut – A low cost solution to 2D animation
Now, part of the game featured a 3D render of the character performing a ‘thumbs up’. Replacing this with an upgraded character would have been impossible within the budget, so I opted to recreate it in 2D.
For this task, I ended up using a neat little application called Anime Studio Debut. It cost me in the region of £20, which is a bargain for what the software does. The application is actually the ‘entry level’ version of the software. And as such, it does have limitations – the most notable being the size of the canvas that it renders, which is just 900×900. Not ideal for epic HD animations. But, it’s an okay size for most game sprites.
It took a bit of experimentation to find the best method of 2D rigging the character. Originally I thought all parts of the character needed to be a separate, but I had problems at the joints of the knees and arms. In the end, I opted to keep some parts of the figure whole – with the software handling the joints – and that worked really well. I was impressed with the feature set, which included Inverse Kinematics and joint limits. You can plant the characters feet on the floor and the legs will buckle automatically if you move the body down. You can also switch the various 2D elements during the animation (different heads for example) – which is great. Definitely a useful tool for doing this sort of thing on a tight budget!
I learned a fair bit from doing this project. I found a number of short-cuts for producing some clean looking cartoon art (cartoon work isn’t one of my strengths!) If I get the opportunity in the coming weeks, I will write about the methods / procedures used for doing this sort of 2D work. You might find some of the ideas useful for your own projects.