Working on Dungeon Sandwich has been an absolute adventure from start to finish. Kongregate has been a pleasure to work with, and it's always a joy to work alongside Richard Hill, a long-time friend, and the first person I made a game with in Australia! This project had a lot of hurdles to overcome for us to hit our deadlines. From the beginning, we made some critical mistakes that would ache throughout the development process. By the end, we may have limped across the finish line, but we accomplished our goals and created something pretty neat!
What Went Well
Our relationship with external parties was a very positive one. As previously mentioned Kongregate was a pleasure to work with from the very beginning. John Cooney, our Kongregate representative, was open to feedback and allowed us a lot of creative freedom. They operated pretty hands-off for most of the project and left it to us to approach them should we have any questions, comments or concerns.
While Richard and I were able to handle the bulk of the programming, design, and art we called upon some contractors to help us with audio and larger pixel art assets. Erika Howell is a friend of mine from my days in Chicago. She's a very talented vector and pixel artist. We brought her on to do the Menu Art, and she did a phenomenal job! On the audio front, we went to Reddit and recruited Shawn Daley, a talented cartoonist and "Chiptunist."
Internally Richard and I worked to our strengths and were optimistic throughout. Even when burnout began to effect our mood we were still able to remain optimistic and keep each other going.
We are both very proud of the amount of work we managed to complete in 5-weeks. Fully animated characters, tons of recipes, nearly a hundred unique dungeon rooms, a fully realized level generator, and more gameplay features than you can shake a stick at! (Unless it's a petite stick and you can avoid exhaustion.)
What Went Wrong
Game Development is a messy business, and this was the first time Rich and I have worked together with a major publisher. We realized halfway through the project that we had grossly overscoped and had low-balled our financial remuneration.
Richard and I work full-time jobs, so juggling domestic responsibilities and work duties should have been manageable were it not for biting off more than we could chew. We had estimated about 100-hours each for the whole month would suffice, but we likely came closer to 150 or 200-hours each. We never were strapped for cash, fortunately, but we should have estimated our costs higher to compensate for this.
Outside financial and project planning issues, one of the major roadblocks we ran into revolved around poor testing time frames. We were unable to apply a lot of polish to the game due to how late some core mechanics and animations were implemented. Design is best when there is time to iterate and improve. We designed the dungeon generation first, then created the character, then monsters. As our game has a focus on combat, we should have designed the character first, then tackled the monsters, and finally dungeon generation.
How to Improve
Next time I am afforded the opportunity to work with a publisher on a project where I have significant creative freedom, I would plan differently.
- Firstly, I'd focus on speaking with the publisher to get their professional advice regarding scope. I would also want to cross-reference the budget and scope with my colleagues.
- Secondly, due to the short time frames on this project and redesigns during the negotiation phase, I was unable to draft up a fleshed out design document from the beginning. This lead to feature creep, which only antagonized our scope issues.
- Finally, focusing on getting placeholder assets and core mechanics blocked out early in the project should be a priority. Especially the mechanics that the player has control over.
With those in mind, I look forward to finishing off this project and diving into the next!
Next Steps Dungeon Sandwich has still got a bit of work to finish up on. Kongregate has allowed us some extra time to finalize a web-build for them. We're pretty exhausted, so we'll be working casually over the next few weeks to finish that. I'll post up some information about that in the not-so-distant future!