Random Ramblings About Making Games and Stuff from Cloud

Posts tagged ‘game development’

Trust is stronger than control

We producers tend trust processes and forget that trust, communication, skill and determination is a huge thing. We tend to rely on Scrum, Kanpan or similar over persons skill and expertise. I know I did. I truly believe that we try to learn from the past projects and even postmortems written by kind folk in Gamasutra, but we usually miss the true reasons why we fail in a game dev projects.

Making software is hard. Making games is damn near impossible. We keep making mistakes and we try to find reasons and that is what we are supposed to do. The things where we usually make mistakes are not the ones that following a processes will solve. Problem might be that we really don’t know how to make a fun game. We know how to make games and we want to believe that we know what a is a fun game, but ultimately it is the players that will tell us how we did. When we fail we tend to take a good look on what went wrong and make adjustments on the processes. While reflecting the past is a good idea we tend to dismiss the persons behind the process. Usually its not that mistake was made because we did not have full control over what happened. More often reason is that direction, goals or communication was not clear. In some cases we just did not have the skills or knowledge. I don’t think that these situations can be solved with adding more and more control by fixing processes.

In my opinion processes should not be treated like silver bullets that work all the time with different companies, projects and team members. No process is a substitute for skill, motivation, communication and ownership. If the team and its members don’t want or know how to or are not allowed to make a fun game I believe it can’t be done with any processes. Using strict processes might tell you that project is not going according to plan, but it will not launch a game. I challenge you to find what works best with your team and with each member of the team and heavily modify the ways you currently work if needed. It will not be easy but as I found it will pay off in the end. The process should give all the people in the team clear goals, improve communication and build trust. In most of companies I used to work we used processes to get control over team, games design direction, get visibility on how well team did their work and usually people actually working in the team did not find them that useful.

I was recently fortunate enough to work with a dream team and it opened my eyes. We managed to launch a top quality game in nine months with next to no pre-production. During that time I had one of the easiest projects in my producer experience. It was odd since we had a hard launch with fixed deadline and a brand that demanded excellence on quality and fun factor. Reason why it was easy was that I was able to truly trust that team was doing their best. Even when the game’s design direction was fluctuating quite a bit, I managed to remain calm because I knew that the team believed it was the best thing to do. I could fully focus on producing: foreseeing possible problems, getting help to the team when needed, reminding the team about goals and schedule, communicate what are we doing and why to the rest of the company and keeping the troops fed, happy and eyes on the big picture.

After our nine months game project I believe I have a hunch on how game projects should be directed and how highly productive team should be built. It is essential to trust the team and that the team is trusted by the company. We should let the professionals to do their work and make the decisions even if you don’t fully agree on everything. That is why you (or someone else) hired them in the first place. When the team works well together, has the skills, motivation and clear direction where to take the game to, then you don’t need heavy processes just communication and shared goals. Just agree the goals with the team, decide who does what and go for it. Follow up on the goals on high level by looking at the daily build. Changing the goals should be as easy as identifying that current ones are no longer relevant and agreeing with the new ones.

How we worked as a team

Word of warning before you read on. Our team had been working for a while together and we knew how we work together. The tools we used and the game genre was familiar to us and team members were highly motivated and skillful professionals. So we knew what we were doing. So don’t apply following ways-of-work without considering if they can work for your team.

Extra warning for the not so indies out there. We really had a creative freedom, peace to operate and we were allowed to make decisions inside the team. The following things will not work if you don’t have the power and freedom to make decisions that your team believes are best for the game.

What worked in our team was one week short sprints with goals like “character abilities for playable demo” and “players are able to purchase gold from shop”. In start of the sprint we had a meeting with leads and we agreed with the goals. Next developers had their own meeting where they agreed on who does what and how to get there. One developer was reserved for designer to code game-play stuff as soon as possible without heavy game design documentation. We had two daily meetings where we showed to the team what we did and what we are planning to do next. Only record of things to do was a whiteboard with huge calendar to remind us how much time we had left. Whiteboard was redesigned couple of times in the project to better match the current state of project. Only thing that remained stable was printed A4 papers labeled Important in top of the white board and not Important in the bottom of the white board. Targets, goals and tasks where put on top and not important ones to lower part of the white board. Usually not important ones were removed from the wall without actually even implementing them at all. And that was the whole process it looks and smells a lot like scrum, but, with two main differences. One the process was designed to give as much information to the team keeping the focus on the end results and not the tasks. Second differences is that we did not measure or monitor how long tasks did to complete. We only looked what was in the game build in each day, at times by each dev build per hour.

What comes to working with the people in the team it varied. One developer preferred to work with a prioritized list so we made it for his tasks. One developer preferred to take one huge goal like “payment system” or “event system” and work with that until it was completed. One developer liked to prototype so we paired him with the designer and they talked together on what would be the next thing to try out. One artist liked to work on the scenes and another wanted to have list of game assets to work with and so on. What I am saying is that only thing that was uniform on how we tracked the progress was two times per day we showed what we had done and we talked a lot to each other and we trusted each others opinions and decisions.

Final words

So anyway this was what I learned in the last nine months on what is important in game development and how highly productive team can be build. Hopefully you find something useful from my ramblings. Start trusting people and see where that takes you.

Thanks for reading and hopefully you got something out of this post.

Also published in Gamasutra as featured blog post.


Tools for decision making

One big problem in game development projects is that we are in constant doubts about the game we are working on. Do we need another game mode? Should we change graphics because they seem too dark? Is this fun? Should we add this new feature? Should we even be making this game in the first place? Is our game good enough because our competitors have much better graphics? We are constantly making decisions based on hunch, fear and panic. To make things worse we are constantly end up revisiting the same problems and decisions over and over again.

I personally believe in the mantra “If your only tool is hammer then every problem looks like a nail”. When you want to stick two pieces of lumber together there are nails, screws and glue just to mention a few that pop into my mind. Same applies when you are making a decision. When it comes to making decisions, there are tools that will help you to make them and have confidence to stand behind them. I don’t have a magical knowledge on when to apply a certain tool on certain problem so it would always result in best decision available. When it comes to making decisions in various game development stages reality is that every game and people making the decisions are different.

So what this blog post means to you and me? Hopefully you will find some new tools or at least new perspectives on how decision-making can be less paralyzing and how to make sanity checks on your decisions. For me the best thing that might come from this blog post is your comments. I want to hear what tools you have in your arsenal and what new perspectives I could learn from you.

First tool that I use in these situations is to ask from someone else who is not working on the game. My animator friend described one of the problem of animating a movie as follows:  “When you are working with animation movie it takes a time. Animating a movie takes so much time that you must have a really good reason to make it into animation in the first place. It would be so much faster to just shoot it into video using live actors or draw a comic book. There will be a time when you will start to hate your story and you want to make changes. This is just because you have worked with the project for so long. When you find your self wanting to change things in mid animation project, the first thing you should do is to ask someone else’s opinion”. In a nutshell this means that you will be fed up by you game project before you have finished it. Sometimes these changes can enhance the game, but sometimes you will want to change things just to amuse yourself. So before you start making changes ask for external opinion. Help can be found from fans, a colleague or a family member just to name a few. As long as those people have not been involved with your project as heavily as you they might have a better understanding of what is new and fun in your game. Of course you should not take their word as an absolute truth because in game development there are a huge amount of uncertainty until the game is polished and it is in the hands of the players.

Second tool is to always remember why you are making the game on and steer the projects towards that goal. One of the most useful piece of knowledge was passed to me just a few weeks ago. “You should always be aware why you are doing something. After that, have an idea how to measure your success when your game is live (or in alpha testing). If everybody in the team understands reasons why this game is being made, they will more likely make the right decisions.” So you should always be aware why you are making a game or an update and everybody in the project team should know those reasons. Reasons may vary from personal ambitions “I want to make this kind of a game” to more business oriented “we need to increase our daily active user count”. Regardless of the origin or the reason it self, it needs to be stated aloud and measured accordingly. If you do that, then it is easier to make decisions. It is always possible to revisit the reasons why you are making something and when you do change the reason don’t be afraid to redesign the whole game or cancel the project.

Third tool in my arsenal is measure and analyze your success and learn from your mistakes. Following piece of helpful encouragement was handed to me by my boss once “You are going to make mistakes. A lot of them. And if you don’t follow the impact of your desired targets, there is high change that you will never learn from your decisions (and how to make them)”. Be prepared to measure your success in some measurable way and learn from your mistakes. If increasing daily active user is your aim, measure the effect post launch, analyze your results and learn from them. If you where making a totally new kind of a game, then check from reviews and comments if the press and gamers really noticed that this is something totally new. You should also try to learn how you and your team make decisions. Keep decision log where you list decisions, dates, people involved and reasons why that decision was made. Use that list when you are facing a situation where you might want to change direction or revert a decision you already made. Also make decision log review a part of your postmortem process and try to learn from it. If you are lucky you even might learn from mistakes of others. Read lots of postmortems and ask how other people make decisions in your workplace or startup sparring circle.

Fourth tool would be fail fast and don’t be afraid to make mistakes. My boss and numerous excellent blog post in Gamasutra (1, 2) already recognize this. You are going to make mistakes and a lot of them. Don’t be afraid to try something out. You should also be equally willing to confess it to your self when the thing you tried out does not work. Don’t force something in to a game just because you made a decision about it. Do, review, iterate and abandon ideas in fast cycles.

Well that was my two cents on how decision-making could be easier and how you and your team decision making could be more efficient. I would love to hear your experiences, comments and ideas. I am always looking opportunities to learn from the best.

Also posted in Gamasutra as a featured blog post.

%d bloggers like this: