At the moment every 7 days more than 100’000 users are downloading my apps, every month I get around 400k to 500k downloads.  Here is a chart how my downloads developed during the last months:

So the big question is: How did I manage it to improve my downloads so much without any marketing and advertising.

Well, the popular app stores like iTunes, google Play or amazon are complex, very dynamic ecosystems.

It’s very hard to make good predictions. Often you’ll hear things like “Make a quality product and you will make money!”.

I’m very sorry to disappoint you but in my experience this is NOT true.

I’ve seen so many developers that worked hard on their game for many, many months. They created a fantastic end result. They released their amazing game and…… NOTHING happend. No Downloads. And “no downloads” equals to “no revenue”. So sure, quality is important, but it doesn’t guarantee you anything.

With this in mind I choose another strategy. My assumption was: Make more apps and you will get more downloads.More downloads = more revenue.

So with this in mind I focused to improve my workflow and tried to cut down the development time of my apps. I don’t want to get too technical but what I’ve mainly did to achieve this was:

• I’ve created a “framework/codebase” for fast prototyping and developing games (reusable components, cross platform, data driven, a scripting language for easy customize existing games/components, tools/editors that integrate well into the workflow)
• An easy workflow with lots of automation to create, maintain and publish an app as well as maintain assets and marketing material (description, keywords, screenshots, icons, resolutions for different stores etc.)
• Re-Skin existing apps for another target audience (with no or less coding involved)

How/Why I started

One evening I played around with Tiled – a great open source map editor. I wrote some classes in Monkey  to load, handle & render those maps (You can find the source here). I played around with some Turrican sprites. When I played my demo game I just had the idea to do a small platform game for iOS and Android and I came up with Sophia’s World. The complete game for iOS and Android was done in three weeks.

 The graphics

I’m very bad at design and graphics. But I had those lovely graphics from a good friend of mine and found it a good idea to use them. I asked him if he could add some animation to some of the sprites and I think he did an amazing job. I also added some nice platform tiles from Kenney.

The code

The coding part was very easy. I did a few platform games in the past (although i haven’t done one for years), but at least I know where to start and how to do it. Thanks to Monkey I were able to run it on various platforms without any additional work. I think the first level with the first enemy, the player sprite, the map and collision was done in one day – then I added additional things from time to time when I had an idea during I designed the levels (for example the mushroom, the bat and the falling blocks).

Of course I had to fix some bugs from time to time. I worked with various collision layers that I can check against each other but this caused some problems later on. I then added some Interfaces so special objects could be notified when various events are occurring. I hardwired this directly in the collision-layer-code so it felt a little bit hacky because I mixed up two separate solutions (data-driven collision-layer-system with object orientated notify-system).

The level design

Tiled is such a great tool. I’ve tried out lots of level editor tools in the past and never found one I REALLY liked. I love Tiled because it’s very easy to handle but because of the customizable properties, vectors- and Rect-Zones it’s very powerful. Also it’s open source, so if I ever need a special feature I can easily add it. I exported that data to JSON, cause this was an easy way to load the data into my monkey application. I now even use it for GUI-Stuff. For all my new apps/games I build the gui stuff and requesters with Tiled, export it as JSON and load it in into my application.

For the level design itself I first started to analyze various parts of well-known platformers like Mario Brothers or Giana Sisters. The levedesign was the part that took most of the time. At the current version there are 24 stages, but I will add more soon.

I decided to only scroll horizontal, although it would have been easy for me to add an 8-way-scrolling-routine. But I decided to go for right-to-left-levels because I want to keep things simple. The graphic style and everything is targeted for kids and although I love huge jump’n'run games where you can explore each level I don’t want to add this complexity to that game.

The audio & music

The complete sound effects and music was done in about two days. I think I could have put a little bit more effort into the sound effects (and I will probably do) but I was very happy how the music turned out (Listen to it: The Night | The Day).

What’s Next

I think I will add some more levels because a lot of players liked it and already asked for more. And I want to improve the “Get-Ready”-Screen (it was added in a hurry before I submitted the game to the app stores). I will replace it, so the player has some sort of progress map and can select each level he reached before directly.

Some stats

The weird thing about this little project is, that it was my first project that reached more success across Android than iOS. Within the first two weeks it was played more than 150 000 times. This is huge for me. Remember: No marketing. No PR. I just submitted it to Google Play and iTunes and hoped for the best. I will prepare ports for the Amazon App Store and for Windows Phone. I have zero experience with these platforms so I’m very interested how they compare.

