A personal post mortem on Depth
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.
Digital Confectioners officially started life in late 2007 when I was still practicing as a pharmacist. I was working on a mod called The Ball that was the brain child of Sjoerd "Hourences" De Jong. The Ball was later released in 2010 as one of the first games powered by UDK on Steam, published by TripWire.
Early in its life, Digital Confectioners was hired by Epic Games to produce documentation and to travel the world to help teach the world Unreal Development Kit. As we grew the team slowly over the years we were able to take on larger projects that were using Unreal Engine 3 or Unreal Development Kit. Digital Confectioners has been very fortunate to be involved in so many successful games such as Q.U.B.E and many other great games.
Late 2012 I started talking to Alex Quick about Depth. Depth had just posted that it was going on a hiatus and I asked Alex a few questions. We started talking about a possible collaboration between Digital Confectioners and Depth Inc and two years later in late 2014 we released Depth on Steam into the wild.
There were a lot of high points and low points during the development of Depth. Personally, I feel that I've grown a lot when it come to developing games, managing a business and more importantly, figuring out how to do it 100% independently. A lot of this post mortem are my personal feelings and opinions.
I want to thank my friends for suggesting these questions for this personal post mortem. I also want to thank my family, especially my wife, for all the support; without it, this would have been impossible for me.
Now then, let's suit up and take the dive into the journey that was the development of Depth.
What was it like being a small studio and having to juggle between many different things?
Insanity. That is definitely the best way to describe it. Digital Confectioners, before the release of Depth, is principally a contracting company which specializes in Unreal Engine. During that time I did not really know whether Depth was going to be successful or not. Rather than hedge my bets and put everything into one basket, we still did contracting work at the same time. While this split our focus away from Depth it did allow us to ensure that our bills were paid, as well as everybody on the team.
Having to juggle between so many tasks simply meant that Sam Evans and I, being the directors of Digital Confectioners, did not usually have forty hour weeks. Often we had sixty or eight hour weeks, as well as a lot of sleepless nights either worrying about Depth or catching up on all of the contract work. While crunch is often seen as a dirty word, it was simply a necessity at the time for us.
Making games is really hard and really time consuming. While forty, sixty or even eighty hours a week may seem like a lot of time, a good chunk of that time can be lost due to context switching. The more things you juggle, the more time you need to context switch.
Time management is always a struggle and I experimented a lot with different tools available: Toodledo, Mantis and JIRA. JIRA for me worked out the best when paired together with AGILE. Work life balance was also a constant struggle and the only real way I found to counter act that was to focus my time on where it mattered the most which was family time, developing Depth and working on contracts.
What surprised me?
Opportunities at the right time, at the right place and from the right people ... as long as I kept a careful eye and ear on what was happening around me.
During the development of Depth, it consumed everything that I had and then some. Sometimes, waking up and getting into the office to deal with a JIRA board full of bugs and tasks was immensely difficult to deal with. There were times when I honestly felt overwhelmed and then there were times when the numbers for a lot of things just didn't stack up.
Fortunately, I made the time to look at what was happening in the outside world by doing talks about Unreal Engine 3 and Unreal Engine 4 at a variety of conferences, listening and watching what other games, other friends and people were doing.
What surprised me in the end, was that sometimes I would make an impact on someone who later changes the direction I would steer Digital Confectioners and or Depth.
What turned out to be a lot harder to manage than I thought?
Project management was and is a problem that we deal with on a daily occurrence. With every change I did, sometimes it felt like taking two steps back. As the team got used to the new step I added to the project management chain, the team would find flaws in it or that there was another new step that we needed to take.
Coming from a modding background, I am a project manager's worst nightmare. I wouldn't schedule bug fixes or tasks. I wouldn't estimate anything. I wouldn't project what core features a game or modification would have. Over time, we adopted many practices, that for a lot of developers would be a no brainer. For the longest time:
- I was not very good at using bug trackers such as JIRA or Mantis.
- I was not very good at clear communication. Skype would often get backed up with random chatter during the weekend.
- I had no idea how Depth was going to work from a game play stand point.
