Friday 15 May 2015

Off the Map Post Mortem



This year’s Off the Map competition was to submit a game based off of Lewis Carroll’s Alice’s Adventures in Wonderland. As a group my team and I decided that we wanted to create a fully playable game that tells the whole story of the book from start to finish, although we did not intend to include every single part of the story because of the timeframe. To decide which parts of the story to include we gathered together after reading the book to pick out the parts we wanted in the game, taking into account possible gameplay mechanics associated with particular parts, significance to the overall story, how memorable a part/character was to us and so on.

The game we created was a side scroller and from the start we decided for it to be primarily story based with some puzzles, but we did not want to include a fail state, no matter what we wanted to player to progress with ease. My role in the project was to block out the whole game, set up the puzzle blueprints, make the character conversation blueprints, add text and create various props. The project was 11 weeks long, the majority of that time I spent visual scripting via blueprints, but fortunately the engine work was shared among my group so I was also able to create some assets for the level.

I think we have some strong parts to our project which went well. The beginning illustrated section ended up looking quite good. Our intention with how our level would start was that the player would start off in a John Tenniel style illustrated book like area, and as you play 3D coloured assets would begin to come into play. The first of the 3 dimensional assets is Alice herself, then the key objects to the key and door puzzle, and then once the puzzle is completed you’re thrown into an almost 3D world where the only remaining 2D illustrated bits are the mouse and the caucus race. The rest of the game is then 3D and in full colour until the end where in the credits Alice wakes up on the river bank in the illustrated style. I think we pulled off this quite well, the quality of the drawings by Max Bostock were very good and I think that the player feels like they’re entering wonderland once the world comes to life in 3D and colour.

The text we used in the game turned out well, we wanted to stick to text from the book as much as we could, I added the occasional sentence and re arranging the position of dialogue to fit the game we were making. We managed to turn Alice’s confrontation with the Caterpillar into a multiple choice conversation, it took me days of working in blueprints to work out how to pull it off as well as the two other character conversations we have in the game, but I really think it paid off. I was happy with how the text boxes themselves turned out as well, there was a point where I wasn’t sure if they’d make it into the game as well as they did, we wanted the text boxes to be in the 3D space rather than the players HUD, but we also wanted the text to always face the camera so the player could read them. We tried out billboard textures, but they didn’t seem to work, (I didn’t find out why, but I think it’s because it’s a sidescroller) so we made them particle effects that were stationary and constantly spawned.

To our delight the mushroom section came out looking quite good, I was concerned that it would be our weakest part because it needed to be created very late into the project by our main character artist(Michael Ollerton) because the team member who was in charge of that area left the course halfway through the project. The section had a very magical feel to it, it was dark and most of the light on the screen was from particle effects and emissive maps on some of the mushrooms.

Our crying statue of Alice turned out well, the feedback from peers has been positive about it. When I pitched the idea to the team of a giant crying statue of Alice which the player falls out of, to symbolise Alice shrinking as she cries, I thought it might’ve been too outlandish of an idea and didn’t think my team would want it in our level. Max Bostock pulled it off, it’s the player’s initial big transition in the fully 3D wonderland, and it’s a weird and wonderful way of saying “Welcome to Wonderland”.

I like the Cheshire Cat section, it’s strange, a little bit creepy and has a jump puzzle. The Cheshire Cat being this multiple sectioned body that floats about to create platforms for the player to use was the first idea I came up with for this project, and I’m really pleased it made it all the way into the final game. It was fun recreating Alice’s famous conversation with the cat and blocking out the matinee that became the cat turning parts of its body into a bridge for the player. Max and I put a lot of planning into this ‘puzzle’ to make sure that the player didn’t feel too intimidated by it, like I stated before we didn’t want a fail state for our game. The section started out as a fall of a cliff and respawn scenario if you failed a jump, but we wanted our game to be one continuous journey without the need for a respawn, so we came up with the ravine with the bouncy mushrooms for the player to try the puzzle again.

