Some of you may have seen that I’ve been participating in a number of 1 hour game jams as of late (held at weekjam.com / #1hgj on AfterNET). They’re super short jams, so there’s not a lot to say about them – I might get around to doing a mass post mortem for them. Maybe. Until then, I am going to look back at a previous jam from earlier in the year and write a bit of a post mortem for it. This time I’m going to talk through my experiences at the 2015 Montreal Global Game Jam, where my team (known as “Not Enough Laptops”) entered for a third year running (this time with an additional programmer).
The Team
Anshul Goyal (Programming)
Sean Noonan (Art and additional design)
Pete Smith (Programming)
Aidan Green (Design and audio)
The Theme
What Do We Do Now?
Here was a theme that filled our minds with possibilities. It almost seemed “too easy”; it was hard not to come up with ideas. However it was not an easy task to decide which of these many ideas we should settle on. It was also when I noticed that we had matured as a group.
This was our third year jamming together and we weren’t as wide eyed, inexperienced or perhaps even as excited as we once were; we basically knew our limits, specialties, and we knew that we could succeed. We weren’t cynical by any means, but the ground felt familiar, and I feel our comforts resulted in an inability to decide on what to make.
Our first night had us pitching, scrawling, debating and scribbling out numerous game ideas; an adventure game of sorts where you wake up alongside a body with the police are banging on your door, a narrative heavy game where you play as a child dealing with the impending divorce of your parents, an isometric prisoner of war escape action game, and a bunch more… but nothing we could all fully agree on.
That is until midnight rolled around and we engaged full panic mode before thinking; what if we go with Pong, and the ball just vanishes? What could we do from there?
Pwrong.
The Jam
In essence, we planned our game as an homage to arcade games roughly spanning the generations; starting with Pong, moving through Pac-Man, switching to Akranoid, then R Type, in to Geometry Wars and finally teasing with a platformer.
Once we had finally decided on a rough concept we took our leave and slept on it some… as well as cramming in some last minute work - for example, I did a couple of colour scheme/interface mock ups, the first of which set the visual direction for the project.
The morning came and went – not that I saw it (I generally don’t rise during daylight on weekends). By the afternoon we realised that we were running behind our planned timeline. We should have had our first playable prototype, but were still not ready. It was around at this point I decided to cut myself off from the team and focus on the art.
I wasn’t used to working at such a high resolution, and the game type was something I’d never built art for. By the evening I’d managed to deliver the bulk of assets – I’d even over delivered by the scope of the game, as we had decided to cut the Geometry Wars game type and severely scale back the platforming section.
We pulled an all-nighter to just get it done. There was a bit of crankiness and we all felt awful, but as the sun rose and we tested the build, it felt worth it.
What went wrong
1. Brainstorm took far too long
As mentioned previously, this really took longer than it should have. I feel like I should shoulder the blame on this one. I personally had a hankering to create something pixel heavy; either something sickeningly cute or fast paced action with lots of explosions and screen shake. I was basically trying to steer all of the brainstorms towards one of these two forms. I simply wasn’t allowing for an organic process.
On top of this I also feel like I may have been on my own in that I was extremely resistant to anything narrative heavy. I wanted to create tight loops with high replayability; score attack, procedural generation, etc.
Basically, I feel my personal agenda got in the way on an early solution during brainstorming. I’ll try to be a little more aware of this in future jams.
2. The game was quite buggy
Though we fixed a bunch of issues post jam, we had a lot of bugs, most likely down to the scope of the game. I’m not sure what we could have done other than shrinking the scope, but with that being our unique feature, as it were, we were resistant to cut any more than we did.
3. Multiplayer only
Same story as with “No Evil”, multiplayer games, especially local, don’t always find an audience. They’re excellent games at the event locations, but once they’re released to the greater public, it’s easy for them to be ignored. Although we were low on time, it might have been preferable to drop the scope of the game a little in order to support solo play.
What went right
1. Two programmers helped with our scope
Our game was a collection of games – the increase in programmer count certainly was one of the driving factors in allowing us to achieve this feat. One of the programmers (Anshul) also wrote a post mortem where he discusses this in more depth – you can read that here.
2. We cut early
We cut the twin stick geometry wars clone early enough for us to continue to have a working flow, and for me to have not spent too much time creating new assets for it (though in the end, I had created most of them anyway).
Realistically it probably wouldn’t have been too difficult to implement the cut twin stick section post jam, but as is the case in most of these group game jams, the project is as it is on the final day, never to be returned to.
3. Tried a different visual style
A very personal positive from the jam – I tried a different style and scale and managed to pull it off. I never work at HD resolutions and always use pixel art as a crutch.
This really highlighted the need for me to start using Illustrator – using raster graphics to fake vectors is a huge time sink, especially when you consider that I have a program as powerful as Illustrator installed on my machine. I have already started learning Illustrator for future work.
