★ TouchArcade needs your help. Click here to support us on Patreon.

Feature Creep and you...

09-04-2009, 01:09 AM
Joined: Jan 2009
Posts: 1,512
Feature Creep and you...

Hi guys...aspiring game designer here with a title being released soon. This will be my first game, and I just wanted to bring up something that has been plaguing me from the beginning of my project. I'm sure some of the seasoned veterans here will know exactly what I'm talking about, even though they're wise enough to avoid this pitfall.

I'm talking about how you start off with something simple...but then a new idea gets you moving off on a tangent...and then you decide to add another feature since it would be "sooo cool!"...and well, you get the picture. What started out as a relatively straightforward concept has turned into this massively complex design.

I guess the easiest way to avoid this scenario is to have a rock solid game design doc BEFORE starting the project, instead of working on it as you go along. Hindsight is always 20/20, lesson learned, lol.

Any devs want to share some of their horror stories? (I'll share mine closer to release)

Front Page Tug Boat Chief Engineer
Check out the awesome Fuzion: Age of Wordcraft website
Follow me and the game on Twitter
preview thread (Fuzion is still in development)
09-04-2009, 04:47 AM
Joined: May 2009
Location: UK
Posts: 761
The biggest factor in feature creep is too many designers. Smaller teams on iPhone mean it's less likely to happen.

My biggest inspiration for Ground Effect is an obscure title I was involved in in the mid '90s called Hi-Octane. For complicated reasons we had to get the game out in 7 weeks on PC. Rising to that challenge was a lot of fun. My contribution was an early version of the graphic engine I was writing for Dungeon Keeper. The guys who put the game in did a great job, after a couple of weeks we had a very fun, simple hover-racing game running on the LAN. Every lunchtime and evening consisted of a queue of people from all the teams in the building to get onto the 8 player LAN game. It was pure racing fun, get the feel of the controls, learn the courses, time your boosts, get better at it, beat your friends.

Then certain people higher up the chain to EA got to play it. Didn't like getting beaten by guys on the LAN, proclaimed it 'no fun' and 'too hardcore', the handling was simplified and the boost downgraded such that you could pretty much make any turn, boost or not. The game was no fun anymore so powerups were added including guns and missiles. This did wonders for the on screen pyrotechnics but turned the game into an almost random experience. Was it more accessible and therefore fun for more people this way than the pure race would have been? I have no way of knowing, all I do know is that the queue of people wanting to play at lunchtime dried up the day after the guns went in.

Ground Effect is almost my attempt at taking the other path. The iPhone seems like a platform where we are able to keep things simple. You are right to be concerned about feature creep.

-=< Fat Owl With A Jetpack >=-
-=< Topia World Builder >=-
-=< Twitter >=-
-=< Blog >=-

Last edited by GlennX; 09-04-2009 at 04:55 AM.

09-04-2009, 05:56 AM
I'm currently writing a feature article for Gamasutra which argues against solid design docs for small teams creating tight, innovative projects on platforms such as the iPhone.

I have a BA (Hons) in Computer Games Design and have production credits for large AAA titles such as Space Siege and The Incredible Hulk, so have seen lots of different projects from both a design and project management perspective at large and small scale.

You shouldn't see adding features as a negative, unless the features aren't required. Feature creep is a common blight for projects, but the best approach, if you want something that is of a good quality in the end, is to only introduce new features where the effort or cost to reward ratio is worthwhile.

Also accept that new features will be added and allow time for it, but get to a point where you lock down and go in to a sprint mode (using Scrum terminology). If you lock everything too early in a GDD you'll likely end up committing yourself to design features that aren't needed whereas if you don't restrain yourself you'll create a monster you never finish!

Creative Director @ Mobile Pie. Visit our website. Follow me on Twitter. Friend me on Plus+ willmp.
09-04-2009, 06:34 AM
Joined: Jan 2009
Location: Toronto, Ontario, Canada
Posts: 5,206
To me, feature creep is never a problem as long as it is something that:

- Doesn't unnecessarily complicate things and won't confuse the user
- Adds value
- Is worth the effort.

Feature creep is often what contributes to the "polish" you always hear about -- the little extras, notions and ideas that make a game more than the sum of its parts. It's all very well to start with a simple idea and execute it simply, but if the things you want to add only make it better from every angle then there's no reason not to add them. If, in the end, it gives people unexpected bonuses that they weren't expecting, or deepens the gameplay, or just makes it more fun and/or visually appealing, it's hard to argue against it.

Design docs to me aren't really that necessary unless you're planning on giving them to one or more people whom you want to follow those docs to the letter. Me, I just outline the important parts -- the ideas I've got, feature creep and all -- and then work with that. As long as I've sketched out the bullet points, the rest can come later as I think about them and decide if it's worth adding.

