Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
August 28, 2014
arrowPress Releases
August 28, 2014
PR Newswire
View All
View All     Submit Event





If you enjoy reading this site, you might also want to check out these UBM Tech sites:


 
Press Copy: A Competitive Evaluation of Flappy Bird and Ironpants
by Seb Long on 06/14/14 08:11:00 am   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.

 

Sebastian Long is a Games User Researcher at Player Research in Brighton, UK. Through playtesting and player experience evaluation, Player Research work to ensure games meet their designers' intent.

I'm sure ‘Flappy’ is not the most commonly-used F-word players use when describing Dong Nguyen’s brutally tough iOS title this year. Love it or loathe it, Flappy Bird caught the attention of many millions of players, hundreds of YouTube casters, developers and journalists alike, sending Nguyen deep into a media frenzy, resulting in the eventual removal of the popular game from app stores worldwide.

Even before Flappy Bird made its unceremonious departure from the App Store, Ironpants, an almost identical game to Flappy Bird but featuring a subtly differing control scheme, had hit the market - but even with a simple formula for success to copy from and improve on, why did Ironpants not reach the same lofty heights as Flappy Bird? How can a clone be less successful than the original, even when the original has been removed from the marketplace?

As an exercise in game usability analysis and player experience evaluation, this article considers some of the factors that might have contributed to this disparity, in order to uncover lessons other titles can learn from these two simple, similar games.

Key Design Successes and Differences

Evaluating games for best practices is a task we perform routinely here at Player Research: extracting the best parts of competitor titles to inform the design of new games - especially for complex UI challenges like in-game stores, loadout management, or real-time feedback for complex or fast-paced games.

Despite both Flappy Bird and Ironpants being such simple games, we’ve identified a number of well-executed UI designs and game design choices, as well as some faux pas, which contribute to explaining the disparity of success, and which we feel other developers could learn from. Lets take a look...

Clear, minimal instructions

Lets start with the basics. Flappy Bird’s visual instructions are perhaps the second-most iconic element of the game - and for good reason: they’re stark, effective and simplistic - exactly as much information as the player requires to start the learning process.

Using animations or stills of fingers can be the most effective way to communicate complex gestures. Showing these finger animations interacting with the game UI itself (e.g. Candy Crush Saga tutorial) or in a dedicated separate animation duplicating the in-game UI (e.g. Angry Birds tutorial) can lead to understandable and accessible text-free introductions to game controls.

Note the use of black and white in indicating a future state of the bird - this serves to better differentiate the animating player character and the assistive ‘future bird’ indicator. The monochrome bird also visually associates with the (similarly monochrome) hand, arrow and bird position, assisting to communicate its message: “your bird flies upwards on tap”. The ‘tap’ indicator is the only use of the colour red in the whole game UI - good for attracting players attention, and reinforcing the importance of the instruction.

Consider Ironpants’ interpretation below. While similar, it lacks the instructional ‘future state’ character, therefore not assisting players in understanding the action that ‘hold’ will have - perhaps increasing the amount of time needed to experiment with the controls - keep this in mind as we move through the rest of Ironpants’ early experience.

Low Pressure Start = Shareable Play

The start of every Flappy Bird level is open and obstacle-free, giving players room to practice tapping between seven and ten times before facing their first pipe obstacle. Players have an opportunity to ‘find the rhythm’ for several seconds before they’re required to navigate the first pipe - helping to avoid too many first-pipe faceplants that will grate on new players, and diminish their ability to practice and master the core controls. This introductory section is entirely missing from Ironpants (see below), where the first obstacle appears immediately - resultantly I’d expect many first-time players’ experience to be unduly frustrating as they’re provided only a fraction of the practice time:

 

Note that in Flappy Bird, both the tutorial image and the introductory open section are repeated for every game - among other benefits, this facilitates a better pass-and-play experience: give the game to a friend and they’re offered the full instructions and a decent practice time, every time - no need for new players to root out a ‘tutorial replay’ button, or suffer a disruptively-steep learning curve (à la Ironpants).

Avoid Exit Points

With both games rich in frustration and failure, they both do well to minimise the ‘down-time’ between levels. Players die fast and often, but neither game prolongs the periods between play sessions with death animations, game over music, or even a significant choice for the player to make - any of these would provide the player with an easy exit opportunity. There are ~2.6 seconds between hitting the pipe and being shown the replay button in Flappy Bird - just enough time for players to acknowledge their death and retry. Unlike Ironpants, Flappy Bird doesn’t have the option to return to the main menu; the design is simpler and better: one button, one tap, one choice, “play again or walk away?”.