We had general ideas of how we envisioned Depth. But we had a lot of prototypes of different game modes, different atmospheres that we were trying to achieve and different directions we thought Depth was going to take. Nothing coalesced together.
There came points in development that I felt we were just burning money. Hind sight now, I realized that all of those things needed to happen in order to high light what the actual problems were. The team felt directionless, which were caused by project management issues from the top.
As we started to adopt practices that I once thought were pointless such as ensuring that bugs and tasks were tracked, estimate with some level of certainty how long bugs and tasks would take and improve on communication, I noticed that things started to just flow together. It felt like we were all on task, trying to bring everything together. It felt like I was actually in control of what was going to happen in the next few weeks.
What I learnt here is that project management is a real skill that you should learn as it affects you on a daily basis. What tasks are you doing? What bugs do you need to fix? What's the best method to communicate with someone about something? Also, managing a project is hard to do on your own!
What was the most effective way to promote Depth?
Sam and Will, at Digital Confectioners, handled the majority of the marketing and promotion of Depth. Everybody pitched in though, which was great as we all had unique ideas and skills. Special thanks to Alexander Bruce, Cliff Harris, Dana Cowley, Rami Ismail and Steve Piggott for all of the advice and help.
Our main focus was YouTube personalities, TwitchTV streamers and the online press. The most important thing we did was plan our strategy and execute the plan accordingly. Planning the strategy helped frame what we needed to do.
To start our planning we focused a lot on what attracts the online press to a game and for them to write a piece. For us, we felt we already had the hook for Depth which was "Become the shark and rip your friends to shreds". We also felt that we already had a simple elevator pitch which was "Divers collect treasure and escort a robot. Sharks use skill and might to rip divers to shreds". The hook and elevator pitch at that time was very raw so we knew we had to work on those.
We knew we had to have a website so that a journalist could quickly find information about the game. Presskit() was an essential element and absolutely had to be available for journalists to look at, so that he or she could get as much information about Depth in the shortest time possible.
So our battle plan looked like this:
- Improve the hook and elevator pitch for Depth.
- Create the website that was easy to use, had high usability and content dense.
- Create the press kit that will provide all of the information that a journalist might need.
- Create a trailer for Depth.
- Create screenshots for Depth.
- Create the Steam Store page for Depth so that journalists have an actionable conclusion to their articles (e.g Go to Depth's Steam Store page). The Steam Store page also allows Steam users to start talking with us about Depth.
- Set up Google analytics so that we can keep track of what is happening. Are people going to the Steam Store page? Are people going to our website? Are journalists going to Presskit()?
We then started to think how we would attract YouTube personalities and TwitchTV streamers. Our current thoughts at the time was that if some good articles were written by popular press then YouTube personalities and TwitchTV streamers would know about Depth, and hopefully start thinking about fitting Depth into their schedule.
What I learnt here was that planning and scheduling is really key when it comes to marketing, rather than leaving it purely to fate and virility.
What were some of the road blocks?
Common road blocks that occurred:
- Running out of funds
- Game design uncertainty
- Time zones
Running out of funds
Contracting is a very lucrative business in the game development space, but I found that there is always constant pressure to expand in order to get more contracts or to be able to take on larger contracts. During the development of Depth, our financials would sometimes get dangerously close to going into the red. Through out Digital Confectioner's existence, we never missed a wage payment for a team member. While I always felt that working in the games industry is a dream job, the realities of bills and employee morale are very real too.
This required us to sometimes stop development on Depth to focus more of our time on contracting. Some team members were constantly switching between Depth and other contracts, with some focused purely on contracts.
The question of crowd funding came up multiple times. I've never been too comfortable with crowd funding. With Depth, we didn't even know what it was back then. Pitching the concept of Depth may have excited a lot of people and we may have even succeeded in getting crowd funded. But the nagging problem was the potential pitfalls that many projects fall into. What happens if the Depth we pitched changed (which it did from its previous concept back in 2009 to what it is now)? What happens if the physical goods we promised aren't possible or cost more than what we thought? What happens if we don't actually get onto an easy to use distribution platform? What happens if we didn't have the technical expertise to pull of a multiplayer game?
While contracting did slow down the development of Depth, I felt this was the best decision because it allowed the game design to mature and have a lot of sink time to think about its mechanics and balance. It also empowered us to do whatever we felt was best for Depth.
Game design uncertainty
2009's version of Depth was pitched as a stealthy cat and mouse style game involving divers and sharks. Divers had to sneak around the map collecting treasure and returning it to the dive cage. Divers were given unique abilities through a class system. Sharks were combat focused and also given unique abilities through a class system.
2014's version of Depth is very different. Initially on release, we only had one shark which was the Great White. The Great White's game play was solely focused on combat and stealth. We dropped the class based system and stealth game play mechanics from the divers. We added an AI companion for the diver team to escort as it traveled through the map cracking treasure chests that contained valuable things.
In between those years there were also many other different prototypes of Depth. There was an experimentation into a horde style game mode. For a very long time we had a game mode that focused on one shark verses four divers. All of those prototypes had issues that we could never solve properly.
The original version of Depth felt oddly paced and sometimes was very confusing on what you were supposed to be doing. The horde style game mode suffered from a variety of different shark species and to fix that was out of scope. The one vs four game mode suffered from inconsistent experiences that sometimes lasted one minute or forty minutes.
All of this tinkering around with game design mechanics was very tiring and sometimes I wondered if we'd ever reach a stage where we knew what Depth was.
Digital Confectioners is located in New Zealand. Alex Quick, Engine Audio and Super Genius were located in both the east and west of the United States. Our level designer, Justyn is located in the United Kingdom.
Getting together and play testing often meant that someone had to suffer with time zones. The only way we really solved it was to have consistent play sessions at an agreed upon time. However, technical difficulties often caused a lot of hiccups there too.
What took the longest time to do on the Depth?
Game design that felt different yet familiar
Game design is extremely difficult. I doubt you need me to remind you of that. Depth did not really have a lot of existing games to use as a theoretical base. While what we ended up with reminds people of a lot of different games, it wasn't because we looked around and took bits and pieces from different games and jammed them together. There was no magical, "It's just X + Y + Z" solution. Thus a lot of the times we were constantly iterating on ideas and refining them and then chucking them out. We did not hold too many ideas close to our hearts as we knew that if they didn't work then they had to be discarded.
At one point the game had a lot of different troublesome metas. These arose because a lot of the mechanics didn't make sense. For example, you used to earn money based on how much gold you were carrying. That didn't make a lot of sense and the troublesome meta that came out of that was that people would hide in a dark corner of the map until they could buy everything.
We also knew that we had to differentiate from what was already on the market. While the divers vs sharks theme could carry the game, I felt that if we didn't then we'd get lost pretty fast. If people could find the same experience in another game, then we'd have issues. However, trying to be extremely different meant that we would also need a lot of tutorials to teach how to play the game.
The only real way we solved this was to just to do a lot of iteration, player testing and listening to the feedback generated by that.
Depth's user interface design originally stemmed from DOTA2's user interface design. This is pretty obvious when looking at the main menu. However, as we continued development we realized that we had a lot of key differences that made copying it difficult. Unfortunately, we were only really able to do anything about it for the in game menus.
We have a few user interface usability guys on the team, so there was a lot of debate over how something should be done or represented.
Because the game design changed a lot and very rapidly, the in game user interface would often have to get revamped a whole lot to accommodate those changes. This ultimately meant less time to tweak and polish the user interface to the level that we wanted.
We also decided to buck the trend with doing a metro user interface, or rather a clean and simple style (we later changed our minds in the current iterations of Depth). We opted to do a very graphical user interface with textures such as metal and wood to convey what you could interact with and what you could not. This was hit and miss with the team members and we also had issues later on when we needed to modify how the art looked as this was handled by Super Genius.
User interfaces are tricky things to get right in any game. Both the art, sound and feedback loops are just as important here as they are in the rest of the game. Depth still doesn't have the perfect UI that we want, but we're making slow incremental steps towards it (I hope!).
