Postmorthem iPhone Games Finger Physics
Finger Physics was developed by PressOK Entertainment (Reaxion) in Moscow and released in September 2009 on the iPhone / iPod platform. The game hit the Top-5 paid applications in the US, and the free version won the honorable first place. Finger Physics at times paid for its budget, was downloaded more than 2.5 million times, received many positive reviews and managed to get into the hands of Steve Jobs himself ( pruflink ) at the January iPad presentation.
Our company has been developing mobile applications for the past 9 years, and this game has become the first iPhone development experience for the team. We always read stories about the success of other developers with interest and want to do the same for those who are just starting to develop games for the iPhone.
Finger Physics began as a personal development of employees in their spare time, eventually becoming a full-fledged company project that brought invaluable development experience on a new platform and dramatically raised the team morale.
')
Team morale
After many years of working on J2ME / BREW applications with all their features and limitations, the iPhone seemed to us to be the ideal platform for our own project - we wanted to “break away”. The main team gathered, the discussion of the concept of a physical puzzle began, the Super Stacker flash game was taken as a model.
The result of the first month was a playable technical demo. In addition to the main game mode (the task of the player is to arrange the issued blocks at the level and achieve equilibrium of the resulting structure), three more were thought out on paper - they were waiting for their turn. A demo with a cover letter went to our US office in anticipation of a “big appruva.” The initial agreement was this - if we make an interesting demo version of the game, our team gets more freedom and can develop it on its own. To our joy, the prototype was approved, and we proceeded directly to the development.
A good symptom of the fact that you have a hit, manifests itself when all developers are happy to do their work. It so happened that for almost every member of the team the work on the project was just that. Artists could “break away” in a visual style after a long and hard work on the previous project, the designers created ingenious physical puzzles, the programmer (who had previously worked exclusively in QA) was finally able to write code, and even master the new platform. In short, it was all very interesting to work on the game. As a result, the motivation of the entire team pushed the project forward as much as possible.
Visual style
Two key words that can describe the visual style of Finger Physics - "tape and cardboard." We did not have a plot or main characters, and for each game mode (there are five of them) a completely abstract level was drawn. Player, in principle, did not care who he was playing, where he was, and we could show him beautiful and diverse landscapes.
Locations - volcano, underwater world, space, spring meadow and night city
We refused to use the accelerometer in the game process (as opposed to the main competitor - Topple, with whom we are constantly compared), but left it for small decorations on the levels. Small stars, pieces of magma and rockets swayed smoothly in time with the movement of the iPhone.
Almost the only art problem that visited us was associated with the game icon. The application icon is the main seller of your game. The magic requirement of the marketing department for the icon is “attractive and descriptive” (attractive and understandable). Given the absence of the protagonist and a rather complex title, we spent several weeks trying to fit the meaning of the word Finger Physics into 57x57 pixels. In addition, in order for the game to further stand out from the rest, the icon had to be made in its original form, and not in the classic iPhone-style - “rounded square”.
Above - prototypes of icons, below - versions of icons
Casual vs Hardcore
When designing levels, we tried to please both polar audiences at the same time - both casual players and hardcore players. It is an extremely difficult task, but judging by how many young fathers write in the comments to the game that they enjoy playing it with their children - the idea was a success.
At the first playtests, quite a lot of players did not understand what was required of them, how the game modes differed. They stubbornly did not read the level goal written in large letters at the start. Therefore, the first levels were made as simple as possible, and new game modes and block types were introduced gradually. Hardcore quickly passed the first elementary levels and received the main portion of the "fan" on, and the casual player gradually delved into the gameplay. Believe me, there is nothing worse than a player who does not understand how to play your game.
Replacability (a competition for the Russian translation of this word is opened in the comments) was achieved by introducing medals. The better the level is, the more impressive is the medal. Getting a gold medal on challenging levels was an extremely difficult task — a great challenge for a hardcore player. There are a few people who received a gold medal at all levels (the most difficult achievement), but they are happy to share their successes on social networks with friends.
Levels were combined into stages (9 levels = 1 stage). The next stage was opened during the passage of 6 of 9 levels of the previous one. Thus, the players did not “hang up” at the levels impassable for them and could continue to play the next ones.
A special contribution to the game was made by the OpenFeint Achievement System. A few hours of integration, a day or two for the implementation of complex Achievements, and you get a free solution for storing your Achievements, records, and notifications of your Achievements on Facebook and Twitter.
Carrying levels to the server side
At the time of development, the levels moved to a server on the Internet. Rebuilding the build with new levels required a mac-machine and time. Having transferred the cards to an external server and implemented the card swapping in the game from the Internet, it was possible to work on updating the maps without rebuilding the build. This saved designers a lot of time and allowed any team member to test the latest versions of the levels.
A wide variety of game modes and blocks
Initially, we wanted to limit ourselves to two game modes - the construction of structures for balance and height. When the physics engine started working, and these modes were implemented, it became clear that, given the possibilities already available, it would not be difficult to add several more game modes. Although in the end we spent a lot of time on them, dramatically increasing the development time, but we got a much more diverse game - each game mode could be brought into a separate full-fledged game.
In addition to various modes, the variability of the game process was achieved due to the different types of blocks with which the player could interact. When designing the blocks, we broke them up into two different properties - how they behave when interacting with the player (touch), and how they behave when interacting with each other. The first property was realized through the texture of the block, the second - through its color.
A - the block can be in motion, but the player cannot interact with it, B is a fixed block, it cannot be moved, C is a movable block
Unlike most competing games, we painted the figures programmatically, rather than rotated the block that the artists had drawn beforehand. This created problems for programmers, but game designers could make blocks of almost any form in the editor.
"Free" terms of development
For about two months we worked on the game in the evenings in the office, outside office hours. This part of our work was not included in the final budget and the project cost was lower than that of similar ones. As a result, colleagues from the American office did not put much pressure on the deadlines, and developers, if necessary, easily switched to other high-priority projects. This increased the overall development time, but made it "invisible" for other projects. As a result, we did almost no cuts in features.
Editor "by programmers, for programmers"
The first version of the level editor was written in a week by one of the programmers. The editor had his whims, and knowing these "glitches" he had to work in a rather specific way. It is difficult to describe this in detail, as an example - some of the things were easier to implement directly in the XML code of the levels than in the editor. As a result, game designers often changed levels directly in the code.
The reason for such an attitude to the editor is the lack of time for a programmer who could spend another few weeks on the editor, but at the same time would stop the production of more critical basic game code.
Physics engine
We spent quite a large part of the development time on the “war” with the Chipmunk physics engine, choosing it instead of the Box2D. Part of the problems managed to win, some - not. As a result, problems with physics passed on to the shoulders of game designers. Simple examples - balls mounted on a plane, began to roll spontaneously, triangles could go into a state of equilibrium when they were installed on one of the vertices, and when combined, they could “clog up in convulsions”. All this created its own limitations, and some of the ideas had to be thrown out of the ready-made maps.
The second headache is randomness. Part of the interactions of physics did not always repeat with the same precision, which put the passage of a number of levels to chance. Through a large number of iterations of the design of each card, this effect could be defeated, and the levels were passed from the first time with a known sequence of actions.
Impassable levels by many players
The thing we paid little attention to is the difficulty of the levels in the middle of the game. Many levels unfairly fall into either simple stages or difficult ones. The reason lies in the small number of playtests, which in a large way consisted of relatives and friends.
Someone will pass the level from 1 time and will not find it difficult, the other player can get stuck for half an hour, even if the other levels pass approximately equally successfully. However, this effect was partially compensated by the need to go through only 6 of 9 levels to proceed to the next stage. Moral - more playtests, good and different.
How Finger Physics Stormed the Top
In the first month after the release of the game, she earned about 10-30 dollars a day and was not in any top. Later, Apple added it to the What's Hot category, but only in the iTunes client. This allowed the game to come out in the Top-10 of Ireland, Sweden and Austria. But the situation is not much better from this, and this gift Apple has not increased sales by an order of magnitude.
The marketing department arranged a promotion by announcing a vote on PressOK on the topic “Which of the company's iPhone games to be free for one day?”. Defeated Finger Physics. For this day, the full version of the game was downloaded about 12 thousand times. The game hit the paid Top-50 (being free, it happens). The campaign ended, Finger Physics quickly flew back from the top. In parallel with this, in Britain, the free version slowly but surely entered the Top-100 and reached first place within a week. Behind her in the first place pulled up and paid. For two weeks, she was in the Top-10, bringing in about 20% of total profits from Foggy Albion.
In the first place in profit were of course the United States. At the same time with Britain, the light version stormed the local Top-100 free games, pulling the paid one. As a result, the light version was in the first place for about a week, and the paid version reached the fourth place and slowly left ten. These days, our office performance dropped significantly, because people were sitting with open iTunes clients and were constantly pressing F5 to be the first to shout that Finger Physics went another line up.
Data
Total
The main contribution to the success of the product was made by the high motivation of the team. Each participant had their own motives to make this game a hit. The second important component is the minimum pressure from the management: our budget, although it was exceeded, was not large, and the developers, if necessary, switched from Finger Physics to other projects of the company. At the same time, extremely “soft” management was used, the team worked on the “same wave”, conflicts were excluded.
As a result, we got a polished product, suitable for a rather wide audience and warmly received by the players. We hope that the accumulated experience will allow us to continue to create high-quality products, and this article will help you make your game better.
Developer and Publisher : PressOK Entertainment (Reaxion)
Release : September 2009
Platform : iPhone / iPod
Team size : from 3 to 7 developers, depending on the stage
Development time : 6 months
Useful links:
Finger Physics game site
An example of a good introduction to the development of the iPhone
Achievement engine and online high score tables
SDK for collecting user statistics
2D engine Chipmunk
2D engine Box2D
Our company has been developing mobile applications for the past 9 years, and this game has become the first iPhone development experience for the team. We always read stories about the success of other developers with interest and want to do the same for those who are just starting to develop games for the iPhone.
Finger Physics began as a personal development of employees in their spare time, eventually becoming a full-fledged company project that brought invaluable development experience on a new platform and dramatically raised the team morale.
')
What was done right
Team morale
After many years of working on J2ME / BREW applications with all their features and limitations, the iPhone seemed to us to be the ideal platform for our own project - we wanted to “break away”. The main team gathered, the discussion of the concept of a physical puzzle began, the Super Stacker flash game was taken as a model.
The result of the first month was a playable technical demo. In addition to the main game mode (the task of the player is to arrange the issued blocks at the level and achieve equilibrium of the resulting structure), three more were thought out on paper - they were waiting for their turn. A demo with a cover letter went to our US office in anticipation of a “big appruva.” The initial agreement was this - if we make an interesting demo version of the game, our team gets more freedom and can develop it on its own. To our joy, the prototype was approved, and we proceeded directly to the development.
A good symptom of the fact that you have a hit, manifests itself when all developers are happy to do their work. It so happened that for almost every member of the team the work on the project was just that. Artists could “break away” in a visual style after a long and hard work on the previous project, the designers created ingenious physical puzzles, the programmer (who had previously worked exclusively in QA) was finally able to write code, and even master the new platform. In short, it was all very interesting to work on the game. As a result, the motivation of the entire team pushed the project forward as much as possible.
Visual style
Two key words that can describe the visual style of Finger Physics - "tape and cardboard." We did not have a plot or main characters, and for each game mode (there are five of them) a completely abstract level was drawn. Player, in principle, did not care who he was playing, where he was, and we could show him beautiful and diverse landscapes.
Locations - volcano, underwater world, space, spring meadow and night city
We refused to use the accelerometer in the game process (as opposed to the main competitor - Topple, with whom we are constantly compared), but left it for small decorations on the levels. Small stars, pieces of magma and rockets swayed smoothly in time with the movement of the iPhone.
Almost the only art problem that visited us was associated with the game icon. The application icon is the main seller of your game. The magic requirement of the marketing department for the icon is “attractive and descriptive” (attractive and understandable). Given the absence of the protagonist and a rather complex title, we spent several weeks trying to fit the meaning of the word Finger Physics into 57x57 pixels. In addition, in order for the game to further stand out from the rest, the icon had to be made in its original form, and not in the classic iPhone-style - “rounded square”.