Compare the placement and design of the replay button in Ironpants and Flappy Bird (see above) - it is clear that Flappy Bird’s replay button has better visibility among the other buttons, and is larger (and therefore easier to hit, especially on smaller devices). Players of Ironpants, by comparison, will need to read through each of the options until they find ‘retry’, increasing the time and effort required on the first few plays. Being nested among other buttons, it also risks being harder to hit in a hurry.

Consider the impact on Ironpants’ early player experience of making it harder to retry, in addition to the increased number of deaths due to the lack of practice time and worse control instructions, and the addition of full-screen adverts after death (discussed later) - not a good start.

Memorable, Meaningful Highscores

Both Flappy Bird and Ironpants benefit from a simple, numeric scoring system. It is easy to remember a simple highscore number - especially when you’re presented with it after every game (what a tease!). Just like Candy Crush Saga’s “Which level are you on?”, the question “What’s your Flappy Bird highscore?” becomes the dominant method of sharing progress. This isn’t common practice throughout other endless runner games - Jetpack Joyride offers a ‘distance travelled’ metric visible at the start of each game, but other endless runners like Subway Surfers and Temple Run have multiple measures of progress (multiplier and highscore, in Subway Surfers’ case) - meaning players have to remember both. When combined with the very large numbers often favoured by these games (my Subway Surfers highscore is 378384, for example), they’re hard to remember and hard to repeat - a mistake Flappy Bird and Ironpants avoid with often single-digit, hard-earned numbers.

The lack of a ‘multiplier’ mechanic also makes players scores more relatable and comparable between players - an admirable score of 90 in Flappy Bird means the player actually passed 90 pipes, not 30 pipes with a 3x multiplier - high scores reflect true performance, not manipulated game mechanics - better for both casual and social players.

Understandable Reward Structure

Flappy Bird players are offered a medal at the end of each session if their score is high enough, following the classic bronze, silver, gold, platinum structure. Using this traditional medal schema renders the reward system understandable to non-gamers. Note that the game doesn’t disclose the specifics of the structure either (it is 10, 20, 30, 40, if you’re interested) - leaving it for players to explore.

Ironpants, by comparison, offers no such reward system at all, relying instead on players’ own highscore as motivation to replay. While an entire article in itself could be written about these two differing approaches, personally I feel Flappy Bird has the right idea - the notion that I’m being graded against a predetermined performance scale - which in itself has the bar set moderately low, you get a bronze medal with a score of just 10 - I consider to be more motivating than trying to beat just my own high score in isolation.

No Intrusive Adverts

Playing any of Flappy Bird’s many clones will quickly expose the original game’s lack of obtrusive advertising, without full-screen ads or pop-ups, and utilising only top-mounted banners away from the players’ fingers. Despite offering fewer adverts, Flappy Bird was still rumoured to be making more than $50,000 per day from ad revenue alone at the height of its success.

Ironpants - by comparison - offers full-screen adverts after every few deaths, even during the potentially replay-rich early experience already discussed. Displaying full-screen adverts at time-critical moments will only contribute to players’ likelihood of exiting the app, and not coming back. Having full-screen adverts which cover buttons in the app also make it far more likely that players will accidentally tap on it - one sure-fire way to get players to quit your app prematurely is to simply do it for them by having them accidentally tap an advert and be whisked away to the App Store or browser.

Clear, Visible and Readable Feedback

Flappy Bird’s UI is consistently clean, readable and minimal - buttons are large when they’re commonly used, are placed in the easy-to-reach lower third of the screen, and have clear iconography - they’re everything a button should be. There is very little text at all: “Tap”, “Game Over”, “Medal”, “Score”, “Best” and “Rate” - again improving accessibility and suitability for young children, and lowering the barriers to entry.

The score indicator in both games is large, readable and uncluttered on the in-game UI - and supported by clear, discrete audio feedback in passing each pipe. It could be argued that the boxy font used in Flappy Bird is less readable than the one used in Ironpants. Flappy Bird’s version suffers less differentiation in the outer shape of the numbers - potentially making it less glance-able in the middle of a frantic session.

Note that Flappy Bird doesn’t take over the screen or distract the player in any way while playing - unlike the latest version of Ironpants, which displays an animated banner over the score when unlocking new costumes. Distracting the player from their task will result in players blaming the game for failure (rather than their own skill), which, in our experience, is a reliable way to have players disengage with your game - particularly with auto-runners.

Clear Audio Feedback

Ironpants and Flappy Bird’s faceplant sound is short, clean and delightfully slapstick - perhaps even funny the first few times. As for passing pipes, a clear (and familiar) ‘buh-ding’ sound is played every time a player successfully flaps through one - being able to ignore the score by utilising audio feedback is a valuable addition in a game that requires a high level of focus on the gameplay. In Flappy Bird’s case It may have helped that this ‘buh-ding’ sound might be similar to the rewarding noise of a coin being collected in the classic Mario series, but in either case, for new players, this reinforces both the core measure of success, and the scoring mechanic: pass between the pipes.