Parts of our project didn’t go as well as we hoped, one big part being the length of the game. To begin with my white box of the level was much larger; we were planning for the game to be much longer. I did strategically plan 3 areas that we were ready to cut out if we needed to, those areas being going into the white rabbit’s house, going through the Duchess’ house and talking to the mock turtle. Both the white rabbit’s and the Duchess’ houses were cut to objects in the background as part of the scenery the player walks past, and the  Mock turtle section was removed from the game entierly. In addition to those planned cuts we made further cuts, in the book after visiting the Caterpillar, Alice runs through trial and error with her size as she eats parts of the mushroom. To represent this I planned for there to be a walking section where the scenery around Alice grows and shrinks repeatedly until she gets to the right size again. This area was cut out, in its place we placed giant pots and pans leading up to the Duchess’ house which get smaller and smaller to symbolise Alice returning to her original size. Another additional cut was the queen’s gardens and the court room, originally our level ended with passing through the queen’s gardens, being arrested and put into court and chased by card men and cut into Alice waking up. Instead we finished our game after the Mad Hatter’s section, telling the player to go hide as the queen is chasing her, and the game ends there.

Another thing that didn’t go so well was my time spent on blueprints. Lots of time needs to be spent on blueprints if you want them to be complex and work well, I spent lots of time, and then found I had to spend lots more due to glitches and problems with our game engine: Unreal Engine 4. I’m going to explain briefly about where you can create blueprints in order to explain the issue I was having:

·         You can choose to build blueprints in the level blueprint, which is what you’d do to make a blueprint that’s a one off for the level and doesn’t need duplicating
·         You can build blueprints in a sub level blueprint if you create sub levels, it’s a lot like the level blueprint but there’s one for each sub level
·         Or you can build individual blueprints by converting something, for example a box collision, into a blueprint which can be duplicated with ease

I chose to build the first puzzle in our game in the level blueprint, it made sense, it did not need duplicating, and making it an individual blueprint would only complicate and already complicated process. I spent a week learning how to create and building our first puzzle in the game, everything was working perfectly until I added the final few parts to the puzzle, then the game started glitching, the player would be able to jump as if there was very little gravity (even though there was absolutely nothing to do with the player or it’s movement in the blueprint), parts of the puzzle were activating prematurely and so on. I eventually worked out that the problem I was facing was that when a blueprint reaches a certain complexity, it begins to break and glitch. So to fix this I rebuilt the puzzle in its own sub level blueprint, and the problem was fixed. But later in production it was decided that we should have text appear when the player activates certain parts of the puzzle, I put the appropriate parts to the blueprint to do this together and then it started breaking and glitching due to reaching this ‘complexity limit’. I had no time to rebuild the blueprint for a third time, but if I did I’d try to make it all in individual blueprints, and maybe that would fix it, but who knows.

It’s unfortunate that the thing stopping us from creating more work and making complex blueprints is down to the tools we use and not our own limitations.

After finishing this project and looking back on it all, I’ve tried to evaluate what I did and how I would change what I did if I were to do it again. First of all I’d try splitting that puzzle into individual blueprints like I said above. At the start of our project we were far too ambitious, and although I planned for parts of our game to be cut, I should have planned more areas to be cut again. Obviously we could not predict the group problems we faced or predict how those problems would affect our whole project, but next time I need to plan for an even worse case scenario again.

I really didn’t enjoy my time as being an engine/blueprint guy, it’s nothing like I intended to do when I joined this course, and I really hope I don’t have to do it again. I was learning a whole new thing which I didn’t want to learn. To me what I’ve made blueprint wise is complex, but if I was to show that to an actual programmer they might see it as primitive work. I wish I just made assets for the game and improved my skills there, instead i made a few assets and I didn’t get much of a chance to learn more about modelling or texturing because I spent very little time making them. Because of this I feel like I’ve wasted 3 months on something I have no interest in doing in the future, and I’ve watched my peers around me advancing 3 months ahead of me in the area that I actually want to work in. This makes me feel absolutely terrible and I’m going to have to work hard to get those 3 months back that I should have spent on my future career.

I wish that our level was more focused on the art and how it looks rather than implementing gameplay and trying to make it a game, I think we would have ended up with a much better outcome if we did this.

My action pan now is to spend some more time on Off the Map for polish it up for the competition hand in, but mostly I want to move onto my projects for the summer. This project has left me feeling pretty rubbish so the quicker I move onto a different project the better I’ll feel. I want to focus mainly on materials, I’m confident in my modelling skills, that’s why I’m going to do material studies, so make basic objects and focus on materials. I also want to perfect my normal map creating skills, I’ve come a long way in the past year in terms of creating normal, but I learned a lot for a lecture I attended recently about creating normal for hard surface assets, and I have not yet had the chance to try out what I learned. I intend to make lots of these small asset studies over the summer and push hard to get back those 3 months I’ve spent in blueprints.

No comments:

Post a Comment