Productivity: The Group of Three
Posted by Rampant Coyote on January 12, 2015
The middle 50%+ of game development is the hardest for most people, I’ve found. At the beginning of the project, the game is being built up from nothing and is making rapid improvements daily. It’s exciting and motivating. And at the end, even though it’s a hard slog, you can see the light at the end of the tunnel. But in the middle, apparent progress is minimal, and there’s just a lot of crap to get through to get it from that early prototype to a finished product.
Back when I was working on Void War, I came up with a methodology – well, a dumb little trick – to help me get through the less exciting phase of game development. I was talking with some people last week about productivity tricks and staying motivated, and thought I’d share this again (since it was lost with the old blog). Honestly, I’ve felt lately that I need to go back to it, just because I’ve been slogging along in a lot of boring work for a while and the game hasn’t made a visible improvement in a while, which can be demoralizing. It works for me. Maybe it will for you.
Maintain a Task List
Number one, you need to maintain a task list. If you aren’t doing that, and have been flying by the seat of your pants, then you’ve already got something critical to fix if you want to get your game done.
I’ve found that simple task lists – occasionally prioritized, but really nothing more complicated than a straight-up list (possibly marked by category – see below) – is the easiest, best tool for me. I’ve used them since the SingleTrac days, and they seem to work better than more sophisticated tools. Ideally, the tasks should be broken down into jobs that will take a short, reasonable time to complete. For me, that means tasks that I can complete in one day or less. If a task is much bigger than that, I try to break it down into smaller sub-tasks. But not too small. If they are less than a quarter of a day in length, I group similar tasks together.
Plan Task Transitions
When I’ve tried to track my progress, I’ve found my biggest waste of time is always on task transitions. Still. I end up spinning my wheels at the end of the task, poking around for little things to tweak, or I end up hunting down a distraction which eats up more time than the original task. It’s a sad commentary on my own laziness, I guess.
This is something I still need to work on. But the “Group of three” helps (at least cuts down the transitions somewhat). Part of it for me is to make sure I already have the next task “queued up” and ready to go, so as soon as I’m finished with my current task, there’s no question about what to do next. I can roll right into the next one. This means taking time at the beginning of my day (which, as a part-time indie, means “night”) and plan out I’m going to work on, in order. I can change the order later, if I choose, but otherwise I have a default plan and no excuse to break my stride and hunt down distractions. Or to waste time gold-plating what I’ve already finished.
The Group of Three
Because I love making games, there are many game-making tasks that I really look forward to doing for one reason or another. They are fun. They are exciting. They are intellectually stimulating. Whatever. We’ll call these the fun tasks.
Then there are some tasks that are really going to visibly improve the game. By “visibly” I really mean “apparent to an outside observer” – so it could be audio or something else, even a layer of polish that will really make things “feel” better. I’m going to call these “visible tasks.”
Then there are some tasks that are just going to be painful or dull, but they have to get done. Often they’ll involve rewriting prototype code or something else that is neither fun nor will make much of a difference to the playable game – they are just things that must get done. We’ll call these “tedious tasks.”
My trick to getting through the long haul of development is to group one task from each category together, and force myself to complete all three before I can move on to the next group. It’s motivating for me, because I want to move on to the next “fun” task… but I have to finish a visible task and a tedious task before I’m allowed to do so.
The other motivational trick is that by grouping the visible tasks in there as well, I force apparent progress. I may not be very excited about them individually, but it does mean that team members and outsiders get to see progress in the game – and I get to feel good about the visible progress. So that’s a longer-term win and motivation.
Now, there are some tasks that might fall into two categories – you can pick a category for them based on whatever list is starting to run low on items. You may need to double up on one category to make it a group of four or whatever. Fortunately, by the time you have to do this, you are getting near the end of development.
Filed Under: Game Development, Production - Comments: 5 Comments to Read
Cuthalion said,
I might try this sometime. I think I remember the old post, too. Although right now my tasks tend to be whatever bug I noticed most recently. 😛
I just spent 5 hours last weekend implementing scrolling and word wrap for my menus. It was, uh… fun? Visible? Maybe tedious. I kind of liked it, but also didn’t. Let’s go with tedious, because then I get to do something fun next. 😉
Rampant Coyote said,
OUCH! And boy, do I feel your pain. I have been working in UI hell for what seems like forever, and that was exactly the sort of thing I was thinking of.
Cuthalion said,
It’s what I get for being so license-paranoid and control-freak as to roll my own menu system. (Which was half-copied from an older one I made in the RPG Toolkit years ago.) On the plus side, YOU CAN SCROLL WITH CTRL+SHIFT+WIN AND THE CONTROLLER’S X BUTTON IF YOU WANT BECAUSE I BELIEVE IN CONFIGURABLE CONTROLS!
*ahem* Might’ve gotten too excited there.
Cuthalion said,
You can also apparently control the game with the microphone, but I haven’t tried it. Part of me hopes that, since the game can see my mic (or just the windows mic volume control?), I could hypothetically control the game by yelling at it.
Xenovore said,
Good stuff!
I use TodoList. It’s got a lot of nice features and is open source if you want to tweak it.