Audience Suitable Controls

And so we reach the critical difference between these games: the controls. Flappy Bird’s mechanic uses a ubiquitous upward ‘flap on tap’, while Ironpants favours a continuous upwards movement while a tap is held anywhere on the screen: a subtle disparity that makes all the difference.

If you’ve played both games (and now is the time to, if you haven’t already) you’ll immediately notice the difference in the way they feel - Flappy Bird is a much less frantic and more rhythmic experience than Ironpants, which requires what feels like a more rapid, nuanced and accurate input. In actual fact, it becomes clear when visualising screen interactions (see below) that the rate of input is broadly similar - only the type of input differs: in Ironpants, managing both the tap and release timing are required to navigate the obstacles, while Flappy Bird requires the player only to execute taps.

(Flappy Bird and Ironpants interactions over time (13s) - click for larger)

This critical difference is the dividing factor between these two titles. In developing an alternative control scheme to its predecessor, did the Ironpants' developers capitalise on their opportunity to improve on the Flappy Bird formula? Unlike many of the designs with poor usability listed throughout this article, there isn’t a ‘right’ and ‘wrong’ design for these control schemes - just suitability for differing audiences. Ironpants’ continuous input is a more complex motor-cognitive task than Flappy Bird’s discrete taps for upward movement, rendering it more suitable for a more game-savvy and skilled core audience.

It is this choice of a more complex control scheme, in combination with the failures in Ironpants' early experience, UI design, player reward and monetisation practice that lies the heart of Ironpants’ failure to reach the mass market engagement that Flappy Bird so famously achieved.

In Summary

Would Ironpants have been more of a success simply by including Flappy Bird’s controls by default? I don’t think so. By failing to understand and (at the very least) copy the good practices in UI design and feedback that Flappy Bird got right, Ironpants would remain the inferior title. I note that the latest version of Ironpants includes a choice of control scheme - including the infamous ‘tap to flap’ - but offering choice will likely result in players blaming their choice of control scheme rather than their own skill or performance - turning the choice into another barrier to engagement. In failing to effectively understand, emulate and improve on the game design and UI design successes of Flappy Bird, I fear that Ironpants’ opportunity for equal success has probably passed.

For game developers, we know that comparing oneself to, and learning from competitor titles in terms of game design is a common task, but taking a step back and considering all aspects of the game - including the game flow, UI design, early experience and how other games have solved UI challenges, is an important and valuable step in developing a game of any kind - if so much can go wrong in a simple clone, then what issues lurk in your own creatively-unique projects?

Of course, one last question remains: what is my Flappy Bird highscore? I’ll let you scroll back through the article and find out…. :-)

 

Author's Note
I’ve based much of this article on a legacy version of Flappy Bird downloaded before it was culled from the store (I resisted the urge to sell my device on eBay), and the latest release version of Ironpants currently in the iOS App Store, ignoring some recently-added features like alternative control schemes. My apologies if, in doing so, I’ve lost version parity along the way.

 

This article was written by Seb Long, Games User Researcher at Player Research in Brighton, UK. With thanks to Alistair Gray and Graham McAllister, also of Player Research, for their contributions.


Related Jobs

Retro Studios - Nintendo
Retro Studios - Nintendo — Austin, Texas, United States
[08.28.14]

Gameplay Engineer
Mixamo
Mixamo — San Francisco, California, United States
[08.28.14]

Animation Outsource Manager
Smilegate West
Smilegate West — Toronto, Ontario, Canada
[08.28.14]

Java Developer
Trion Worlds
Trion Worlds — Redwood City, California, United States
[08.28.14]

Senior Gameplay Engineer






Comments


Phil Maxey
profile image
I think the key question is whether those small differences led to the huge difference in success between the 2 games. I would argue not. I'm not saying that small differences to a games gameplay mechanic cannot lead to success or failure, of course they can, I'm just saying that Flappy Birds success was so great that I don't think those small differences would of made that kind of difference in success.

Essentially it comes down to make your game as simple as possible, and this is especially true on mobile because of the control method and lack of time people have to play games on them. In the words of Einstein "Everything should be as simple as possible but not simpler". That simplicity is not just about the control method it's also about the UI, it's about every aspect of your mobile game. There are exceptions to this of course, if you genre is a strategy board game then you don't have to make it as simple as possible, but if it's a simple action games it needs to be.

The above is all trumped though by the social and viral nature of Flappy bird, ultimately that aspect of it's success is more important than it's graphics, gameplay or anything else.