And here is why:

• I don’t want to drive my users to twitter or facebook. Instead I want them to spend more time within my game.
• If the app is worth talking about it, my users will do. If not, they won’t.
• I don’t want to force my users to spread the word because it won’t work.

I have facebook fan pages for my apps because it’s easy to setup and it’s nice to spread the word around my social network. But I won’t advertise it anymore within my games.

Here are some numbers of a friend of mine. Michael Contento is owner of Cora Games. His latest game is a word puzzle app for kids. Within the first week people played it 12 000 times. Those plays generated 90 clicks on the Facebook Page and 122 on the Twitter Button. From those who clicked nearly nobody liked the Facebook page or followed the Twitter account.

In my game Captain Backwater I included integration so each player can share their achievements. Nearly nobody makes use of it. I see other games forcing the players sharing their stories by offering them something in return like special items or ingame currency. Come on, guys! That’s lame. Make a better game and people will talk about it.

I’m also working on a flash-based version for Kongregate and Mochi right now. I really hope that a browser-based port will help to spread the word. We’ll see…

For the monetization I decided to go the freemium way. It’s possible to play through the whole game for free. But to unlock the later levels the player need to collect coins. So if he has to less coins to unlock a level he can either go back and play for more coins or can buy some via In-App-Purchases in the shop.

Controls

After a few tweaks the controls turned out great on my iPad. However, as the original game was designed for desktop resolutions in mind, controls are a little bit tricky and frustrating on mobile devices due to the small screen.

I tried several things and also thought about making it available for tablets only, but now I’ve found a solution with that I’m very satisfied. On mobile devices as soon as you enter the game it zooms in the play field. The hit boxes of the items are bigger then and the controls are very smooth. The player is also able to zoom in and out using the typical two-finger touch gesture.

The new look of a mighty pirate

Another thing we changed is the look of the main character. We didn’t like the look of him anymore, so Emmanuel recreated him from scratch. This caused that we had to replace him on some places in-game as well as on the games’ icon. I’m proudly present: The MIGHTY NEW Captain Backwater:

And here how it turned out in lightwave:

Monetization

A lot of time and effort also went into the balancing. As I told in an earlier post, I decided to go for a freemium model. The whole game can be played for free. During the game the player is able to explore hidden coins. To unlock upcoming levels the player has to spent those coins. If he hasn’t enough, he could go a few levels back playing for some more coins or he can order some via In-App-Purchases.

The first attempt was very unbalanced and it forced the player already in the beginning to play the first levels over and over again or spent some real money to buy the required coins. So now I changed it that there are much, much more coins to discover. Also I dropped the prices for the levels to unlock. I’m feeling it’s also much more motivating and rewarding to find lots of hidden coins.

Bugs, bugs, bugs

The rest of time during the last days I spent on hunting bugs, mainly in the GUI System. As I ported the code 1:1 from the desktop version, a lot of things didn’t work on touch devices anymore and OnMouseOut, OnMouseHit and OnMouseOver Events were triggered by mistake. This did take quite some time to figure it out how to fix it, cause the gui system is quite complex and I wrote it two years ago or so .

Performance issues

I had also performance issues on older devices, and the reason for that was the rendering of the in game texts. I’m using Bitmap Fonts for all ingame text. Each bitmap font consists of three layers: Main chars, borders and shadows. Now each frame, each char was rendered in three passes. And this slowed down things quite a bit. I’ve now baked the fonts onto one layer, still rendering each char on each frame – but it’s fast enough at this time. A better solution would probably to pre-render the text that doesn’t change each frame – but for now I’ll leave it that way.

I still have some problems with the sound, but I think I will get those issues addressed soon.
At all Linux users: Watch the trailer now to know what awaits you soon

If you want to support me, you can preorder now. You will get instant access to the Windows & Mac version. And you will get the Linux port as soon as I finished it!

First of all I was in holidays in Mallorca with some good friends for 10 days. It was planned to stay there for 7 days but we missed our plane back home (note to myself: never book a return flight again that starts off early in the morning). Here is a picture we took in the Megapark:
Mallorca 2011 – I’m the one with Dennis’ hand on my face.

Back home I first analysed and organized the feedback of the nearly 200 beta-testers. The feedback was great, but it was also a lot of work to put every bug report and every feature idea into my issue tracker.

Then I worked on a lot of minor and major bugfixes and thought about which features and ideas I wanted to implement into the final game. On the map screen the player now has two paths he can select from.

The main path contains the path to treasure and fame. During game progression the player can collect hidden coins and catch some bananas. With those the player can buy/unlock levels on the second path, where he can rescue Captain Backwaters big love Mary’o'Malley.

I also worked many, many hours on level design. The final game will contain 100 levels (80 + 20), but I will hopefully release level-packs later on and I also hope that the players will create lots of cool user-generated-levels using the easy-to-use level editor.

Whats left is that I have to design the 20 bonus levels, then re-balance some of them and whats left after that is some polishing.

Last but not least Emmanuel from Graveyard Dogs is working right now on the last missing graphics.

1. Follow Captain Backwater on twitter or become a fan of his facebook group.
2. All followers/fans will get a personal message at sunday evening (CET).
3. You play the game and you will have a lot of fun being a mighty pirate on treasure hunting.
4. If you have suggestions, ideas or problems you send me a message. (I will take every single reported issue or idea very serious!!!)

Oh, and by the way: Did I mention that ALL beta testers will get their name in the credits section of the final game? Imagine for a few seconds: YOUR name in a damn COOL game! True story!

To be honest: This will be the best game I’ve ever made and I’m already very proud of it. But I really need your help to make it even better.

Now it’s your turn:

• Follow Captain Backwater on twitter.
• Become a fan on facebook.

NEW: Follow the mighty pirate @CptBackwater on twitter!

History lesson

Emmanuel had the idea for the game quite some years ago. During this time I started my new job at gameforge and I stopped development.

Later, during my spare time I started to work on a version for Nintendo DS. When it was nearly finished, I contacted several DS Publishers. Although they loved the game, most publisher told me, that the market for DS is flooded and they don’t want to publish games for DS without big licences. Indeed JoWood was the only publisher that was interested, but they went bankrupt.

Because I really loved the concept and Emmanuel’s graphics, we decided to reanimate our old idea and bring it to Mac OS, Linux and Windows (this way we can offer it for download, without the need for a publisher). We started to work on it at the begin of 2011.

The gear behind

Captain Backwater for DS was developed in C++ with the Nintendo SDK (TWL, Nitro) and BitEngine from Bitfield (great Framework). Here is a video how an early DS version looked like:

For the Mac OS, Linux and Windows versions I decided to use BlitzMax and my own (free) Horizon-Framework. BlitzMax is an easy-to-use yet very powerful programming language. We are also using TimelineFX – a great particle editor/engine to do all those fancy effects.

Emmanuel and I are also talking about an iPhone/iPad/Android version from time to time, but we haven’t decided yet. Probably if people like the PC-Version we’ll make a decision.

For the graphics, Emmanuel uses pencils, Lightwave3D and Photoshop. As you can see from the screenshots, he does a great job on the graphics.

I will do the whole soundtrack for Backwater in my home studio with lots of external synths and effect units as well as VST-Instruments and Audio-Units. Everything is controlled by Renoise – a great music tracker/sequencer.

Is it just another puzzle / match-3-game?

When you have a look at the screenshots, you might think it’s just another match-3 game or something like that. But that’s not true. The gameplay is very different. The basic rules are simple, but it will get very challenging during the later levels. We have also a lot of different special items implemented already, like the teleporter, the skull, the holy gray, the dancing monkey and much, much more.

When is it done?

We do the project in our spare time, so we won’t announce a release date. We want to do our perfect puzzle game, so if we need more time, we’ll take it. I work a lot on small details, which slows down development a little bit, but the results are great so far. Also, we have a very agile development process: If we have a new idea, we’ll try it and if we like it we’ll keep it in. If we are bored of Backwater, we’ll do other stuff. We don’t want to force ourselves to get work done. I think this will ensure, that the final game will be of great quality.

I will probably release a free work-in-progress download if enough people are interested in it. And if you like it you can probably give us feedback, suggestions or ideas for new features we haven’t thought of yet.

Keep me up to date

Here are some more resources for you, if you want to keep informed about development on Backwater:

• Follow me on twitter or facebook to get the latest updates
• Watch the latest screenshots at Picasa

Completed Features:

(last update: April, 3rd, 2011)
Basic gameplay, ingame gui, level-editor, level format (save/load routines), secret coins, game event system, dialogs, tutorial system, particle system, teleporters, undo/restart buttons, talking Backwater avatar, bones, holy grail, dancing monkey/bananas, arrows, skull (bomb), blur of color, score system / balancing, dynamite, player profiles, auto save, hammer, ice block, paint brush, rotation arrow, stop sign, ingame message system, all level- and bonus-level-design, …

Features left to do:

polish level complete stats, final testing

• Download Aiolos for Windows
• Download Aiolos for Mac OS X