Featured Blog | This community-written post highlights the best of what the game industry has to offer. Read more like it on the Game Developer Blogs or learn how to Submit Your Own Blog Post
Postmortem - Mobsferatu, our game for the LudumDare 34
A look into how we made Mobsferatu, the problems we had during the our first LudumDare and how much we worked and how little we slept in those 72 hours.
LudumDare 34 has been a quite curious experience for us, CremaGames. Not only it was the first time we worked on a jam as a team, it was also a really strange challenge since two themes were chosen for it.
We spent three days working a lot of extra hours on a side project and it was really worth it: not only because we loved what we had achieved with Mobsferatu, but also because we ranked in a very, very good place: we are the 5th game in the Overall category! That means our game was the fifth most liked of all the jam! Amazing!
Before you start reading this long, long post, have you played Mobsferatu? Here’s a link if you haven't. We recommend you download it, the webplayer version is not as smooth as intended. Played it already? Wicked. Let’s go on.
As it was our first jam ever, we wanted to take part in the event not only because we liked the challenge, but because we love working together, even in sleep deprivation, and because it’s a great tool for learning. Adri, Carlos, Dani, Javi, Alex, Guille and Enrique – seven people in total – worked 60 hours in three days. Two more joined in the final hours, Ana and Paloma; and we also had help of two great audio design students: Vicent Tobar (Music) and Juan Novella (SFX). So, everyone except the person writing this (hah) worked a lot and against the clock.
This postmortem is an attempt to remember that both great and awful weekend, to show how we worked and to share what we learned.
The idea
It’s important to point that this was our first LudumDare. It was also a very special one for everyone because two themes tied and the organization decided to make both available to develop.
With that in our minds, we used the first three hours, from 3.00 AM to 6:00 AM in Spain to brainstorm. Yeah, the time zone sucked, but we were full of energy and we had great ideas to start with. We had a lot more ideas before the voting ended and there was a tie between two of the themes, but hey, we came here for the challenge!
We knew most people would choose something about plants, vegetation, rockets, planes and such since the themes were so obviously easy to implement with them. We avoided those things in particular and we also chose 3D over 2D because we love to swim against the current and because most games always are in 2D.
The challenge was everything for as at this moment. Silly people…
The idea we started working with at first was about an architect building a tower. The player had to carefully choose materials and build as fast as he could. It had both themes and it was a neat idea. It ended up being too complex for the time we had. We decided to go on with half this concept: we said goodbye to the architect and the tower remained.
Soon enough we agreed that a group of people would be running around it, going up as they could and collecting more people as they progressed. A always growing angry mob was just the perfect concept to add, and the two buttons control worked great as they would jump, change direction and dash easily. It would also look really complex, because they would run and jump following the leaders.
As huge Simpsons dorks, we imagined the mob in a very specific way, so making everyone somehow different and noticeable would be a part of the game, too.
The first steps and missteps
With all of this in mind, we started working in different layouts for the characters. We needed them to be swapped easily every time you played. Also, we made a first attempt with the tower design.
We used the following tools:
– Unity 5.3
– Photoshop
– 3DS Max
– Substance Designer / Painter
The art guys, Javi and Alex, did a terrific job since the beginning, to the point that they felt a little relaxed by the last day – lucky guys. The first concept art Javi did of the tower was so great it helped everyone focus on exactly what we needed to do and how to do it. Alex started making the first five villagers, with a focus on making them different. Even if the point was to make a angry mob, we wanted some individuality within it.
We draw a lot from inspiration from Nebulus, an awesome NES game that inspired some great titles in the future. At that moment, we didn’t know about Mr. Crab, a game for smartphones that had some of the same ideas we brought to the table for Mobsferatu, but a lot of people pointed the resemblance once we posted the game online. It seems everything is invented after all, isn’t it?
This was the first prototype.
At the end of the first day, we had this already:
The inspiration for the characters was The Legend of Zelda: Wind Waker. It seems strange, we know, but we wanted something that looked a little like celshading. And which is the best celshading game ever? You don’t have to answer, everybody knows Wind Waker is ‘da best’.
We all agree now that we went a little too far with the characters. We started with five of them: three with funny hats, two of them hatless. Two days later, we had a editor to randomly build and combine them. Take a look at some examples:
t this moment, sleepiness started to kick in and we made our first mistake: taking too much caffeine. Did you know coffee, RedBull and Burn are not a great mix? We discovered it the bad way and one of the members of the time had to go home and rest. Yeah, next time we will probably buy more fruit and vegetables and less junk food. Staying kind of healthy, not only awake, really affects what you can produce in a jam. Thank god this happened in the first day, anyway.
Art and programming were working against the clock to include the main stuff at the end of the second day. Design was thinking about what kind of challenges they could implement. After almost 40 hours of making more characters, models and textures, being sure the code worked as intended and the bugs were, at least, noted to be solved.
The ‘wooo’ noise for the the dash was something we did not plan initially, but our awesome sound designers included it. We loved as much as everybody online. And yes, this video has a lot a bugs, but that’s the way it is when you only have 72 hours for a game.
This is what we had at that moment:
The music was already on, too. We loved the work they did instantly: it helped the atmosphere a lot and it even changed a little when you got higher and higher.
The game was looking good, but we needed to implement the lighting and all the details so it looked great. In the end, we had a lot of little stuff: two textures for the floor and the walls, doors on the walls, stained glass windows, rocks for the edges, armors to decorate the tower and, of course, the traps. We had doors, doors with spikes, rolling trunks with spikes and lava and poisonous floors.
We really think that all these little details are what make the game really eye-catching. The lighting, the shaders, the glow; all this stuff is really helpful, but having a crowded game with all this little parts working together makes a great impression. And only in three days.
Even after all of this, we still didn’t know if the tower would be randomly generated or a fixed design. We really wanted to make something different for each player, but after acknowledging the little time we had, Enrique started making the tower final design. The first one was a little too hard, so he made some changes and included arrows to point where to go. Still, it felt kind of hard to some people, but we wanted it to be a little challenging, anyway.
There was just something left: the ending. Even if we would have loved to make a final boss, the time was so scarce we opted for a joke. A very well known one: at the top of Nosferatu’s tower, a little servant with a red and white hat tells the player that Nosferatu is in another tower, and the game ends.
This silly Super Mario Bros. joke seemed like a neat idea and a way to make game end without a boss, but not in an anticlimactic way. Most people found it funny online and no one thought the game was incomplete. Oh, and the music in this precise moment is outstanding, making it even more ridiculous and the reference more obvious, yet.
The streaming
The whole jam was streamed in Twitch. We thought it would be really funny to show how we worked, how tired we were at the end and what each one of us was doing every time. We had seven different points of view, one for each one of the members plus a general view of the office. We had some fun when someone was sleeping or using the main camera as a joke, too.
Overall, it was a good idea, one that we’ll try again in another circumstances – like not being sleepy while working. Maybe a live devlog? We don’t know yet, but having people watching us made our work a little more self-aware and made us think that we had to end what we had in hands because people wanted to see it. Also, the voting system to choose each camera, or a combined view worked really well and a lot of people took part in how the streaming worked.
Actually, those viewers where really helpful when we needed a name for Mobsferatu. Even if one of us come with such a cool name, it was possible thanks to the online brainstorming with our viewers. So thanks to everyone who watched us.
What it didn’t work
We learned a lot of stuff working in Mobsferatu, so only for that, this was worth it. Let’s take a look at what went wrong.
We learned the hard way that WebGL and Unity3D do not like shaders. This was a really hard thing to solve, we didn’t have the time to do it and, sincerely, we were exhausted. We sacrificed WebGL support, so no Chrome version of the game online. This is something we are sure affected the likeness of Mobsferatu: making everyone change to Firefox or download the game after seeing a ‘this web browser is not supported’ screen is bad.
The Webplayer version we made for Firefox also had a huge problem: it worked neat on Windows, but only on Windows. The Mac version was broken and a black screen appeared when you executed the game. We tried to solve this by changing from linear to gamma correction, but this made things worse: now all versions looked too dark online, something we didn’t notice until the players told us about it. Again, we tried to solve it, but the Webplayer version looked kind of overexposed after all. The best way to play Mobsferatu is to download it.
After this two missteps, we could say that if you downloaded Mobsferatu, everything worked as intended and the game looked great. The bad first impression was there in some cases, but people forgave us for it.
Now, the little mistakes. For example, choosing Z and X as the two buttons for the game. You see, Spain has a QWERTY layout in all their PCs, although it’s not the same as the english one – we have another letter, the Ñ. We didn’t consider other keyboard layouts, like AZERTY, where Z and X are not together. Mobsferatu control was awkward in some parts of the world.
More rookie mistakes: working on unnecessary stuff. Don’t get us wrong: having windows, good shadows, a nice lighting and such was the best thing about Mobsferatu and everyone loved that. The problem is that we made a lot of layouts for the villagers, most of them expendable because you couldn’t really notice the different faces once you started playing.
Oh, and don’t forget about the strange bug that made the moaning of the villagers sound constantly when one of them was left behind. We tried to emulate that time Peter Griffin hurt his knee, but instead it sounded like, ehem, porn. That was a really funny/disturbing bug, but we solved it in the following hours, though some streamers had some fun with it…
Lastly, a lesson for the future: the ending is not that important and not everyone is as skillful as us playing our game. We don’t think Mobsferatu was that hard. We wanted a game that required a few tries before you could reach the end, but when there’s a thousand pending games to play, no one finished every single one of them. The pun at the top of the tower was good, but most people didn’t see it. And that make us kind of sad…
The happy ending
As we already said, only the experience and the lessons learned were enough reward for us, but it seems Mobsferatu was a even greater game that we thought because it ranked fifth in the Overall category of the LudumDare. That means it’s the 5th most liked game in the whole jam, the first one if you only consider those made in 3D!
This is something amazing and we didn’t expect to rank so high. We made a little bet as an inside joke in the studio and almost no one thought we would get so high. We are really, really humbled be the response, by so many nice comments and because so many people has played something we build from scratch in so little time. It’s like a late Christmas gift!
Right now, we don’t know what we will make with Mobsferatu. It’s a good game for a jam, but we are not really sure if it would fit a more wide launch, even if we worked on it. We are knee deep into Immortal Redneck’s development, so spending resource on it right now it’s very unlikely, but never say never.
Once again, thank you so much for the response, for playing Mobsferatu and for voting it. We’ll be back for another jam, we promise. Maybe not for the April LudumDare, but maybe in August…
Read more about:
Featured BlogsAbout the Author
You May Also Like