Seb Long
profile image
Hi Phil, thanks for commenting. For clarity, I haven't tried here to conclusively explain why Flappy Bird was a success in its own right - there are many more factors than covered above, of course.

I agree that focusing on simplicity is key; having an interaction design and UX background I'd refer to this as focusing on 'usability' - whereby players' own needs, expectations and experiences are taken into account when designing - not just 'simplicity' per se.

Concerning 'viral and social nature': I'd like to see a discussion about how these factors have been implemented and affected Flappy Bird (and clones). Flappy Bird - at least the version I have - contains no Twitter or Facebook integration, not even a 'tweet your score' button. This seems to fly against the traditional notion of 'social' - perhaps highlighting that games can be truly socially successful without 'social features'.

I would also extend this to 'virality'; as a researcher who has worked on a number games which have achieved very high social engagement, I would be wary of divorcing a focus on sharability of scores, accessibility and early experience design (as mentioned above) from the pursuit of 'virality' in the abstract.

TC Weidner
profile image
who the hell knows what makes things go viral theses days, be better off talking about chaos theory, but if your talking about Flappy birds, and if I had to guess why it took off, I'd say " the pipes". The mario pipes and art style is instantly familiar and non threatening. I would suggest if flappy birds was the exact same game without borrowing the mario art style, it would simply be lost among all the other games that come out every week.

Lets face facts, unless you have a few million to pay a PR company, and/or if you are part of one of the 6 media conglomerates ( who can then cross promote the hell out of you and your game) then its all a crap shot for the most part now.

Arthur Hulsman
profile image
Thanks for this post!

We've tried to incorporate some of this study to make our own version of Flappy. Now that we finished its easy to recognize the mistakes to be made and pro's to be done :) http://youtu.be/YuVcaLk6eOE

Alexander Jhin
profile image
Thanks for this article. The differences may appear small, but they're very important for simple games that a player repeats dozens of times in one sitting. I played a repeatable game recently where the "Retry" button was far from "home" position for my fingers. I realized my hand was cramping because of bad "Retry" button placement and I physically had to stop playing the game.

Curtiss Murphy
profile image
Great article - loved the nuanced exploration of difference. Well done.

Jason Wilson
profile image
The biggest difference between the two games is time. Ironpants came out after Flappy Bird became a huge hit. It existed in a world where it needed to compete against not only Flappy Bird (while it was still on the store) but also every other Flappy Bird clone on the market. One cannot discount the advantage of being first.

These other factors certainly contribute to Ironpants not being as successful as it could be. Perhaps a comparison of Ironpants against other Flappy Bird clones would be much more interesting as they would both be competing in a post Flappy Bird world.

Ron Dippold
profile image
That's it for me. I played Flappy Bird just to see what was up. Seb does an in depth job of of analyzing what Flappy does better than Ironpants, but I've never bothered with a single clone (the deluge of them quickly led to ennui) so actual mechanics don't even matter at that point, even if they're superior.

Other people's mileage obvious varies, but novelty was a huge factor.

Lucas Rowe
profile image
Excellent analysis of what made Flappy Bird fundamentally sound. Like some of the other comments said, it feels like we aren't able to really compare the two in a vacuum so everything is speculation as to why one is a hit and the other not. I'm in the camp that attributes it to first mover advantage. Without Flappy Bird being successful I doubt Ironpants would have had any media coverage of note.

Lucas Rowe
profile image
Also not mentioned is the fact that the audio feedback for the Flappy Bird jump is very stark while Ironpants is entirely silent during press. To me that makes a huge difference in usability. I'd also point out that Iron Pants' controls are only somewhat more obtuse than Jetpack Joyride, which has had no problem reaching casual audiences.

Ian Uniacke
profile image
You're clearly also forgetting the most important metric, "Time To Crate", which ironpants clearly fails upon.

Seb Long
profile image
A 'TTC' you have to measure in milliseconds. Might be a new record?

ganesh kotian
profile image
Thank you for sharing the post.

Bob Johnson
profile image
I think the art work and simplicity and feel of the game are what helped make it a hit.

Flappy Bird is more visually appealing to a larger audience I think. The pipes just add familiarity because they were in Mario. The Bird is larger than the super hero and has big cute eyes.

The super hero and crates just doesn't have that same zing to it. The super hero is too small and nothing stands out about him. Generic super hero 101.

The Flappy Bird has a familiar feel to anyone who played a Mario game. Yet not too familiar. It looks like those fish that fly above the bridge in the game.

Bob Johnson
profile image
Oh and yeah that other game is a clone. It was doomed from the start just because of that. No one wants to play a cash grab copy cat game.

Stephen Etheridge
profile image
Great article Seb & Co! I should have guessed it was the minds at Player Research behind it!

Seb Long
profile image
Thanks Stephen!


none
 
Comment: