Learning to handle the “just 1 more feature” disease
I have this disease that probably a lot of enthusiastic game developers also have, the “just 1 more feature”. At first it isn’t that bad and you’re even enjoying it but when you’re not careful about it and let it spread it can really get nasty.
I have the luxury of not having to wait for my games to make money to pay the mortgage (I try to see it that way instead of saying that I don’t have the luxury to quit my job …) so there’s always the danger of loosing focus when working on a game. I don’t have a supervisor to tell me that it’s enough and that I must ship so I have to play that role as well. When I don’t the “just 1 more feature” strikes.
At first the things you’re adding are just really making a lot of sense if not just required. You don’t feel bad about it because you’re telling yourself “well if I don’t do that it will be no good”. Even though I’m using the same framework as Golemizer for Dungeon of Loot I had to change the way the inventory system works. The first reason is that if the game was to know some nice success the inventory system would explode. It works just fine for Golemizer but for Dungeon of Loot it was flawed. Long story short version, Golemizer = unique items with different properties for multiple records, Dungeon of Loot = only stacks of generic type of items for 1 record each.
Then come the things “I can fix that with…”. Maybe you thought of those before and they turned out to take much more time than expected or it’s really something you missed and requires you to change back some of the work already done. In my case for Dungeon of Loot that’s the whole creation of each levels of the dungeon. I thought that since I already had the sandbox power of Golemizer that it was not a problem but then I realized that all levels are 100×100 (Golemizer 20×10) and that I was going to have to build nearly 100 to start with … Oops!
It’s true that I already had all the tools to create each level manually but it would take countless hours. While manually creating such content could be a nice touch it’s really just pointless for Dungeon of Loot. So I spent quite some time working on tools to automatically create new levels, automatically populate them with monsters and automatically spread chests to loot. So I started with a first version, loved how it was making the work so much easier than tweaked for another version and another and another …
Of course I don’t let the machine do all the work as there’s always some tweaking to do but these tools are saving me a lot of work. That’s stupid but I never really thought about the way I would be building the content until I had to. That’s bad planning on my part. While this was needed for the game I almost lost control of my time invested on that feature. I had to stop and accept the imperfections of the tool. It’s already doing fine and some things are missing but I’ll be more than able to release the game as it is right now.
And then comes the “oh look, it’s so cool and easy to add”. Those are terribly sneaky. When you’re passionate and doing what you like you can get carried away by your own amazement at yourself. That’s just narcissistic in a way but you need to feel like that from time to time otherwise you might feel bad that nobody around you is there to say “wow! you’re good!”.
So I had this cool playable first level and then thought “what about boss encounters?”. Easy! Just put a bigger monster with bigger reward! Hmm but it would be cool if players had to team up to beat it. But how will I deal with XP and loot if 10 players fight the same boss?!?! Just making the last blow count wouldn’t be really interesting (other monsters work like that but that’s just fine for a quick fight). So then I started to work on a boss system. It can’t be that hard after all. I already have a powerful AI system, all my combat logic is clearly contained in a nice structure and … Well that looks easy and that looks essential but I didn’t took the time to see if it was REALLY needed or if I could replace that with something else. Now that it’s done I’m not sure if it was time well invested. Sure it provides a different challenge to players but maybe that’s also a bad thing? Maybe if I just …
That’s when I stopped. Boss system is there and not changing. Required or not it do what I wanted it to do which is share XP between players based on damage inflicted and loot for players that dealt 10% or more damage. Maybe that should have been part of my initial plan but it wasn’t (besides maybe 1 or 2 unique encounter). Working on that new cool thing lead me to think that I should have a way to allow players to group together, deal with group XP, bring back the guild feature from Golemizer, bring back guild chat, friends’ list and so on. That’s all nice and maybe needed to some point but that’s not the initial plan I had for Dungeon of Loot.
Even though it’s a multiplayer dungeon I see it mostly as a casual solo adventure with chat. So it’s not quite an MMO but I’m still keeping the tagline “A previous-gen MMO” though because it makes me smile and me smiling makes me look good like my wife says
Overall I’ve been able to keep control over the “just 1 more feature” for Dungeon of Loot. This weekend was the first days I was spending only to build content and it went amazingly fast so now that I have a taste of it I’ll be keeping up that way until I’m done. No more new features until the content is completed.
The danger of keeping control over the “just 1 more feature” is to maybe cut something that really should have been part of the project. I’m not an all-seeing game design expert so often some nice stuff comes past the initial design. Dungeon of Loot could have at least 20 more obvious features that could “make it better” in a way but I would be missing my initial plan which was to release a simple game that would require no expense besides the server it runs on and that would require as little as possible new development on the framework Golemizer was built with.
The only big new thing I’m spending more time is on the Facebook and Heyzap integration because I want these to be the power driving new players to it. I’m not using all the features offered by Facebook yet as right now I have decided to complete the content first but that’s still more than I had done for Golemizer. It’s way easier to get something done by taking small bites than to try to fit the whole thing in your mouth in one big chunk. I can always complete the work later and that will be a head start for a future project.
Golemizer is a good example of the “just 1 more feature” disease. There’s so much stuff in there going in very different directions that it’s no wonder why it took me so much time to build and why I have spent so much time on it in its first year of release. For example I’m not quite sure why I really added a weapon system in there. The whole point was to use golems for combat yet I found a way to sneak weapons in there. I’m sure some players suggested that at some point and I must have said something like “sure, that’s cool and easy to add”. Now that I look back that’s something that should have never been there. While some players might appreciate it I now know that it wasn’t required and that the time spent there should have been spent elsewhere.
The nice thing is that now I know!