[Relax with Galactic Chill] [Let me tell you a story.]
Currently working on: Music for Spirit Hunter Mineko
09-04-2009, 08:04 AM
Our whole game is based on feature creep!
Its part of the fun, but knowing when to say no is good too.
Set deadlines and you will realize that some features just won't make it in.
09-04-2009, 02:03 PM
Joined: Mar 2009
Posts: 659
The iphone is the first platform I've ever tried making a full game for.
Once the ball got rolling and the game was playable I started getting sucked into the feature creep trap. Then I backed off and took a look at the bigger picture.
Im making an arcade style game so it's going to be very tight featurewise. The approach I'm taking now is for everything there to have a very specific purpose and not just fluff.
I might even strip out some features if I find that they aren't that useful.

This also comes down to the style of games that I like to play usually dont have a ton of features. Just a specific set of rules and focus on the core gameplay.
09-04-2009, 02:07 PM
Simplicity is King!

I have designed a lot of games and have a golden rule. Strip a game down until it is naked. We had six control buttons for 1 to 25. Re- Do Un-Do Erase etc....... I got so mad at the complexity that I realized that all we need was to see if the puzzle was being filled in correctly. One button and we got to finally use the word" Purge". I love that word it is never used in game design but has such a strong meaning. It really worked!

Pharmx I would be glad to beta test for you> glad to help.

Simple ideas sell better since to fit the category of mass market better.
09-04-2009, 03:00 PM
One of my favorite quotes:

"Perfection is not achieved when there is nothing left to add, but when there is nothing left to remove."

Xenovoid - FREE 4X Galactic Conquest MMO for iOS, PC, OSX, and Linux
JOIN THE FREE BETA @ Xenovoid.com
09-08-2009, 01:07 AM
This applies better to big companies or hobbyists who have other means of income:

There are certain features that you can never have enough, but there's also features that slow down the user experience and you can make playing the game a chore. Imagine a tutorial that just doesn't know when to call it quits. Or character creation screen that seems to never end. Or when you have so many character abilities that you're using every key on the keyboard, then you know you have gone too far

Eyecandy is usually the stuff that you can never have enough, unless it starts affecting the frame rate. Assassin's Creed 2 and Uncharted 2 have thousands of animations and it only makes the play experience better and smoother. Operation Flashpoint has lots of different guns and they're nicely integrated: you pick them up from killed enemies! The game won't drown you in a huge gun catalogue that you have to wade through. You can choose naturally along the route if you want to pick up that RPG or jump into that humvee.


For small companies where your income depends on your work:

I've learned my lessons about feature creep many years ago. At a small company you have be VERY careful about what you choose to do. Wrong feature choices can pretty much kill the company. When your income depends on your games you can't fiddle around forever, you need results and they need to be decent results. I'm not a very good animator or the best texture artist around, but the one thing I can do pretty well is plan a game that's cheap & fast to do - I know a bit about everything (cost of art assets, cost of programming, cost of audio work) and it allows me to keep all the variables in mind when doing design choices.

If in your game prototype you can't move your character but there's animated birds singing on the fence you're guilty of the worst kind of feature creep.

Naughty Dog programmer had a good analogy for processing power which I think applies to features nicely: You have an empty glass in the desk. You drop big pieces of rock into the glass (player movement & actions, environment & objects to interact etc.), then add a smaller set of rocks (textures, lighting, animation, sounds, music etc.), then you add very small rocks (tutorial, menus etc.) and the last thing you add is fine sand (custom music support, flip screen support, small variations of sound effects etc.). Once the deadline for submission has arrived you should be pouring sand! If you're pouring something else the game is not ready or you're doing it in wrong order

Design documents are a must if you work in a bigger team. In a small team you don't need those thick bibles but you have to communicate clearly. Written text is slower in the short term, but trust me, it's much better in the long term because it's searchable. At Mountain Sheep we use a lot Skype text chat even though we could just talk. It allows edits to messages, we also try to write everything correctly so it's searchable later.

2009 is the year of the indie developer (last year was even better), it's gonna be harder to jump in next year. You can already see the App Store lists filling with big publisher brands and it's only gonna get worse... until one day it looks like the Gamestop wall: Activision, EA, Nintendo, Sony, Capcom, Activision, EA ad infinitum.


KingHunt - The Next Generation Slicing Game Lead designer
Bike Baron artist & co-creator
Minigore artist & co-creator
Ice Rage artist & co-creator
Death Rally game designer
09-08-2009, 08:21 AM
Joined: Jan 2009
Location: Toronto, Ontario, Canada
Posts: 5,206
I'd agree. There are different kinds of feature creep. The kind I like (and which I advocate) is the stuff that is not critical to gameplay and does not get in the way of the user experience. The way I look at it, the game should be packed, but the interface and the mechanics should be as simple as you can make them. To me it's kind of like a ratio: The most output (entertainment) from the game for the least input (work) from the user.

[Relax with Galactic Chill] [Let me tell you a story.]
Currently working on: Music for Spirit Hunter Mineko