Above - prototypes of icons, below - versions of icons
Casual vs Hardcore
When designing levels, we tried to please both polar audiences at the same time - both casual players and hardcore players. It is an extremely difficult task, but judging by how many young fathers write in the comments to the game that they enjoy playing it with their children - the idea was a success.
At the first playtests, quite a lot of players did not understand what was required of them, how the game modes differed. They stubbornly did not read the level goal written in large letters at the start. Therefore, the first levels were made as simple as possible, and new game modes and block types were introduced gradually. Hardcore quickly passed the first elementary levels and received the main portion of the "fan" on, and the casual player gradually delved into the gameplay. Believe me, there is nothing worse than a player who does not understand how to play your game.
Replacability (a competition for the Russian translation of this word is opened in the comments) was achieved by introducing medals. The better the level is, the more impressive is the medal. Getting a gold medal on challenging levels was an extremely difficult task — a great challenge for a hardcore player. There are a few people who received a gold medal at all levels (the most difficult achievement), but they are happy to share their successes on social networks with friends.
Levels were combined into stages (9 levels = 1 stage). The next stage was opened during the passage of 6 of 9 levels of the previous one. Thus, the players did not “hang up” at the levels impassable for them and could continue to play the next ones.
A special contribution to the game was made by the OpenFeint Achievement System. A few hours of integration, a day or two for the implementation of complex Achievements, and you get a free solution for storing your Achievements, records, and notifications of your Achievements on Facebook and Twitter.
Carrying levels to the server side
At the time of development, the levels moved to a server on the Internet. Rebuilding the build with new levels required a mac-machine and time. Having transferred the cards to an external server and implemented the card swapping in the game from the Internet, it was possible to work on updating the maps without rebuilding the build. This saved designers a lot of time and allowed any team member to test the latest versions of the levels.
A wide variety of game modes and blocks
Initially, we wanted to limit ourselves to two game modes - the construction of structures for balance and height. When the physics engine started working, and these modes were implemented, it became clear that, given the possibilities already available, it would not be difficult to add several more game modes. Although in the end we spent a lot of time on them, dramatically increasing the development time, but we got a much more diverse game - each game mode could be brought into a separate full-fledged game.
In addition to various modes, the variability of the game process was achieved due to the different types of blocks with which the player could interact. When designing the blocks, we broke them up into two different properties - how they behave when interacting with the player (touch), and how they behave when interacting with each other. The first property was realized through the texture of the block, the second - through its color.
A - the block can be in motion, but the player cannot interact with it, B is a fixed block, it cannot be moved, C is a movable block
Unlike most competing games, we painted the figures programmatically, rather than rotated the block that the artists had drawn beforehand. This created problems for programmers, but game designers could make blocks of almost any form in the editor.
"Free" terms of development
For about two months we worked on the game in the evenings in the office, outside office hours. This part of our work was not included in the final budget and the project cost was lower than that of similar ones. As a result, colleagues from the American office did not put much pressure on the deadlines, and developers, if necessary, easily switched to other high-priority projects. This increased the overall development time, but made it "invisible" for other projects. As a result, we did almost no cuts in features.
What mistakes were made
Editor "by programmers, for programmers"
The first version of the level editor was written in a week by one of the programmers. The editor had his whims, and knowing these "glitches" he had to work in a rather specific way. It is difficult to describe this in detail, as an example - some of the things were easier to implement directly in the XML code of the levels than in the editor. As a result, game designers often changed levels directly in the code.
The reason for such an attitude to the editor is the lack of time for a programmer who could spend another few weeks on the editor, but at the same time would stop the production of more critical basic game code.
Physics engine
We spent quite a large part of the development time on the “war” with the Chipmunk physics engine, choosing it instead of the Box2D. Part of the problems managed to win, some - not. As a result, problems with physics passed on to the shoulders of game designers. Simple examples - balls mounted on a plane, began to roll spontaneously, triangles could go into a state of equilibrium when they were installed on one of the vertices, and when combined, they could “clog up in convulsions”. All this created its own limitations, and some of the ideas had to be thrown out of the ready-made maps.