Did we have a design document?
Yes and no. Alex Quick originally wrote a Depth guideline on what it was about. We followed that pretty closely at the beginning but some areas started to deviate. However, the core concepts stayed the same such as feeling like a powerful shark but with limitations of being a living animal, being a diver with an arsenal of equipment to help you.
Did we use any contractors, and for what part? How did we find them?
Digital Confectioners is composed principally of engineers with a wide variety of specialties. However, we can't really do art to save ourselves. Thus we contracted out:
- Art to Super Genius
- Sound to Engine Audio
- Level Design to Justyn Dagg
We did attempt to contract out user interface art to another contractor, but we decided to pull that into Super Genius in the end. We also tried to contract out other things such as sys admin, QA and a few other smaller roles. However, these didn't work out and we pulled those in house.
We met Super Genius when we were developing another project called Forge. We really enjoyed working with them and their team that they were always on our minds. Thus, we didn't really find them but rather we met on another project.
Engine Audio and Justyn Dagg worked on Depth prior to our involvement, so it seemed only natural to keep them onboard as they had experience working with Depth's environment and needs.
Did we write any of our own tools?
Yes, but not for developing the game itself.
Unreal Engine 3 already comes with Unreal Editor which handles most of the aspects of the game development such as managing game assets, creating levels and so forth.
Some of the tools we wrote were things such as a build + deployment tool. While Unreal Engine 3 comes with the Unreal Frontend, we needed a tool that would perform the same duties as the Unreal Frontend but also package the game, upload it to Amazon, Dedicated Server Hosts and Steam and then perform some other cron jobs as a series of steps.
Did we do much research about shark behavior?
Not a lot. The reason for this is because divers and sharks are Hollywood-ized. In particular, sharks were extremely fantasy based. This is mostly because we had set of mechnics that worked particularly well for the game that we were making. Some of the mechanics did derive from the theme of divers and sharks, but purely from a "what if" observation rather than "well sharks do this in real life ... so...".
Depth was never, ever intended to be Shark Simulator 2014.
What was the biggest obstacle that we solved?
Figuring out the game type! When we started making Depth, we had no idea what the game type was going to be. We initially worked from Alex's original game design document which had divers as predominantly a stealth based team that was lurking, gathering treasure and remaining undetected. Sharks were very powerful creatures that were all about full frontal attacks and generally being the dominant species in the water. Unfortunately that game mode just didn't work as it was too easy to "troll" when being a diver by simply hiding and prolonging the game. Sharks were also extremely powerful and getting the balance for it was right.
While I would have liked to have written that we had an a-ha moment and figured it out in an instant, we simply just keep prototyping and testing game mechanics over and over again until we felt like we made progress towards a game type that would work for Depth.
What were we overly nervous about?
Whether Depth was actually something new and interesting. We submitted several times to the IGF (Independent Games Festival) but we were never picked. The comments that we did receive were positive, but over all most places that we submitted the game to responded with a 'meh'. So there was an internal conflict within myself and I wondered whether if what we were doing was actually interesting, innovative and new.
Post release, we found that most playerss who did play Depth really enjoyed it because was fresh, new and exciting.
Why did we decide to make Depth?
This was largely due to the relationship that I had with Alex Quick. Back in the modding days of UT2003 / UT2004 I technically rated his mod Killing Floor. We had some discussions about it and before I knew it, he and Tripwire made it into the success that it is today. At a later point he contacted us about contracting to work on AI for Depth.
When Depth went on hiatus, Digital Confectioners was a four man company, with Kenneth Churcher joining us that year. Depth had a lot of assets already built and we felt that it would be a shame to see such an interesting idea potentially disappear. We had built a few prototypes of our own, but we knew we didn't really have the resources to make a game the size of Depth at the time. I also wanted to build Depth independently as my game development career started in places such as TigSource.
Depth has been an incredible journey. The game is still going strong which seems to increasingly a rare thing for indie multiplayer games these days. It's been a roller coaster ride and one I would repeat again!
Please feel free to email me at [email protected] or comment here and I'll reply as soon as I can. Thank you for reading!