EXNIHILO STUDIOS – Spelunk! – Post Mortem

Posted by (twitter: @@exnihilodev)
April 23rd, 2015 8:37 am

spelunk0

So this is where we share with you a little of the ‘behind-the-scenes’ crazy that happened during our Jam as we developed Spelunk!

Getting Ready

Ex Nihilo Studios does not have a full-time artist, so we generally subcontract artists on a project basis. Ludum Dare was just such a project and so I placed an ad on PixelJoint.com looking to hire a pixel artist for a 72 hour project. Six artists responded and after evaluating their work, I contacted ShroomArts to let him know we wanted to work with him. We quickly worked out a communication plan as he and I are separated by an 8 hour time difference, Shroom in Istanbul, Turkey and Exnihilo in Colorado, USA. This would be a critical piece of the development cycle given the quick turnaround we would both need to work well together under a stress-full 72 hour time limit.

As the voting week began, I began tracking the voting of the themes after the first round closed and it was clear by the end of the third round that the three top voted themes had some cool potential – “Companion”, “An Unconventional Weapon”, and “Deeper and Deeper”. Within a few hours I had a pretty solid game concept that incorporated all of the top three themes. Now the wait for Friday evening to see if our hedged bet would bear fruit.

Background

Coincidentally, a week prior to the voting, my oldest son and I had been talking about some fun game concepts that we wanted to try and flesh out, and one of those was a caving, spelunking, exploration idea that used algorithmically generated maps of immense size that would go down thousands of tiles deep, giving the player a huge underground world to explore – different every time you played.

One of the ideas we felt was important was that the hero could cut through the earth, ala Dig-Dug and make their own path. This would lift the requirement off of our map generator to make connected cave systems and paths, whereby the hero could traverse the whole map. With the ability to dig into almost anything underground, the whole map was available to explore, whatever the generator might spit out. With that design and some initial algorithm tests under my belt, I entered the voting week with a hopeful anticipation that a theme would emerge that might allow the use of our idea. Much to my delight, “Deeper and Deeper” was a top contender for the LD32 theme!

spelunk1

Theme Announced and Game Design

“An Unconventional Weapon” was the official theme for the 32nd Ludum Dare game Jam and Exnihilo Studios was ready to go! Shroom and I had already begun talking about the likely outcome of the theme votes and we had a solid idea of how we could incorporate all three of the top themes into our game. We already knew the setting would be in the future with the need for light-weight rock-cutting technology that could double as our “Unconventional Weapon”, and to fulfill the “Companion” theme, we would incorporate a hover-bot that would accompany the hero wherever he went, assisting him in completing his adventures. Of course, with the central game mechanic revolving around digging deep into the earth, we had “Deeper and Deeper” already covered. We were off!

Right away I decided I wanted a heroine as the hero in the game and I was leaning towards an Alita:Battle Angel kind of look. Our initial inspiration for the hover -bot was V.I.N.CENT from Walt Disney’s ‘The Black Hole”. With our two central characters fleshed out we turned to some of the finer points in the game like – what’s the objective of the game? Being an adventuress, our heroine would naturally be looking for ancient artifacts and priceless gems to help finance her spelunking addiction and the costly tech it requires, so it was decided that gems would be sprinkled liberally throughout the underground map. She had to collect the gems for credits – simple enough.

Since I wanted her hover-bot, now called a ‘Spelo-bot’, to have a more central role in helping the player, we decided that the hero would give the gems she collected to him and he would store them inside himself until he was full. At that point he would begin the journey back to the surface to sell the players gems for credits and then visit the local spelo shop to allow the player to buy any needed supplies or upgrades. All communication with Spelo-bot would happen through a detailed HUD that Shroom was designing. The HUD would also be used to identify important objects and explain their importance or use to the player through a series of mini-tutorials. The game was starting to take shape.

spelunk2

What Went Right

The whole LD32 experience was not new to either of us. Shroom has been in several other time-limited game jams, and I used to work for an independent game developer in CA, so working around the clock to hit a deadline within a few days was nothing new. We both knew what to expect in terms of mental and physical stresses of working round-the-clock and were prepared to meet that challenge. Below are a few highlights of what went right for us.

  1. We had a solid initial game design to begin with and both programmer and artist were committed to that design.
  2. Shroom was able to capture much of the visual vision that I had for the look of the characters and visual elements in the game and responded well to my direction, freeing me to concentrate on the code.
  3. I had a core base of code to develop from that I had written from scratch, so understanding the game engine and how it worked (and didn’t) and where to go to effect the change I wanted was critical throughout the development cycle.

What Went Wrong

What was new for me was dealing with the logistical complications of the eight hour time difference between programmer and artist. But I think overall that was not a big negative factor in our development cycle. Below are a few highlights of what went wrong.

  1. We over-estimated what could be accomplished given the time-frame we had and feature creep added a bit more to the initial design that added more to our plates. As design lead, I should have clamped down on the scope and kept us a bit more focused – but I like to dream big too!
  2. Although I had foundational working code for a basic tiled-based map engine, my initial design was focused more on compactness and saving memory (for huge world maps) and thus added a layer of complexity in the packing and unpacking of data in the map. This complexity was finally scrapped for quick and simple access to the map data. Having all this worked out ahead of time would have freed me up to implement more features.
  3. We changed our method of lighting the hero in a dark, underground environment mid-game and this added some time and work in implementing a new method. While the final result is closer to our preferred look, the initial implementation was acceptable and the cost-benefit was likely a loss.

Final Thoughts

Shroom did an excellent job of keeping me supplied with fresh artwork to implement into the game and sadly, some of it did not make it into our final version due to time constraints. High on the list was the cool tutorials he designed that would have added a nice in-game ‘instruction manual’ to aid players in quickly figuring out what the game was about.

Additionally, he created some cool monsters that we both wanted to see provide some action moments in the game for our heroine, but I simply ran out of time to code the AI for their behaviors. To his credit, Shroom was cool about the missing features but I know he was as disappointed as I was that his baddies and tutorials did not make it into the game. Here are a few stills of what went missing…

baddies

Overall, I think we were both impressed with what we were able to accomplish. There is a certain amount of polish that the game shows, despite being whipped together so quickly. Being a perfectionist, I can point to many places in the game where I would like to improve the experience or look of a feature (code implementation, not artwork), but for what we had to work with, the time constraints, the distance and 8 hour difference in our working day, and the fact that Shroom and I have never collaborated on anything prior to LD32, I think we did a smashing job!

My hats off to all the other contestants on a great and fun LD32!

Cheers!

Gil


Leave a Reply

You must be logged in to post a comment.

[cache: storing page]