The Game
Pwrong was well received, especially in it’s scope – we didn’t “win” anything as such but were name dropped amongst a list of honourable mentions. It was definitely one of the best jam games I’ve been a part of and it set me on a path of exploring vector art for future games.
Pwrong can be played over at GameJolt here: http://gamejolt.com/games/pwrong/45763
Final Words
Over the last couple of years “Not Enough Laptops” have followed up the Global Game Jam with the Ubisoft Game Jam under the name “Titanic Conspiracy”. However, with half of the members leaving Ubisoft and the other half wanting to participate in Ludum Dare instead, we decided to retire the name and sit out of the third Ubisoft jam.
Perhaps this was the final “Not Enough Laptops”/”Titanic Conspiracy” game, but I owe a lot to the various members for the opportunity to make games that I would never have on my own.
Here are the developers I’ve had the pleasure of working alongside in the jams over the last few years:
Aidan Green
I met Aidan at Ubisoft when working alongside him on Watch_Dogs; it was here where he introduced me to Construct 2, as well as the Montreal IGDA demo night. The demo night acted as an inspiring kick up the arse to participate in the global game jam. I strongly recommended anyone reading this to go to one of these IGDA nights in your area, or even travel if needed. Very inspiring stuff.
In our game jams, Aidan was our primary designer and sometimes audio creator. Though we often had conflicting opinions on where to focus our time on design, I think resulted in richer experiences – with Aidan’s focus on mechanics and mine on details, we created games I don’t think that we’d have been able to create alone.
Aidan left Ubisoft Montreal after shipping Watch_Dogs and a stint on Far Cry 4. He’s now at Eidos Montreal working on Deus Ex - Mankind Divided.
Anshul Goyal
A coding machine, Anshul was another Ubisoft employee. I never had the pleasure to work professionally alongside him, but rather met him through Aidan. Anshul was responsible for the code side of all of our jams.
Seriously, a machine. We couldn’t have done any of our jams without him.
Anshul is still at Ubisoft Montreal.
Scott Morin
Scott and I started Ubisoft on the same day and quickly became design pals, despite his love for Blizzard and World of Warcraft (what a massive nerd).
Scott was present for our Ubisoft Montreal jams and responsible for design and art duties.
Scott left Ubisoft Montreal after shipping Watch_Dogs and now teaches level design at Vancouver Film School. Scott and I are also working on this…
Pete Smith
I met Pete through Aidan just before the 2015 global game jam. I don’t think we’d have been able to finish Pwrong without him. Not only did he bring code, but really solid design feedback.
Pete left Ubisoft Montreal a couple months ago and is now going it alone.
Jean-Sébastien Caron
JS joined us for the 2013 Global Game Jam providing art. There was a wee language barrier at times, but that’s not unusual for game development in Montreal.
I’ve not heard from JS in a while, but a quick Google shows that he’s currently an animator at FAKE Digitial Entertainment. Nice.
David Baron
Indie warrior and fellow level designer on Watch_Dogs, David was present during our first couple of jams, providing design input and feedback.
David left Ubisoft in the later days of Watch_Dogs and is now at Eidos Montreal working on Deus Ex - Mankind Divided.
Many thanks to you all.
This first version was simply an auto runner and represented those first 5 hours from my 0h Game Jam tests; complete with fixed jump height, fixed speed, and randomly generated platforms with varying widths and heights. The player’s distance was displayed on screen and the Game Boy aesthetic already in place.
With the base in place I needed to add the attack. I had already decided that the input method had to be a single tap or key press, so I thought the best implementation was to add an attack that happened whilst the player was mid jump. This is where I started to think about Castlevania with it’s strict jump heights and whip mechanics. So I gave my character a whip and something to hit, in the form of candles.
After sending the build around some friends and coworkers I realised my mistake in making the game to suit my personal difficulty level – it was barely playable for other people and nobody really understood how the whip worked or what the candles were for. So I added a tutorial in the title screen. I was determined to avoid throwing up controls or instructions for the player. I wanted players to work out how to play for themselves. The tutorial forced players to jump due to it being their only possible action. Then at the height of their jump they were required to hit a candle to start the game. It was pretty successful for the most part.
At this point my confidence was growing and I knew where I was going to go with it next. I added jump heights based upon the amount of time the player held their jump input for and made the game increase in speed over time. This resulted in the game feeling pretty tight - the player had an increased level of control but was also being challenged over time. It started to feel like a game.
I updated the football headed main character with something a little easier to animate (the hat in particular), the placeholder “FIRE” effect was replaced with a small explosion and an animated melted candle, and I added four new layers of parallax and an animated background to look as though there was lightning flashing at different points.
At this point minor changes started to take a lot longer than expected. All vector fonts were removed and replaced with pixel fonts. That process took a lot longer than I had hoped it would, but it was uncharted territory.
Posts