The second headache is randomness. Part of the interactions of physics did not always repeat with the same precision, which put the passage of a number of levels to chance. Through a large number of iterations of the design of each card, this effect could be defeated, and the levels were passed from the first time with a known sequence of actions.
| Chipmunk | Box2d |
|---|---|
| + Easy to implement - Little documentation - No advanced functionality | + Good documentation and community + Advanced functionality - Relatively difficult to implement |
Impassable levels by many players
The thing we paid little attention to is the difficulty of the levels in the middle of the game. Many levels unfairly fall into either simple stages or difficult ones. The reason lies in the small number of playtests, which in a large way consisted of relatives and friends.
Someone will pass the level from 1 time and will not find it difficult, the other player can get stuck for half an hour, even if the other levels pass approximately equally successfully. However, this effect was partially compensated by the need to go through only 6 of 9 levels to proceed to the next stage. Moral - more playtests, good and different.
How Finger Physics Stormed the Top
In the first month after the release of the game, she earned about 10-30 dollars a day and was not in any top. Later, Apple added it to the What's Hot category, but only in the iTunes client. This allowed the game to come out in the Top-10 of Ireland, Sweden and Austria. But the situation is not much better from this, and this gift Apple has not increased sales by an order of magnitude.
The marketing department arranged a promotion by announcing a vote on PressOK on the topic “Which of the company's iPhone games to be free for one day?”. Defeated Finger Physics. For this day, the full version of the game was downloaded about 12 thousand times. The game hit the paid Top-50 (being free, it happens). The campaign ended, Finger Physics quickly flew back from the top. In parallel with this, in Britain, the free version slowly but surely entered the Top-100 and reached first place within a week. Behind her in the first place pulled up and paid. For two weeks, she was in the Top-10, bringing in about 20% of total profits from Foggy Albion.
In the first place in profit were of course the United States. At the same time with Britain, the light version stormed the local Top-100 free games, pulling the paid one. As a result, the light version was in the first place for about a week, and the paid version reached the fourth place and slowly left ten. These days, our office performance dropped significantly, because people were sitting with open iTunes clients and were constantly pressing F5 to be the first to shout that Finger Physics went another line up.
Data
- It is impossible to say with certainty what exactly was the impetus for getting into the Top-100;
- The game has a rather unusual description in AppStor, in which there is not a word about the gameplay;
- For three weeks in the Top-100, the budget was recaptured twice;
- Sales between the 15th and 5th place in the top do not differ significantly;
- Conversion from free to full version was 2.6%;
- On weekends, sales increase by 20–30%. In the Christmas holidays - 200-500%;
- Adding code RateMe increased the number of daily reviews of players several times and raised the average score of the game to 4.5;
- The results of the connection of advertising in the light version - CTR clicks on banners ~ 3%, conversion of clicks into purchases ~ 1%. Those. With 10,000 banner impressions you will have 3 purchases. Think twice if you want to make money on advertising in free applications;
- The price of the game is $ 0.99, it has never increased;
- After the release of Finger Physics from the main sales tops, a one-day price reset to zero raised sales, but did not return the game to the top.
Total
The main contribution to the success of the product was made by the high motivation of the team. Each participant had their own motives to make this game a hit. The second important component is the minimum pressure from the management: our budget, although it was exceeded, was not large, and the developers, if necessary, switched from Finger Physics to other projects of the company. At the same time, extremely “soft” management was used, the team worked on the “same wave”, conflicts were excluded.
As a result, we got a polished product, suitable for a rather wide audience and warmly received by the players. We hope that the accumulated experience will allow us to continue to create high-quality products, and this article will help you make your game better.
Developer and Publisher : PressOK Entertainment (Reaxion)
Release : September 2009
Platform : iPhone / iPod
Team size : from 3 to 7 developers, depending on the stage
Development time : 6 months
Useful links:
Finger Physics game site
An example of a good introduction to the development of the iPhone
Achievement engine and online high score tables
SDK for collecting user statistics
2D engine Chipmunk
2D engine Box2D
Source: https://habr.com/ru/post/85566/
All Articles