Inspired by the comments of some other Android developers, I added a “rate this app” button to my slot machine game in the last version. The invitation to rate the app appears whenever the player earns an achievement medal and this seems to work well in terms of encouraging good ratings and comments.

Cherry Chaser performed really well, far better than I expected, at one point it was number 12 in the top free cards and casino games on the Android market and has now been downloaded over 500,000 times. I'm really pleased with the success and feel really proud of it.

It has now slipped down to 50th in it's category and what I'm finding a little worrying however is the nature of it's slide back down from prominence in the market. The point at which it stopped being so popular can be seen quite visibly on the graph of downloads on it's market page. I also know what happened at this point.

At that time two reviews appeared on the game's page saying that the game was unstable and/or crashed a lot. It also appears that a lot of Android users marked these reviews as being "helpful" using the market app on their phones. That meant that these two reviews became the first ones seen by people browsing the market looking for a game to download.

There are a couple of reasons that I find this suspicious. The first is that I have never seen the game crash or heard any trusted reports of it crashing. I get crash reports through the market and the crashes that I get represent less than 1% of all user plays (which I can estimate from ad impressions). Both of these negative "helpful" reviews were from Galaxy S2 owners and I have also confirmed that the game runs fine on this device.

The second reason that I am suspicious is that both reviews that complain of the game crashing constantly award the game four stars. That seems very weird to me. However if the market search algorithm favours higher star reviews to highlight then it would make a lot of sense to someone with untoward intentions.

I'm not sure if there is anything I can do about this. Cherry Chaser has still outperformed my hopes and still earns more than I expected so I am not going to get angry about it. If anybody has any ideas or knows anybody who they could share this story with I would be really interested to hear from them. 

Cherry Chaser Slot Machine is doing far better than I could have imagined when I released it a few weeks ago.  There have been 223K downloads and it is currently installed on 135K devices.  As a "test" of the Android market it has surpassed itself.

The adverts are doing well even with a relatively low average ecpm of about $0.30 per thousand adverts.  The sheer quantity of adverts the game is requesting more than makes up for it - 765K requests yesterday.

I’ve been very lucky and found a small gap in the market for a “european style” slot machine game. It also helps to be in the less competitive “cards and casino” section of the market. I think my relatively low ecpm is largely due to me wanting to make the ads in my game somewhat unobtrusive, I feel that has helped with its popularity which is a good compensation.

Next on my agenda is to port Plasma Duel across to Android as well.  The porting process from C++ to java was less painful than expected and bodes well for finishing Plasma Duel in a timely manner.  After that I will probably do another update for Cherry Chaser to keep it fresh.  I'm seriously considering using www.odesk.com to outsource the artwork and get some really good looking new graphics. 

It's been a week since I launched my Android app and I now have over 9000 downloads.  The things I'm taking from this so far are:

*It is possible to make money as an Android apps developer.*  You just need to find the right niche, or ideally a number of niches.

*App store optimization is important.*  Having the right keywords in your description is obvious.  Also think about other search terms used in different countries, for Cherry Chaser I added "one armed bandit" for the UK market and "pokie" for the Australian market.  Adding keywords out of context to your description is somewhat crass but you can be clever and lever the desired search terms in there without being too obvious: "If you like one armed bandits or pokies, you'll love this game."

*Consider your app store icon.*  You can give more information about your product with your icon.  My original icon was the same as my app icon, a couple of cherries, it looked okay but didn't say anything about the app to a casual browser.  My new icon looks like the type of slot machine that my app emulates.  It gives the user a good idea of what to expect from the game.  After changing to this image my app store downloads more than doubled.

*Android fragmentation is a myth.*  At least the idea that it is problematic to developers is a myth.  My app is running on every version of Android from 1.5 to 4.03 without any problems.  I've even been told that it works well on a tablet, despite being targeted at lower resolution displays. 

Cherry Chaser slot machine is now available on the Android Market:
https://market.android.com/details?id=net.exobyte.cherrychaser 

Cherry Chaser has now been uploaded to the Android Market.

I've decided to launch Cherry Chaser slot machine on to the Android market in its current state.  I had plans to add some features such as multi-currency support and the ability to save the game state over multiple sessions before releasing it.  I've changed my mind based upon the philosophy of releasing early and iterating quickly.  These added features can easily be added in future updates to the app.  I think it is much better to get the app out there where people can use it and start getting feedback.

I added a feature yesterday to make particles pop out of the buttons as the player presses them.  I showed it to Jen this morning and her advice was that it didn't look very good and it was unnecessary.  Looking again with objective eyes I agree with her.  This means that some of the work I did yesterday to add the feature is wasted but sometimes you have to try things just to find out that they do not work.  Anyway the hardest bits of code I wrote can almost certainly be used later so all is not lost.

I'm now adding adverts to the game using the Admob advert service from Google.  This will be a little experiment as my first game with adverts to see if it generates any revenue.  The original Cherry Chaser game I wrote for Windows Mobile had 60,000+ downloads as a freeware app, that was back in the days when the mobile market was much smaller.  Against that, the number of apps relative to the size of the market these days is probably even higher than then, so hopefully my app won't get lost in the masses. 

I've completed all the programming on the game code now and everything all seems to work well.  That means I'm about half way through the development process by my reckoning.

Things still to do include:

Nice animations for jackpots and easter eggs.
More interesting animation of lights during game.
Help and credits screens.
Game options screen.
Support for multiple currencies.
Storage of persistent data on game close.
Adverts for free version.
Animated cash win tray for premium version.

Then there is also a lot of testing and bug fixing before I can upload it to the market.

As part of the job of porting Cherry Chaser to Android I want to make a number of improvements to the game.  One feature that I felt needed improvement was how the nudge feature worked.

In the Windows Mobile version of the game you would win nudges and then nudge the reels manually using the buttons beneath each reel.  There was no facility to nudge upwards and visibility is limited to only the fruit directly above the winline.  That meant that it was very difficult to get a decent win with nudges.

In Cherry Chaser for Android I've just implemented an auto-nudge feature that will entirely replace manual nudges.  Once the amount of nudges awarded has been established the machine looks for the best possible win attainable with those nudges and then performs the nudges automatically.

I've now got all the cash win and credit functionality working and Cherry Chaser is slowly turning into a playable game.  The gamble feature is also in place although I need to alter the timing code to make it independent of processor speed.  I've also put in vibration feedback as each of the reels stop which feels really nice.